Home
  By Author [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Title [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Language
all Classics books content using ISYS

Download this book: [ ASCII | HTML | PDF ]

Look for this book on Amazon


We have new books nearly every day.
If you would like a news letter once a week or once a month
fill out this form and we will give you a summary of the books for that week or month by email.

Title: A Text-book of Assaying: For the Use of Those Connected with Mines.
Author: Beringer, Cornelius, Beringer, J. J. (John Jacob), 1857-1915
Language: English
As this book started as an ASCII text book there are no pictures available.
Copyright Status: Not copyrighted in the United States. If you live elsewhere check the laws of your country before downloading this ebook. See comments about copyright issues at end of book.

*** Start of this Doctrine Publishing Corporation Digital Book "A Text-book of Assaying: For the Use of Those Connected with Mines." ***

This book is indexed by ISYS Web Indexing system to allow the reader find any word or number within the document.

USE OF THOSE CONNECTED WITH MINES.***


      includes the original illustrations and in which the chemical
      equations are easier to read.
Transcriber's Note:

   Text:

      Words surrounded by a tilde such as ~this~ means the word is
      in bold face.

      Words surrounded by underscores like _this_ means the word is
      in italics in the text.

      Letters in brackets with an = sign before it means that the
      letters have a macron over them, e.g. H[=A=c] signifies that
      the Ac has a macron over it.

   Numbers and equations:

      Parentheses have been added to clarify fractions.

      Underscores before bracketed numbers in equations denote a
      subscript.

      Superscripts are designated with a caret and brackets, e.g.
      11.1^{3} is 11.1 to the third power.

      The symbol .'. designates the symbol usually used for therefore
      (three periods in a triangle shape).

      A down arrow is represented by a vertical line over a capital
      V. Like this:  |
                     V

   Minor typographical errors have been corrected. Footnotes have
   been moved to the end of the chapter, and all advertisements
   have been moved to the end of the book.



A TEXT-BOOK OF ASSAYING:
FOR THE USE OF THOSE CONNECTED WITH MINES.

by

C. AND J. J. BERINGER.

Revised by

J. J. BERINGER,

Assoc. of the Royal School of Mines; Fellow of the Chemical Society and
of the Inst. of Chemistry; Principal of the Camborne Mining School; and
Late Public Analyst for the County of Cornwall.

With numerous Diagrams and Tables.

Ninth Edition.



London:
Charles Griffin and Company, Limited,
Exeter Street, Strand.
1904.
[All rights reserved.]



PUBLISHER'S NOTE TO THE NINTH EDITION


The continued popularity of the present work, the last edition of which
was published only a little over a year ago, continues to be a source of
gratification to the publishers, who have much pleasure in issuing the
present edition.

_January 1904._


PREFACE TO THE SIXTH EDITION

The principal changes in this edition are additions to the articles on
Gold, Cyanides, and Nickel, and a much enlarged Index. The additional
matter covers more than forty pages.

J. J. BERINGER.

CAMBORNE,
_January 1900._



PREFACE.


The Text-book now offered to the public has been prepared to meet the
existing want of a practical "handy book" for the Assayer.

To mining men the word "assaying" conveys a sufficiently clear meaning,
but it is difficult to define. Some writers limit it to the
determination of silver and gold, and others imagine that it has only to
do with "furnace-work." These limitations are not recognised in
practice. In fact, assaying is becoming wider in its scope, and the
distinction between "assayers" and "analysts" will in time be difficult
to detect. We have endeavoured rather to give what will be of use to the
assayer than to cover the ground within the limits of a faulty
definition.

At first our intention was to supply a description of those substances
only which have a commercial value, but on consideration we have added
short accounts of the rarer elements, since they are frequently met
with, and occasionally affect the accuracy of an assay.

Under the more important methods we have given the results of a series
of experiments showing the effect of varying conditions on the accuracy
of the process. Such experiments are often made by assayers, but seldom
recorded. Statements like those generally made--that "this or that
substance interferes"--are insufficient. It is necessary to know under
what conditions and to what extent.

Students learning any particular process cannot do better than repeat
such a series of experiments. By this means they will, at the same time,
acquire the skill necessary for performing an assay and a confidence in
their results based upon work under different conditions.

The electrolytic method of copper assaying given under _Copper_ is a
modification of Luckow's; it was introduced by us into the offices of
the Rio Tinto Copper Company, and has been in use for many years with
success. This modification is now employed in copper-works in Spain,
Germany, and England, and is used in place of the dry assay for the
commercial valuation of copper ores.

We have adhered to the gram and the "c.c." as the units of weight and
volume. Those who prefer working with grains and grain-measures can use
the figures given, multiplied by ten. For example:--When 1 gram is
mentioned, 10 grains should be used, and 10 grain-measures take the
place of 1 "c.c." It is not advisable to mix the two systems, as by
using gram weights and grain-measures.

We have intentionally to a large extent omitted to mention the names of
those who have originated or modified the various processes. The
practice of naming a process after its discoverer has developed of late
years, and is becoming objectionable. It is a graceful thing to name a
gas-burner after Bunsen, or a condenser after Liebig; but when the
practice has developed so far that one is directed to "Finkenerise" a
residue, or to use the "Reichert-Meissl-Wollny" process, it is time to
stop.

We are indebted to the standard works of Allen, Crookes, Fresenius,
Lunge, Michell, Percy, and Sutton, and wish to express our sense of
special indebtedness to Mr. Richard Smith, of the Royal School of Mines.
One or two of the illustrations are taken from Mr. Sexton's excellent
little book on _Qualitative Analysis_. Our obligation to some others is
mentioned in the text.

Finally, we have to thank for assistance in the experimental work
Messrs. Bailey, Beswick, Clarke, Grant, Higgins, and Smith.

THE AUTHORS.

CAMBORNE, _Nov. 1889_.



CONTENTS.


PART I.


CHAPTER I.

INTRODUCTORY.
                                                       Page
Object of assaying                                        1
Sampling                                                  1
Drying: determination of moisture                         5
Calculation and statement of results                      7
Laboratory books and report forms                         9
Quantity to be taken for an assay                        11
Exercises                                                14


CHAPTER II.

METHODS OF ASSAYING.--DRY GRAVIMETRIC METHODS.

Methods of assaying                                      15
Gravimetric methods                                      15
Mechanical separations                                   16
Dry assays                                               16
  (a) Fluxes                                             16
  (b) Reducing agents                                    21
  (c) Oxidising agents                                   22
  (c) Apparatus                                          24


CHAPTER III.

WET GRAVIMETRIC METHODS.

Wet gravimetric methods                                  27
  (a) Solution                                           29
  (b) Precipitation                                      30
  (c) Filtration                                         31
  (c) Drying and igniting                                32


CHAPTER IV.

VOLUMETRIC ASSAYS.

Titrometric assays                                       35
  (a) Standard solutions                                 36
  (b) Standardising                                      37
  (c) Methods of working                                 42
  (c) Indirect titration                                 43
Colorimetric assays                                      44
Gasometric assays                                        44


CHAPTER V.

WEIGHING AND MEASURING.

Weighing                                                 47
Measuring liquids                                        49
  (a) Graduated flasks                                   49
  (b) Pipettes                                           50
  (c) Burettes                                           51
Measuring gases                                          52


CHAPTER VI.

REAGENTS.

Acids, &c.                                               54
Bases, salts, &c.                                        59


CHAPTER VII.

Formulæ, equations, &c.                                  68


CHAPTER VIII.

SPECIFIC GRAVITY.

Introductory                                             75

Determination of specific gravity--
  (a) Hydrometers                                        76
  (b) Specific gravity bottles                           78
Calculations depending on specific gravity               84


PART II.


CHAPTER IX.

SILVER, GOLD, PLATINUM, CYANIDES, MERCURY.

SILVER--Detection                                        87
  Dry assay                                              87
    (1) Scorification                                    88
    (2) Pot assays, average ores                         90
           "        ores with metallic oxides            91
           "        ores with metallic sulphides         91
  Explanatory notes on the fusion                        93
  The effect of charcoal, flour, &c.                     94
  The effect of nitre                                    95
  The effect of mineral sulphides                        95
    (3) Cupellation                                      98
      The loss of silver                                101
      Condition affecting the loss                      102
      Methods of correction                             103
      Lead required for cupellation                     105
    (4) Calculation of the results in ounces to the
      ton of 2240 lbs. Table                            107
      Ores with metallic particles                      108
    (5) Explanatory notes                               110
    (6) Examples of dry silver assays                   113
  Wet assays                                            116
    Gravimetric method                                  117
    Gay-Lussac's method                                 119
    Volhard's method                                    121
    A modified Gay-Lussac                               123
    Volhard's method applied to arsenic                 124
GOLD--Detection                                         126
  Amalgamation assay                                    126
  Dry assay                                             127
    (1) Size of charges                                 127
    (2) Sampling                                        127
    (3) Assay tons                                      131
    (4) Small buttons, weighing                         131
          "      "     measuring                        133
    (5) Concentration in lead                           136
      Quartz ores                                       136
      Ores with oxide of iron                           138
      Ores with metallic sulphides                      139
    (6) Cyanide charges, residues, &c.                  140
    (7) Cupellation                                     142
      Cupels                                            142
      Cupellation temperature                           143
      Cupellation loss                                  145
    (8) Inquartation                                    146
    (9) Flatting                                        149
   (10) Parting, in flasks                              151
          "      in test tubes                          152
          "      in glazed crucibles                    153
          "      Loss, &c.                              154
   (11) Check assays, surcharge                         154
   (12) Bullion assays in special apparatus             156
      Silver, &c., in gold bullion                      157
   (13) Sampling of base bullion, &c.                   157
CYANIDES--Commercial cyanides                           160
  Double cyanides                                       161
  Prussic acid                                          162
  Gold-dissolving power of cyanide liquor               162
  Assay for cyanide strength                       163, 165
  Assay of commercial cyanide                           167
  Alkalinity of cyanides                                167
  Acidity of ores                                       168
  Metals in cyanide liquors                             169
  Cyanicides                                            169
PLATINUM                                                170
IRIDIUM                                                 171
MERCURY                                                 171
    Dry assay                                           172
    Wet method                                          173


CHAPTER X.

COPPER, LEAD, THALLIUM, BISMUTH, ANTIMONY.

COPPER--Introductory                                    175
    Dry assay                                           176
    Valuation of copper ores                            181
  Wet methods                                           183
    (1) Electrolytic assay                              184
  Volumetric methods                                    194
    (1) Cyanide method                                  194
    (2) Iodide method                                   199
    (3) Colorimetric method                             203
  Examination of commercial copper                      205
LEAD                                                    211
  Dry assay                                             211
  Wet assay                                             213
    (1) Gravimetric method                              213
    (2) Volumetric method                               214
    (3) Colorimetric method                             218
THALLIUM                                                219
BISMUTH                                                 220
    Dry assay                                           221
    Wet method                                          221
      (1) Gravimetric determination                     222
      (2) Colorimetric assay                            223
ANTIMONY                                                225
  Dry assay                                             225
  Wet method                                            227
    (1) Gravimetric assay                               228
    (2) Volumetric method                               229


CHAPTER XI.

IRON, NICKEL, COBALT, ZINC, CADMIUM.

IRON                                                    231
  Gravimetric determination                             233
  Permanganate and bichromate methods                   234
  Stannous chloride method                              244
  Colorimetric determination                            247
NICKEL                                                  251
  Dry assay                                             251
  Electrolytic assay                                    254
  Titration by cyanide                                  255
COBALT                                                  259
ZINC                                                    261
  Gravimetric method                                    262
  Volumetric method                                     263
  Gasometric method                                     266
CADMIUM                                                 269


CHAPTER XII.

TIN, TUNGSTEN, TITANIUM.

TIN                                                     271
  Vanning                                               273
  Dry assay                                             276
  Detection, &c.                                        279
  Gravimetric determination                             281
  Volumetric determination                              282
  Examples                                              284
TITANIUM                                                292
TUNGSTEN                                                295
NIOBIC AND TANTALIC OXIDES                              297


CHAPTER XIII.

MANGANESE, CHROMIUM, ETC.

MANGANESE                                               298
  Gravimetric determination                             300
  Volumetric determination                              300
    Ferrous sulphate assay                              301
    Iodine assay                                        302
  Colorimetric determination                            306
CHROMIUM                                                307
VANADIUM                                                310
MOLYBDENUM                                              311
URANIUM                                                 312


CHAPTER XIV.

EARTHS, ALKALINE EARTHS, ALKALIES.

ALUMINA                                                 314
THORIA                                                  317
ZIRCONIA                                                317
CERIUM                                                  318
LANTHANUM AND DIDYMIUM                                  319
YTTRIA                                                  319
BERYLLIA                                                319
LIME                                                    320
STRONTIA                                                324
BARYTA                                                  326
MAGNESIA                                                328
THE ALKALIES                                            330
  SODIUM                                                334
  POTASSIUM                                             336
  LITHIUM                                               338
  CÆSIUM                                                339
  RUBIDIUM                                              340
  AMMONIUM                                              340


PART III.


CHAPTER XV.

OXYGEN AND OXIDES--THE HALOGENS.

OXYGEN                                                  344
OXIDES                                                  345
WATER                                                   350
THE HALOGENS                                            358
  CHLORINE                                              359
  BROMINE                                               361
  IODINE                                                362
  FLUORINE                                              363


CHAPTER XVI.

SULPHUR AND SULPHATES.

SULPHUR                                                 367
  Gravimetric determination                             369
  Volumetric determination                              370
SULPHATES                                               377
SELENIUM                                                379
TELLURIUM                                               379


CHAPTER XVII.

ARSENIC, PHOSPHORUS, NITROGEN.

ARSENIC                                                 381
  Gravimetric determination                             383
  Volumetric method, "iodine"                           384
      "        "     "uranic acetate"                   389
PHOSPHORUS                                              394
  Gravimetric determination                             396
  Volumetric determination                              397
NITROGEN AND NITRATES                                   400


CHAPTER XVIII.

SILICON, CARBON, BORON.

SILICON AND SILICATES                                   405
CARBON AND CARBONATES                                   414
  COALS                                                 418
  SHALES                                                420
  CARBONATES                                            424
BORON AND BORATES                                       429


APPENDIX A.

Table of atomic weights and other constants             433
Table for converting degrees of the centigrade
  thermometer into degrees of Fahrenheit's scale        435
Tables showing strengths of aqueous solutions of nitric
  and hydrochloric acids, of ammonia and of sulphuric
  acid                                                  436


APPENDIX B.

Estimation of small quantities of gold                  440
Practical notes on the iodide process of
   copper assaying                                      441
Method of separating cobalt and nickel                  442


APPENDIX C.

A lecture on the theory of sampling                     444

INDEX                                                   450



A TEXT-BOOK OF ASSAYING.



CHAPTER I.

INTRODUCTORY.


Assaying has for its object the determination of the quantities of those
constituents of a material which add to or detract from its value in the
arts and manufactures. The methods of assaying are mainly those of
analytical chemistry, and are limited by various practical
considerations to the determination of the constituents of a small
parcel, which is frequently only a few grains, and rarely more than a
few ounces, in weight. From these determinations calculations are made,
which have reference to a mass of material of, perhaps, hundreds of
tons. But in all cases, whether the mass under consideration be large or
small, whether the material be obtained by mining, grown, or
manufactured, the assayer is supposed to receive a small quantity,
called "the sample," which is, or ought to be, the exact counterpart of
the mass of material that is being dealt with. The taking and making of
this sample is termed "sampling"; and the men whose special work it is
to select such samples are "the samplers."

But although "sampling" is thus distinct from "assaying," the assayer
should be familiar with the principles of sampling, and rigorous in the
application of these principles in the selecting, from the sample sent
him, that smaller portion upon which he performs his operations.

~Sampling.~--_In the case of gases_, there is absolutely no trouble in
mixing. The only difficulty is in drawing off a fair sample where, as in
flues, the body of the gas is in motion, and varies a little in
composition from time to time. In this case, care must be taken to draw
off uniformly a sufficient volume of the gas during a prolonged period;
any portion of this larger volume may then be taken for the analytical
operation.

_In the case of liquids_, which mix more or less easily--and this class
includes metals, &c., in the state of fusion--more or less severe
agitation, followed by the immediate withdrawal of a portion, will yield
a fairly representative sample.

_In the case of solids_, the whole mass must be crushed, and, if not
already of fairly uniform quality, mixed, before sampling can take
place. Most of the material which a sampler is called upon to deal with,
is, however, in a more or less divided state and fairly uniform. In
practice it is assumed that 5 per cent. of the whole (= 1/20th), if
taken in portions of equal weight and at frequent and regular intervals,
will represent the mass from which it was taken. Taking a heap of ore,
A, and selecting one out of every twenty spade-, bag-, barrow-, or
wagon-fuls, according to the quantity of stuff in the heap, there is
obtained a second heap, B, containing one-twentieth of the stuff of the
heap A. If we crush the stuff in B until this heap contains
approximately the same number of stones as A did--which means, crushing
every stone in B into about twenty pieces--B will become the counterpart
of A. Selecting in the same manner 5 per cent. of B, there is got a
third heap, C. This alternate reduction and pulverising must be carried
on until a sample of suitable size is obtained. This may be expressed
very clearly thus:--

     A  =  1000     tons of rocks and lumpy ore.
     B  =    50      "   "  rough stones,  1/20th of A.
     C  =     2.5    "   "  small stones,  1/20th of B.
     D  =     0.125  "   "  coarse powder, 1/20th of C.

[Illustration: FIG. 1.

CONE

PARTLY REDUCED CONE

PLAN OF FRUSTRUM DIVIDED.

ELEVATION OF FRUSTRUM DIVIDED.]

If the material to be sampled is already a dry powder, 5 per cent. of it
should be heaped in a cone; each lot being added on the apex of the
cone already formed, so that it may distribute itself by falling evenly
in all directions. When the cone is completed, convert it into a low
frustrum of a cone by drawing stuff uniformly and in a direct line from
the centre to the circumference. Draw two diameters at right angles to
each other, and reserving any two alternate quarters, reject the others.
Mix; and form another cone, and proceed until a sample is got of the
bulk required.

This is the usual plan, and all samples should be treated in this way
when the stuff is fine enough to fall evenly down the sides of a cone.

Samples as they reach the assay office are seldom in a fit state for the
work of the assayer; they are generally too coarse, and ought always to
be more than he wants for any particular determination. The portion he
requires should never be taken at hap-hazard; the sample must be reduced
systematically to the quantity required.

1. _If the sample is a liquid:_ it is sufficient to shake the bottle,
and take out a measured or weighed quantity for the assay.

2. _If a liquid with a solid in suspension:_ measure the whole of it.
Filter. Make up the filtrate with the wash-water or water to the
original bulk. Assay it. Dry and weigh the residue, and make a separate
assay of it.

3. _If of a creamy consistency, free from heavy particles:_ mix well;
spread out evenly on a glazed tile. Take up equal portions at equal
distances. Mix and assay.

4. _If a mud of coarse and fine particles, or of particles of unequal
density:_ weigh and transfer to a porcelain dish, or weigh in the dish.
Dry at 100° C., weigh. Treat the residue as a solid capable of being
powdered.

5. _If a solid capable of being powdered, or already powdered:_ heap up
into a cone; flatten with a spatula; divide along two diameters at right
angles, and carefully reject the whole of two alternate quarters,
brushing away any fine powder. Mix the other quarters, and repeat (if
necessary). For small quantities a fine state of division is essential.

6. _If a solid with metallic particles:_ powder and pass through a
sieve; the metallic particles will not pass through. Weigh both portions
and assay separately. _Sifting should be followed by a very thorough
mixing._

7. _If a metal or alloy in bar or ingot:_ clean the upper surface of the
bar, and bore through the bar. Use the borings. If the ingot or bar is
small, cut it through and file the section. Filings must be freed from
fragments of the file by means of a magnet; and from oil, if any be
present, by washing with a suitable solvent.[1] Where practicable,
metals and alloys are best sampled by melting and granulating. The
student must carefully avoid any chance of mixing dirt or particles of
other samples with the particular sample which he is preparing. One ore
should be done at a time, and when finished, it should be labelled and
wrapped up, or bottled, before starting on a fresh sample.

When an ore requires to be very finely ground in an agate mortar, it is
often advisable to mix with a little pure alcohol and rub until free
from grit; dry at 100° C. and mix well before weighing.

When an assay is required of a quantity of ore made up of parcels of
different weight and quality, each parcel should be separately sampled
and parts of each sample, bearing to each other the same proportion by
weight as the original parcels, should be taken and mixed. For example,
a lot of ore is made up of one parcel of A, 570 tons, one of B, 180
tons, and another of C, 50 tons; a sample representing the whole may be
got by mixing 57 parts of a sample of A with 18 parts of a sample of B,
and 5 parts of a sample of C.

[Illustration: FIG. 2.]

A bruising plate, like that in fig. 2, is convenient for general office
work. The slab is of cast iron, about an inch thick. It is firmly
supported on a solid block of wood, and pivoted for convenience in
emptying. The bruising-hammer is steel-faced, about 4 inches square, and
1-1/2 inch thick. The block is firmly fixed to a small table or tressel,
so that the slab is about 2 feet 6 inches from the ground. The slab is
cleaned, and the sample collected with the help of a stiff-haired brush.

~Drying: Determination of Moisture.~--In practice, the moisture is
generally determined by the samplers, and the proportion is specified in
grains per pound on the label attached to the sample when it reaches the
assay office. The method adopted is usually to dry 1 lb. = 7000 grs. of
the ore in a frying-pan heated over a gas flame, or in an ordinary oven,
until a cold bright piece of metal or glass is no longer damped when
held over it. The loss of weight in grains = moisture.

Properly, however, this work should be done by the assayer, if only for
the following reason. It is assumed that the dry ore of the sampler and
of the assayer are the same thing; according to the nature of the ore,
this may or may not be the case. The assayer, however, uses the sample
which he has dried for his moisture-determination, as the dry ore on
which he makes his other assays, and no variation in moisture would
influence the other and more important determinations. Some ores are
sent to the smelter with from 5 to 15 per cent. of adherent water. In
these cases it is best to spread out the sample, and taking equal
portions fairly at regular intervals, weigh into a Berlin dish 20 grams.
This should then be dried over a sand-bath, or if the ore is likely to
be injured by excess of heat, over a water-bath until the weight is
constant. The loss of weight multiplied by 5 gives the percentage of
water present.

Example:--

  Weight of dish + wolfram      32.67 grms.
    "    "  dish                12.67   "
                                -----
    "    "  wolfram             20.00   "

    "    "  dish + wolfram      32.67   "
    "    "    "    dried        30.15   "
                                -----
    "    "  water                2.52   "

  2.52 × 5 = 12.6                        ~12.6%.~

There are other ores which are not apparently wet, but in the state
called "air-dried." It is easier to take fair samples of these, and,
consequently, it is not necessary to use so large a quantity as 20
grams. But with a smaller quantity, extra precautions must be taken. All
dry solids at ordinary temperatures absorb moisture from the air. The
amount varies with the nature of the material and with the quantity of
surface exposed. Light bulky powders absorb more than heavy ones,
because of the greater condensing surface. It is on this account that it
is well to weigh substances, which have been dried, between
close-fitting watch-glasses. The method of determining moisture is to
weigh out into the glasses 5 grams of ore, and dry in the water-oven
until there is no further loss of weight. On taking the glasses out of
the oven, they should be at once closed, the clip put on, and after
cooling in a desiccator weighed. If after a second trial the loss is the
same, or only increased by a milligram, the determination is finished.

Example:--

  Weight of glasses + pyrites                    31.0470 grms.
    "    "  glasses                              26.0470   "
                                                 -------
    "    "  pyrites                               5.0000   "
    "    "  glasses + pyrites, dried 1 hour      30.8965   "
    "    "    "         "      dried 1-1/2  "    30.8957   "
    "    "    "         "                        31.0470   "
    "    "    "         "      dried             30.8957   "
                                                 -------
    "    "  moisture                              0.1513   "

    0.1513 × 20 = 3.026                                   ~3.02%.~

[Illustration: FIG. 3.]

Sometimes it may be advisable to dry 10 grams, in which case multiplying
the loss by 10 will give the percentage. The dried ore should be
transferred to a weighing-tube (fig. 3), and reserved for the subsequent
determinations. The weighing-tube with the ore must be marked, and kept
in a desiccator.

Most ores and inorganic substances can be dried, and their moisture
determined by the loss in this way. When, however, the substance
contains another somewhat volatile ingredient, it is exposed over
sulphuric acid in a desiccator for two days (if _in vacuo_, all the
better), and the loss determined. Moisture in dynamite should be
determined in this way.

When water is simply mechanically mixed with a substance it presents but
little difficulty. The combined water is a different matter. Slaked
lime, even when perfectly dry, contains much water; and if the water of
soda crystals were separated and frozen, it would occupy a volume equal
to that of the original crystals. Perfectly dry substances may contain
much water, and this combined water is retained by different materials
with very unequal vigour. Sodium sulphate and sodium phosphate crystals
lose water even when exposed under ordinary conditions to dry air. Soda
crystals when heated melt, and at a moderate temperature give off their
water with ebullition. The temperature at which all the water is given
up varies with each particular salt; the actual determination of the
water in each case will require somewhat different treatment. Such
determinations, however, are seldom required; and from a practical
point of view this combined water causes no trouble.

_In assaying ores_, we term "moisture" all water which is lost by
exposure in a water-oven at 100° C., and the "dry ore" is the ore which
has been dried at this temperature. No advantage, but rather endless
confusion, would be caused by varying the temperature with the object of
estimating the whole of the water which a hydrated salt may contain. The
results of the assay of the other components should be calculated on the
"dry ore." One advantage of this is obvious:--The dry ore has a constant
composition, and the results of all assays of it will be the same, no
matter when made; the moisture, however, may vary from day to day, and
would be influenced by a passing shower of rain. It is well to limit
this variability to the moisture by considering it apart, and thus avoid
having the percentage, say, of copper rising and falling under the
influence of the weather.

In the case of certain salts, however, such as soda crystals and
hydrated sulphate of copper (when these constitute the bulk of the
substance to be assayed), it is as well to perform the assay on the
moist, or at any rate air-dried, substance.[2] It would be equally
convenient to calculate on the substance dried at 100° C.; but in this
case it would be well, in order to avoid a somewhat shallow criticism,
to replace the term "moisture" by the longer but equivalent phrase
"water lost at 100° C."

~Calculation and Statement of Results.~--By far the most generally
convenient method of stating the results of an assay is that of the
percentage or parts in a hundred, and to avoid a needlessly troublesome
calculation it is well to take such a quantity of ore for each assay as
by a simple multiplication will yield the percentage. In these
calculations decimals are freely employed, and students should make
themselves familiar with the methods of using them.

Other methods of statement are in use, and have advantages in certain
special cases. With bullion the parts in a thousand are given, and in
those cases in which the percentage is very small, as in water analysis,
it is convenient to report on parts in 100,000, or even on parts per
1,000,000. These are easily got from the corresponding percentages by
shifting the decimal point one, three, or four places to the right. Thus
92.5 per cent. is 925 per thousand; and 0.0036 per cent. is 3.6 per
100,000, or 36 per million.

With ores of tin, silver, and gold, the result is stated as so many
cwts., lbs., or ozs., in the ton. With dressed tin ores as they are
sent to the smelter, the produce is given in cwts. and quarters to the
ton. The corresponding percentage may be obtained by multiplying by
five; or, inversely, if the percentage is given, the produce may be got
by dividing by five. A produce of 13-1/2 equals a percentage of 13.5×5
= 67.5; and a percentage of 70.0 equals a produce of 70/5 = 14. With
tin ores as raised (in which the percentage is small) the reduction must
be carried to pounds per ton. One per cent. equals 22.4 lbs. to the ton;
consequently, if we multiply the percentage by 22.4, the produce will be
given. Thus, if an ore contains 6.7 per cent. of oxide of tin, the
produce is 6.7×22.4 = 150 lbs. (or 1 cwt., 1 quarter, and 10 lbs.) to
the ton. With gold and silver ores, the proportion of precious metal is
small, and it is necessary to carry the reduction to ozs. and dwts. to
the ton; and since gold and silver are sold by troy weight, whilst the
ton is avoirdupois, it is of importance to remember that the ounces in
the two systems are not the same. A ton contains 15,680,000 grains,
which equal 653,333.3 dwts. or 32,666.6 ozs. (troy). The following rules
are useful:--

  To get ozs. (troy) per ton, multiply parts per 100,000 by 0.327;
  To get dwts. per ton, multiply parts per 100,000 by 6.53;
  To get grains per ton, multiply parts per 100,000 by 156.8.

Where liquids are being assayed, cubic centimetres are held to be
equivalent to grams, and the usual method of statement is, "so many
parts by weight in so many by measure." Where the statement is made as
grams per litre or grains per gallon, there can be no doubt as to what
is meant; and even if it be expressed in parts per 100,000, parts by
weight in a measured volume must be understood unless the contrary is
expressly stated.

In some cases, where the density of the solution differs greatly from
that of water, the percentage by weight may be given; and in others,
mixtures of two or more liquids, the percentages may be given by volume
or by weight; as so many c.c. in 100 c.c., or as so many grams in 100
grams, or even as so many grams in 100 c.c. In such cases it must be
distinctly shown which method of statement is adopted.

One grain per gallon means 1 grain in 70,000 grain-measures, or one part
in 70,000. Dividing by 7 and multiplying by 10 will convert grains per
gallon into parts per 100,000. Inversely, dividing by 10 and multiplying
by 7, will convert parts per 100,000 into grains per gallon.

Grams per litre are parts per 1000; multiplying by 100 will give parts
per 100,000, and multiplying by 70 will give grains per gallon.

Among foreign systems of weights, the French is by far the best.
Kilograms (2.205 lbs.) per quintal (220.5 lbs.) are parts per cent.; and
grams (15.43 grs.) per quintal are parts per 100,000. From the rule
already given, grams per quintal may be converted into ounces to the ton
by multiplying by 0.327.

The German loths per centner (1/2 oz. (avoirdupois) to 100 lbs.) equal
parts per 3200; they are converted into parts per cent. by dividing by
32, or into ounces (troy) per ton by multiplying by 10.208.

In the United States, as a sort of compromise between the avoirdupois
and metric systems, a ton is taken as 2000 lbs. There, too, the custom
is adopted of reporting the gold and silver contents of an ore as so
many dollars and cents to the ton. In the case of gold, an ounce is
considered to be worth 20.6718 dollars. With silver, the _nominal_ value
is 1.2929 dollars per ounce, but frequently in assay reports it is taken
as one dollar. The practice is objectionable. The prices of metals vary
with the fluctuations of the market, and if the assayer fixed the price,
the _date_ of his report would be all important; if, on the other hand,
he takes a fixed price which does not at all times agree with the market
one, it leaves a path open for the deception of those unacquainted with
the custom. American "dollars on the ton of 2000 lbs." may be converted
into "ounces in the ton of 2240 lbs." by dividing by 1.1544 in the case
of silver, and by 18.457 in the case of gold.

~Laboratory Books and Report Forms.~--The record which the assayer makes
of his work must be clear and neat, so that reference, even after an
interval of years, should be certain and easy. One method should be
adopted and adhered to. Where there are a large number of samples, three
books are required.

_Sample Book._--This contains particulars of the samples (marks, &c.),
which are entered by the office-clerk as they arrive. He at the same
time puts on each sample the distinguishing number.

            EXAMPLE OF PAGE OF SAMPLE BOOK.

 +----------+----------+--------------------------+----------------+
 |  Date.   |  Number. |        Sample.           |      Remarks.  |
 +----------+----------+--------------------------+----------------+
 |  Feb. 1  |   482    | Tough Copper             | For Arsenic.   |
 |   "   2  |    X     | Piece of Metal           | For Ni and Cu. |
 |   "      |   483    | Tough Copper.            |                |
 |   "      |    73    | Silver Precipitate,      | With Letter.   |
 |          |          |   4 casks, 24 cwt. 1 qr. |                |
 |   "      |   494    | Purple Ore, 200 tons.    |                |
 |   "      |  1 J.T.  | Lead Ore, 1 J.T.         | From Corsica.  |
 |   "      |  2 J.T.  |    "      2 J.T.         |                |
 +----------+----------+--------------------------+----------------+

_Laboratory Book._ This is the Assayer's note-book, in which he enters
clearly the particulars of his work--the results obtained, as well as
how these results were arrived at. The calculations should be done on
scrap-paper, and should not be entered, although, of course, detail
enough must be shown to enable the results to be recalculated.

            EXAMPLE OF PAGE OF LABORATORY BOOK.
______________________________________________________________

  Purple Ore                                     5 grams
    19/10/89          0.0042 grm.
                      0.0021  "
                      ------
       Colorimetric   0.0063 × 20         =  0.13% Copper
______________________________________________________________

      482
  Tough Copper                              10 grams
    Feb. 1/89        10.5 c.c. Uranium.
                                          =  0.52% Arsenic
______________________________________________________________

     2082
  Tough Copper                              10 grams
                     12.7 c.c. Uranium.
                                          =  0.63% Arsenic
______________________________________________________________

      491                                   10 grams
  Tough Copper       13.7 c.c. Uranium
    Feb. 1/89
                                          =  0.68% Arsenic
______________________________________________________________

  Standard of Uranium acetate.
            0.150 gram As_{2}O_{3} = 23.3 c.c. Uranium.
              .'. 100 cc. Uranium  =  0.5 gram As.
______________________________________________________________

     10071                                    5 grams
   Tin Ore          Cruc. and SnO_{2} 9.6065 grms.
  Feb. 3/89         Cruc. and Ash     9.4235   "
                                   ------
                         SnO_{2} = 0.1830  = 2.88% Tin
______________________________________________________________

_The Assay Book._--This is the Official book, and is a combination of
the Sample and Laboratory books. It corresponds with the report-forms.
Without being loaded with detail, it should contain sufficient to
characterise each sample.

Key to following example page of Assay book:
DR = DATE REPORTED.
Not Det. = Not detected

            EXAMPLE OF PAGE OF ASSAY BOOK.

-------------------------------------------+----+-------+---------------+----
     DESCRIPTION OF SAMPLE.                |    | Water | Assay on      |
------+--------------------+---------------|    |Lost at|   the Dry     |
Date. |    Material.       |    Weight.    |No. |100° C.|Material.      | DR
------+--------------------+---+---+---+---+----+-------+---------------+----
1889  |                    |ton|cwt|qrs|lbs|    |       |               |
Feb. 1|Tough cake copper   |   |   |   |   | 482|       |Arsenic, 0.52% |  7
  "   |Tough cake copper   |   |   |   |   |2082|       |Arsenic, 0.63% |  7
  "   |Tough cake copper   |   |   |   |   | 491|       |Arsenic, 0.68% |  7
      |                    |   |   |   |   |    |       |               |
Feb. 2|Nickel disc for C.R.|   |   |   |   | X  |       |Copper, 73.75  |  7
      |                    |   |   |   |   |    |       |Nickel, 24.34  |
      |                    |   |   |   |   |    |       |Iron,    2.18  |
      |                    |   |   |   |   |    |       |        -----  |
      |                    |   |   |   |   |    |       |       100.27  |
      |                    |   |   |   |   |    |       |       ------  |
  "   |Silver precipitate, |   | 24|  1|  0|  73| Not   |               |
      | 4 casks            |   |   |   |   |    | det.  |Silver,  4.851 | 10
      |                    |   |   |   |   |    |       |Gold,    0.0215|
      |                    |   |   |   |   |    |       |Lead,   19.37  |
      |                    |   |   |   |   |    |       |Zinc,    2.00  |
      |                    |   |   |   |   |    |       |Silver, 1584.7 |
      |                    |   |   |   |   |    |       |  ozs. per ton |
      |                    |   |   |   |   |    |       |Gold, 7.0      |
      |                    |   |   |   |   |    |       |  ozs. per ton |
  "   |Purple ore          |200|   |   |   | 494| Not   |Copper,  0.13% | 11
      |                    |   |   |   |   |    | det.  |Sulphur  0.15% |
------+--------------------+---+---+---+---+----+-------+---------------+----

When the number of samples is small, the Sample Book may be omitted, and
the entries made in the Assay Book as the samples arrive.

_Report-forms._ These should entail as little writing as possible in
making out the report. For general purposes the form given on p. 12 is
useful.

~The quantity of substance~ to be taken for any particular assay depends
largely upon the method of assay adopted. There are, however, some
general considerations which should be remembered, and some devices for
simplifying the calculations which should be discussed.

The smaller the percentage of the substance to be determined, the larger
should be the amount of the ore taken. The following table will give a
general idea as to this:--

Percentage of the substance    Amount of ore, &c. to
     to be determined.             be weighed.
        100-10                      1 gram.
         10-5                       2 grams.
          5-1                       5   "
          1-0.1                    10   "
        0.1-0.01                   20   "

[Illustration: ASSAY NOTE]

The rougher the method of assay adopted, the larger should be the
quantity of ore taken. If the degree of accuracy attainable with the
methods and instruments at the assayer's service is known, it is easy to
calculate what quantity should be taken for any particular case. If the
results are good within 0.001 gram, then, taking 1 gram of ore we can
report within 0.1 per cent., or if they are good within 0.0002 gram,
taking 20 grams of ore, we can report within 1 part per 100,000, or very
closely within 6-1/2 dwt. to the ton. If it is wished to be yet more
particular in reporting, larger quantities must be taken. The difficulty
of manipulating very small or very large precipitates, &c., must be
borne in mind. So, too, must the fact that the greater the weight of the
final product of an assay, the less, as a rule, is the percentage error.
The distinction between absolute and percentage error, often overlooked,
is important. If 0.5 gram of silver be cupelled with 20 grams of lead,
there may be obtained a button of 0.495 gram; the absolute loss is 0.005
gram, and this equals 1 per cent. of the silver present. Similarly,
cupelling 0.1 gram, the resulting button may be 0.098; the absolute loss
is only 0.002 gram, but this equals 2 per cent. of the silver present.
In the same way the student should see that the two results, 91.5 per
cent. and 92.0 per cent., are really more concordant than the results
9.1 per cent. and 9.2 per cent.

A device often adopted in practice where a large number of assays of one
kind are made, and the report is given as so many ounces or pounds to
the ton, is that known as the _assay ton_. The assay ton may be any
arbitrary and convenient weight, but its subdivisions must bear to it
the same relations as pounds and ounces bear to the actual ton. On the
other hand, in a laboratory where many kinds of work are performed,
different sets of weights of this kind would only tend to confusion,
even if they were not unnecessary. With a set of gram weights and its
subdivisions anything may be done. If it is desired to report as pounds
to the ton, then, since there are 2240 lbs. to the ton, a weight of
2.240 grams may be taken as the assay ton, and each 0.001 gram yielded
will equal 1 lb., or 22.4 grams may represent the ton, and each 0.01
gram a pound. Similarly, since there are 32,666.6 ozs. troy to the ton;
if we take 32.6667 grams as the assay ton, each 0.001 gram will equal 1
oz. to the ton. In some cases it may be convenient to have, in addition
to the usual gram weights, one or other of the "assay tons" mentioned
above, but generally it is better to work on a purely decimal system,
and convert when required into ounces per ton, &c., either by actual
calculation or by reference to a set of tables.


PRACTICAL EXERCISES.

The student should practise such calculations as the following:--

1. Calculate the percentages in the following cases:--
  (a) Ore taken, 2 grams; copper found, 0.2155.
  (b)    "       1.5 gram; iron found, 0.8340.
  (c)    "       30 grams; lead found, 23.2.

2. Calculate the parts per thousand in the following:--
  (a) Bullion taken, 1.1 gram; silver found, 1.017.
  (b)        "       1.14 gram; silver found, 1.026.
  (c)        "       0.6 gram; gold found, 0.5500.

3. Calculate parts per 100,000 in the following:--
  (a) Ore taken, 20 grams; silver found, 0.0075.
  (b)    "       50 grams; gold found, 0.0026.
  (c) Water taken, 500 c.c.; solids found, 0.1205.

4. Calculate cwts. to the ton in the following:--
  (a) Ore taken, 5 grams; tin found, 2.816.
  (b)    "       5 grams; tin found, 3.128.
  (c) An ore with 68.2 per cent. of tin.

5. Calculate lbs. to the ton in the following:--
  (a) An ore with 3.28 per cent. oxide of tin.
  (b) Ore taken, 20 grams; oxide of tin found, 1.67.

6. Calculate ozs. (troy) to the ton in the following:--
  (a) Ore taken, 50 grams; gold found, 0.0035.
  (b)    "       20 grams; silver found, 0.0287.
  (c)    "       25 grains; silver found, 0.0164.

7.  Calculate in grains per gallon:--
  (a) 0.51 gram per litre.
  (b) 24.6 parts per 100,000.
  (c) Solution taken, 100 c.c.; copper found, 0.0045 gram.
  (c)         "        50 c.c.; iron found, 0.165 gram.

8. Convert into ozs. (troy) per ton:--
  (a) 7 loths per centner.
  (b) 30 grams per quintal.
  (c) 15 parts per 100,000.

FOOTNOTES:

[1] Ether or carbon bisulphide.

[2] Such substances are best dried by pressing between folds of dry
filter-paper.



CHAPTER II.

METHODS OF ASSAYING.--DRY GRAVIMETRIC METHODS.


The methods of assaying are best classed under two heads, Gravimetric
and Volumetric, in the former of which the final results are weighed,
whilst in the latter they are measured. A commoner and older division is
expressed in the terms much used in practice--wet assays and dry assays.
Wet assays include all those in which solvents, &c. (liquid at the
ordinary temperature), are mainly used; and dry assays, those in which
solid re-agents are almost exclusively employed. Dry assays form a
branch of gravimetric work, and we shall include under this head all
those assays requiring the help of a wind furnace. Wet assays, as
generally understood, would include not only those which we class as wet
gravimetric assays, but also all the volumetric processes.

~Gravimetric Methods~ aim at the separation of the substance from the
other matters present in the ore, so that it may be weighed; and,
therefore, they must yield the _whole_ of the substance in a pure state.
It is not necessary that a metal should be weighed as metal; it may be
weighed in the form of a compound of definite and well known
composition. For example, one part by weight of silver chloride contains
(and, if pure, always contains) 0.7527 part of silver; and a quantity of
this metal can be as exactly determined by weighing it as chloride as by
weighing it in the metallic state. But in either case the metal or its
chloride must be pure.

Exact purity and complete separation are not easily obtained; and
methods are used which are defective in one or both of these respects.
It is well to note that an impure product increases the result, whilst a
loss of the substance decreases it; so that if both defects exist in a
process they tend to neutralise each other. Of dry methods generally, it
may be said that they neither give the whole of the substance nor give
it pure; so that they are only calculated to show the amount of metal
that can be extracted on a manufacturing scale, and not the actual
quantity of it present. Their determinations are generally rough and
always low. The gold and silver determinations, however, will compare
very favourably with any of the other processes for the estimation of
these metals in their ores.

The calculation of the results of a gravimetric assay has already been
referred to. If the result is to be stated as percentage, it may always
be done by the following rule:--_Multiply the weight of the substance
got by the percentage of metal it contains, and divide by the weight of
ore taken._

Gravimetric methods are divided into three groups: (1) mechanical
separations; (2) dry methods; and (3) wet methods.

~Mechanical Separations.~--Under this head are classed the method of
assaying tin ores, known as vanning, and the amalgamation assay for
gold. A set of sieves to determine the relative proportion of powders of
different degrees of fineness is sometimes useful. A set with 10, 20, 40
and 80 meshes to the inch is convenient.

~Dry Assays.~--An important distinction between wet and dry methods of
assaying is, that in the former the substance is got into the liquid
state by solution, whilst in the latter fusion is taken advantage of.

The difference between solution and fusion is easily illustrated: a lump
of sugar heated over a candle-flame melts or fuses; suspended in water
it dissolves. Many substances which are insoluble or infusible of
themselves, become soluble or fusible when mixed with certain others;
thus, in this way, solution is got with the aid of reagents, and fusion
with the help of fluxes. For example, lead is insoluble in water, but if
nitric acid be added, the metal rapidly disappears. It is convenient,
but somewhat inaccurate, to say that the acid dissolves the lead. If the
lead be acted on by nitric acid alone, without water, it is converted
into a white powder, which does not dissolve until water is added; in
this case it is obvious that the water is the solvent. The function of
the acid is to convert the lead into a soluble compound.

~Fluxes~ may act as true solvents. Fused carbonate of soda dissolves
baric carbonate, and perhaps in many slags true solution occurs; but in
the great majority of cases a flux is a solid reagent added for the
purpose of forming a fusible _compound_ with the earthy or stony
minerals of the ore. Few of the minerals which occur in the gangue of an
ore are fusible; and still fewer are sufficiently fusible for the
purposes of the assayer, consequently the subject is one of importance,
and it ought to be treated on chemical principles. An idea of the
composition of some of the more frequently occurring rocks may be
gathered from the following table, which represents rough averages:--

  -----------------------------------------------------------------
                       |       |        |Oxide|Lime and |
                       |Silica.|Alumina.| of  |Magnesia.|Alkalies.
                       |       |        |iron |         |
  -----------------------------------------------------------------
                       |   %   |    %   |  %  |      %  |   %
  Sandstone, grit,     |       |        |     |         |
    quartzite, &c.     |80-100 |   --   | --  |     --  |  --
  Granite, gneiss,     |       |        |     |         |
    quartz-porphyry,   |       |        |     |         |
    fire-clay, &c.     | 70-75 | 13-20  |  2  |      2  | 5-8
                       |       |        |     |         |Less in
                       |       |        |     |         |fire-clay.
  Mica-schist          |    65 |    18  |  5  |      3  |   3
  Trachyte, syenite    |    60 |    17  |  7  |    4-7  | 6-9
  Clay-slate           |    60 |    18  | 10  |      8  |   3
  Diorite              |    54 |    17  | 12  |      9  | 3-4
  Horneblende-rock     |    50 |    18  | 15  |     12  | 3-4
  Brick-clay           |    50 |    34  |  8  |      6  |  --
  China-clay           |    47 |    39  | --  |     --  |  --
  Basalt, dolerite, &c.|    50 |    15  | 15  |     16  |   3
  Serpentine           |    44 |    --  | --  |     44  |  --
  Chalk, limestone,    |       |        |     |         |
    dolomite, &c.      |    -- |    --  | --  |  45-55  |  --
  -----------------------------------------------------------------

Silica itself, and the silicates of alumina, of lime, and of magnesia,
are practically infusible; the silicates of soda, of potash, and of iron
are easily fusible if the base (soda, potash, or oxide of iron) be
present in sufficient quantity, and if, in the case of the iron, it is
present mainly as lower oxide (ferrous silicate). The addition of lime,
oxide of iron, or alkali to silicate of alumina results in the formation
of a double silicate of alumina and lime, or of alumina and iron, &c.,
all of which are easily fusible. Similarly, if to a silicate of lime we
add oxide of iron, or soda, or even alumina, a fusible double silicate
will be formed. Thus lime, soda, oxide of iron, and clay, are _fluxes_
when properly used; but since lime, clay (and oxide of iron if there be
any tendency to form peroxide), are of themselves infusible, any excess
of these fluxes would tend to stiffen and render pasty the resulting
slag. So, too, soda, which is a very strong base, may act prejudicially
if it be in sufficient excess to set free notable quantities of lime and
magnesia, which but for that excess would exist in combination as
complex fusible silicates. There are many minerals which with but little
soda form a glass, but with more yield a lumpy scoriacious mass. There
are many minerals, too, which are already basic (for example, calcite),
and which, when present, demand either a less basic or an acid flux
according to the proportions in which they exist. For purposes of this
kind borax, or glass, or clay with more or less soda may be used, and of
these borax is by far the most generally useful. An objection to too
basic a slag (and a very important one) is the speed with which it
corrodes ordinary crucibles. These crucibles, consisting of quartz and
clay, are rapidly attacked by lime, soda and bases generally.

[Illustration: FIG. 4.]

In considering what is and what is not a good slag, certain chemical
properties are of importance. If a mixture of many substances be fused
and allowed to solidify in a crucible, there will be found some or all
of the following. At the bottom of the crucible (fig. 4) a button of
metal, resting on this a speise; then a regulus, next a slag made up of
silicates and borates and metallic oxides, and lastly, on the top
another layer of slag, mainly made up of fusible chlorides and
sulphates. In assaying operations the object is generally to concentrate
the metal sought for in a button of metal, speise or regulus, and to
leave the earthy and other impurities as far as possible in the slag;
whether there be one or two layers of slag is a matter of
indifference;[3] but the chemical action of the lower layer upon the
speise, or regulus, or metal, is of great importance.

A _regulus_ is a compound of one or more of the metals with sulphur; it
is usually brittle, often crystalline, and of a dull somewhat greasy
lustre. It is essential that the slag, when solid, shall be so much more
brittle than the regulus, that it shall be easy to crumble, and remove
it without breaking the latter; and it must not be basic. The effect of
fusing a regulus with a basic slag is well seen when _sulphide of lead_
is fused with _carbonate of soda_; the result is a button of metal (more
or less pure), and a slag containing sulphides of lead and sodium; and
again, if sulphide of lead be fused with an excess of oxide of lead, a
button of lead will be got, and a slag which is simply oxide of lead
(with whatever it may have taken up from the crucible), or if a
sufficient excess has not been used, oxide of lead mixed with some
sulphide. When (as is most frequently the case) the desire is to prevent
the formation of regulus, these reactions may be taken advantage of, but
otherwise the use of a flux having any such tendency must be avoided. A
good slag (from which a regulus may be easily separated) may be obtained
by fusing, say, 20 grams of ore with borax 15 grams, powdered glass 15
grams, fluor spar, 20 grams, and lime 20 grams; by quenching the slag in
water as soon as it has solidified, it is rendered very brittle.

Sulphide of iron formed during an assay will remain diffused through
the slag, instead of fusing into a button of regulus, if the slag
contain sulphide of sodium. The same is true of other sulphides if not
present in too great a quantity, and if the temperature is not too high.

_Speises_ are compounds of a metal or metals with arsenic. They are
chiefly of interest in the metallurgy of nickel, cobalt, and tin. They
are formed by heating the metal or ore in covered crucibles with arsenic
and, if necessary, a reducing agent. The product is fused with more
arsenic under a slag, consisting mainly of borax. They are very fusible,
brittle compounds. On exposure to the air at a red heat the arsenic and
the metal simultaneously oxidize. When iron, cobalt, nickel, and copper
are present in the same speise, they are eliminated in the order
mentioned.

_Slags_ from which metals are to be separated should not be too acid; at
least, in those cases in which the metal is to be reduced from a
compound, as well as separated from earthy impurities. Where the object
is simply to get a button of metal from a substance in which it is
already in the metallic state, but mixed with dross (made up of metallic
oxides, such as those of zinc or iron), from which it is desired to
separate it, an acid flux like borax is best; or, if the metal is easily
fusible, and there would be danger of loss of metal by oxidation or
volatilising, it may be melted under a layer of resin or fat. Common
salt is sometimes used with a similar object, and is often useful. Under
certain conditions, however, it has a tendency to cause the formation of
volatile chlorides with a consequent loss of metal.

In the great majority of cases, the fusion of the metal is accompanied
by reduction from the state of oxide; in these the slag should be basic.
It is not easy to reduce the whole of a reducible oxide (say oxide of
copper or of iron) from a slag in which it exists as a borate or
silicate; there should be at least enough soda present to liberate it.
When the object is to separate one metal, say copper, without reducing
an unnecessary amount of another (iron) at the same time, a slag with a
good deal of borax is a distinct advantage. The slag then will probably
not be free from copper, so that it will be necessary to powder and mix
the slag with some soda and a reducing agent, and to again fuse the slag
in order to separate this residual metal. In all those cases in which
the slag retains an oxide of a heavy metal, this cleaning of the slag is
advisable, and in the case of rich ores necessary. Slags containing
sulphides are especially apt to retain the more easily reducible metals.

The following are the ordinary and most useful fluxes:--

~Soda.~--The powdered bicarbonate, sold by druggists as "carbonate of
soda," is generally used. It gives off its water and excess of carbonic
acid readily and without fusion. Where the melting down is performed
rapidly, the escaping gas is apt to cause trouble by frothing, and so
causing waste of the material. Ordinary carbonate of soda, when hydrated
(soda crystals), melts easily, and gives off its water with ebullition.
It is unfit for use in assaying, but when dried it can be used instead
of the bicarbonate. One part of the dried carbonate is equivalent to
rather more than one and a half parts of the bicarbonate. From two to
four parts of the flux are amply sufficient to yield a fluid slag with
one part of earthy matter. This statement is also true of the fluxes
which follow.

~Borax~ is a hydrated biborate of soda, containing nearly half its
weight of water. When heated it swells up, loses its water, and fuses
into a glass. The swelling up may become a source of loss in the assay
by pushing some of the contents out of the crucible. To avoid this,
_fused_ or _dried borax_ may be used, in which case a little more than
half the amount of borax indicated will suffice. Borax will flux almost
anything, but it is especially valuable in fluxing lime, &c., and
metallic oxides; as also in those cases in which it is desired to keep
certain of the latter in the slag and out of the button of metal.

~Oxide of Lead~, in the form of red lead or litharge, is a valuable
flux; it easily dissolves those metallic oxides which are either
infusible or difficultly fusible of themselves, such as oxides of iron
or copper. The resulting slag is strongly basic and very corrosive; no
crucible will long withstand the attack of a fused mixture of oxides of
lead and copper. With silicates, also, it forms very fusible double
silicates; but in the absence of silicates and borates it has no action
upon lime or magnesia. Whether the lead be added as litharge or as red
lead, it will exist in the slag as monoxide (litharge); the excess of
oxygen of the red lead is thus available for oxidising purposes. If this
oxidising power is prejudicial, it may be neutralised by mixing the red
lead with 1 per cent. of charcoal.

~Glass~: broken beakers and flasks, cleaned, dried, and powdered will
do. It should be free from lead.

~Fluor~: fluor-spar as free as possible from other minerals, powdered.
It helps to flux phosphate of lime, &c., and infusible silicates.

~Lime~: should be fresh and powdered. It must not be slaked. Powdered
white marble (carbonate of lime) will do; but nearly double the quantity
must be taken. One part of lime produces the same effect as 1.8 parts of
the carbonate of lime.

~Tartar~ and "black flux," are reducing agents as well as fluxes. The
"black flux," which may be obtained by heating tartar, is a mixture of
carbonate of potash and charcoal.

REDUCING AGENTS.--The distinction between reducing agents and fluxes
(too often ignored) is an important one. Fluxes yield slags; reducing
agents give buttons of regulus or of metal. The action of a reducing
agent is the separation of the oxygen or sulphur from the metal with
which it is combined. For example, the mineral anglesite (lead sulphate)
is a compound of lead, sulphur, and oxygen; by carefully heating it with
charcoal the oxygen is taken away by the charcoal, and a regulus of lead
sulphide remains. If the regulus be then fused with metallic iron the
sulphur is removed by the iron, and metallic lead is left. The charcoal
and the iron are reducing agents. But in defining a reducing agent as
one which removes oxygen, or sulphur, from a metallic compound so as to
set the metal free, it must be remembered that sulphur itself will
reduce metallic lead from fused litharge, and that oxygen will similarly
set free the metal in fused lead sulphide. There is no impropriety in
describing sulphur as a reducing agent; but it is absurd to call oxygen
one. Some confusion will be avoided if these substances and those which
are opposite to them in property be classed as oxidising and
de-oxidising, sulphurising, and de-sulphurising agents. Most oxidising
agents also act as de-sulphurisers.

_The de-oxidising agents_ most in use are the following:--

~Charcoal.~--Powdered wood charcoal; it contains more or less
hygroscopic moisture and about 3 or 4 per cent. of ash. The rest may be
considered carbon. Carbon heated with metallic oxides takes the oxygen;
at low temperatures it forms carbon dioxide, and at higher ones, carbon
monoxide. Other conditions besides that of temperature have an influence
in producing these results; and as the quantity of charcoal required to
complete a definite reaction varies with these, it should be calculated
from the results of immediate experience rather than from theoretical
considerations.

~Flour.~--Ordinary wheat flour is convenient in use. On being heated it
gives off inflammable gases which have a certain reducing effect, and a
residue of finely divided carbon is left. It is likely to vary in the
quantity of moisture it contains. Two parts of flour should be used
where one part of charcoal would be otherwise required.

~Tartar.~--This is crude hydric potassic tartrate; the purified salt,
cream of tartar, may be used. On being heated it gives off inflammable
gases, and leaves a residue formed of potassic carbonate mixed with
finely divided carbon. Five parts of tartar should be used in the place
of one of charcoal.

~Anthracite~ or ~Culm~ is a kind of coal containing 90 per cent. or more
of carbon. It gives off no inflammable gas. It is denser, and takes
longer in burning, than charcoal. Its reducing effect is little inferior
to that of charcoal. Almost any organic substance can be used as a
reducing agent, but it is well not to select one which melts, swells up,
or gives off much water and gas when heated in the furnace.

~Potassic Cyanide~ is an easily fusible and somewhat volatile salt,
which, when fused, readily removes oxygen and sulphur from metallic
compounds, and forms potassic cyanate or sulphocyanate as the case may
be. Commercial samples vary much in purity; some contain less than 50
per cent. of the salt. For assaying, only the better qualities should be
used.

~Iron~ is a de-sulphurising rather than a de-oxidising agent. Iron is
used in the form of rods, 1/2-inch in diameter, or of nails, or of hoop
iron. In the last case it should be thin enough to be bent without
difficulty. Wrought iron crucibles are very useful in the processes
required for making galena assays.

_The chief oxidising agents (which are also de-sulphurisers)_ are the
following:--

~Nitre~, or Potassic Nitrate.--This salt fuses very easily to a watery
liquid. It oxidises most combustible substances with deflagration, and
thereby converts sulphides into sulphates, arsenides into arsenates, and
most metals into oxides. In the presence of strong bases, such as soda,
the whole of the sulphur is fully oxidised; but in many cases some
arsenic is apt to escape, and to give rise to a peculiar garlic-like
odour. The sulphates of soda and potash are thus formed, and float as a
watery liquid on the surface of the slag.

~Red lead~ is an oxide of lead. About one-quarter of its oxygen is very
loosely held, and, hence, is available for oxidising purposes, without
any separation of metallic lead. The rest of the oxygen is also
available; but for each part of oxygen given off, about 13 parts of
metallic lead are deposited. In silver assays this power of readily
giving up oxygen is made use of. The residual oxide (litharge) acts as a
flux.

~Hot air~ is the oxidising agent in roasting operations. The sulphur and
arsenic of such minerals as mispickel and pyrites are oxidised by the
hot air and pass off as sulphur dioxide and "white arsenic." The metals
generally remain in the form of oxide, mixed with more or less sulphate
and arsenate. The residue may remain as a powdery substance (a calx), in
which case the process of roasting is termed calcination; or it may be a
pasty mass or liquid. In the calcination of somewhat fusible minerals,
the roasting should be done at a low temperature to avoid clotting;
arsenic and sulphur being with difficulty burnt off from the clotted
mineral. A low temperature, however, favours the formation of sulphates;
and these (if not removed) would reappear in a subsequent reduction as
sulphides. These sulphates may be decomposed by a higher temperature
towards the end of the operation; their removal is rendered more certain
by rubbing up the calx with some culm and re-roasting, or by strongly
heating the calx after the addition of solid ammonic carbonate. In
roasting operations, as large a surface of the substance as possible
should be exposed to the air. If done in a crucible, the crucible should
be of the Cornish type, short and open, not long and narrow. For
calcinations, _roasting dishes_ are useful: these are broad and shallow,
not unlike saucers, but unglazed. In those cases in which the products
of the roasting are liquid at the temperature used, a _scorifier_ (fig.
38) is suitable if it is desired to keep the liquid; but if the liquid
is best drained off as quickly as it is formed, a _cupel_ (fig. 5)
should be used.

[Illustration: FIG. 5.]

A scorifier is essentially a roasting dish sufficiently thick to resist,
for a time, the corrosive action of the fused metallic oxides it is to
contain. The essential property of a cupel is, that it is sufficiently
porous to allow the fused oxide to drain into it as fast as it is
formed. It should be large enough to absorb the whole of the liquid; and
of course must be made of a material upon which the liquid has no
corrosive action. Cupels do not bear transport well; hence the assayer
generally has to make them, or to supervise their making. A quantity of
bone ash is carefully mixed with water so that no lumps are formed, and
the mixture is then worked up by rubbing between the hands. The bone ash
is sufficiently wet when its cohesion is such that it can be pressed
into a lump, and yet be easily crumbled into powder. Cupel moulds should
be purchased. They are generally made of turned iron or brass. They
consist of three parts (1) a hollow cylinder; (2) a disc of metal; and
(3) a piston for compressing the bone ash and shaping the top of the
cupel. The disc forms a false bottom for the cylinder. This is put in
its place, and the cylinder filled (or nearly so) with the moistened
bone ash. The bone ash is then pressed into shape with the piston, and
the cupel finished with the help of three or four smart blows from a
mallet. Before removing the piston, turn it half-way round upon its axis
so as to loosen and smooth the face of the cupel. The cupel is got out
by pressing up the disc of metal forming the false bottom; the removal
is more easily effected if the mould is somewhat conical, instead of
cylindrical, in form. The cupels are put in a warm place to dry for two
or three days. A conveniently sized cupel is 1-1/4 inches in diameter
and about 3/4 inch high. The cavity of the cupel is about 1/4 inch deep,
and something of the shape shown in fig. 5.

[Illustration: FIG. 6.]

[Illustration: FIG. 7.]

[Illustration: FIG. 8.]

There are two kinds of furnaces required, the "wind" and "muffle"
furnaces. These are built of brick, fire-brick, of course, being used
for the lining. They are connected with a chimney that will provide a
good draught. Figure 6 shows a section of the wind furnace, fig. 7 a
section of the muffle furnace, and fig. 8 a general view of a group
comprising a muffle and two wind furnaces suitable for general work.
When in operation, the furnaces are covered with iron-bound tiles. The
opening under the door of the muffle is closed with a loosely fitting
brick. The floor of the muffle is protected with a layer of bone-ash,
which absorbs any oxide of lead that may be accidentally spilt. The fire
bars should be easily removable.

Few tools are wanted; the most important are some cast-iron moulds,
tongs (fig. 9), stirrers for calcining (fig. 10), and light tongs of a
special form for handling scorifiers and cupels (_see_ SILVER).

[Illustration: FIG. 9.]

[Illustration: FIG. 10.]

The coke used should be of good quality; the formation of a fused ash
(clinker), in any quantity, causes ceaseless trouble, and requires
frequent removal. The coke should be broken into lumps of a uniform size
(about 2 in. across) before being brought into the office. The furnace
should be well packed by stirring, raising the coke and not ramming it,
and it should be uniformly heated, not hot below and cold above. In
lighting a furnace, a start is made with wood and charcoal, this readily
ignites and sets fire to the coke, which of itself does not kindle
easily.

In commencing work, add (if necessary) fresh coke, and mix well; make
hollows, and into these put old crucibles; pack around with coke, so
that the surface shall be concave, sloping upwards from the mouths of
the crucibles to the sides of the furnace; close the furnace, and, when
uniformly heated, substitute for the empty crucibles those which contain
the assays. It is rarely advisable to have a very hot fire at first,
because with a gradual heat the gases and steam quietly escape through
the unfused mass, while with too strong a heat these might make some of
the matter in the crucible overflow. Moreover, if the heat should be too
strong at first, the flux might melt and run to the bottom of the
crucible, leaving the quartz, &c., as a pasty mass above; with a gentler
heat combination is completed, and the subsequent fiercer heat simply
melts the fusible compound into homogeneous slag.

The fused material may be left in the crucible and separated from it by
breaking when cold. It is generally more convenient to pour it into
cast-iron moulds. These moulds should be dry and smooth. They act best
when warmed and oiled or black-leaded.

Air entering through the fire-bars of a furnace and coming in contact
with hot coke combines with it, forming a very hot mixture of carbonic
acid and nitrogen; this ascending, comes in contact with more coke, and
the carbonic acid is reduced to carbonic oxide; at the top of the
furnace, or in the flue, the carbonic oxide meeting fresh air, combines
with the oxygen therein and re-forms carbonic acid. In the first and
third of these reactions, much heat is evolved; in the second, the
furnace is cooled a little. It must always be remembered, that the
carbonic oxide of the furnace gases is a reducing agent. When these
gases are likely to exert a prejudicial effect, and a strongly oxidising
atmosphere is required, the work is best done in a _muffle_.

FOOTNOTES:

[3] There is an exception to this, as when the slag is liable to be
acted on when exposed to the air and to the gases of the furnace. In
this case a layer of fused common salt floating on the slag, so as to
protect it from the air and furnace gases, is a distinct advantage.



CHAPTER III.

WET GRAVIMETRIC METHODS.


In _dry assays_ the metal is almost always separated and weighed as
metal; in _wet_ gravimetric assays the metal is more usually weighed in
the form of a definite compound of known composition. The general
methods of working resemble those of ordinary chemical analysis, and
their successful working is greatly helped by a knowledge of, at any
rate, those compounds of the metal which enable it to be separated, and
of those which are the most convenient forms in which it can be weighed.
But the work of the assayer differs from that of the analyst, inasmuch
as the bulk of his estimations are made upon material of practically the
same kind, varying only in richness; consequently in assaying, it is
possible (and necessary) to work on such a definite plan as will involve
the least amount of labour in weighing and calculating.

The assayer connected with mining has generally two classes of material
to deal with: those comparatively rich and those comparatively poor. For
example, silver in bullion and in ores; copper precipitates or regulus,
and copper ores and slags; and "black tin" and tin ores. He is only
occasionally called on to assay the intermediate products. It is
indispensable that he should have an approximate knowledge of the
substance to be determined. With new ores this information is best got
by a qualitative testing. Knowing that only certain bodies are present,
it is evident that the number of separations can be reduced, and that
simple methods can be devised for arriving at the results sought for.
The best method is that which involves the least number of separations.
The reactions must be sharp and complete, and yet not be liable to error
under varying conditions.

To bring the richer and poorer materials under the same conditions for
the assay, a small weight, say 1 gram of the richer, and a larger weight
(5 or 10 grams) of the poorer, substance is weighed up. A method is then
adopted which will concentrate the whole of the metal (either during or
after solution) in a product which need not necessarily be pure. The
work on this product is comparatively easy. In separating small
quantities of a substance from a large bulk of impurities, the group
separations must not as a rule be too much relied on. Very large
precipitates carry down small quantities of bodies not belonging to the
group, more especially when there is a tendency to form weak double
compounds. The re-dissolving and re-precipitating of bulky precipitates
should be avoided.

When a large number of assays of the same kind have to be carried out, a
plan something like the following is adopted:--The samples, after having
been dried, are placed in order on a table at the left hand of the
assayer. He takes the first, marks it with a number, samples and weighs
up the quantity required, and transfers it to a flask, which is
similarly marked. As the weighings are finished, the samples are placed
in the same order on his right hand. The assistant takes the flasks in
batches of four or five at a time to the fume cupboard, where he adds a
measured quantity of acid. When solution has been effected, dilution
with a measured volume is generally necessary. The assayer sees to this
and (whilst the funnels and filters are being prepared) makes any
separation that is necessary. The filters are arranged in order on a
rack (fig. 11), and need not be marked unless the precipitates or
residues have subsequently to be dried. The filters are washed with hot
water, and if the filtrates are wanted flasks are placed beneath, if
not, the solution is drained off down the sink. Precipitation or
reduction (or whatever it may be) is now made; the assistant filters the
prepared samples, one at a time, whilst the assayer is engaged with the
others. The same style of work is continued until the assays are
completed. If one should be spoiled, it is better to allow it to stand
over for assaying along with the next batch. If one filters slowly or is
in any way less forward than the rest, it may lessen the accuracy of the
other assays, owing to oxidation, &c., it should, therefore, be put on
one side. The assays are dealt with in batches of ten or twenty, so that
a large quantity of work can be quickly finished.

[Illustration: FIG. 11.]

When the assays are finished, it is the duty of the assistant to clean
the apparatus (with reagents, if necessary), and to put the vessels in
the place set apart for them. Flasks are best kept inverted on a rack,
so that they may be dry and clean by the next morning. Berlin crucibles
must be cleaned and ignited.

The amount of apparatus employed should be as little as is feasible. The
assay should be carried out as much as possible in the same flask. The
bench must be clean, and altogether free from apparatus not in actual
use. Crucibles and dishes in which weighings are made should be marked
with numbers or letters; and their weights recorded, together with the
date of weighing, in a small ledger, which is kept in the drawer of the
balance. By this means a record of the "wear" of each piece of apparatus
is obtained, and, what is more important, much weighing is saved, and
increased confidence is gained. The weight of each piece of apparatus
need not be taken daily. It will be seen from the record in the book and
a knowledge of the use it has been put to how often a checking of the
weight is necessary. The entries are made in black lead as follows:--

  Dish, A.  Feb. 3    9.4210 grams.
                 5    9.4225
                 6    9.4230
                 7    9.4200

Platinum vessels and apparatus lose, and porcelain ones slightly gain,
weight with continued use.

The special details of the work is given under each assay; certain
general instructions will be given here.

~Solution.~--It is not always necessary to get the whole of the mineral
in solution, provided the body sought for is either completely dissolved
or altogether left in the residue. It is often only by a qualitative
examination of the solution (or residue, as the case may be) that the
assayer can satisfy himself that it is free from the substance sought.
But previous experience with the same kind of ore will show to what
extent this testing is necessary.

Solution is generally best effected in flasks; but where the resulting
liquid has afterwards to be evaporated to dryness and ignited,
evaporating dishes (fig. 12) are used. With them clock glasses are used
as covers during solution to avoid loss through effervescence.
Evaporating dishes are also best when an insoluble residue has to be
collected, since it is difficult to wash out most residues from a flask.
Bumping occurs less frequently in dishes than in flasks.

[Illustration: FIG. 12.]

After the addition of the acid, and mixing by agitation, the vessel
containing the substance is heated. This is best done on the "hot
plate" (fig. 13). This consists of a slab of cast iron about half or
three-quarters of an inch thick, supported on loose fire bricks, and
heated by two or three ring burners (figs. 14 and 15). The burners are
connected to the gas supply by means of _lead_ tubing, to which they are
soldered. Flasks and dishes after being put on the plate are not further
handled until solution is complete or the evaporation is carried to
dryness. The hot plate is contained in a cupboard so as to be out of the
reach of cold draughts.

[Illustration: FIG. 13.]

[Illustration: FIG. 14.]

[Illustration: FIG. 15.]

The action of the acids and other solvents is described in the chapter
on Reagents.

~Precipitation.~--In precipitating add sufficient of the reagent to
complete the reaction. The student must be on his guard against adding a
very large excess, which is the commoner error. In some reactions the
finishing point is obvious enough; either no more precipitate is formed,
or a precipitate is completely dissolved, or some well-marked colour or
odour is developed or removed.

In those cases in which there is no such indication, theoretical
considerations should keep the use of reagents within reasonable limits.
The solutions of the reagents (_see_ REAGENTS) are generally of five or
ten per cent. strength. A small excess over that demanded by theory
should be sufficient.

[Illustration: FIG. 16.]

[Illustration: FIG. 17.]

[Illustration: FIG. 18.]

[Illustration: FIG. 19.]

~Filtration.~--Solutions are best filtered hot whenever the assay allows
of this being so done. The precipitate should be allowed to settle, and
the clear liquid decanted on the filter with the aid of a glass rod if
necessary. The filter-paper must not be too large, but at the same time
it must not be overloaded with the precipitate. There should be ample
room for washing. For general use three sizes of filter-paper are
sufficient. Common quick filtering-paper (English) is best for most work
in assaying. The specially prepared paper (Swedish or Rhenish) is used
for collecting those precipitates which have to be weighed. The papers
are folded as shown in fig. 16, and should not project above the funnel.
The filter-paper works better if damped with hot water. In special cases
filtering is hastened by means of an air-pump. The apparatus used
consists of a water-jet (fig. 17), which is connected with the tap, as
also with a bottle fitted as shown in fig. 18. The pump draws the air
out from the bottle, and atmospheric pressure forces the liquid through
the filter-paper. The bottom of the funnel is provided with a platinum
cone, which supports the filter-paper, and prevents its breaking. The
pump is only used in exceptional cases; nearly all the filtrations
required by the assayer can be made without it. The usual methods of
supporting the funnel during filtration are shown in fig. 19. Where the
filtrate is not wanted, pickle bottles make convenient supports. After
the precipitate has been thrown on the filter, it is washed. In washing,
several washings with a small quantity of water are more effective than
a few with a larger quantity of that fluid. The upper edge of the
filter-paper is specially liable to escape complete washing. Excessive
washing must be avoided; the point at which the washing is complete is
found by collecting a little of the filtrate and testing it. The
precipitate is removed from the filter-paper for further treatment by
opening out the paper and by washing the precipitate with a jet of water
from a wash-bottle into a beaker, or back through the funnel into the
flask. In some cases, when the precipitate has to be dissolved in
anything in which it is readily soluble, solution is effected in the
filter itself allowing the liquid to run through as it is formed.

~Drying and Igniting.~--Precipitates, as a rule, require drying before
being ignited. With small precipitates the filter-paper may be opened
out, and placed on a warm asbestos slab till dry; or the funnel and the
filter with the precipitate is placed in a warm place, and supported by
any convenient means. The heat must never be sufficient to char the
paper. Some precipitates must be dried at a temperature not higher than
100° C. These are placed in the water-oven (fig. 20), and, when
apparently dry, they are taken from the funnel, placed between glasses,
and then left in the oven till they cease to lose weight. Such
precipitates are collected on tared filters. Those precipitates which
will stand a higher temperature are dried in the hot-air oven at a
temperature of from 120° to 150°. The drying is continued until they
appear to be free from moisture, and until the precipitate ceases to
adhere to the filter. In drying sulphides the heat must not be raised to
the melting point of sulphur, since, if there is any free sulphur
present, it fuses and filters through.

[Illustration: FIG. 20.]

The precipitate, having been dried, is transferred to a watch-glass. The
filter-paper is opened out over a sheet of note-paper, and, with a
camel-hair brush, the precipitate is gently brought into the glass. Most
precipitates come away easily, and the transfer can be made without
apparent loss. The watch-glass is covered by the funnel, and the
filter-paper (folded into a quadrant) held by the tweezers and set fire
to with the flame of a Bunsen burner. It is allowed to burn over the
crucible, into which the black bulky ash is allowed to drop, and two or
three drops of nitric acid are then added. The crucible is placed on a
pipe-stem triangle (fig. 21), supported on a tripod. It is at first
heated gently with a Bunsen burner, and afterwards more strongly, until
the residue is free from carbon. It is cooled, and treated with any acid
necessary to convert the small amount of precipitate into the state in
which it is to be weighed; heated again, and cooled. The main
precipitate is transferred to the crucible, and the heating repeated
very gently at first, but more strongly towards the end of the
operation. It is next placed in the muffle, and, after two or three
minutes at a red heat, it is removed and allowed to cool in the
desiccator before weighing. This is for bodies that will bear a red
heat; for those compounds that require a lower temperature the heating
in the muffle is omitted. The muffle used for this purpose must not be
used at the same time for cupelling; a gas muffle (fig. 22), such as one
of Fletcher's, is best. A desiccator (fig. 23) is an air-tight vessel
which prevents access of moisture, &c., to the substance. Usually the
air in it is kept dry by means of a basin containing sulphuric acid.

[Illustration: FIG. 21.]

[Illustration: FIG. 22.]

[Illustration: FIG. 23.]

The crucible is removed from the muffle with the tongs and carried to
the desiccator. It is best, in an office, to have a large desiccator
permanently fixed alongside the balance, into which all substances may
be put before being weighed. The substance is removed from the bench or
muffle in the small hand apparatus generally sold, and carried to the
balance room to be transferred to the large desiccator, where it is
allowed to become thoroughly cold before being weighed. Twenty minutes
is generally the time allowed after ignition before it is advisable to
weigh. Bodies allowed to cool in the air after they have been ignited will
absorb moisture, and hot bodies placed in the balance-pan will disturb
the equilibrium and show false results. Compounds that absorb moisture
must be weighed quickly; they should, therefore, be weighed in covered
vessels. Such compounds are detected by their continually-increasing
weight. They should be ignited and weighed again
in a well-covered dish.

Substances that have been washed with alcohol, ether, or any readily
volatile liquid are dried in the water oven. They quickly dry if there
is no water present, and are generally fit for weighing in less than one
hour. Sometimes drying for a few minutes only will be sufficient.

The weight of the crucible and precipitate having been obtained, the
weight of the crucible and ash is deducted; for example--

  Crucible and precipitate    10.183 grams.
  Crucible and ash             9.527   "
                              ------
                               0.656   "

The weight of the ash is best added to that of the crucible. The amount
of ash in filter-papers must not be neglected, although papers are now
made almost free from ash, and the amount to be deducted is found by
taking eight or ten papers and burning them until they become white, and
then weighing the ash. The amount varies from 0.004 to 0.0005 gram for
different papers. Having determined the ash, place in the balance-drawer
three of the filter-papers pinned together, with the weights marked on
them in the way shown in fig. 24, so as to be readily seen when there is
occasion to refer to them.

[Illustration: FIG. 24.]

It must be remembered that the determination of small quantities of
substances generally involves the use of reagents which are often
contaminated, as an impurity, with the body sought for. Thus, in
assaying silver, the oxide of lead or metallic lead used is rarely free
from silver; and in the case of arsenic, the acids, zinc or ferric
chloride are sure to contain arsenic. The same observation applies to
the precipitation of lead by zinc, &c. The errors caused by these
impurities are more marked in the determination of material having small
quantities of metal than in that of ores which contain larger
quantities. Errors of this kind are counteracted or neutralised by
"blank" or "blind" determinations. These consist in carrying out by the
side of and during the assay a duplicate experiment with the reagents
only, which are thereby subjected to the same processes of solution,
evaporation, filtration, &c. The final result thus obtained is deducted
from that given by the assay, the difference gives the corrected result.
In some cases, where it is desired or necessary to have a tangible
residue or precipitate, some _pure_ inert material is added.



CHAPTER IV.

VOLUMETRIC ASSAYS.


These have been already described as those in which the results are got
by measuring, either--(1) the volume of a reagent required to complete
some reaction, or (2) the volume of the resulting product. For example,
if a permanganate of potash solution be added to a solution containing a
weighed amount of iron, dissolved in sulphuric acid, the strong colour
of the permanganate of potash will be removed until a certain quantity
of it has been added. Repeating the experiment, it will be found that
the same amount of iron decolorises the same volume of the permanganate
solution within certain narrow limits of variation, known as "error of
experiment." This error is due to variation in the method of working and
to slight differences in the weighings and measurings; it is present in
all experimental methods, although the limits of variation are wider in
some than in others. Apart from this error of experiment, however, it is
certain that a given volume of the permanganate of potash solution
corresponds to a definite weight of iron, so that if either is known the
other may be calculated. Similarly, if a known weight of zinc (or of
carbonate of lime) be dissolved in hydrochloric acid, a gas will be
given off which can be measured, and so long as the conditions of the
experiment do not vary, the same weight of zinc (or of carbonate of
lime) gives off the same volume of gas. The weight of the one can be
determined from the volume of the other.

Or, again, the quantity of some substances may be measured by the colour
of their solutions, on the principle that, other things being equal, the
colour of a solution depends upon the quantity of colouring matter
present. So that if two solutions of the same substance are equally
coloured they are of equal strength. In this way an unknown may be
compared with a known strength, and a fairly accurate determination may
be made. These three illustrations serve as types of the three chief
classes of volumetric assays--titrometric, gasometric, and colorimetric.

~Titrometric Assays.~--Within the limits of the error of experiment, a
definite volume of a solution or gas represents a certain weight of
metal or other substance, hence the exact weight may be determined by
experiment. The error of experiment may be reduced to insignificant
dimensions by repeating the experiment, and taking the mean of three or
four determinations. This will at the same time show the amount of
variation. Thus, if 0.5 gram of iron were dissolved and found to require
50.3 cubic centimetres of the solution of permanganate of potash, and if
on repeating, 50.4, 50.2, and 50.3 c.c. were required, the experimenter
would be justified in saying that 50.3 c.c. of the permanganate solution
represent 0.5 gram of iron, and that his results were good within 0.2
c.c. of the permanganate solution. So that if in an unknown solution of
iron, 50.5 c.c. of the permanganate solution were used up, he could
state with confidence that it contained a little more than 0.5 gram of
iron. With a larger experience the confidence would increase, and with
practice the experimental error will diminish.

But supposing that the unknown solution required, say, 100.5 instead of
50.5 c.c., he would not be justified in saying that, since 50.3 c.c. are
equivalent to 0.5 gram, 100.6 c.c. are equivalent to twice that amount;
and that, consequently, the unknown solution contained a little less
than 1 gram of iron; or, at least, he could not say it except he (or
some one else) had determined it by experiment. But if on dissolving 1
gram of iron, he found it to require 100.6 c.c. of the solution, and in
another experiment with 0.8 gram of iron that 80.5 c.c. of the solution
were required, he would be justified in stating that _the volume of
solution required is proportional to the quantity of metal present_.
There are a large number of volumetric assays of which this is true, but
that it is true in any particular case can only be proved by experiment.
Even where true it is well not to rest too much weight upon it, and in
all cases the quantity of metal taken, to determine the strength of the
solution used, should not differ widely from that present in the assay.
There are certain terms which should be explained here. When the
solution of a reagent is applied under such conditions that the volume
added can be correctly determined, the operation is called "titrating,"
the solution of the reagent used the "standard solution," and the
process of determining the strength of the standard solution is
"standardising." The "standard" is the quantity of metal equivalent to
100 c.c. of the standard solution.

~Standard Solutions.~--In making these the salt is accurately weighed
and transferred to a litre flask, or to the graduated cylinder, and
dissolved. The method of dissolving it varies in special cases, and
instructions for these will be found under the respective assays.
Generally it is dissolved in a small quantity of liquid, and then
diluted to the mark. For those substances that require the aid of heat,
the solution is made in a pint flask, cooled, and transferred; after
which the flask is well washed out. After dilution, the liquids in the
measuring vessel must be thoroughly mixed by shaking. This is more
easily and better done in the cylinder than in the litre flask. The
solution is next transferred to a dry "Winchester" bottle and labelled.
The label may be rendered permanent by waxing it.

Standard solutions should not be kept in a place exposed to direct
sunlight. Oxidising and reducing solutions, such as those of
permanganate of potash, ferrous sulphate, iodine, hyposulphite of soda,
&c., gradually weaken in strength; the solutions of other salts are more
stable; while those of potassium bichromate and baric chloride are
almost permanent. Solutions of potassium permanganate may be kept for a
month or so without much change. The solutions of hyposulphite of soda
and of iodine should be examined weekly. Ferrous sulphate solutions, if
acidulated with sulphuric acid, may be depended on for two or three
weeks without fresh standardising. Before filling the burette, the
"Winchester" bottle should be well shaken and a portion of about 50 or
100 c.c. poured into a dry beaker or test-glass. Besides the standard
solutions, which are required for titrating an assay, permanent
solutions of the metal or acid of equivalent strength are very useful.
When the finishing point of a titration has been overstepped (_i.e._,
the assay has been "overdone"), a measured volume, say 5 or 10 c.c., of
a solution containing the same metal may be added. The titration can
then be continued, but more cautiously, and the value in "c.c." for the
quantity added be deducted from the final reading.

~Standardising.~--Suppose the object is to standardise a solution of
permanganate similar to that referred to above. A convenient quantity of
iron (say 0.5 gram) would be weighed out, dissolved in dilute sulphuric
acid, and the solution titrated. Suppose 49.6 c.c. of the permanganate
solution are required, then

        49.6 : 0.5 :: 100 : _x_
                            _x_ = 1.008 gram.

This result, 1.008 gram, is the "_standard_." When a gas is measured,
the standard may be calculated in the same way. For example: with 0.224
gram of zinc, 75.8 c.c. of gas were obtained. Then the quantity of zinc
equivalent to 100 c.c. of the gas is got by the proportion.

        75.8 : 0.224 :: 100 : _x_
                              _x_ = 0.2955 gram.

Using the term "standard" in this sense, the following rules hold
good:--

To find the weight of metal in a given substance:--_Multiply the
standard by the number of c.c. used and divide by 100._ For example: a
piece of zinc was dissolved and the gas evolved measured 73.9 c.c. Then
by the rule, 0.2955×73.9/100 should give the weight of the piece of
zinc. This gives 0.2184 gram.

To find the percentage of metal in a given substance:--_Multiply the
standard by the number of c.c. used and divide by the weight of
substance taken._ For example: if 2 grams of a mineral were taken, and
if on titrating with the permanganate solution (standard 1.008) 60.4
c.c. were required, then (1.008×60.4)/2 = 30.44. This is the
percentage.

If the standard is exactly 1 gram, and 1 gram of ore is always taken,
these calculations become very simple. The "c.c." used give at once the
percentage, or divided by 100 give the weight of metal.

If it is desired to have a solution with a standard exactly 1.0 gram, it
is best first to make one rather stronger than this, and then to
standardise carefully. Divide 1000 by the standard thus obtained and the
result will be the number of c.c. which must be taken and be diluted
with water to 1 litre. For example: suppose the standard is 1.008, then
1000/1.008 gives 992, and if 992 c.c. be taken and diluted with water to
1000 c.c. a solution of the desired strength will be obtained. The
standard of this should be confirmed. A simpler calculation for the same
purpose is to multiply the standard by 1000; this will give the number
of c.c. to which 1 litre of the solution should be diluted. In the above
example a litre should be diluted to 1008 c.c.

It has been assumed in these rules that the titration has yielded
proportional results; but these are not always obtained. There can be no
doubt that in any actual re-action the proportion between any two
re-agents is a fixed one, and that if we double one of these then
exactly twice as much of the other will enter into the re-action; but in
the working it may very well be that no re-action at all will take place
until after a certain excess of one or of both of the re-agents is
present. In titrating lead with a chromate of potash solution, for
example, it is possible that at the end of the titration a small
quantity of the lead may remain unacted on; and it is certain that a
small excess of the chromate is present in the solution. So, too, in
precipitating a solution of silver with a standard solution of common
salt, a point is reached at which a small quantity of each remains in
solution; a further addition either of silver or of salt will cause a
precipitate, and a similar phenomenon has been observed in precipitating
a hydrochloric acid solution of a sulphate with baric chloride. The
excess of one or other of the re-agents may be large or small; or, in
some cases, they may neutralise each other. Considerations like these
emphasise the necessity for uniformity in the mode of working. Whether a
process yields proportional results, or not, will be seen from a series
of standardisings. Having obtained these, the results should be arranged
as in the table, placing the quantities of metal used in the order of
weight in the first column, the volumes measured in the second, and the
standards calculated in the third. If the results are proportional,
these standards will vary more or less, according to the delicacy of the
process, but there will be no apparent order in the variation. The
average of the standards should then be taken.

  +-------------+---------------+----------+
  |   Weight.   | Volume found. | Standard |
  +-------------+---------------+----------+
  | 0.2160 gram |   72.9 c.c.   |  0.2963  |
  | 0.2185  "   |   73.9  "     |  0.2957  |
  | 0.2365  "   |   79.9  "     |  0.2959  |
  | 0.2440  "   |   82.3  "     |  0.2964  |
  | 0.2555  "   |   85.9  "     |  0.2974  |
  +-------------+---------------+----------+

Any inclination that may be felt for obtaining an appearance of greater
accuracy by ignoring the last result must be resisted. For, although it
would make no practical difference whether the mean standard is taken as
0.2961 or 0.2963, it is well not to ignore the possibility that an error
of 0.4 c.c. may arise. A result should only be ignored when the cause of
its variation is known.

In this series the results are proportional, but the range of weights
(0.216-0.2555 gram) is small. All processes yield fairly proportional
results if the quantities vary within narrow limits.

As to results which are not proportional, it is best to take some
imaginary examples, and then to apply the lesson to an actual one. A
series of titrations of a copper solution by means of a solution of
potassic cyanide gave the following results:--

  +---------------+---------------+-----------+
  | Copper taken. | Cyanide used. | Standard. |
  +---------------+---------------+-----------+
  |   0.1 gram    |   11.9 c.c.   |   0.8403  |
  |   0.2  "      |   23.7  "     |   0.8438  |
  |   0.3  "      |   35.6  "     |   0.8426  |
  |   0.4  "      |   47.6  "     |   0.8403  |
  +---------------+---------------+-----------+

These are proportional, but by using a larger quantity of acid and
ammonia in the work preliminary to titration, we might have had to use
1 c.c. of cyanide solution more in each case before the finishing point
was reached. The results would then have been:

  +---------------+---------------+-----------+
  | Copper taken. | Cyanide used. | Standard. |
  +---------------+---------------+-----------|
  |    0.1 gram   |   12.9 c.c.   |   0.7752  |
  |    0.2   "    |   24.7  "     |   0.8097  |
  |    0.3   "    |   36.6  "     |   0.8191  |
  |    0.4   "    |   48.6  "     |   0.8230  |
  +---------------+---------------+-----------+

It will be noted that the value of the standard increases with the
weight of metal used; and calculations from the mean standard will be
incorrect.

By subtracting the lowest standardising from the highest, a third result
is got free from any error common to the other two; thus:--

    0.4 gram = 48.6 c.c. "cyanide."
    0.1  "   = 12.9   "     "
    ---        ----
    0.3  "   = 35.7   "     "

And the standard calculated from this corrected result is 0.8404.
Further, if 0.3 gram requires 35.7 c.c., then 0.1 gram should require
11.9 c.c., or 1.0 c.c. less than that actually found.

We may therefore use the following rules for working processes which do
not yield proportional results. Make a series of two or three
titrations, using very different quantities of metal in each. Subtract
the lowest of these from the highest, and calculate the standard with
the remainder. Calculate the volume required by this standard in any
case, and find the excess or deficit, as the case may be. If an excess,
subtract it from the result of each titration; if a deficit, add it; and
use the standard in the usual way. The following table shows an actual
example:--

  +--------------+---------------+-----------+
  | Chalk taken. | Gas obtained. | Standard. |
  +--------------+---------------+-----------+
  |  0.0873 gram |   17.8 c.c.   |  0.4904   |
  |  0.1305  "   |   27.3  "     |  0.4780   |
  |  0.1690  "   |   35.8  "     |  0.4721   |
  |  0.1905  "   |   40.4  "     |  0.4715   |
  |  0.2460  "   |   52.5  "     |  0.4686   |
  |  0.3000  "   |   64.0  "     |  0.4687   |
  +--------------+---------------+-----------+

It will be seen that the standard decreases as the quantity of chalk
increases; this points to a deficiency in the quantity of gas evolved.

Then

    0.3000 = 64.0 c.c.
    0.0873 = 17.8  "
    ------ = ----
    0.2127 = 46.2  "

and 0.2127×100/46.2 = 0.4604. Then, multiplying the weight of chalk
taken by 100, and dividing by 0.4604, we get the calculated results of
the following table:--

  +--------------+------------+-----------------+-------------+
  | Chalk taken. | Gas found. | Gas calculated. | Difference. |
  +--------------+------------+-----------------+-------------+
  | 0.0873 gram  | 17.8 c.c.  |    18.9 c.c.    | -1.1 c.c.   |
  | 0.1305  "    | 27.3  "    |    28.3  "      | -1.0  "     |
  | 0.1690  "    | 35.8  "    |    36.7  "      | -0.9  "     |
  | 0.1905  "    | 40.4  "    |    41.4  "      | -1.0  "     |
  | 0.2460  "    | 52.5  "    |    53.4  "      | -0.9  "     |
  | 0.3000  "    | 64.0  "    |    65.1  "      | -1.1  "     |
  +--------------+------------+-----------------+-------------+

By adding 1 c.c. to the quantity of gas obtained, and taking 0.4604 as
the standard, the calculated results will agree with those found with a
variation of 0.1 c.c. When a large number of assays of the same kind are
being made, this method of calculation is convenient; when, however,
only one or two determinations are in question, it is easier to make a
couple of standardisings, taking quantities as nearly as possible the
same as those present in the assays.

Sometimes it is necessary to draw up a table which will show, without
calculation, the weight of substance equivalent to a given volume of gas
or of solution. The substance used for standardising should be, whenever
possible, a pure sample of the substance to be determined--that is, for
copper assays pure copper should be used, for iron assays pure iron, and
so on; but when this cannot be got an impure substance may be used,
provided it contains a known percentage of the metal, and that the
impurities present are not such as will interfere with the accuracy of
the assay. Including compounds with these, the standard may be
calculated by multiplying the standard got in the usual way, by the
percentage of metal in the compound or impure substance, and dividing by
100. If, for example, the standard 1.008 gram was obtained by using a
sample of iron containing 99.7 per cent. of metal, the corrected
standard would be 1.008×99.7/100 = 1.005.

In volumetric analysis the change brought about must be one in which the
end of the reaction is rendered prominent either by a change of colour
or by the presence or absence of a precipitate. If the end of the
reaction or finishing-point is not of itself visible, then it must be
rendered visible by the use of a third reagent called an indicator.

For example, the action of sulphuric acid upon soda results in nothing
which makes the action conspicuous; if, however, litmus or
phenolphthalein be added the change from blue to red in the first case,
or from red to colourless in the second, renders the finishing-point
evident. Some indicators cannot be added to the assay solution without
spoiling the result; in which case portions of the assay solution must
be withdrawn from time to time and tested. This withdrawal of portions
of the assay solution, if rashly done, must result in loss; if, however,
the solution is not concentrated, and if the portions are only withdrawn
towards the end of the titration, the loss is very trifling, and will
not show-up on the result. The usual plan adopted is to have a solution
of the indicator placed in drops at fairly equal intervals distributed
over a clean and dry white porcelain-plate: a drop or two of the
solution to be tested is then brought in contact with one of these and
the effect noted. Another plan is to have thin blotting-paper, moistened
with a solution of the indicator and dried; a drop of the solution to be
tested placed on this shows the characteristic change. When the assay
solution contains a suspended solid which interferes with the test, a
prepared paper covered with an ordinary filter-paper answers very well;
a drop of the solution to be tested is placed on the filter-paper, and,
sinking through, shows its effect on the paper below.

Except when otherwise stated, all titrations should be made at the
ordinary temperature; cooling, if necessary, by holding the flask under
the tap. When a titration is directed to be made in a boiling solution,
it must be remembered that the standard solution is cold, and that every
addition lowers the temperature of the assay.

On running the solution from the burette into the assay, do not let it
run down the side of the flask. If a portion of the assay has to be
withdrawn for testing, shake the flask to ensure mixing, and then take
out a drop with the test-rod; the neglect of these precautions may give
a finishing-point too early. This is generally indicated by a sudden
finish, in which case on shaking the flask and again testing no reaction
is got. Do not remove the drop on the point of the burette with the
test-rod; let it remain where it is or drop it into the solution by
carefully opening the clip.

Generally the methods of working are as follows:--

(1) _When the finishing-point depends on a change of colour in the
solution._--Increase the bulk of the assay up to from 100 to 150 c.c.
with water. Boil or cool, as the case may be. Run in the standard
solution from a burette speedily, until the re-agent appears to have a
slower action, and shake or stir all the time. Then run 1 c.c. or so at
a time, still stirring, and finally add drops until the colour change is
got.

(2) _When an outside-indicator is used._--Pour the standard solution
from a burette into the assay until 5 or 6 c.c. from the
finishing-point; then run in 1 c.c. at a time (stirring and testing on
the plate between each) until the indicator shows the change wanted, and
deduct 0.5 c.c. for excess. When greater accuracy is sought for a
duplicate assay is made. In this case the standard solution is run in
close up to the end, and the operation is finished off with a few drops
at a time.

(3) _Where the finishing-point depends upon the absence of a precipitate
and no outside-indicator is used._--As in the last case, run in the
standard solution up to within a few c.c. of the end, then run in 1 c.c.
at a time until a precipitate is no longer formed, but here 1.5 c.c.
must be deducted for excess, since it is evident that the whole of the
last "c.c." must have been, and a portion of the previous one may have
been, in excess.

~Indirect Titration.~--The action of permanganate of potash upon a
ferrous solution is one of oxidation, hence it is evident that if any
other oxidising agent is present it will count as permanganate. In such
a case the titration can be used (indirectly) to estimate the quantity
of such oxidising agent, by determining how much less of the
permanganate is used. For example, suppose that 1 gram of iron dissolved
in sulphuric acid requires 100 c.c. of standard permanganate to fully
oxidise it, but that the same amount of iron only requires 35.6 c.c. of
the same standard permanganate if it has been previously heated with 0.5
gram of black oxide of manganese. Here it is evident that 0.5 gram of
black oxide does the work of 64.4 c.c.[4] of the permanganate solution,
and that these quantities are equivalent; moreover, if 64.4 c.c.
correspond with 0.5 gram, then 100 c.c. correspond with 0.7764 which is
the standard. On theoretical grounds, and by a method of calculation
which will be explained further on (under the heading "Calculations from
Formulæ"), it can be found that if the standard for iron is 1 gram, that
for the black oxide will be 0.7764 gram.

The principles of these indirect titrations become clearer when
expressed in a condensed form. Thus, in the example selected, and using
the formulæ Fe = Iron, KMnO_{4} = permanganate of potash, and MnO_{2} =
oxide of manganese, we have:--

(1)                1 gram Fe = 100 c.c. KMnO_{4}

(2)                1 gram Fe = 35.6 c.c. KMnO_{4} + 0.5 gram MnO_{2}
       .'. 100 c.c. KMnO_{4} = 35.6 c.c. KMnO_{4} + 0.5 gram MnO_{2}
    (100-35.6) c.c. KMnO_{4} = 0.5 gram MnO_{2}
          64.4 c.c. KMnO_{4} = 0.5 gram MnO_{2}

The iron does not enter into the calculation if the same quantity is
present in the two experiments.

An indirect titration thus requires three determinations, but if more
than one assay is to be carried on, two of these need not be repeated.
The standard is calculated in the usual way.

~Colorimetric Assays.~--These are assays in which the colour imparted to
a solution by some compound of the metal to be determined is taken
advantage of; the depth of colour depending on the quantity of metal
present. They are generally used for the determination of such small
quantities as are too minute to be weighed. The method of working is as
follows:--A measured portion of the assay solution (generally 2/3, 1/2,
1/3, or 1/4 of the whole), coloured by the substance to be estimated, is
placed in a white glass cylinder standing on a sheet of white paper or
glazed porcelain. Into an exactly similar cylinder is placed the same
amount of re-agents, &c., as the portion of the assay solution contains,
and then water is added until the solutions are of nearly equal bulk.
Next, a standard solution of the metal being estimated is run in from a
burette, the mixture being stirred after each addition until the colour
approaches that of the assay. The bulk of the two solutions is equalised
by adding water. Then more standard solution is added until the tints
are very nearly alike. Next, the amount added is read off from the
burette, still more is poured in until the colour is slightly darker
than that of the assay, and the burette read off again. The mean of the
readings is taken, and gives the quantity of metal added. It equals the
quantity of metal in the portion of the assay. If this portion was
one-half of the whole, multiply by two; if one-third, multiply by three,
and so on. When the quantity of metal in very dilute solutions is to be
determined, it is sometimes necessary to concentrate the solutions by
boiling them down before applying the re-agent which produces the
coloured compound. Such concentration does not affect the calculations.

~Gasometric Assays.~--Gasometric methods are not much used by assayers,
and, therefore, those students who wish to study them more fully than
the limits of this work will permit, are recommended to consult Winkler
and Lunge's text-book on the subject. The methods are without doubt
capable of a more extended application. In measuring liquids, ordinary
variations of temperature have but little effect, and variations of
atmospheric pressure have none at all, whereas with gases it is
different. Thus, 100 c.c. of an ordinary aqueous solution would, if
heated from 10° C. to 20° C., expand to about 100.15 c.c. 100 c.c. of a
gas similarly warmed would expand to about 103.5 c.c., and a fall of one
inch in the barometer would have a very similar effect. And in
measuring gases we have not only to take into account variations in
volume due to changes in temperature and atmospheric pressure, but also
that which is observed when a gas is measured wet and dry. Water gives
off vapour at all temperatures, but the amount of vapour is larger as
the temperature increases.

By ignoring these considerations, errors of 3 or 4 per cent. are easily
made; but, fortunately, the corrections are simple, and it is easy to
construct a piece of apparatus by means of which they may be reduced to
a simple calculation by the rule of three.

The volume of a gas is, in practice, usually reduced to that which it
would be at a temperature of 0° C., when the column of mercury in the
barometer is 760 mm. high. But, although convenient, this practice is
not always necessary. The only thing required is some way of checking
the variations in volume, and of calculating what the corrected volume
would be under certain fixed conditions.

Suppose that at the time a series of standardisings is being made, 100
c.c. of air were confined in a graduated tube over moist mercury. These
100 c.c. would vary in volume from day to day, but it would always be
true of them that they would measure 100 c.c. under the same conditions
as those under which the standardisings were made. If, then, in making
an actual assay, 35.4 c.c. of gas were obtained, and the air in the tube
measured 105 c.c., we should be justified in saying, that if the
conditions had been those of the standardising, the 105 c.c. would have
measured 100 c.c., and the 35.4 c.c. would have been 33.7; for 105:
100:: 35.4: 33.7. The rule for using such a piece of apparatus for
correcting volumes is:--_Multiply the c.c. of gas obtained by 100, and
divide by the number of c.c. of air in the apparatus._

If it is desired to calculate the volumes under standard conditions
(that is, the gas dry, at 0° C. and 760 mm. barometric pressure) the
calculations are easily performed, but the temperature and pressure must
be known.

_Correction for Moisture._--The "vapour tension" of water has been
accurately determined for various temperatures, and it may be looked
upon as counteracting the barometric pressure. For example, at 15° C.
the vapour tension equals 12.7 millimetres of mercury; if the barometer
stood at 750 mm., the correction for moisture would be made by
subtracting 12.7 from 750, and taking 737.3 mm. to be the true
barometric pressure.

The vapour tensions for temperatures from 0° C. to 20° C. are as
follows:--

  -------+----------++-------+----------++-------+----------
   Temp. | Tension. || Temp. | Tension. || Temp. | Tension.
  -------+----------++-------+----------++-------+----------
     0°  | 4.6 mm.  ||   7°  |  7.5 mm. ||  14°  | 11.9 mm.
     1°  | 4.9 mm.  ||   8°  |  8.0 mm. ||  15°  | 12.7 mm.
     2°  | 5.3 mm.  ||   9°  |  8.6 mm. ||  16°  | 13.5 mm.
     3°  | 5.7 mm.  ||  10°  |  9.2 mm. ||  17°  | 14.4 mm.
     4°  | 6.1 mm.  ||  11°  |  9.8 mm. ||  18°  | 15.3 mm.
     5°  | 6.5 mm.  ||  12°  | 10.5 mm. ||  19°  | 16.3 mm.
     6°  | 7.0 mm.  ||  13°  | 11.2 mm. ||  20°  | 17.4 mm.
  -------+----------++-------+----------++-------+-----------

The _correction for pressure_ is:--Multiply the volume by the actual
pressure and divide by 760.

The _correction for temperature_:--Multiply the volume by 273 and divide
by the temperature (in degrees Centigrade) added to 273.

For all three corrections the following rules hold good. _To reduce to
0° C. and 760 mm. dry._

                        Volume × 0.3592 × (Pressure-tension)
  Corrected volume = --------------------------------------
                              Temperature + 273

To find the volume, which a given volume under standard conditions would
assume, if those conditions are altered.

                      Volume × 2.784 × (Temperature + 273)
  Resulting volume =  ------------------------------------
                               Pressure - tension

As an example, we will suppose that it is desired to enclose in the
apparatus referred to on p. 45, a volume of air, which, when dry (at 0°
C. and 760 mm.), shall measure 100 c.c., whilst the actual temperature
is 15° C., and the pressure 750 mm.

The second formula is the one to be used, and we get 108.7 c.c.

                       100 c.c.×2.784×288
  Required volume =  ----------------------
                           750-12.7

                    80179.2
                  = -------
                     737.3

                  = 108.7 c.c.

FOOTNOTES:

[4] 100-35.6 = 64.4.



CHAPTER V.

WEIGHING AND MEASURING.


~Weighing.~--The system of weights and measures which we have adopted is
the French or metric system; in this the gram (15.43 grains) is the unit
of weight; the only other weight frequently referred to is the
milligram, which is 0.001, or 1/1000 gram. The unit of volume is the
cubic centimetre, which is approximately the volume of 1 gram of water,
and which thus bears to the gram the same relation as grain-measures
bear to grains. It is usual to write and even pronounce cubic centimetre
shortly as c.c., and the only other denomination of volume we shall have
occasion to use is the "litre," which measures 1000 c.c., and is roughly
1-3/4 pints.

The weights used are kept in boxes in a definite order, so that the
weights on the balance can be counted as well by noting those which are
absent from the box as by counting those present on the scale-pan. The
weights run 50, 20, 10, 10, 5, 2, 1, 1 and 1 grams, and are formed of
brass. The fractions of the gram are generally made of platinum or of
aluminium, and are arranged in the following order:--0.5, 0.2, 0.1, 0.1,
and 0.05, 0.02, 0.01, 0.01. These may be marked in this way, or they may
be marked 500, 200, 100, 100, 50, 20, 10, 10; the 500 meaning 500
milligrams.

Some makers send out weights in the series 50, 20, 20, 10, &c.

Weights of less than 0.01 gram are generally present in a box, but it is
much more convenient to work with a rider. This is a piece of wire which
in the pan weighs 0.01 gram; it is made in such a form that it will ride
on the beam, and its effective weight decreases as it approaches the
centre. If the arm of the beam is divided into tenths, then each tenth
counting from the centre outward equals 0.001 gram or 1 milligram, and
if these tenths be further subdivided the fractions of a milligram are
obtained; and these give figures in the fourth place of decimals. A
fairly good balance should be sensitive to 0.0001 gram. The weights must
never be touched with the fingers, and the forceps for moving them is
used for no other purpose. When not in actual use the box is kept
closed. The weights must not be allowed to remain on the pan of the
balance. The balance-case must not be open without some reason. It must
be fixed level, and, once fixed, must not be needlessly moved. The bench
on which it stands should be used for no other purpose, and no one
should be allowed to lean upon it.

[Illustration: FIG. 25.]

When using a balance sit directly in front of it. Ordinarily the
substance to be weighed is best put on the pan to the user's left; the
weights and the rider are then easily manipulated. Powders, &c., should
not be weighed directly on the balance; a counterpoised watch-glass or
metal scoop (fig. 25) should be used. In some cases it is advisable to
use a weighing-bottle. This is a light, well-stoppered bottle (fig. 3)
containing the powdered ore. It is first filled and weighed; then some
of the substance is carefully poured from it into a beaker or other
vessel, and it is weighed again; the difference in the two weighings
gives the weight of substance taken. A substance must always be cold
when weighed, and large glass vessels should be allowed to stand in the
balance-box a little while before being weighed. Always have the balance
at rest when putting on or taking off anything from the pans. Put the
weights on systematically. In using the rider (except you have a reason
to the contrary), put it on at the 5; if this is too much, then try it
at the 3; if then the weights are too little, try at the 4, if still not
enough, the correct weight must be between the 4 and 5; try half-way
between.

It is best to work with the balance vibrating; equilibrium is
established when the vibration to the left is the mean of the preceding
and succeeding vibrations to the right. For example, if it vibrates 6
divisions to the right on one swing, and 5 divisions on the next, the
intermediate vibration to the left should have been 5-1/2.

Note whether the substance increases in weight whilst on the balance. If
it does it may be because it was put on warm, and is cooling, or it may
be because it is taking up moisture from the air. Substances which take
up moisture rapidly should be weighed in clipped watch-glasses or in
light-weighing bottles or tubes.

Students, in recording the weights, should first read off those missing
from the box, writing down each order of figures as determined; first
tens, then units, and so on. Remember that the first four platinum
weights give the figures of the first place of decimals, the second four
give the second place, and that the third and fourth places are given by
the rider. Having taken down the figures, confirm them by reading off
the weights as you put them back into the box. Do not rest a weight on
the palm of your hand for convenience in reading the mark upon it.
Remember one weight lost from a box spoils the set. Do not take it for
granted that the balance is in equilibrium before you start weighing:
try it.

[Illustration: FIG. 26.]

~Measuring Liquids.~--For coarse work, such as measuring acids for
dissolving ores, graduated glasses similar to those used by druggists
may be used. It is well to have two sizes--a smaller graduated into
divisions of 5 c.c. (fig. 26), and a larger with divisions equal to 10
c.c. No measurement of importance should be made in a vessel of this
kind, as a slight variation in level causes a serious error.

~Graduated flasks~ must be used when anything has to be made up to a
definite bulk, or when a fixed volume has to be collected. If, for
example, a certain weight of substance has to be dissolved and diluted
to a litre, or if the first 50 c.c. of a distillate has to be collected,
a flask should be used. Each flask is graduated for one particular
quantity; the most useful sizes are 1000 c.c., 500 c.c., 200 c.c., 100
c.c., and 50 c.c. The mark should be in the narrowest part of the neck,
and should be tangential to the curved surface of the liquid when the
flask _contains_ the exact volume specified. The level of a curved
surface of liquid is at first somewhat difficult to read: the beginner
is in doubt whether the surface should be taken at A, B, or C (fig. 27).
It is best to take the lowest reading C. In some lights it is difficult
to find this; in such cases a piece of white paper or card held behind
and a little below, so as to throw light up and against the curved
surface, will render it clear. In reading, one should look neither up at
nor down upon the surface, but the eye should be on the same level with
it. It must be kept in mind that flasks _contain_ the quantity
specified, but deliver less than this by the amount remaining in them
and damping the sides. If it is desired to transfer the contents say of
a 100 c.c. flask to a beaker, it will be necessary to complete the
transfer by rinsing out the flask and adding the washings; otherwise
there will be a sensible loss. Graduated cylinders (fig. 28) are
convenient for preparing standard solutions.

[Illustration: FIG. 27.]

[Illustration: FIG. 28.]

[Illustration: FIG. 29.]

~Pipettes~ and burettes are graduated to _deliver_ the quantities
specified. The principle of the pipette, and the advantages and
disadvantages of its various forms, may be understood by considering the
first form shown in fig. 29. It is essentially a bulbed tube drawn out
to a jet at its lower end, and having on each side of the bulb a mark so
placed that when the surface of the liquid falls from the upper to the
lower mark the instrument shall deliver exactly 100 c.c. The bore of the
jet should be of such a size as will allow the level of the liquid to
fall at the rate of about one foot in two minutes. If it runs more
quickly than this, an appreciable error arises from the varying amount
of liquid remaining, and damping the sides of the bulb. The flow of
liquid from a pipette must not be hastened by blowing into it. The lower
tube or nose of the pipette should be long enough to reach into the
bottle or flask containing the liquid about to be measured. The pipette
is filled by sucking at the open end with the mouth; this method of
filling renders the use of the instrument dangerous for such liquids as
strong acids, ammonia, and such poisonous solutions as that of potassic
cyanide. One attempt with a fairly strong solution of ammonia will teach
the beginner a very useful lesson. As soon as the liquid rises above the
upper mark in the pipette, the mouth is withdrawn, and the pipette
quickly closed by pressing the upper aperture with the index finger of
the right hand; it is well to have the finger slightly moist, but not
damp. The neck of the pipette should be long enough to allow its being
firmly grasped by the fingers and thumb of the right hand without
inconvenience. The pipette is first held in a vertical position long
enough to allow any moisture outside the tube to run down, and then the
liquid is allowed to run out to the level of the upper mark; this is
easily effected by lessening the pressure. If the finger is wet, the
flow will be jerky, and good work impossible. The pipette is next held
over the vessel into which the 100 c.c. are to be put, and the liquid
allowed to run out. When the bulb is nearly empty, the flow should be
checked by replacing the finger, and the liquid allowed to escape slowly
until the lower mark is reached. The pipette is then withdrawn; it is in
the withdrawing that the disadvantage of this particular form[5] makes
itself felt. It must be withdrawn very steadily, as the slightest shock
causes the remaining column of liquid to vibrate, whereby air is drawn
in and the liquid is forced out.

This disadvantage is got rid of by making the mouth of the jet the lower
limit, or, in other words, allowing the instrument to empty itself.
There are two forms of such pipettes; in the one generally recommended
in Gay-Lussac's silver assay (the last shown in fig. 29) the nose is
replaced by a jet. This is most conveniently filled by stopping the jet
with the finger, and allowing the liquid to flow in a fine stream into
the neck until the pipette is filled, and then working as just
described. The other form is the one in general use; in fact, a long
nose to a pipette is so convenient that it may almost be said to be
necessary. But the accuracy is slightly diminished; a long narrow tube
makes a poor measuring instrument because of the amount of liquid it
finally retains. A defect possessed by both forms is the retention of a
drop of varying size in the nozzle. Whatever method is adopted for
removing this drop must be always adhered to. The most convenient form
is the one last described, and the most useful sizes are 100 c.c., 50
c.c., 20 c.c., 10 c.c., and 5 c.c. Ten c.c. pipettes graduated into
tenths of a cubic centimetre are very useful: those are best in which
the graduation stops short of the bottom.

All measurements should be made at the ordinary temperature; and, before
being used, the pipette should be rinsed out with a cubic centimetre or
so of the solution to be measured. After using, it should be washed out
with water.

~Burettes~ differ mainly from pipettes in having the flow of liquid
controlled from below instead of from above. The best form is that known
as Mohr's, one kind of which is provided with a glass stopcock, while
the other has a piece of india-rubber tube compressed by a clip. The
latter cannot be used for solutions of permanganate of potash or of
iodine, or of any substance which acts on india-rubber; but in other
respects there is little to choose between the two kinds. A burette
delivering 100 c.c., and graduated into fifths (_i.e._, each division =
0.2 c.c.), is a very convenient size. For some kinds of work, 50 c.c.
divided into tenths (_i.e._, each division = 0.1 c.c.) may be selected.

Burettes may be fixed in any convenient stand; they must be vertical and
should be so placed that the assayer can read any part of the graduated
scale without straining. When not in use, they should be kept full of
water. When using a burette, the water must be run out; the burette is
next rinsed with some of the solution to be used, and drained; and then
it is filled with the solution. Next squeeze the india-rubber tube so as
to disentangle air-bubbles and, by smartly opening the clip, allow the
tube and jet to be filled; see that no bubbles of air are left. Then
run out cautiously until the level of the liquid in the burette stands
at zero. In reading the level with very dark-coloured liquids it is
convenient to read from the level A (fig. 27), and, provided it is done
in each reading, there is no objection to this. The accuracy of the
reading of a burette is sensibly increased by the use of an Erdmann
float. This is an elongated bulb, weighted with mercury, and fitting
(somewhat loosely) the tube of the burette. It floats in the solution,
and is marked with a horizontal line; this line is taken as the level of
the liquid. If the burette is filled from the top, the float rises with
aggravating slowness, and this is its chief disadvantage. The float must
come to rest before any reading is made.

[Illustration: FIG. 30.]

A convenient plan for filling a burette from below is shown in fig. 30.
The diagram explains itself. The bottle containing the standard solution
is connected with the burette by a syphon arrangement through the glass
tube and T-piece. The flow of liquid into the burette is controlled by
the clip. When this clip is opened, the burette fills; and when it is
closed, the burette is ready for use in the ordinary way.

~Measuring Gases.~--Lange's nitrometer (fig. 69) is a very convenient
instrument for many gasometric methods. It requires the use of a fair
quantity of mercury. In fig. 31, there is a representation of a piece of
apparatus easily fitted up from the ordinary material of a laboratory.
It is one which will serve some useful purposes. It consists of a
wide-mouthed bottle fitted (by preference) with a rubber cork. The cork
is perforated, and in the perforation is placed a glass tube which
communicates with the burette. The burette is connected by a rubber tube
and a Y-piece, either with another burette or with a piece of ordinary
combustion-tube of about the same size. The wide-mouthed bottle contains
either a short test-tube or an ordinary phial with its neck cut off. In
working the apparatus the weighed substance is put in the bottle and the
re-agent which is to act on it, in the test-tube; the cork is then
inserted. The liquid in the two burettes is next brought to the same
level, either by pouring it in at A or running it out at B. The level of
the liquid in the apparatus for correcting variation in volume is then
read and noted. Next, after seeing that the level of the liquid in the
burette has not changed, turn the bottle over on its side so that the
re-agent in the test-tube shall be upset into the bottle. Then, as the
volume of the gas increases, lower the liquid in the burette by running
it out at B, and at the same time keep the level in A half an inch or so
lower than that in the burette. When the action has finished bring the
liquid in the two vessels to the same level and read off the burette.
This part of the work must always be done in the same manner.

[Illustration: FIG. 31.]

_The volume corrector for gas analysis_ is a graduated glass tube of 120
c.c. capacity inverted over a narrow glass cylinder of mercury. It
contains 0.2 or 0.3 c.c. of water and a volume of air, which, if dry and
under standard conditions, would measure 100 c.c. The actual volume
varies from day to day, and is read off at any time by bringing the
mercury inside and outside to the same level. This is done by raising or
lowering the tube, as may be required. Any volume of gas obtained in an
assay can be corrected to standard temperature and pressure by
multiplying by 100 and dividing by the number of c.c. in the corrector
at the time the assay is made.

FOOTNOTES:

[5] It is best to use this form with a glass stopcock, or with an
india-rubber tube and clip, after the manner of a Mohr's burette.



CHAPTER VI.

RE-AGENTS.--ACIDS, ETC.


~Acetic Acid~, H[=A=c] or C_{2}H_{4}O_{2}. (sp. gr. 1.044, containing 33
per cent. real acid).--An organic acid, forming a class of salts,
acetates, which are for the most part soluble in water, and which, on
ignition, leave the oxide or carbonate of the metal. It is almost always
used in those cases where mineral acids are objectionable. To convert,
for example, a solution of a substance in hydrochloric acid into a
solution of the same in acetic acid, alkali should be added in excess
and then acetic acid. Many compounds are insoluble in acetic acid, which
are soluble in mineral acids, such as ferric phosphate, ferric arsenate,
zinc sulphide, calcium oxalate, &c., so that the use of acetic acid is
valuable in some separations. The commercial acid is strong enough for
most purposes, and is used without dilution.

~"Aqua Regia"~ is a mixture of 1 part by measure of nitric acid and 3
parts of hydrochloric acid. The acids react forming what is practically
a solution of chlorine.[6] The mixture is best made when wanted, and is
chiefly used for the solution of gold and platinum and for "opening up"
sulphides. When solutions in aqua regia are evaporated, chlorides are
left.

~Bromine~, Br. (sp. gr. 3.0). Practically pure bromine.--It is a heavy
reddish-brown liquid and very volatile. It boils at 60° C., and,
consequently, must be kept in a cool place. It gives off brown
irritating vapours, which render its use very objectionable. Generally
it answers the same purpose as aqua regia, and is employed where the
addition of nitric acid to a solution has to be specially avoided. It is
also used for dissolving metals only from ores which contain metallic
oxides not desired in the solution.

~"Bromine Water"~ is simply bromine shaken up with water till no more is
dissolved.

~Carbonic Acid~, CO_{2}.--A heavy gas, somewhat soluble in water; it is
mainly used for providing an atmosphere in which substances may be
dissolved, titrated, &c., without fear of oxidation. It is also used in
titrating arsenic assays with "iodine" when a feeble acid is required
to prevent the absorption of iodine by the alkaline carbonate. It is
prepared when wanted in solution, by adding a gram or so of bicarbonate
of soda and then as much acid as will decompose the bicarbonate
mentioned. When a quantity of the gas is wanted, it is prepared, in an
apparatus like that used for sulphuretted hydrogen, by acting on
fragments of marble or limestone with dilute hydrochloric acid.

~Citric Acid~ (H_{3}[=C=i] or C_{6}H_{8}O_{7}.H_{2}O) is an organic acid
which occurs in colourless crystals, soluble in less than their weight
of water. The solution must be freshly prepared, as it gets mouldy when
kept. It forms a comparatively unimportant class of salts (citrates). It
is used in the determination of phosphoric acid, chiefly for the purpose
of preventing the precipitation of phosphates of iron and alumina by
ammonia, and in a few similar cases. The commercial crystals are used;
they should be free from sulphuric acid and leave no ash on ignition.

~Hydrochloric Acid~, HCl in water, (sp. gr. 1.16. It contains 32 per
cent. of hydrogen chloride).--It is sometimes called "muriatic acid,"
and when impure, "spirit of salt." The acid solution should be
colourless and free from arsenic, iron, and sulphuric acid. It forms an
important family of salts, the chlorides. It is the best acid for
dissolving metallic oxides and carbonates, and is always used by the
assayer when oxidising agents are to be avoided. The acid is used
without dilution when no directions are expressly given to dilute it. It
has no action on the following metals: gold, platinum, arsenic, and
mercury; it very slightly attacks antimony, bismuth, lead, silver, and
copper. Tin is more soluble in it, but with difficulty; whilst iron,
zinc, nickel, cobalt, cadmium, and aluminium easily dissolve with
evolution of hydrogen and the formation of the lower chloride if the
metal forms more than one class of salts. All the metallic oxides,
except a few of the native and rarer oxides, are dissolved by it with
the formation of chlorides of the metal and water.

~Dilute Hydrochloric Acid~ is made by diluting the strong acid with an
equal volume of water. This is used for dissolving precipitates obtained
in the general course of analysis and the more easily soluble metals.

~Hydrofluoric Acid, HF.~--A solution in water may be purchased in
gutta-percha or lead bottles. It is of variable strength and doubtful
purity. It must always be examined quantitatively for the residue left
on evaporation. It is used occasionally for the examination of
silicates. It attacks silica, forming fluoride of silicon, which is a
gas. When the introduction of another base will not interfere with the
assay, the substance may be mixed in the platinum dish with fluoride of
ammonium, or of potassium, or of calcium, and hydrochloric acid,
instead of treating it with the commercial acid. It is only required in
special work. The fumes and acid are dangerous, and, of course, glass or
porcelain vessels cannot be used with it.

~Iodine, I.~--This can be obtained in commerce quite pure, and is often
used for standardising. It is very slightly soluble in water, but
readily dissolves in potassium iodide solution. It closely resembles
chlorine and bromine in its properties, and can be used for dissolving
metals without, at the same time, attacking any oxide which may be
present. It is chiefly used as an oxidizing agent in volumetric work,
being sharp in its reactions and easily detected in minute quantities.
It cannot be used in alkaline solutions, since it reacts with the
hydrates, and even with the carbonates, to form iodides and iodates.
Iodine is soluble in alcohol.

~Nitric Acid, HNO_{3}.~ (Sp. gr. 1.42; boiling point 121° C.; contains
70 per cent. by weight of hydrogen nitrate).--It is convenient to
remember that one c.c. of this contains 1 gram of real acid. It combines
the properties of an acid and of an oxidising agent. One c.c. contains
0.76 gram of oxygen, most of which is very loosely held, and easily
given up to metals and other oxidisable substances. Consequently it will
dissolve many metals, &c., upon which hydrochloric acid has no action.
All sulphides (that of mercury excepted) are attacked by it, and for the
most part rendered soluble. It has no action on gold or platinum, and
very little on aluminium. The strong acid at the ordinary temperature
does not act on iron or tin; and in most cases it acts better when
diluted. Some nitrates being insoluble in nitric acid, form a protecting
coat to the metal which hinders further action. Where the strong acid
does act the action is very violent, so that generally it is better to
use the dilute acid. When iron has been immersed in strong nitric acid
it not only remains unacted on, but assumes a _passive_ state; so that
if, after being wiped, it is then placed in the dilute acid, it will not
dissolve. Tin and antimony are converted into insoluble oxides, while
the other metals (with the exception of those already mentioned)
dissolve as nitrates. During the solution of the metal red fumes are
given off, which mainly consist of nitrogen peroxide. The solution is
often coloured brown or green because of dissolved oxides of nitrogen,
which must be got rid of by boiling. Generally some ammonium nitrate is
formed, especially in the cases of zinc, iron, and tin, when these are
acted on by cold dilute acid. Sulphur, phosphorus, and arsenic are
converted into sulphuric, phosphoric, and arsenic acids respectively,
when boiled with the strong acid.

~Dilute Nitric Acid.~--Dilute 1 volume of the strong acid with 2 of
water.

~Oxalic Acid~, H_{2}[=O] or (H_{2}C_{2}O_{4}.2H_{2}O.)--This is an organic
acid in colourless crystals. It forms a family of salts--the oxalates.
It is used in standardising; being a crystallised and permanent acid, it
can be readily weighed. It is also used in separations, many of the
oxalates being insoluble. For general use make a 10 per cent. solution.
Use the commercially pure acid. On ignition the acid should leave no
residue.

[Illustration: FIG. 32.]

~Sulphuretted Hydrogen.~ Hydrosulphuric acid, SH_{2}.--A gas largely
used in assaying, since by its action it allows of the metals being
conveniently classed into groups. It is soluble in water, this liquid
dissolving at the ordinary temperature about three times its volume of
the gas. The solution is only useful for testing. In separations, a
current of the gas must always be used. It is best prepared in an
apparatus like that shown in fig. 32, by acting on ferrous sulphide with
dilute hydrochloric acid. When iron has to be subsequently determined in
the assay solution, the gas should be washed by bubbling it through
water in the smaller bottle; but for most purposes washing can be
dispensed with. The gas is very objectionable, and operations with it
must be carried out in a cupboard with a good draught. When the
precipitation has been completed, the apparatus should always be washed
out. The effect of this acid on solutions of the metals is to form
sulphides. All the metallic sulphides are insoluble in water; but some
are soluble in alkaline, and some in acid, solutions. If sulphuretted
hydrogen is passed through an acid solution containing the metals till
no further precipitation takes place, a precipitate will be formed
containing sulphides insoluble in the acid. On filtering, adding ammonia
(to render the filtrate alkaline), and again passing the gas, a further
precipitate will be obtained, consisting of sulphides insoluble in an
alkaline solution, but not precipitable in an acid one; the filtrate may
also contain sulphides not precipitable in an acid solution, which are
soluble in an alkaline one; these will be thrown down on neutralising.
Again, the metals precipitated in the acid solution form sulphides which
may be divided into groups, the one consisting of those which are
soluble, and the other of those which are not soluble, in alkalies. This
classification is shown in the following summary:--

1. _Precipitable in an acid solution._

(a) Soluble in Alkalies.--Sulphides of As, Sb, Sn, Au, Pt, Ir, Mo, Te,
and Se.

(b) Insoluble in Alkalies.--Sulphides of Ag, Pb, Hg, Bi, Cu, Cd, Pd, Rh,
Os, and Ru.

2. _Not precipitated in an acid solution, but thrown down in an alkaline
one._

Sulphides of Mn, Zn, Fe, Ni, Co, In, Tl, and Ga.

These can again be divided into those which are dissolved by dilute
acids and those which are not.

3. _Not precipitated in an acid or alkaline solution, but thrown
down on neutralising the latter._

Sulphides of V and W.

Sulphuretted hydrogen is a strong reducing agent. Ferric salts are
thereby quickly reduced to ferrous; in hot solutions nitric acid is
decomposed. These changes are marked by a precipitation of sulphur, and
the student must be careful to pass the gas sufficiently long, and not
be too hasty in concluding that no sulphide will form because it does
not at once make its appearance. The best indication that it has been
passed long enough is the smell of the gas in the solution after
shaking.

~Sulphurous Acid~, H_{2}SO_{3}.--The reagent used may be regarded as a
saturated solution of sulphur dioxide in water. It may be purchased, and
keeps for a long time. It may be made by heating copper with sulphuric
acid and passing the gas formed into water. The heat should be withdrawn
when the gas is coming off freely. It is used as a reducing agent, and
should not be diluted.

~Sulphuric Acid~, H_{2}SO_{4}. (Sp. gr. 1.84, containing 96 per cent. of
real acid, H_{2}SO_{4}.)--This acid forms insoluble sulphates with salts
of lead, strontium, and barium. It has a high boiling point, 290° C.,
and, when evaporated with salts of the more volatile acids, converts
them into sulphates. When nitrates or chlorides are objectionable in a
solution, evaporation with sulphuric acid removes them. In working with
this acid caution is necessary, since, on mixing with water, great heat
is evolved; and, if either the acid or water has been previously heated,
a serious accident may result. In diluting the acid it should be poured
into cold water. Glass vessels containing boiling sulphuric acid should
be handled as little as possible, and should not be cooled under the
tap. The action of diluted sulphuric acid on metals closely resembles
that of dilute hydrochloric acid. Magnesium, aluminium, iron, zinc,
nickel, cobalt, manganese, and cadmium dissolve, with evolution of
hydrogen, in the cold acid, or when warmed. The action of hot and strong
sulphuric acid is altogether different; it acts as an oxidising agent,
and is itself reduced to sulphur dioxide or even to sulphur. The
following metals are attacked in this way:--copper, bismuth, mercury,
silver, antimony, tin, and lead. Gold, platinum, and arsenic are not
affected. This property is made use of in parting silver from gold and
platinum. Metallic sulphides are similarly attacked; but this method of
opening up minerals has the disadvantage of giving rise to the formation
of anhydrous sulphates of iron, &c., which are not readily dissolved
when afterwards diluted. The use of sulphuric acid in assaying is (for
these reasons) to be avoided. Its chief use is as a drying agent, since
it has a strong affinity for water. Air under a bell jar may be kept dry
by means of a basin of sulphuric acid, and gases bubbled through it are
freed from water-vapour.

~Dilute Sulphuric Acid.~--This is made by diluting 1 volume of the
strong acid with 4 of water.

~Tartaric Acid~, H_{2}[=T] or C_{4}H_{6}O_{6}.--A crystallised organic
acid, soluble in less than its own weight of water, or in less than
three parts of alcohol. It is used for the same purposes as citric acid
is. The solution is made when required.


BASES, SALTS, &c.

~Alcohol~, C_{2}H_{6}O. (Commercial alcohol of sp. gr. 0.838; it
contains 84 per cent. by weight of alcohol.)--It should burn with a
non-luminous flame and leave no residue. It is used for washing
precipitates where water is inapplicable, and for facilitating drying.

~Ammonia~, NH_{3}. (Commercial ammonia, a solution having a sp. gr. of
0.88 to 0.89, and containing about 33 per cent. of ammonia.)--It is used
as an alkali (more commonly than soda or potash), since an excess of it
is easily removed by boiling. The salts of ammonium formed by it may be
removed by igniting, or by evaporating in a porcelain dish with an
excess of nitric acid. It differs in a marked way from soda or potash in
its solvent action on the oxides or hydrates of the metals. Salts of the
following metals are soluble in an ammoniacal solution in the presence
of ammonic chloride:--copper, cadmium, silver, nickel, cobalt,
manganese, zinc, magnesium, sodium, potassium, and the alkaline earths.

~Dilute Ammonia~ is made by diluting 1 vol. of commercial ammonia with 2
of water. The dilute ammonia is always used; but in assays for copper a
stronger solution (1 of strong ammonia to 1 of water) is required.

~Ammonic Carbonate~ (Am_{2}CO_{3}) is prepared by dissolving one part of
the commercial sesquicarbonate of ammonia in four parts of water, and
adding one part of strong ammonia.

~Ammonic Bicarbonate~ (HAmCO_{3}) is prepared by saturating a solution
of the sesquicarbonate of ammonia with carbon dioxide.

~Ammonic Chloride~, AmCl.--Use the commercial salt in a 20 per cent.
solution in water. The salt should leave no residue on ignition.

~Ammonic Molybdate.~--The solution is prepared as follows:--Dissolve 100
grams of the powdered commercial salt in 200 c.c. of dilute ammonia, and
pour the solution in a slow stream into 750 c.c. of dilute nitric acid;
make up to 1 litre, and allow the mixture to settle before using. It is
used for the purpose of separating phosphoric oxide from bases and from
other acids, and also as a test for phosphates and arsenates. In using
this solution the substance must be dissolved in nitric acid, and a
considerable excess of the reagent added (50 c.c. is sufficient to
precipitate 0.1 gram P_{2}O_{5}); when the phosphate is in excess no
precipitate will be got. The precipitate is phospho-molybdate of
ammonia.

~Ammonic Nitrate~ (AmNO_{3}) is used in the separation of phosphoric
oxide by the molybdate method, and occasionally for destroying organic
matter. It is soluble in less than its own weight of water. The solution
is made when wanted.

~Ammonic Oxalate~ (Am_{2}C_{2}O_{4}.2H_{2}O) is used chiefly for the
separation of lime. The solution is made by dissolving 15 grams of the
salt in 100 c.c. of water.

~Ammonic Sulphide~ may be purchased in the state of a strong solution.
It is yellow, and contains the disulphide, S_{2}Am_{2}. It serves the
same purpose as is obtained by passing a current of sulphuretted
hydrogen through an ammoniacal solution; but has the disadvantage of
loading the solution with sulphur, which is precipitated when the
solution is subsequently acidified. It is useful for dissolving the
lower sulphide of tin (SnS).

~Baric Carbonate~ (BaCO_{3}) is sometimes used for precipitating the
weaker bases. It should be prepared when wanted by precipitating a
solution of baric chloride with ammonic carbonate and washing. The moist
precipitate is used without drying.

~Baric Chloride~, BaCl_{2}.2H_{2}O.--A crystallised salt, soluble in
2-1/2 parts of water. It is used for the detection and separation of
sulphates. Make a 10 per cent. solution.

"~Black Flux.~"--A mixture of finely divided carbon with carbonate of
potash or with carbonates of potash and soda. It is prepared by heating
tartar or "rochelle salt" until no more combustible gas is given off.
One gram will reduce about 2 grams of lead from litharge.

~Borax~, Na_{2}B_{4}O_{7}.10H_{2}O.--It is chiefly used as a flux in dry
assaying, as already described. It is also used in testing before the
blowpipe; many metallic oxides impart a characteristic colour to a bead
of borax in which they have been fused.

~Calcium Chloride.~--The crystallised salt is CaCl_{2}.6H_{2}O; dried at
200° C. it becomes CaCl_{2}.2H_{2}O, and when fused it becomes
dehydrated. The fused salt, broken into small lumps, is used for drying
gases. It combines with water, giving off much heat; and dissolves in a
little more than its own weight of water. Strong solutions may be used
in baths in which temperatures above the boiling-point of water are
required. One part of the salt and 2 of water give a solution boiling at
112°, and a solution of 2 parts of the salt in 1 of water boils at 158°.
The salt is very little used as a reagent.

~Calcium Fluoride~ or "~Fluor Spar~," CaF_{2}.--The mineral is used as a
flux in dry assaying; it renders slags which are thick from the presence
of phosphates, &c., very fluid. Mixed with hydrochloric acid it may
sometimes be used instead of hydrofluoric acid.

~Calcium Carbonate~, CaCO_{3}.--It is precipitated in a pure state by
ammonic carbonate from a solution of calcium chloride. It is used for
standardising. In the impure state, as marble or limestone, it is used
in the preparation of carbonic acid.

~Calcium Hydrate~ or ~"Lime Water."~--This is used in testing for carbon
dioxide and in estimating the amount of that gas present in air. It may
be made by slaking quicklime and digesting the slaked lime with water.
One hundred c.c. of water at 15° C. dissolves 0.1368 grams of the
hydrate (CaH_{2}O_{2}), and hot water dissolves still less. "_Milk of
lime_" is slaked lime suspended in water.

~Cobalt Nitrate~ (Co(NO_{3})_{2}.6H_{2}O) is used in a 10 per cent.
solution for the detection of oxides of zinc, aluminium, &c.; on
ignition with which it forms characteristically coloured compounds.

~Copper~, Cu.--Pure copper, as obtained by electrolysis, can be
purchased. This only should be used.

~Copper Oxide~, CuO.--It occurs as a black, heavy, and gritty power, and
is used for the oxidation of carbon and hydrogen in organic substances.
It should be ignited and cooled out of contact with air just before
using, since it is hygroscopic. Oxide of copper which has been used may
be again utilised after calcination.

~Copper Sulphate~ (CuSO_{4}.5H_{2}O) contains 25.4 per cent. of copper.
It is used in the outer cell of a Daniell-battery. The commercial salt
is used for this purpose. The re-crystallised and pure salt is used for
preparing the anhydrous sulphate, which is used for detecting moisture
in gases. For this purpose it is dried at 200° C. till no trace of
green or blue colour remains. It must be prepared when wanted. It may be
conveniently used in the form of pumice-stone, saturated with a solution
of the salt and dried. Traces of moisture develop a green colour.

~Ferric Chloride~, Fe_{2}Cl_{6}. (When crystallised,
Fe_{2}Cl_{6}.6H_{2}O.)--The solution is prepared as described under
iron. The commercial salt contains arsenic, and, since the chief use of
ferric chloride is for the determination of this substance, it must be
purified (_see_ under ARSENIC).

~Ferric Sulphate~ (Fe_{2}(SO_{4})_{3}) is a yellowish white deliquescent
salt. It is used as an indicator in volumetric silver assaying, and for
the separation of iodine from bromine. It may be purchased as iron alum,
Am_{2}Fe_{2}(SO_{4})_{4}.24H_{2}O. But it is best prepared by adding
strong sulphuric acid to ferric hydrate in equivalent proportions. Use
it as a solution containing 2 or 3 per cent. of iron.

~Ferrous Sulphate~, FeSO_{4}.7H_{2}O.--The granulated form is best, and
can be purchased pure. It is used for standardising. It keeps better in
crystals than in solution. It is readily soluble in water, but the
solution is best made with the help of a little free acid. As a re-agent
use a 10 per cent. solution. The crystals should be clear bluish-green;
if their colour is dark green, brown, or blue, they should be rejected.

~Ferrous Sulphide~ (FeS) is used for the preparation of sulphuretted
hydrogen. It may be purchased and broken in small lumps, nut-size, for
use.

"~Fusion Mixture~" (K_{2}CO_{3}.Na_{2}CO_{3}) is a mixture of potassic
and sodic carbonates in the proportions of 13 of the former to 10 of the
latter, by weight. It is hygroscopic. A mixture of the bicarbonates is
better, being purer and less apt to get damp.

~Gallic Acid~ (C_{7}H_{6}O_{5}.H_{2}O) is an organic acid, occurring as
a pale fawn-coloured crystalline powder, soluble in 100 parts of cold
water, or in 3 parts of boiling water. It is used for the determination
of antimony. A 10 per cent. solution in warm water is made when
required.

~Hydrogen~ (H) is a gas. It is obtained by acting on zinc with dilute
hydrochloric or sulphuric acid. It is used as a reducing agent, and for
providing an atmosphere free from oxygen. It reduces metallic oxides at
a high temperature. It must be freed from water; and special precautions
should be taken to prevent an admixture with air. It is generally
required in a current which can be continued for an hour or more without
interruption. The preparation can be conveniently carried out in the
apparatus shown (fig. 33). A quart bottle is half filled with sheet
zinc, and connected with bulbs filled with sulphuric acid, and with a
calcium chloride tube. The last is connected with the apparatus through
which the gas has to be passed. Dilute hydrochloric acid mixed with a
few cubic centimetres (20 c.c. to 1 pint) of stannous chloride sol. to
fix any dissolved oxygen, is placed in the funnel, and let into the
bottle by opening the stopcock when required. Care must be taken to let
the hydrogen escape for some time before starting the reduction.

[Illustration: FIG. 33.]

~Gold~, Au.--Gold, obtained by cupelling and "parting," is for most
purposes sufficiently pure. It is best kept in the shape of foil. When
the purer metal is required, gold should be dissolved in aqua regia, the
solution evaporated to a paste, diluted, allowed to stand, and filtered.
The filtered solution is acidified with hydrochloric acid, warmed, and
precipitated with sodium sulphite. The precipitate is collected, washed,
and fused on charcoal.

~Iron~, Fe.--The soft wire (thin) is used for standardising. Rods are
used in dry assays as a desulphurising agent. Steel must not be used,
since it is not pure, and contains a variable amount of iron.

~Lead~, Pb.--Granulated lead or lead-foil is used in the dry assay for
silver and gold, and in the preparation of lead salts. It can be
obtained very pure, but always contains more or less silver, 1 or 2
milligrams in 100 grams. The amount of silver it contains must be
determined and recorded.

~Lead Acetate~ (Pb[=A=c]_{2}.3H_{2}O, or
Pb(C_{2}H_{3}O_{2})_{2}.3H_{2}O) is used as a test, specially for the
detection and estimation of sulphuretted hydrogen. Prepare a 10 per
cent. solution for use.

~Lead Nitrate~ (Pb(NO_{3})_{2}) can be purchased pure. It is used for
standardising.

~Lead Dioxide~ (PbO_{2}) occurs as a dark-brown powder. It is used as an
oxidizing agent and for absorbing sulphurous oxide. It can be prepared
by digesting red lead with warm dilute nitric acid; washing and drying
the residue.

"~Litharge~," PbO.--It can be purchased as a yellow heavy powder. It is
used in dry assaying as a flux, as a desulphurising agent, and also as a
source of lead. It always contains some silver, the amount of which must
be determined.

~Litmus.~--This is an organic colouring matter which is turned red by
acids and blue by alkalies. For ordinary purposes it is best used as
litmus paper, which may be purchased in small books. A solution is
prepared by digesting 15 or 20 grams of the commercial litmus in 100
c.c. of water on the water bath. After being allowed to settle, it is
filtered and made just faintly red with acetic acid. Then there is
added a drop or two of a solution of soda and 10 c.c. of alcohol. It
should be kept in a loosely-covered bottle.

~Magnesia~, MgO.--It may be purchased as "calcined magnesia." It is used
for making "magnesia mixture," and should be kept in a corked
wide-mouthed bottle.

"~Magnesia Mixture.~"--Dissolve 22 grams of magnesia in about a quarter
of a litre of dilute hydrochloric acid, avoiding excess. Add 5 grams of
magnesia, boil, and filter. Add 300 grams of ammonic chloride, and 250
c.c. of strong ammonia; and dilute with water to 2 litres. It should be
kept in a stoppered winchester.

~Magnesium Sulphate~, MgSO_{4}.7H_{2}O.--It can be purchased very pure,
and is occasionally used as a standard salt.

~Manganese Dioxide~, MnO_{2}.--It is used in the preparation of
chlorine. The commercial article is not pure, but is sufficiently so for
this purpose.

~Marble~, CaCO_{3}.--Fragments of the white crystalline variety only
should be used. It is used as a source of lime and of carbon dioxide.

~Mercury~, Hg.--This can be purchased pure. It should have a bright
surface, flow without a tail, and leave no residue on ignition. It is
used as a standard; for amalgamation; and as a confining liquid in gas
analysis.

~Mercuric Chloride~ (HgCl_{2}) may be purchased pure. Make a 5 per cent.
solution in water. It is used for destroying an excess of stannous
chloride; for removing sulphuretted hydrogen from solution; and as a
test for stannous salts.

~Microcosmic Salt~, HAmNaPO_{4}.8H_{2}O.--When fused NaPO_{3} is formed.
It is used in testing for metallic oxides and silica before the
blowpipe. The crystals are sometimes used as a standard for phosphoric
acid.

"~Nessler's Solution.~"--Mode of preparation: Dissolve 35 grams of
potassium iodide in 100 c.c. of water; dissolve 17 grams of mercuric
chloride in 300 c.c. of water, and pour this solution into that of the
iodide till a permanent precipitate is produced; make up to 1 litre with
a 20 per cent. solution of potash; add mercuric chloride till a
precipitate is again formed; allow to settle and decant. It is used for
detecting ammonia.

~Nitre.~--This is potassium nitrate.

~Platinum Chloride~, 2HCl.PtCl_{4}. (In the crystallised form it has
6H_{2}O).--It may be made as follows:--Take 5 grams of clean platinum
scrap and dissolve in a flask at a gentle heat in 50 c.c. of
hydrochloric acid with the occasional addition of some nitric acid;
evaporate to a paste; and then dissolve in 100 c.c. of water. It is used
for separating and determining potassium.

~Phenolphthalein~ is an organic compound used as an indicator; more
especially in determining the weaker acids, it cannot be used in the
presence of ammonia. Dissolve half a gram in 100 c.c. of dilute alcohol.

~Potassium Bicarbonate~, KHCO_{3}.--It may be purchased pure; on
ignition it leaves the carbonate, K_{2}CO_{3}, which may be used as a
standard.

~Potassium Cyanide~, KCN.--It is used in the dry assay as a reducing
agent. The commercial salt is very impure. Purchase that sold as
potassic cyanide (gold) which contains about 95 per cent. of KCN. It is
used for copper assaying and occasionally in separation. Make a 10 per
cent. solution when wanted.

~Potassium Bichromate~, K_{2}Cr_{2}O_{7}. It may be purchased nearly
pure. It is used as an oxidising agent, for determining iron; and as a
test solution. For this last purpose a 10 per cent. solution is
prepared.

~Potassium Chlorate~ (KClO_{3}) can be purchased pure. It is used with
hydrochloric acid as a substitute for aqua regia.

~Potassium Ferrocyanide~ (K_{4}Fe(CN)_{6}.3H_{2}O), or "yellow prussiate
of potash," is used as a test; as an indicator; and for the
determination of zinc. Make a 5 per cent. solution.

~Potassium Ferricyanide~ (K_{6}Fe_{2}(CN)_{12}), or "red prussiate of
potash," is used for testing; and as an indicator. Make a 5 per cent.
solution when wanted, as it decomposes on keeping.

~Potassium Hydrate~, KHO. Purchase that purified with alcohol. It is an
alkali, and is used for absorbing carbonic acid, &c.

~Potassium Iodide~, KI. It may be purchased nearly pure. It is used as a
test and for dissolving iodine. It should be used in a 10 per cent.
solution freshly made. The solution decomposes on exposure to light,
with separation of iodine.

~Potassium Nitrate~ (KNO_{3}) can be purchased pure. It is used in the
dry way as an oxidizing agent. It is very fusible. It decomposes at a
low temperature into potassium nitrite (KNO_{2}) and free oxygen; and at
a higher temperature leaves potash (K_{2}O). It oxidizes sulphur and
carbon with explosive violence. This action may be moderated by mixing
the nitre with carbonate of soda, common salt, or some other inert body.

~Potassium Nitrite~, KNO_{2}.--The commercial article is not pure, but
is sufficiently so for the purpose required. A saturated solution is
used in the separation of cobalt; the solution is made when wanted.

~Potassium Permanganate~, KMnO_{4}.--This salt can be purchased
sufficiently pure. It is much used as an oxidizing agent.

~Potassium Bisulphate~ (KHSO_{4}) is used as a dry reagent for opening
up minerals. It fuses; and at a much higher temperature is converted
into potassium sulphate with loss of sulphuric acid.

~Potassium Sulphocyanate~ (KCNS) is used for the detection and
determination of traces of ferric iron; as also in the separation of
silver and copper from some of the other metals. Make a 10 per cent.
solution. It should show no colour on the addition of hydrochloric acid.

"~Red Lead~" (Pb_{3}O_{4}) is used in the dry assay as a flux instead of
litharge, from which it differs in containing a little more oxygen. When
acted on by nitric acid a brown residue of lead dioxide is left, nitrate
of lead going into solution. Like litharge it always carries silver;
about 2 milligrams in 100 grams.

~Silver~, Ag.--Pure silver in foil is required as a standard. It may be
prepared as follows:--Dissolve scrap silver in dilute nitric acid and
decant off from any residue; dilute the solution with hot water and add
hydrochloric acid until there is no further precipitate, stir; allow the
precipitate to settle; decant and wash; dry the precipitate, mix it with
twice its bulk of carbonate of soda and fuse the mixture in a crucible
until tranquil; clean the button and roll or hammer it into foil.

~Sodium Acetate~, NaC_{2}H_{3}O_{2}.3H_{2}O.--The crystals may be
purchased sufficiently pure. Make a 20 per cent. solution in water. It
is used for replacing mineral acids by acetic acid.[7]

~Sodium Acetate and Acetic Acid.~--A solution is used in the
determination of phosphates and arsenates; 100 grams of the salt is
dissolved in 500 c.c. of acetic acid, and diluted with water to one
litre.

~Sodium Bicarbonate~ (NaHCO_{3})is used as a flux in dry methods. On
ignition it leaves the carbonate (Na_{2}CO_{3}), which is used as a
standard reagent. Make a 20 per cent. solution of the carbonate for use.
It should be free from chlorides or sulphates, or if impure the amount
of impurities must be determined.

~Sodium Hydrate~, NaHO. It may be purchased in sticks, which should be
kept in a well-corked bottle. It is sometimes called "caustic soda." It
is a strong alkali. It is used for neutralizing acid solutions and for
separations where ammonia is unsuitable. Make a 5 per cent. solution for
use.

~Sodium Hyposulphite~, Na_{2}S_{2}O_{8}.5H_{2}O.--It may be purchased
pure. It is generally known as "hypo." It is used as a standard.

~Sodium Sulphite~ (Na_{2}SO_{3}.7H_{2}O) is used as a reducing agent.

~Sodium Phosphate~, Na_{2}HPO_{4}.12H_{2}O. The crystals may be
purchased pure, but they effloresce in dry air with loss of water. It
is used as a standard and for precipitating magnesia, &c. Make a 10 per
cent. solution.

~Stannous Chloride~, SnCl_{2}.2H_{2}O.--The crystals are best purchased.
If kept dry and free from air they are fairly permanent. A solution is
made by dissolving 20 grams in 10 c.c. of hydrochloric acid and diluting
to 1 litre. The solution is not permanent. It is a strong reducing
agent, and is chiefly used in solution for this purpose.

~Tin~, Sn.--Grain tin should be purchased. It is not pure, but contains
99.5 per cent. of the metal. The chief impurity is copper. It can be
used as a standard. When acted on with hot hydrochloric acid it slowly
dissolves (more rapidly in contact with platinum) and forms stannous
chloride.

~Uranium Acetate~, UO_{2}(C_{2}H_{3}O_{2})_{2}.H_{2}O.--It is best
purchased in crystals. The solution is used for the determination of
phosphates and arsenates. A solution of 3 per cent. strength is
occasionally used as an indicator.

~Uranium Nitrate~, UO_{2}(NO_{3})_{2}.6H_{2}O.--This salt is very
soluble in water and is sometimes used instead of the acetate, which is
somewhat difficult to dissolve.

"~Water~," H_{2}O.--Spring or well water is sufficiently pure for most
purposes, 100 c.c. will leave a residue of from 10 to 30 milligrams, so
that where a salt has to be dissolved out, evaporated, and weighed it
should be replaced by distilled water. Rain water, melted snow, &c.,
always leave less residue than spring water; but in other respects they
are often dirtier. Distilled water is best prepared in the office, a
glass or tin condenser being used.

~Zinc~, Zn.--It is sold in a granulated form or in sticks. It generally
contains over 1 per cent. of lead, with a little iron and arsenic. It is
used for separating metals from their solutions, and generally as a
reducing agent. For the preparation of hydrogen, and in most other
cases, scrap sheet zinc may be used.

~Zinc Oxide~, ZnO.--The commercial oxide sometimes contains carbonate.

~Zinc Sulphate~, ZnSO_{4}.7H_{2}O.--It is occasionally used as a
standard, and can be purchased nearly pure.

FOOTNOTES:

[6] 3HCl + HNO_{3} = Cl_{2} + NOCl + 2H_{2}O.

[7] NaC_{2}H_{3}O_{2} + HCl = H_{4}C_{2}O_{2} + NaCl.



CHAPTER VII.

FORMULÆ, EQUATIONS, ETC.


Formulæ and equations are a kind of short hand for expressing briefly
and in the language of the atomic theory the facts of chemical
composition and reaction. The convenience of this method of expressing
the facts justifies a short description of it here.

On comparing the percentage composition of a series of compounds the
proportions in which the elements combine appears to be regulated by no
simple law. For example:

             Realgar.     Orpiment.     Mispickel.     Pyrites.
  Arsenic     71.4          60.9          46.0            --
  Sulphur     28.6          39.1          19.6           53.3
  Iron         --            --           34.4           46.7
             ------        ------        ------         ------
             100.0         100.0         100.0          100.0

But if in these examples the composition is calculated, not on 100
parts, but on 107, 246, 163, and 120 parts respectively, evidence of a
simple law becomes apparent.

             Realgar.     Orpiment.     Mispickel.     Pyrites.
  Arsenic     75.0          150.0         75.0           --
  Sulphur     32.0           96.0         32.0          64.0
  Iron         --             --          56.0          56.0
             ------         ------       ------        ------
             107.0          246.0        163.0         120.0

It will be seen that the proportion of arsenic is 75 or twice 75, that
of iron is 56, and that of sulphur 32 or some simple multiple of 32. The
series of examples might be extended indefinitely, and it would still be
found that the "combining proportions" held good. The number 75 is
spoken of as the "combining weight," or, more frequently, as the "atomic
weight" of arsenic. Similarly 56 is the atomic weight of iron, and 32
the atomic weight of sulphur. The importance of this law of chemical
combination is altogether independent of the atomic theory; but this
theory furnishes the simplest explanation of the facts. According to it
a chemical compound is made up of exactly similar groups of particles.
The particles of each elementary substance are all alike, but differ
from those of other elements in weight. Ultimate particles are called
_atoms_, and the groups of atoms are called _molecules_. The atomic
weight of any particular element is the weight of its atom compared with
the weight of an atom of hydrogen. The atom of sulphur, for instance, is
32 times as heavy as the atom of hydrogen, and the atomic weight of
sulphur is 32. The _molecular weight_ is the sum of the atomic weights
of the group. The molecule of pyrites contains two atoms of sulphur and
one of iron: on referring to the table of atomic weights it will be seen
that the atomic weights are--sulphur 32, and iron 56. The molecular
weight, therefore, is 32+32+56--that is, 120. The meaning of this
is, 120 parts by weight of iron pyrites contain 64 parts of sulphur and
56 parts of iron; and this is true whether the "parts by weight" be
grains or tons.

_The symbol or formula of an atom_ is generally the initial letter or
letters of the Latin or English name of the substance. The atom of
hydrogen is written H, that of oxygen O, of sulphur S, of iron (ferrum)
Fe, and so on. A list of these symbols is given in the table of atomic
weights.

_The formula of a molecule_ is obtained by placing together the symbols
of the contained atoms. Thus, Fe represents an atom of iron, S an atom
of sulphur, while FeS represents the molecule of sulphide of iron as
containing one atom of each element.

When more than one atom of an element is present this is shown by
writing a figure under and after the symbol; thus, FeS_{2} represents a
molecule with one atom of iron and two atoms of sulphur, Fe_{2}S_{3}
similarly shows one with two atoms of iron and three of sulphur. When a
group of atoms is enclosed in brackets, a figure after and under the
bracket multiplies all within it; for example, Pb(NO_{3})_{2} is another
way of writing PbN_{2}O_{6}. Sometimes it is convenient to represent the
atoms of a molecule as divided into two or more groups; this may be done
by writing the formulæ of the groups, and separating each simple formula
by a full stop. Slaked lime, for instance, has the formula CaH_{2}O_{2};
or, as already explained, we may write it Ca(HO)_{2}; or, if for
purposes of explanation we wished to look on it as lime (CaO) and water
(H_{2}O), we could write it CaO.H_{2}O. A plus sign (+) has a different
meaning; CaO + H_{2}O indicates quantities of two substances, water and
lime, which are separate from each other. The sign of equality (=) is
generally used to separate a statement of the reagents used from another
statement of the products of the reaction; it may be translated into the
word "yields" or "becomes." The two statements form an equation.

Ignoring the quantitative relation, the meaning of the equation CaO +
H_{2}O = CaO.H_{2}O is: "lime and water yield slaked lime." By referring
to a table of atomic weights we can elicit the quantitative relations
thus:--

    CaO    +    H_{2}O       =     CaH_{2}O_{2}
     |             |                   |
     V             V                   V
  Ca = 40    H_{2} =  2 = 1×2      Ca  = 40
  O  = 16       O  = 16          H_{2} =  2 = 1×2
       --            --          O_{2} = 32 = 16×2
       56            18                  --
                                         74

Or, putting it in words, 56 parts of lime combine with 18 parts of water
to form 74 parts of slaked lime. This equation enables one to answer
such a question as this:--How much lime must be used to produce 1 cwt.
of slaked lime? for, if 74 lbs. of slaked lime require 56 lbs. of lime,
112 lbs. will require (56×112)/74, or about 84-3/4 lbs.

As another example having a closer bearing on assaying take the
following question:--"In order to assay 5 grams of 'black tin' (SnO_{2})
by the cyanide process, how much potassic cyanide (KCN) will be
required?" The reaction is

      SnO_{2}   +   2KCN    =    Sn + 2KCNO
        |             |
        V             V
     Sn = 118        K = 39
  O_{2} =  32        C = 12
          ---        N = 14
          150            --
                         65×2 = 130

What is sought for here is the relation between the quantities of
SnO_{2} and KCN. Note that a figure before a formula multiplies all that
follows up to the next stop or plus or equality sign. The question is
now resolved to this: if 150 grams of oxide of tin require 130 grams of
cyanide, how much will 5 grams require?

        150 : 130 :: 5 : _x_
                         _x_ = 4.33 grams.

A problem of frequent occurrence is to find the percentage composition
of a substance when its formula has been given. For example: "What
percentage of iron is contained in a mineral having the formula
2Fe_{2}O_{3}.3H_{2}O?" Bringing this formula together we have
Fe_{4}H_{6}O_{9}. Find the molecular weight.

  Fe_{4}  =   224 = 56×4
  H_{6}   =     6 = 1×6
  O_{9}   =   144 = 16×9
              ---
              374

Then we get: 374 parts of the mineral contain 224 of iron. How much will
100 contain?

        374 : 224 :: 100 : _x_
                           _x_ = 59.89.

And the answer to the question is 59.89 per cent.

Again, suppose the question is of this kind:--"How much crystallised
copper sulphate (CuSO_{4}.5H_{2}O) will be required to make 2 litres of
a solution, 1 c.c. of which shall contain 0.0010 gram of copper?"

A litre is 1000 c.c., so, therefore, 2 litres of the solution must
contain 0.001 gram × 2000, or 2 grams. How much crystallised copper
sulphate will contain this amount of metal?

  Cu      = 63.3
  S       = 32.0
  O_{4}   = 64.0 = 16×4
  5H_{2}O = 90.0 = 18×5
           -----
           249.3

If 63.3 grams of copper are contained in 249.3 grams of sulphate, in how
much is 2 grams contained.

        63.3 : 249.3 :: 2 grams : _x_
                                  _x_ = 7.8769 grams.

The answer is, 7.8769 grams must be taken.

As a sample of another class of problem similar in nature to the last
(but a little more complicated) take the following:--"What weight of
permanganate of potash must be taken to make 2 litres of a solution, 100
c.c. of which shall be equivalent to 1 gram of iron?" In the first place
the 2 litres must be equivalent to 20 grams of iron, for there are 20 ×
100 c.c. in two litres. In the titration of iron by permanganate
solution there are two reactions. First in dissolving the iron

Fe + H_{2}SO_{4} = FeSO_{4} + H_{2}
 |
 V
 56

and second, in the actual titration,

10FeSO_{4} + 2KMnO_{4} + 9H_{2}SO_{4}= 2MnSO_{4}
               |       + 5Fe_{2}(SO_{4})_{3} + 2KHSO_{4} + 8H_{2}O
               V
             K = 39
            Mn = 55
          O_{4}= 64
                ---
                158 × 2 = 316

As before, attention is confined to the two substances under
consideration--viz., Fe and KMnO_{4}. In the second equation, we find
316 parts of the permanganate are required for 10 molecules of FeSO_{4};
and in the first equation 56 parts of iron are equivalent to one
molecule of FeSO_{4}, therefore 560 of iron are equivalent to 316 of
permanganate; and the question is, How much of the permanganate will be
equivalent to 20 grams of iron?

        560 : 316 :: 20 grams : _x_.
                                _x_= 11.286 grams.

The answer is 11.286 grams.

Very similar to this last problem is the question suggested under the
head "Indirect Titration" (p. 43). "If 100 c.c. of the standard
permanganate solution are equivalent to 1 gram of iron, how much
peroxide of manganese will they be equivalent to?" The equation for
dissolving the iron is already given; the second equation is

2FeSO_{4} + MnO_{2} + 2H_{2}SO_{4}
              |      = Fe_{2}(SO_{4})_{2} + MnSO_{4} + 2H_{2}O
              |
              V
           Mn = 55
        O_{2} = 32
                --
                87

It will be seen that 87 grams of peroxide of manganese are equivalent to
112 grams of iron. How much then is equivalent to 1 gram of iron?

        112 : 87 :: 1 gram : _x_
                             _x_ = 0.7767 gram.

It is sometimes convenient to calculate the formula of a substance from
its analysis. The method of calculating is shown by the following
example. Required the formula of a mineral which gave the following
figures on analysis:--

  Cupric oxide (CuO)          10.58
  Ferrous oxide (FeO)         15.69
  Zinc oxide (ZnO)             0.35
  Sulphuric oxide (SO_{2})    28.82
  Water (H_{2}O)              44.71
                             ------
                             100.15

First find the molecular weights of CuO, FeO, &c., and divide the
corresponding percentages by these figures. Thus, CuO = 63.3+16 = 79.3
and 10.58 divided by 79.3 gives 0.1334. Similarly FeO = 56+16 = 72 and
15.69 divided by 72 gives 0.2179. Treated in the same way the oxide of
zinc, sulphuric oxide and water give as results 0.0043, 0.3602 and
2.484.

Classify the results as follows:--

    Bases.         Acids.         Water.

  CuO 0.1334    SO_{3} 0.3602    H_{2}O 2.484
  FeO 0.2179
  ZnO 0.0043
  ----------    -------------    ------------
   RO 0.3556    RO_{3} 0.3602    R_{2}O 2.484

The figures 0.3556, 0.3602 and 2.484 should be then divided by the
lowest of them--_i.e._, 0.3556; or where, as in this case, two of the
figures are very near each other the mean of these may be taken--_i.e._,
0.3579. Whichever is taken the figures got will be approximately 1, 1
and 7. The formula is then RO.SO_{3}.7H_{2}O in which R is nearly 2/5ths
copper, 3/5ths iron and a little zinc.

This formula requires the following percentage composition, which for
the sake of comparison is placed side by side with the actual results.

                 Calculated.   Found.
  Cupric oxide     11.29       10.58
  Ferrous oxide    15.37       15.69
  Zinc oxide        nil         0.35
  Sulphuric oxide  28.47       28.82
  Water            44.84       44.71
                   -----      ------
                   99.97      100.15

Trimming the results of an analysis to make them fit in more closely
with the calculations from the formula would be foolish as well as
dishonest. There can be no doubt that the actual analytical results
represent the composition of the specimen much more closely than the
formula does; although perhaps other specimens of the same mineral would
yield results which would group themselves better around the calculated
results than around those of the first specimen analysed. It must be
remembered that substances are rarely found pure either in nature or in
the arts; so that in most cases the formula only gives an approximation
to the truth. In the case of hydrated salts there is generally a
difficulty in getting the salt with exactly the right proportion of
water.


PRACTICAL EXERCISES.

The following calculations may be made:--

1. Calculate standards in the following cases--
  (a) Silver taken, 1.003 gram. Standard salt used, 100.15 c.c.
  (b) Iron taken, 0.7 gram. Bichromate used, 69.6 c.c.

2. Calculate percentages:--
  (a) Ore taken, 1 gram. Solution used, 65.2 c.c. Standard, 0.987
    gram.

  (b) Ore taken, 1 gram. Barium sulphate got, 1.432 gram. Barium
    sulphate contains 13.73 per cent. of sulphur, and the percentage
    of sulphur in the ore is wanted.

  (c) Barium sulphate is BaSO_{4}. Calculate the percentage of sulphur
    it contains, for use in the preceding question.

3.  A method of estimating the quantity of peroxide in a manganese ore
is based on the following reactions:--

    (1) MnO_{2} + 4HCl = MnCl_{2} + Cl_{2} + 2H_{2}O.

    (2) Cl + KI = KCl + I.

To how much MnO_{2} is 1 gram of Iodine (I) equivalent?

4. A mineral has the following composition:--

  Carbonic acid (CO_{2})     19.09
  Copper oxide (CuO)         71.46
  Water (H_{2}O)              9.02

What is its formula?

5. How much copper is contained in 1.5 gram of crystallized copper
sulphate (CuSO_{4}.5H_{2}O)? How much of these crystals must be taken to
give 0.4 gram of copper?

6. How much ferrous sulphate crystals (FeSO_{4}.7H_{2}O) must be taken
to yield 2 litres of a solution, 100 c.c. of which shall contain 0.56
gram of iron?

7. Galena is PbS, and hæmatite Fe_{2}O_{3}. What percentages of metal do
these minerals contain?



CHAPTER VIII.

SPECIFIC GRAVITY.


The relation of the weight of a substance to its volume should be kept
in mind in all cases where both weight and volume are dealt with.
Students are apt to imagine that on mixing equal volumes of, say,
sulphuric acid and water, an acid of half the strength must be obtained.
If the statement of strength is in parts by weight this will lead to
considerable error. For example, 100 c.c. of sulphuric acid containing
98 per cent. by weight of real acid, will, if diluted with 100 c.c. of
water, yield a solution containing not 49 per cent. by weight, but about
63.5 per cent. of the acid. The reason is this: the 100 c.c. of
sulphuric acid weighs 184 grams, and contains 180.32 grams of real acid,
while the 100 c.c. of water weighs only 100 grams; the mixed water and
acid weighs 284 grams, and contains 180.32 of real acid, which is
equivalent to nearly 63.5 per cent. by weight. If, however, the method
of statement be volumetric, it would be correct to say that doubling the
volume halves the strength: if 100 c.c. of brine contains 10 grams of
salt, and is diluted with water to 200 c.c., it would be of one-half the
former strength, that is, 100 c.c. of the solution would contain 5 grams
of salt.

This confusion is avoided by always stating the strengths as so many
grams or "c.c." in 100 c.c. of the liquid. But obviously it would be
advantageous to be able to determine quickly the weight of any
particular substance corresponding to 1 c.c. or some other given volume.
Moreover, in descriptions of processes the strengths of acids and
solutions are frequently defined neither by their gravimetric nor
volumetric composition, but by a statement either of specific gravity or
of the degrees registered by Twaddell's or Beaumé's hydrometer. Thus, in
the description of the process of gold parting, one writer gives: "The
acid should be of 1.2 specific gravity"; and another says: "The acid
must not be stronger than 32° Beaumé."

These considerations justify an account of the subject in such a work as
this. And on other grounds the determination of a specific gravity is
one of the operations with which an assayer should be familiar.

The meaning of "specific gravity" is present in the mind of every one
who uses the sentence "lead is heavier than water." This is meaningless
except some such phrase as "bulk for bulk" be added. Make the sentence
quantitative by saying: "bulk for bulk lead is 11.36 times heavier than
water," and one has the exact meaning of: "the specific gravity of lead
is 11.36." A table of the specific gravities of liquids and solids shows
how many times heavier the substances are than water.

It is better, however, to look upon the specific gravity (written
shortly, sp. g.) as the weight of a substance divided by its volume. In
the metric system, 1 c.c. of water at 4° C. weighs with sufficient
exactness 1 gram; consequently, the sp. g., which states how many times
heavier than water the substance is, also expresses the weight in grams
of one c.c. of it. So that if a 100 c.c. flask of nitric acid weighs,
after the weight of the flask has been deducted, 120 grams, 1 c.c. of
the acid weighs 1.2 gram, and the sp. g. is 1.2. The specific gravity,
then, may be determined by dividing the weight of a substance in grams
by its volume in c.c.; but it is more convenient in practice to
determine it by dividing _the weight of the substance by the weight of
an equal volume of water_. And since the volumes of all substances,
water included, vary with the temperature, the temperature at which the
sp. g. is determined should be recorded. Even then there is room for
ambiguity to the extent that such a statement as the following, "the
specific gravity of the substance at 50° C. is 0.9010," may mean when
compared with water at 50° C. or 4° C., or even 15.5° C. For practical
purposes it should mean the first of these, for in the actual
experiments the water and the substance are compared at the same
temperature, and it is well to give the statement of results without any
superfluous calculation. In the metric system the standard temperature
is 4° C., for it is at this point that 1 c.c. of water weighs exactly 1
gram. In England, the standard temperature is 60° F. (15.5° C.), which
is supposed to be an average temperature of the balance-room. The
convenience of the English standard, however, is merely apparent; it
demands warming sometimes and sometimes cooling. For most purposes it is
more convenient to select a temperature sufficiently high to avoid the
necessity of cooling at any time. Warming to the required temperature
gives very little trouble.

~Determination of Specific Gravity.~--There is a quick and easy method
of determining the density or sp. g. of a liquid, based upon the fact
that a floating body is buoyed up more by a heavy liquid than by a light
one. The method is more remarkable for speed than accuracy, but still
is sufficiently exact. The piece of apparatus used for the purpose is
endowed with a variety of names--sp. g. spindle, hydrometer, areometer,
salimeter, alcoholimeter, lactometer, and so on, according to the
special liquid upon which it is intended to be used. It consists of a
float with a sinker at one end and a graduated tube or rod at the other.
It is made of metal or glass. Generally two are required, one for
liquids ranging in sp. g. from 1.000 to 2.000, and another, which will
indicate a sp. g. between 0.700 and 1.000. The range depends on the size
of the instrument. For special work, in which variations within narrow
limits are to be determined, more delicate instruments with a narrower
range are made.

[Illustration: FIG. 34.]

In using a hydrometer, the liquid to be tested is placed in a cylinder
(fig. 34) tall enough to allow the instrument to float, and not too
narrow. The temperature is taken, and the hydrometer is immersed in the
fluid. The mark on the hydrometer stem, level with the surface of the
liquid, is read off. With transparent liquids it is best to read the
mark under and over the water surface and take the mean.

The graduation of hydrometers is not made to any uniform system. Those
marked in degrees Baumé or Twaddell, or according to specific gravity,
are most commonly used. The degrees on Baumé's hydrometer agree among
themselves in being at equal distances along the stem; but they are
proportional neither to the specific gravity, nor to the percentage of
salt in the solution. They may be converted into an ordinary statement
of specific gravity by the following formulæ:--

  Sp. g. = 144.3/(144.3-degrees Baumé.)

or putting the rule in words, subtract the degrees Baumé from 144.3, and
divide 144.3 with the number thus obtained. For example: 32° Baumé
equals a sp. g. of 1.285.

  144.3/(144.3-32) = 144.3/(112.3) = 1.285

This rule is for liquids heavier than water; for the lighter liquids the
rule is as follows:--

  Sp. g. = 146/(136 + degrees Baumé.)

or in words divide 146 by the number of degrees Baumé added to 136. For
example: ammonia of 30° Beaumé has a sp. g. of 0.880 (nearly).

  146/(136+30) = 146/166 = 0.8795

A simple series of calculations enables one to convert a Beaumé
hydrometer into one showing the actual sp. g. Graduation, according to
sp. g. is the most convenient for general purposes. In these instruments
the distances between the divisions become less as the densities
increase.

Twaddell's hydrometer is graduated in this way: Each degree Twaddell is
0.005 in excess of unity. To convert into sp. g. multiply the degrees
Twaddell by 0.005, and add 1. For example: 25° Twaddell equals a sp. g.
of 1.125.

  25×.005 = 0.125; + 1.000 = 1.125.

There is a practice which ignores the decimal point and speaks of a sp.
g. of 1125 instead of 1.125. In some cases it is convenient, and
inasmuch as no substance has a real sp. g. of much over 20, it can lead
to no confusion. The figures expressed in this way represent the weight
of a litre in grams.

Some hydrometers are graduated so as to show at a glance the percentage
composition of the liquid they are intended to be used with. Gay-Lussac
designed one to show the alcoholic strength of mixtures of alcohol and
water; the construction of others upon the same principle is easy and
perhaps useful. But when the principle is applied to complex liquids and
mixed solutions, it is misleading.

The various methods of graduation ought all to give place to one showing
a simple statement of the sp. g.

The method of determining sp. g. with the hydrometer is obviously
inapplicable to the case of solids, and in the case of liquids it should
not be used where exact figures are required. There are several other
methods which may be used, but on the whole those with the specific
gravity bottle are most convenient.

[Illustration: FIG. 35.]

~The specific gravity bottle~ (fig. 35) is a light flask of about 25
c.c. capacity, provided with a well-fitting perforated stopper. It is
essentially a graduated flask, which measures a constant volume, but it
does not much matter what the volume is.

_In taking the sp. g. of a liquid_ (_or, what is the same thing, a fused
solid_) there is wanted the weights (1) of the flaskful of water and (2)
of the flaskful of the liquid. Dividing the second by the first gives
the required sp. g. The actual weighings required are--

  (1) of the dry and empty flask,

  (2) of the flask filled with water, and

  (3) of the flask filled with the liquid.

The weighing of the flask once made need not be often repeated. It is
well to do so now and then for safety's sake; but one weighing will
serve for a large number of determinations. The same remarks apply to
the weighing of the bottle filled with water. The bottle is dried by
rinsing out first with alcohol and afterwards with ether; ether is very
volatile, and a short exposure in a warm place will soon drive off the
little remaining about the sides. The ether vapour should be sucked out
through a glass tube. See that the bore of the stopper is dry as well as
the bottle. Let the dry bottle stand in the box of the balance for a
minute or two before weighing. The weight is, strictly speaking, not
that of the empty bottle, but of the bottle filled with air. The empty
bottle would weigh from 20 to 30 milligrams less. Correcting for this
would, in most cases, only make a difference in the fourth place of
decimals,[8] so that it is better to ignore the error.

The weight of the flask filled with water is got by filling it with
distilled water, and inserting the stopper. The excess of water will
overflow at the margin and through the bore. The bottle is wiped with a
soft, dry cloth, taking care not to squeeze or warm the bottle. The
bottle will remain filled to the top of the stopper. It is allowed to
stand in the balance box for a minute or two, and then weighed.

Distilled water, as stated, should be used; the use of ordinary water
may increase the weight by 5 or 6 milligrams. Many waters, if they have
not previously been boiled, give off bubbles of air which render the
weighing worthless.

The temperature of the water is of greater importance; lowering the
temperature 2° will increase the weight by 10 or 12 milligrams. A beaker
of water may be warmed or cooled to the required temperature; then the
bottle is filled from it, and quickly weighed. If the balance-room is
cooler than the water, the latter will draw back into the bottle, and a
few small bubbles of air will enter; but even in extreme cases this will
only increase the weight by a very small fraction of a milligram. There
is more trouble caused when the room is warmer, for the liquid then
expands and protrudes as a drop resting on the top of the stopper.
There will in this case be loss by evaporation, which in the case of the
more volatile liquids, such as alcohol, is serious. To prevent this
loss, as well as any that may arise by overflow, the stopper should be
dilated above into a small cup, A (fig. 36), which may itself be
stoppered. In a bottle of this kind the neck of the stopper is
graduated, and the bottle is considered full when the liquid stands at
the level of the mark in the neck. On inserting the stopper, the liquid
rises into the cup, and is reduced to the level of the mark by
absorption with pieces of filter-paper.

[Illustration: FIG. 36.]

For most purposes, however, there is no need for cooling and allowing
room for subsequent expansion. The assayer, as a rule, can select his
own standard temperature, and may choose one which will always
necessitate warming. It will be handier in this case to have a bottle
with a thermometer stopper. Of the two types shown in fig. 37, that with
the external thermometer tube (A) is more generally useful.

[Illustration: FIG. 37.]

The bottle is filled at a lower temperature, and is then gently warmed
so as to slowly raise the temperature to the required degree. The
superfluous liquid is then at once wiped off, and the bottle cooled and
weighed.

The weight of the flask filled with the liquid whose sp. g. has to be
determined is ascertained in a similar way. Of course the temperature
must be the same. If the liquid does not mix with water, the bottle
should be dried before filling, but otherwise the flask need only be
rinsed out two or three times with the liquid.

Having obtained the three weighings, deduct the weight of the bottle
from each of the others to get the weights of the water and liquid
respectively. Divide the latter by the former, the result shows the sp.
g. As an example, take the following, in which a rather large sp. g.
bottle was used:--

  1. Weight of bottle               39.299 gram

  2. Weight of bottle and water     81.884  "

  3. Weight of bottle and paraffin  73.146  "

By subtracting 1 from 2 and 3 the result is as follows:--

  81.884 grams         73.146 grams
  39.299   "           39.299   "
  ------               ------
  42.585 of water.     33.847 of paraffin.

Divide the weight of the paraffin by that of the water--

  42.585)33.8470(0.7948
         29.8095
         -------
       .......

The sp. g. of the paraffin is 0.7948.

_The sp. g. of a fusible solid_ may be obtained in the same way at a
temperature some degrees above its fusing point.

_The sp. g. of a solid in powder or gravel sufficiently fine to pass
through the neck of the bottle_ is easily determined. If the bottle
filled with water weighs 50 grams, and there is placed on the pan
alongside of it 20 grams of a sand, the weight of the two together will
of course be 70 grams. But if the sand is put in the bottle, it
evidently displaces its own bulk of water; and if, on again weighing,
the weight is found to be 62 instead of 70 grams, it is because the 20
grams of sand has displaced 8 grams of water. Bulk for bulk, the sand is
2-1/2 times as heavy.

In practice, the weight of the bottle filled with water will probably be
already known; if not, it must be determined. A certain quantity, say 20
grams, of the powdered substance is then transferred carefully to the
bottle. The bottle need not be dry inside, but its neck and outside must
be. In making this transference a careful worker will make no loss, and
the mode of working saves a little time. But it is better to weigh the
dry flask; put into it 10 to 20 grams of the powder, and weigh again.
The increase in weight gives accurately the weight of powder in the
bottle. About two-thirds fill the bottle with distilled water, and mix
with the powder by gentle shaking. Air bubbles will disentangle
themselves, and rise to the surface of the water. Wash back anything
adhering to the stopper with a jet of water, and fill the bottle almost
to overflowing. Allow it to stand for a minute or so; replace the
stopper; warm to the required temperature; take off the superfluous
moisture; wipe and weigh. As an example, take the following:--

  1. Weight of bottle                         12.681 grams
  2.   "    "  bottle filled with water       37.708   "
  3.   "    "  bottle with wolfram            40.821   "
  4.   "    "  bottle with wolfram and water  61.199   "

Subtract (1) from (3) to get the weight of wolfram taken:

  40.821 grams
  12.681   "
  ------
  28.140   "

add the weight of the wolfram to the weight of the bottle filled with
water:

  28.140 grams
  37.708   "
  ------
  65.848   "

subtract (4) from this to get the weight of water displaced:

  65.848 grams
  61.199   "
  ------
   4.649   "

Divide the weight of the wolfram by the weight of the water displaced to
get sp. g.:

  4.649)28.140(6.053
        27.894
        ------
        ......

_If the solid is soluble in water, or has a tendency to float_, some
liquid other than water is used. Paraffin oil or oil of turpentine will
do. The process is as follows:--The weight of the dry and empty bottle
having been determined, add a sufficiency of the substance and weigh
again to find how much has been added. Fill up with paraffin oil and
weigh again. Clean out the substance by rinsing with paraffin; fill up
and weigh. Calculate the sp. g. as if water had been used, and multiply
by the sp. g. of the paraffin.

For example:

  1. Weight of bottle                         39.299 grams
  2.    "    " bottle and nitre               57.830   "
  3.    "    " bottle and paraffin            73.146   "
  4.    "    " bottle and paraffin and nitre  84.665   "
  5.    "    " bottle and water               81.884   "

First from (1),(3), and (5), calculate the sp. g. of the paraffin as
already shown. It will be 0.7948. Deduct (1) from (2) to get the weight
of the nitre:

  57.830 grams
  39.299   "
  ------
  18.531   "

add this to (3):

  18.531 grams
  73.146   "
  ------
  91.677   "

and deduct (4) to find the weight of the equal bulk of paraffin.

  91.677 grams
  84.665   "
  ------
   7.012   "

divide the weight of the nitre by the weight of the paraffin:

  7.012)18.531(2.6427
        ------
        ......

The sp. g., taking paraffin as the standard instead of water, is 2.6427.
Multiply this by the sp. g. of paraffin, 0.7948, and the result is
2.1004 as the sp. g. of nitre compared with water.

Similarly, a sp. g. compared with water at say 50° C. can be converted
into one compared with water at standard temperature, by multiplying by
the sp. g. of water at 50° C. The following table gives the sp. g. of
water at various temperatures:--

  -----------+------++-----------+------++-----------+-------
   Degrees   |      || Degrees   |      || Degrees   |
  Centigrade.|Sp. G.||Centigrade.|Sp. G.||Centigrade.|Sp. G.
  -----------+------++-----------+------++-----------+-------
       4°    |1.0000||    20°    |0.9982||    40°    |0.9923
      10°    |0.9997||    25°    |0.9971||    50°    |0.9881
      15°    |0.9991||    30°    |0.9957||   100°    |0.9586
  -----------+------++-----------+------++-----------+-------

If, for example, a substance at 50° C. has a sp. g. of 0.9010 as
compared with water at 50° C., it will have (compared with water at 4°
C.) a sp. g. of 0.9010×0.9881; or 0.8903. The figures 0.8903 represent
the sp. g. of the substance at 50° C. compared with water at 4° C.
Except in comparing the sp. gravities of the same substance at different
temperatures, a calculation of this kind serves no useful purpose.

_In taking the specific gravity of a solid not in powder_, a lump of it
is freed from loose particles and its exact weight determined. By means
of a horse hair with a slip knot it is suspended to the balance, and
beneath it is placed, out of contact with the balance pan, a beaker of
distilled water. The horse hair must be long enough to keep the mineral
well beneath the surface of the water so as to allow the balance to
vibrate. Air bubbles are removed by touching with a camel-hair pencil.
Whilst the mineral is suspended in water the weight is again taken. It
will weigh less than before, and the difference between the two
weighings gives the weight of water (and consequently the volume)
displaced by the mineral. The weight in air divided by the difference is
the specific gravity. Thus

  Weight in air     3.2170 grams
  Weight in water   2.7050   "
                    ------
      Difference    0.5120 gram

  3.2170/0.5120 equals 6.28, the sp. g.

The sp. g. of a substance depends mainly on its composition, but is
affected by certain conditions. The effect of temperature has been
already considered. Air holes and empty spaces lessen the specific
gravity of otherwise solid bodies; and metals, which after fusion become
imperfect solids, have their density increased by hammering or rolling.
But metals when free from pores have their density diminished when
rolled, without annealing. The effects of these conditions are slight
when compared with those due to the presence of impurities.

For simple substances, or mixtures of only two substances, a
determination of sp. g. is a sufficient check on the composition for
many practical purposes; and with more complex mixtures, such as slags
and some of the products of dressing operations in which the material
does not differ much in its nature from time to time, such a
determination will yield information of considerable value, and afford a
check upon the proper working of a process.

When the mixing of two substances is accompanied by a change in volume,
the sp. g. of the mixture can only be learnt by experiment. But when the
substances have no such action on each other the resulting sp. g. can be
calculated. Some of these calculations have a practical interest as well
as an educational value. Students should practise them so as to become
familiar with the relations between weight and volume.

_When substances are mixed by volume_, the sp. g. of the mixture is the
mean of those of its constituents, and may be calculated in the usual
way for obtaining averages. 1 c.c. of a substance having a sp. g. of 1.4
mixed with 1 c.c. of another having a sp. g. of 1.0 will yield 2 c.c. of
a substance having a sp. g. of 1.2. If, however, we write gram instead
of c.c. in the above statement, the resulting sp. g. will be 1.16. The
simplest plan is to remember that the sp. g. is the weight divided by
the volume (sp. g. = w/v) and the sp. g. of a mixture is the sum of the
weights divided by the sum of the volumes (sp. g. = (w + w' + w",
&c.)/(v + v' + v", &c.)). In the above example the sum of the volumes is
2 c.c.; the weights (got by multiplying each volume by its
corresponding sp. g.) are 1.4 gram and 1 gram. The sum of the weights
divided by the sum of the volumes is 2.4/2 or 1.2.

The sp. g. of a mixture of 10 c.c. of a substance having a sp. g. of
1.2, with 15 c.c. of another having a sp. g. of 1.5 may be thus found:--

  sp. g. = (12+22.5)/(10+15) = 1.38

multiply each volume by its sp. g. to get its weight:

  10×1.2 = 12   15×1.5 = 22.5

add these together (12+22.5 = 34.5) and divide by the sum of the volumes
(10+15 = 25):

   25)34.5(1.38
      25
      --
      95, &c.

The sp. g. will be 1.38, provided the mixture is not accompanied by any
change of volume.

The same formula will serve when the proportion of the ingredients is
given by weight. A mixture of 4 parts by weight of galena (sp. g. 7.5)
with 5 parts of blende (sp. g. 4) will have a sp. g. of 5.06:

  sp. g. = (4+5)/(0.53+1.25) = 9/1.78 = 5.06

It is necessary in this case to calculate the volumes of the galena and
of the blende, which is done by dividing the weights by the sp.
gravities: thus, 4 divided by 7.5 gives 0.53 and 5 divided by 4 gives
1.25.

The converse problem is a little more difficult. Given the sp. g. of a
mixture and of each of the two ingredients, the percentage by weight of
the heavier ingredient may be ascertained by the following rule, which
is best expressed as a formula. There are three sp. gravities given; if
the highest be written H, the lowest L and that of the mixture M, then:

  Percentage of heavier mineral = (100×H×(M-L))/(M×(H-L))

Suppose a sample of tailings has a sp. g. of 3.0, and is made up of
quartz (sp. g. 2.6) and pyrites (sp. g. 5.1): then the percentage of
pyrites is 27:

  (100×5.1×(3-2.6))/(3×(5.1-2.6)) = (510×0.4)/(3×2.5) = 204/7.5 = 27.2

The same problem could be solved with the help of a little algebra by
the rule already given, as thus: the sp. g. of a mixture equals the sum
of the _weights_ of the constituents divided by the sum of the
_volumes_. Then 100 grams of the tailings with _x_ per cent. of pyrites
contain 100-_x_ per cent. of quartz. The sum of the weights is 100. The
volume of the pyrites is _x_/5.1 and of the quartz (100-_x_)/2.6.

Then we have by the rule

  3 = 100/((_x_/5.1)+(100-_x_)/2.6)
  3 = 1326/(510-2.5_x_)
  204 = 7.5_x_
    and _x_ = 27.2

If the percentage (P) and sp. g. (H) of one constituent and the sp. g.
(M) of the mixture are known, the sp. g. of the other constituent may be
calculated by the following formula, in which _x_ is the required sp.
g.:

  _x_ = ((100-P)×M×H)/((100×H)-(P×M))

For example, "tailings" (sp. g. 3.0) containing 27.2 per cent. of
pyrites (sp. g. 5.1) will contain (100-27.2), 72.8 per cent. of earthy
matter having a mean sp. g. of _x_:

  _x_ = ((100-27.2)×3×5.1)/((100×5.1)-(27.2×3))
      = 1113.84/428.4 = 2.6

The differences in sp. g. corresponding to differences in strength have
been carefully determined and tabulated in the case of the stronger
acids and of many other liquids. Such tables are given at the end of
this book.

_To Calculate the Weight of a Measured Volume of Mineral or
Rock._--Multiply the cubic feet by 62.4 and then multiply by the sp. g.
of the stuff, the answer gives the weight in pounds. For example, 100
cubic feet of quartz weighs 100×62.4×2.6 = 16,224 lbs. The weight of any
mass of mineral of known extent and sp. g. is ascertained in this way.

The following table gives the specific gravities of some of the commoner
minerals.

  Barytes              4.5
  Blende               4.0
  Calcite              2.6
  Cassiterite          6.9
  Chalybite            3.8
  Copper pyrites       4.2
  Fluor                3.1
  Galena               7.5
  Hæmatite             5.0
  Mispickel            6.2
  Pyrites              5.0
  Quartz               2.6

FOOTNOTES:

[8] The difference of 20 or 30 milligrams is disregarded here because it
detracts equally from the actual weight of the water and liquid to be
determined. If the liquid is a heavy one the difference shows itself in
the third or second place of decimals. The correction may be made by
deducting from the weight of the flask 0.0012 grams for each gram of
water it holds.



PART II.--THE METALS.



CHAPTER IX.

SILVER, GOLD, CYANIDES, PLATINUM, MERCURY.


SILVER.

Silver is widely diffused, and has been found in most mining districts.
It occurs native in sufficient quantity to constitute one of the chief
ores of the metal. It also occurs combined with sulphur (as in
argentite), with sulphur and antimony (as in stephanite or brittle
silver ore, and in pyrargyrite or ruby silver), and with copper,
sulphur, antimony, and arsenic, as in polybasite. Chloride of silver
occurs native as horn silver or kerargyrite. Silver is found in the ores
of other metals, such as fahlerz, which sometimes contains from two to
ten per cent. of the metal, and galena, which is an important source of
it; in fact, galena is never found entirely free from silver. It is
present also in greater or less quantity in the ores of copper and zinc.

Silver dissolves readily in nitric acid, forming silver nitrate. It only
forms one family of salts, and of these the chloride and nitrate are of
chief importance to the assayer. The formation of the chloride of silver
on the addition of hydrochloric acid or a soluble chloride to the nitric
acid solution, serves for the recognition and separation of silver. The
precipitated chloride is white (becoming violet on exposure to light),
insoluble in nitric acid, soluble in ammonia, hyposulphite of soda, or
concentrated solutions of chlorides. The best confirmatory test is made
by wrapping the precipitate in a little sheet lead, and cupelling, when
the silver will be left in the metallic state, and is easily recognized.

~Dry Assay.~--This assay is made up of two parts: (1) the concentration
of the silver in a button of lead; and (2) the cupellation of the
resulting alloy. The concentration of the button of lead may be
effected either by scorification or by fusion in a crucible.

The scorification assay is performed in a scorifier, which is a shallow
open-mouthed dish about 2-1/2 inches across, with a very thick bottom to
enable it to withstand the corrosive action of the slag. A charge of
more than 3 or 5 grams of the ore cannot be worked in one, and with such
small charges the unavoidable variations have a serious effect on the
figures reported. A difference of one milligram on the weight of the
button of silver got represents a difference of 6 or 10 ounces per ton.
With rich ores such variation is unavoidable under any conditions, and
the only safe plan is to take the mean of several assays. But with
poorer ores the accuracy of the assay, as well as convenience in
working, is much increased by working in a crucible with larger charges.

In scorification the proportion of lead required for scorifying 1 gram
of ore is in average cases from 10 to 15 grams, sinking in the case of
galena to 2 grams, and rising with earthy and refractory substances to
from 30 to 40 grams. But by fusing in a crucible with well-selected
fluxes, a proportion of 4 of flux to 1 of ore is generally sufficient;
and not only is the proportion of added matter less, but it is also
easier to manipulate large quantities in crucibles, so that, although in
some cases the crucible assay is more troublesome and less satisfactory,
yet with poor and earthy ores it is the best method of dealing with
them; while when properly worked it yields results as accurate as
scorification does. As a general rule, if more than 5 grams of ore must
be taken, the crucible assay should be adopted.

[Illustration: FIG. 38.]

~Scorification Assay.~--The charge of ore is usually 3 grams, sometimes
5; the lead varies from 30 to 70 grams, and the quantity of soda, borax,
or powdered glass added varies from 0.3 to 3 or 4 grams. It is generally
recommended to have the lead granulated,[9] and to mix the ore with
about half of it in the scorifier; then to put on the rest of the lead;
and finally to sprinkle the borax or glass on the top. It answers just
as well, however, to use the lead in the shape of foil, and wrap the ore
up in it; and if the ore contains much sulphur, the borax may with
advantage be added (wrapped in a little tissue paper) some five or ten
minutes after the operation has started.

[Illustration: FIG. 39.]

The process of scorification is as follows:--A scorifier (fig. 38) of
convenient size having been selected (one 2-1/2 inches across is most
generally useful), it is dried at a gentle heat for about ten minutes.
The charge is then put into it, and it is introduced, with the help of a
scorifier tongs (fig. 39), into a muffle heated considerably above
redness. The muffle is then closed, and when the metal has melted down,
it is opened, but the temperature is kept up. A ring of slag will, after
a time, form around the metal, and when this appearance (known as the
eye) presents itself, the temperature may be lowered. When the eye has
disappeared--that is, when the layer of slag has quite closed in--a
pinch of powdered culm wrapped in tissue paper is added. As soon as the
slag has again become tranquil, the scorifier is taken out, and its
contents are poured into a mould (fig. 40), the slag is detached, and
saved. If the button of metal weighs more than 30 grams, its size is
reduced by another scorification in the same scorifier, which should
have been replaced in the muffle immediately after the contents had been
poured out. If the ore is not a very rich one, the button of lead will
carry practically all the silver; but with rich ores it is more
satisfactory to save the slag, and subsequently to melt it down with the
cupel on which the lead has been treated, so as to recover the silver
lost in the slag, together with that absorbed in the cupel, at one
operation. Or, if the cupellation loss is neglected or calculated in
some other manner, the slag or slags from the scorifier may be powdered
and mixed with 20 grams of oxide of lead, 5 grams of borax, and 1 gram
of charcoal. This should be melted down in a small crucible, and the
resulting button of lead cupelled.

[Illustration: FIG. 40.]

If the scorification has been unsatisfactory, the quantity of silver
obtained from the slag will be by no means inconsiderable. The usual
explanation is that with sulphury ores compounds of metallic oxides and
sulphides (oxysulphides) are formed, which remain in the slag, retaining
considerable quantities of the precious metal. It is said that under
certain conditions such a slag may contain as much as 10 per cent. of
silver. An excess of lead and a high temperature prevents the formation
of these oxysulphides. But if much silver is present in the ore, the
slag cannot be safely thrown away, even if sulphur is absent, and the
process has been satisfactorily performed.

If the crust which appears on the surface of the lead does not clear,
add a small lump of borax and 20 grams more lead; then close the muffle,
and keep the temperature as high as possible. If the slag forms
properly, but shows unfused or only half-fused lumps, even when the
scorification has proceeded for some time, add more borax, and stir with
an iron rod. The slag adhering to the rod must be detached by hammering,
and replaced in the scorifier.

If the ore consists largely of quartz, soda should be added instead of
borax; or, if it contains much copper, powdered quartz may be used. If
the scorifier at the end of an operation is more than usually corroded,
the borax should be replaced in subsequent assays on similar ores by
powdered glass or quartz.

If a fairly fluid slag is formed which does not clear from the metal and
show the eye, more lead and a higher temperature is wanted.

As a general rule, it may be stated that when a scorification is
unsatisfactory, what is wanted is more heat, more lead, or more borax.

It is a safe plan when work has to be done on a strange ore, to make
three or four assays with varying quantities of lead. The proportion of
lead is right when a further addition does not yield a higher result.
The proper proportion having been found, a note of it should be made for
future use.


POT ASSAYS.

The object of the fusion in a crucible, like that of scorification, is
to concentrate the silver in a button of lead which is to be
subsequently cupelled; and to retain the earthy and waste matters in the
slag. It is necessary to consider the quality of the slag and the weight
and quality of the lead. The slag when fused should be liquid and
homogeneous, and not too corrosive on the crucible. The button of lead
should be soft, malleable, and free from a coating of regulus.[10] In
weight it should not differ much from the ore taken. With 20 grams of
ore, for example, a button of lead weighing from 18 to 25 grams will be
satisfactory: less than this would leave an undue proportion of silver
in the slag; and more would be unnecessarily large for cupelling, and
would increase the loss in that operation.

With average ores, take 20 grams of the powdered ore and mix with 30
grams of "soda," 40 grams of red-lead or litharge, 5 grams of borax, and
from 2 to 2.5 grams of flour, and place in an E crucible (Battersea
round). Put these in the furnace at a red heat, cover the crucible, and
gradually raise the temperature until the whole charge has melted down
and is in a state of tranquil fusion. Pour into a mould, and replace the
crucible in the furnace. As soon as the lead is solid, detach the slag
and put it back into the crucible; and when it is again fluid, charge on
to it with a copper scoop a mixture of 20 grams of oxide of lead, and 1
gram of charcoal: when fusion has again become tranquil, pour and detach
the button of lead. The lead buttons should be hammered into discs with
rounded edges, and be freed from slag; if too big for a cupel they may
be scorified together in a small scorifier, but it is better to cupel
them separately.

~Ores containing Metallic Oxides.~--Peroxides of iron, manganese, and
copper interfere by counteracting the effect of the charcoal or flour,
and thus reducing the size of the lead button. Peroxide of iron will
reduce the weight of lead by a little more than its own weight; and
peroxide of manganese has about twice this effect. When these oxides are
present an additional quantity of flour must be used, and precautions
must be taken to prevent reoxidation of the slag by the furnace gases.
This may best be prevented by using a layer of common salt as a cover to
the charge. When the ores contain a good deal of quartz or stony matter,
the fluxes just given (for average ores) will do; but the proportion of
soda should be diminished, and that of the borax, oxide of lead, and
flour increased as the quantity of metallic oxides become greater. If
the ore contains practically no quartz, the soda may be altogether
omitted, and some glass or powdered quartz added. The following charge
may be taken as an example: weigh up 20 grams of the powdered ore, 15
grams each of "soda" and borax, 60 grams of oxide of lead, and 5 grams
of flour. Mix and place them in an E crucible, and cover with a layer of
from a quarter to half an inch of common salt. Place in the furnace as
before. The salt will give off a considerable amount of fume, which
will, to a certain extent, conceal the state of the charge: when the
crucible has been in the furnace for about 25 minutes remove it and pour
out the contents immediately. With ores that produce a thick slag the
addition of 5 grams of fluor spar will be an advantage. It may happen
that with an unknown ore the first assay will be more or less
unsatisfactory: but from it the necessity for adding more or less flour
will be learnt, and a second assay, with the necessary modification of
the charge, should give a good result.

~Ores containing much Sulphides.~--Ores of this class may be easily
recognized, either by the appearance of the minerals they contain or by
the odour of sulphurous oxide (SO_{2}) which they evolve when roasted on
a spatula. The sulphides most commonly present, in addition to the
sulphurized minerals of silver, are pyrites, galena, blende, and
mispickel. When they are present in only a moderate amount, their effect
is simply to increase the weight of the button of lead; and this is
easily counteracted by reducing the amount of flour, or by omitting it.
When in larger amounts, they not only yield large buttons, but also
render the metal sulphury, sometimes even giving a button of regulus
instead of lead. This last evil may be remedied (1) by putting in a rod
of iron as soon as the charge has fused, or (2) it may be counteracted
by a proper addition of nitre, or (3) when the sulphides present are
only those of iron or copper the sulphur may be removed by calcining,
and the ore converted into one of the class containing metallic oxides.
The calcination is effected as follows:--Weigh up 20 grams of the
powdered ore and place it in a wide-mouthed crucible sufficiently large
to perform the subsequent melting down in. The roasting must be done at
a gentle heat at first, so as to avoid clotting: the mouth of the
crucible should project considerably above the coke, and should slope
forward towards the worker. The charge must be occasionally stirred with
the stirrer (fig. 10) so as to expose fresh surfaces to the action of
the air, and to prevent adhesion to the sides of the crucible. The
stirrer should not be removed till the calcination is finished. The
temperature should be raised at the end to a good red heat; and (to
ensure the decomposition of any sulphate that may be formed) the roasted
ore should be rubbed up in a mortar with a pinch of anthracite, and
again calcined. It is then mixed with fluxes as described, and fused in
the same crucible.

The calcination of an ore is a work occupying a good deal of time, and,
in most cases, it is better to take advantage of the desulphurizing
power of red lead or nitre. Red lead by itself will do, but a large
quantity of it will be required; 1 part of a metallic sulphide needs
from 20 to 50 parts of red lead to yield a button free from sulphur;
whereas at most from 2 to 2-1/2 parts of nitre are sufficient. There is
sometimes an advantage in having a considerable excess of oxide of lead
in the slag, but where there is no such reason, 2 parts of red lead to 1
of ore is enough. A charge which will do for most sulphides is the
following: 20 grams of ore, 40 to 100 grams of red lead, 20 grams of
"soda," 5 of borax, and sufficient nitre (or perhaps flour) to give a
button of about 25 grams of lead. How much this must be (if not already
known) may be approximately determined by fusing 3 grams of the ore and
3 grams of "soda" in a small crucible (C) with 50 grams of litharge (not
red lead) under a cover of salt, and weighing the resulting button of
lead. Subtract 3 from the weight of lead obtained, and the difference
multiplied by 1.3 will give the quantity in grams of nitre required. If
the button of lead weighs less than 3 grams flour must be added. If this
is not satisfactory repeat the assay, adding an extra gram of nitre for
each 4 grams of lead in excess of that required, or 1 gram of flour for
a 12-gram deficiency.

In the method in which iron is used as a de-sulphurising agent, only as
much oxide of lead should be added as will give a button of lead of the
required size. Rather a large button of lead should be got, and the slag
should be strongly alkaline; if the ore does not already carry a large
amount of sulphur some should be added. The fusion should be performed
at a low temperature (similar to that for a galena assay), and should be
continued for some time after it has become tranquil. Take 20 grams of
the ore, 40 grams of "soda," 40 grams of oxide of lead, and 5 or 10
grams of borax; place this mixture in a crucible (with a rod of iron, as
in the galena assay), cover, and fuse for about half an hour. Take out
the rod, washing it in the slag, and, in a minute or two, pour. Clean
and cupel the button of lead.

~General Remarks on the Fusion.~--Other things being equal, the smaller
the quantity of the slag the better, provided there is sufficient to
cover the metal. The presence of peroxides of the heavy metals is
prejudicial, since they tend to increase the quantity of silver retained
in the slag. It may be given as a general rule that when iron, copper,
manganese, &c., are present, there is a more than ordinary need for
cleaning the slags, and care must be taken to keep these metals in the
state of lower oxide.

In selecting the fluxes, it should be remembered that soda is the best
for quartz, and borax for lime and metallic oxides. And that with ores
almost free from gangue some quartz or glass should be added to protect
the crucible. Two parts of soda are enough to flux 1 part of quartz;
whilst of borax, or oxide of lead, 4 parts are barely sufficient. Oxide
of lead has the advantage of being heavy and so does not occupy much
space in the crucible; on the other hand, if the melting down be
performed too quickly, or if oxide of lead only is used, this high
specific gravity is a disadvantage, for the lighter earthy matter floats
as a pasty mass on the more fluid oxide of lead, and thus escapes its
action.

When metallic sulphides are present in the ore, an excess of oxide of
lead helps to keep the sulphur out of the button of metal. In addition
to the oxide of lead required as a flux, some will be required to
provide the lead in which the silver is to be collected. Oxide of lead,
mixed with charcoal or flour, yields, when heated, a multitude of minute
buttons of metal uniformly distributed through the mass of the charge;
as the charge melts down these run together and fall to the bottom; this
shower of lead collects the silver more easily than a single button at
the bottom of the crucible could do. Only that portion of the oxide of
lead which remains in the slag can be considered as a flux; very often
the first indication of an excessive reduction of lead is the pastiness
of the slag rendered thick by the withdrawal of the oxide of lead which
would have kept it fluid. If, in an assay, it is found that 5 parts of
flux are not sufficient for 1 part of ore, the remedy lies in using a
different flux rather than in taking a larger quantity.

_On the Reducing Effect of Charcoal, Flour, and Tartar._--The weight to
be got from a given charge will depend (provided sufficient oxide of
lead is present) upon the proportion of the reducing agents in it. We
have thought it well to illustrate this part of the subject by a series
of experiments which the learner will do well to practise for himself
before proceeding to the assay of actual ores. Take 80 grams of litharge
and 20 grams of a mixture of borax and soda. Fuse three lots (1) with
1.5 gram of charcoal, (2) with 3 grams of flour, and (3) with 7.5 grams
of tartar. Weigh the buttons of lead obtained, and divide each by the
weight of reducing agent used. The results will differ somewhat with the
dryness and quality of the flour, etc., used; in one series of
experiments they were as follows:--

  Gram.             Grams.        Gram.        Grams.
  1.5 charcoal gave 34.0 lead .'. 1 charcoal = 22.6 lead.
  3.0 flour      "  33.5   "  .'. 1 flour    = 11.2   "
  7.5 tartar     "  38.0   "  .'. 1 tartar   =  5.0   "

The use of flour as a reducing agent has many advantages, and it is well
to remember that _1 gram of flour reduces about 11 grams of lead_; and
that charcoal has twice, and tartar one-half, this reducing effect.

_On the Reducing Effect of Charcoal, &c., on Red Lead._--It is often
easier to obtain red lead of good quality than it is litharge, and by a
large number of assayers red lead is the form of oxide of lead always
used. Red lead, however, contains an excess of oxygen which will use up
some of the reducing agent before lead separates out. On making a series
of experiments (similar to the last, but using 80 grams of red lead
instead of the litharge) the results were, with the same quantities of
the reducing agents:--

  With charcoal, 18 grams of lead.
   "   flour,    18   "      "
   "   tartar,   22   "      "

Comparing these with the results with litharge, in the previous table it
will be seen that the same quantity of reducing agent has in each case
brought down 16 grams less of lead, so that a larger amount of the
reducing agent must be added to get a button of the same weight as that
obtained with litharge. To get a button of a desired weight, say 22
grams, we must add reducing agent sufficient to throw down 22 + 16 or 38
grams of lead, which would require 3.4 grams of flour. If this amount of
flour is fused with 80 grams of red lead, a button of lead weighing 22
grams will be formed, the other 16 grams being kept up by the oxygen of
the red lead.

If the quantity of red lead differs from 80 grams, this rule must be
modified. With 40 grams of red lead, for example, we should add an
excess of reducing agent sufficient to throw down 8 grams of lead
instead of 16. Similarly, with 160 grams of red lead, we should add
enough to throw down 32 grams.

The following rule will enable one to calculate the weight of flour
required to produce a button of lead of any desired weight from any
given quantity of red lead. Each 5 grams of red lead present diminishes
the weight of the lead by 1 gram. If then we _divide the weight of red
lead in a charge by 5, and add this to the weight of lead required, the
sum divided by 11 will give the weight of flour which must be added_.
Using 80 grams of red lead and wanting a button of 20 grams, we should
add 3.3 grams of flour.

  80/5 = 16; 16+20 = 36; 36/11 = 3.3 nearly.

The following are some results obtained which will illustrate the
rule:--

  Red Lead used.    Flour used.    Lead got.
   40 grams         3 grams        25.0 grams
  100   "           3   "          13.5   "
   80   "           4   "          30.0   "
   80   "           5   "          40.0   "

_On the Reducing Effect of Metallic Sulphides, and the Counteracting
Effect of Nitre._--The sulphides found in ores will reduce a button of
lead from oxide of lead just as flour does; and, as charcoal, flour and
tartar differ in their reducing power, so equal weights of the different
mineral sulphides throw down different weights of lead.

One gram of iron pyrites yields about 11 grams of lead. One gram of
copper pyrites, blende, fahlerz, or mispickel, yields 7 or 8 grams of
lead, whilst 1 gram of antimonite will give 6, and 1 gram of galena only
a little over 3 grams. It is evident that if an ore carries much of
these sulphides, the quantity of lead reduced will be very much larger
than that required for an assay. To counteract this effect nitre is
added; _1 gram is added for each 4 grams of lead in excess of that
required_. For example: with a 20-gram charge of an ore containing 50
per cent. of pyrites, if no nitre were added, 110 grams of lead would be
got; or, if there was not sufficient oxide of lead to yield this
quantity of metal, the button would be sulphury. To reduce the weight of
the button by 80 grammes, we should add 20 grams of nitre, if litharge
were used; or if red lead were used, we should add 16 grams of nitre,
_since the oxidizing effect of 20 grams of red lead is equivalent to
that of 1 of nitre_, and since 80 grams of red lead are generally used
in a charge. Two assays of an ore of this kind with these quantities of
nitre gave 26.0 grams of lead with litharge, and 22.5 grams with red
lead.

It is best to use in these assays 80 grams of red lead, 20 of soda, and
5 of borax, with 20 grams of the ore. If the lead got by the preliminary
fusion in a small crucible with litharge (described under "_ores
containing much sulphides_") is known, the following table will indicate
the quantity of nitre, or flour, to be added with this charge:--

  -------------------------------+-------------------+------------------
  Lead got in Preliminary Fusion | Flour to be added | Nitre to be added
        with 3 grams of Ore.     |   to the Assay.   |   to the Assay.
  -------------------------------+-------------------+------------------
             0.0 gram            |     3.3 grams     |      none
             3.0 grams           |     1.3 gram      |       --
             6.0   "             |       none        |     4.0 grams
             9.0   "             |        --         |     9.0   "
            12.0   "             |        --         |    14.0   "
            15.0   "             |        --         |    19.0   "
            18.0   "             |        --         |    24.0   "
            21.0   "             |        --         |    29.0   "
  -------------------------------+-------------------+------------------

If litharge is used in the assay instead of red lead 4 grams more nitre,
or 1.5 gram less flour must be used. When more than a few grams of nitre
are added to a charge the proportion of "soda" and borax should be
increased, because one of the products of the reaction of nitre upon
sulphides in the presence of soda is sulphate of soda, and because the
"soda" thus used up no longer serves as a flux; more borax should be
added, as it is the best flux for the metallic oxides which are formed
in the process. If in an assay too large a button of lead is got, even
after this calculation has been made, and the assay is repeated, add 1
gram more nitre for each 4 grams of lead in excess. Sometimes the assay
appears tranquil before the nitre has produced its full effect; in such
cases it is well to seize the crucible with the tongs and mix its fused
contents by rotating them; if this causes an effervescence, the crucible
should be replaced in the fire and the fusion continued. The following
experiments will illustrate the extent to which the above rules may be
relied on. In all of them the standard flux was used, viz.:--80 grams of
red lead, 20 of soda, and 5 of borax.

  _Pyrites_          5     5     5     5     2.5   5    10    15    20
  Quartz            --    20    --    20    17.5  15    10     5
  Nitre             --    --     5     5    --     4    16    28.5  41
  Lead got          42.5  36.0  16.0  19.0  11.5  22.5  22.5  26.5  27.5

  _Copper Pyrites_   8     8     8     8
  Quartz            --    12    --    12
  Nitre             --    --     4     4
  Lead got          47.5  34.0  33.0  26.0

  _Antimonite_       8     8     8     8
  Quartz            --    12    --    12
  Nitre             --    --     4     4
  Lead got          29.0  26.0  13.0  13.0

  _Galena_          10     10    10    10    15    20
  Quartz.           --     15    --    15     5    --
  Nitre             --     --     3     3     3.5   7
  Lead got          17.0   19.0   8.0   8.0  18.5  18.5

A similar set of experiments, with 80 grams of litharge instead of 80
grams of red lead, gave:--

  _Pyrites_          4     4     4     4     7    10
  Quartz            --    15    --    15    13    10
  Nitre             --    --     5     5    12.5  20
  Lead got          46.5  40.5  25.5  24.5  27.0  26.5

  _Copper Pyrites_   5     5     5     5
  Quartz            --    15    --    15
  Nitre             --    --     5     5
  Lead got          44.5  32.5  23.0  25.0

  _Blende_           5     5     5     5    10
  Quartz            --    15    --    15    10
  Nitre             --    --     5     5    15
  Lead got          41.5  38.5  21.5  22.5  21.6

  _Antimonite_       5     5     5     5    10
  Quartz            --    15    --    15    10
  Nitre             --    --     5     5    10
  Lead got          31.0  32.5  11.5  12.5  18.7

  _Galena_          10    10    10    10    15    20
  Quartz            --    15    --    15     5    --
  Nitre             --    --     5     5     7.5  11
  Lead got          33.5  33.5  13.0  14.0  19.5  22.7

The variation in some of these experiments, in which we might have
expected similar results, is due to the fact that the sulphur, and in
some cases the metals, are capable of two degrees of oxidation. For
example: theoretically 1 gram of iron pyrites (FeS_{2}) would yield 8.6
grams of lead if the sulphur were oxidised to sulphurous oxide (SO_{2}),
and the iron to ferrous oxide (FeO); whilst if the sulphur were oxidised
to sulphate (SO_{3}), and the iron to ferric oxide, 12.9 grams of lead
will be thrown down. Similarly the yield with copper pyrites would be
7.5 or 11.6; with blende, 6.4 or 8.5; with antimonite, 5.5 or 8; and
with galena, 2.6 or 3.4. As regards the metals, the lower oxide will
always be formed if the assay is carried out properly (fused under a
cover, and with a sufficiency of reducing agent). But the proportion of
sulphur oxidised completely will vary with the conditions of the assay.
With a slag containing much soda the tendency will be to form sulphate,
and, in consequence, a big reduction of lead; whilst with an acid slag
containing much quartz the tendency will be for the sulphur to go off as
sulphurous oxide (SO_{2}). In a fusion with litharge alone all the
sulphur will be liberated as the lower oxide, whilst with much soda it
will be wholly converted into sulphate. For example: 3 grams of an ore
containing a good deal of pyrites and a little galena, gave, when fused
with litharge, 16.5 grams of lead. A similar charge, containing in
addition 20.0 grams of soda, gave 22.5 grams of lead.

It will be noted from the experiments that 1 gram of nitre kept up on
the average 4 grams of lead; the range being from 3.2 with acid slags to
5.3 with very basic ones. These facts serve to explain some apparently
irregular results got in practice.


CUPELLATION.

The process is as follows:--The cupels, which should have been made some
time before and stored in a dry place, are first cleaned by gentle
rubbing with the finger and blowing off the loose dust; and then placed
in a hot muffle and heated to redness for from 5 to 10 minutes before
the alloy to be cupelled is placed on them. The reasons for this are
sufficiently obvious: the sudden evolution of much steam will blow a
cupel to pieces; and, if the whole of the water has not been removed
before the cupel is filled with molten lead, the escaping steam will
bubble through, and scatter about particles of the metal. If some
particles of unburnt carbon remain in the bone ash, a similar result
will be produced by the escape of bubbles of carbonic acid as soon as
the fused litharge comes in contact with them. The cupels having been
prepared are arranged in a definite order in the muffle, and the assay
buttons are arranged in a corresponding order on some suitable tray
(cupel tray, fig. 41); the heat of the muffle being at bright redness.
Then with the help of the tongs (fig. 42) the assay buttons should be
placed each in its proper cupel; a note having been previously made of
the position it is to occupy, and the door of the muffle closed.

[Illustration: FIG. 41.]

This part of the work should be done promptly, so as not to unduly cool
the muffle: the start requires a fairly high temperature, and is a
critical part of the process. A black crust forms at once on the surface
of the lead; but this ought soon to fuse and flow in greasy drops from
off the face of the metal, so as to leave the latter fluid with a
well-defined outline, and much brighter than the cupel. If this clearing
does not take place, the buttons are said to be frozen; in which case
the temperature must be raised, some pieces of charcoal put in the
muffle, and the door closed. If they still do not clear, the heat must
have been much too low, and it is best to reject them and repeat the
assays.

[Illustration: FIG. 42.]

When the buttons have cleared it is well to check the draught of the
furnace, and to partly open the door of the muffle, so as to work at as
low a temperature as is compatible with the continuation of the
process.[11] Too low a temperature is indicated by the freezing of the
buttons and the consequent spoiling of the assays. Experience soon
enables one to judge when the heat is getting too low. A commoner error
is to have the heat too high: it should be remembered that that which
was high enough to clear the buttons at starting is more than sufficient
to keep the process going. At the finish a higher temperature is again
required: therefore the door of the muffle should be closed and the
furnace urged. The finish is easily recognised. The drops of litharge
which in the earlier stages flow steadily from the surface of the alloy,
thin off later to a luminous film. At the end this film appears in
commotion, then presents a brilliant play of colours, and, with a sudden
extinction, the operation is finished. The metal again glows for an
instant whilst becoming solid.

If the button is a small one the cupel is withdrawn at once and placed
on that square of the cupel tray which corresponds to the position it
occupied in the muffle. If, however, it is fairly large precautions must
be taken to prevent spirting.

Molten silver dissolves oxygen from the air and gives it off on
solidifying; the escape of the gas on sudden cooling is violent and, by
throwing off particles of the metal, may cause loss. This is called
"vegetation" or "spirting." The silver is apparently solid when spirting
takes place; the crust breaks suddenly and some of the metal is forced
out. The evil is best guarded against by slow cooling and avoiding
draughts. With large buttons of silver precautions should never be
omitted. One plan is to allow the cupels to cool in the muffle itself,
the mouth being closed with hot charcoal. Another is to cover the cupel
with another cupel previously heated to redness; in this case the silver
cools between two hot cupels, and, of course, cools slowly. A third plan
is to withdraw the cupel to the door of the muffle, holding it until it
begins to get solid and then immediately to put it back into the hotter
part of the muffle.

Silver remains after cupellation in flattened elliptical buttons,
adhering but only slightly to the cupel. Its upper surface should show
faint markings as if it were crystalline. The presence of platinum
renders it still more crystalline, but removes the characteristic lustre
and renders the metal dull and grey. Copper, if not completely removed,
has a very marked effect on the appearance of the button: the metal is
spread out, damping, as it were, and firmly adhering to the cupel, which
latter in the neighbourhood of the metal is almost black with oxide of
copper. Sometimes the silver button is globular, or even more sharply
rounded on its under than on its upper surface; it is said that this is
due to the presence of lead. Gold may be present even to the extent of
50 per cent. without showing any yellow colour.

The appearance of the cupel affords some useful information. The
presence of cracks evidently due to shrinkage indicates a badly made
cupel. If, however, they are accompanied by a peculiar unfolding of the
cupel, the margin losing its distinctness, it is because of the presence
of antimony. When lead is the only easily oxidisable metal present, the
stained portion of cupel is yellow when cold. A greenish tint may be due
to small quantities of copper or, perhaps, nickel, cobalt, or platinum.
Larger quantities of copper give a greenish grey or almost black colour.
A dark green and corroded cupel may be due to iron. Rings of
pale-coloured scoria may be due to tin, zinc, antimony, or arsenic. When
the cupel shows signs of the presence of these metals in objectionable
quantity, it is well to repeat the assay and scorify so as to remove
them before cupellation.

The button should be detached from the cold cupel by seizing with a
pair of pliers: the under surface should be distorted by squeezing or
hammering the button so as to loosen the adhering bone ash. The cleaning
is easily completed by rubbing with a clean hard brush. After cleaning
the buttons are best put on a tray of marked watch-glasses, and then
taken to the balance and weighed. The weight of silver got needs a small
correction; (1) by deducting for the amount of silver introduced by the
lead or oxide of lead used in the assay;[12] and (2) by adding for the
cupellation loss.

~Loss in Cupellation.~--During the whole process of cupelling a silver
lead alloy a more or less abundant fume may be observed rising from the
cupel. This furnishes an evident loss of lead and a possible loss of
silver; for although silver at the temperature of cupellation gives off
no appreciable vapour, it is known that such fume formed on a large
scale contains silver. It is, however, difficult to believe that the
small amount of lead vapourised carries with it a weighable amount of
silver. That it does not do so in the ordinary way of working is shown
by the fact that a button of silver equal in weight to the silver lost
in cupelling may be got by smelting the cupel and cupelling the
resulting button of lead. The loss of silver by volatilisation is
altogether inconsiderable, unless the temperature at which the operation
is performed is much too high.

Another possible source of loss is the infiltration of small particles
of alloy into the cupel. The cupel is necessarily porous, and particles
of metal may perhaps drain into it, more especially if the bone ash is
not in fine powder; but if this is the main source of loss it is hard to
see why, in cupelling equal weights of silver and gold, the loss is not
equal in each case. It is not easy to believe that the mere filtration
of the fused alloy will effect such a change in the proportion of the
metals as that which actually occurs. For example: a cupel on which an
alloy consisting of 0.80 gram of silver, 0.47 gram of gold, and 25 grams
of lead had been cupelled, was found to contain 7-1/2 milligrams of
silver, and rather less than half a milligram of gold. Assuming, for the
sake of argument, that the gold present had filtered into the cupel in
the form of small drops of alloy, it would have been accompanied by less
than a milligram of silver, and the presence of the extra 6 or 7
milligrams of silver must have been due to a different cause. There can,
thus, be little doubt that the cause of the greater part of the
"cupellation loss" is a chemical one and cannot be counteracted by a
mechanical contrivance.[13] In cupellation, then, there is a loss,
apart from imperfect working, inherent in the process itself; and as the
amount of this loss varies under different conditions, it is necessary
to study it somewhat in detail.

The following experiments are taken without selection from the work of
one student. Three experiments were made for each determination, and the
mean result is given. By "range" is meant the difference between the
highest and lowest result and the percentage loss is calculated on the
silver present. The silver added in the lead used has been deducted.

~Effect of Varying Lead.~--In each experiment 0.4 gram of silver was
taken and cupelled with the lead. The silver loss and "range" are
expressed in milligrams.

  ------------+--------------+--------+------------------
   Lead Used. | Silver Lost. | Range. | Percentage Loss.
  ------------+--------------+--------+------------------
     Grams.   |              |        |
       10     |      6.5     |   1.0  |       1.62
       20     |      7.0     |   1.0  |       1.75
       40     |     12.0     |   1.5  |       3.00
       60     |     12.7     |   0.5  |       3.17
  ------------+--------------+--------+------------------

The loss increases with the lead used.

~Effect of Varying Temperature.~--0.4 gram of silver was cupelled with
20 grams of lead.

  Temperature.    Silver Lost.    Range.    Percentage Loss.

  Bright red          7.0          1.0            1.75
  Clear yellow       17.3          1.7            4.32

The difference in temperature in these experiments was much greater than
would occur even with careless work.

~Effect of Varying Silver.~--20 grams of lead were used in each
cupellation.

  ---------------+--------------+--------+------------------
   Silver Taken. | Silver Lost. | Range. | Percentage Loss.
  ----------------------------------------------------------
    Milligrams.  |              |        |
       12.5      |      0.7     |   0.2  |       5.6
       25.0      |      1.4     |   0.1  |       5.6
       50.0      |      1.6     |   0.4  |       3.2
      100.0      |      2.9     |   0.3  |       2.9
      200.0      |      5.6     |   0.5  |       2.8
      400.0      |      7.0     |   1.0  |       1.7
      800.0      |      9.7     |   1.0  |       1.2
  ---------------+--------------+--------+------------------

It will be seen that, although the quantity of silver lost increases
with the silver present, the percentage loss is greater on the smaller
buttons.

The following results are often quoted:--Cupelling 1 grain of silver
with 10 grains of lead, the loss was 1.22 per cent.; 10 grains of silver
with 100 grains of lead, loss 1.13 per cent.; 25 grains of silver
cupelled with 250 grains of lead, lost 1.07 per cent. The proportion of
silver to lead was the same in the three experiments, and the largest
button gave the best result. Evidently, if the quantities of lead had
been the same in the three experiments (say, 250 grains in each case),
the loss on the smaller quantities of silver would appear worse in the
comparison.

In judging these results, it must be borne in mind that it is difficult
to regulate the temperature, &c., in consecutive experiments so as to
get exactly similar results, so that the range in consecutive
cupellations is greater than that in a batch cupelled side by side.

~Effect of Copper and Antimony.~--0.1 gram of silver was cupelled with
20 grams of lead, and to one batch 0.5 gram of antimony, and to another
0.5 gram of copper was added.

                                                Loss in
                     Silver Lost.    Range.   Percentage.

  Without addition        2.9         0.3         2.9
  With antimony           3.2         0.2         3.2
  With copper             4.9         1.7         4.9

Perhaps the antimony has so small an effect because it is eliminated in
the earlier part of the process, while the silver is still alloyed with,
and protected by, a large proportion of lead; whilst the copper on the
other hand makes its fiercest attack towards the close, when the silver
is least capable of resisting it. The ill effects of copper are most
strongly felt when the quantity of lead present is not sufficient to
remove it: the coppery button of silver got under these conditions is
very considerably less than the weight of silver originally taken.

Although the above is a fair statement of the loss attending average
work, it will not do in very important and exact work to place too much
reliance on the figures given, or, indeed, on any other set of figures,
with the object of correcting the result of an assay. Each man must rely
on his own work.

It is easy to determine what allowance must be made for the loss in
cupellation by cupelling side by side with the assay piece an alloy of
similar and _known_ composition. For, if the two pieces are very nearly
alike, we may justly conclude that the loss on each will be the same;
and if, further, we take the average of three or four such
determinations we shall get results accurate within 0.1 per cent. The
method of getting such results may be best explained by one or two
illustrations. This method of working is termed "assaying by checks."

Suppose we have an alloy of silver and lead in unknown proportions and
that by cupelling two lots of 10 grams each there is got from I. 0.1226
gram of silver, and from II. 0.1229 gram. We should know from general
experience that the actual quantity of silver present was from 2 to 4
milligrams more than this. To determine more exactly what the loss is,
the following plan is recommended:--The two silver buttons are wrapped
up each in 10 grams of lead, and cupelled side by side with two other
lots of 10 grams of the original alloy. If now the two buttons I. and
II. weigh 0.1202 and 0.1203, they will have suffered in this second
cupellation an average loss of 2.5 milligrams. Suppose the two fresh
lots of alloy gave 0.1233 and 0.1235 of silver, the average loss on
these would also be 2.5 milligrams. Add this loss to each result, and
take the mean; which is in this case 0.1259.

If copper is present in the alloy as well as silver, it is necessary to
add about the same quantity of copper to the checks as is supposed, or
known, to be present in the assays. If the substance to be assayed is an
alloy of silver and copper, first cupel 0.5 gram of it, with, say, 10
grams of lead, and weigh the resulting button of silver, in order to get
an approximate knowledge of its composition. Suppose the button weighs
0.3935 gram. We know that this is below the truth: for the sake of round
numbers take it as 0.4, and assume that the rest of the alloy (0.1 gram)
was copper. Two check pieces are then weighed out, each containing 0.4
gram silver and 0.1 gram of copper wrapped in 5 grams of lead. Of course
the silver must be pure. And there is also weighed out two (or better,
four) assay pieces each containing half a gram of the alloy wrapped in 5
grams of lead. The whole lot are then cupelled as nearly as possible
under the same conditions. With four assay pieces, the cupels should be
placed close together in two rows of three across the muffle; the two
check pieces are put in the middle cupels. Suppose the buttons of silver
got weighed as follows:--

  Check pieces      I.  0.3940    II.  0.3945
  Assay pieces      I.  0.3905    II.  0.3912
                  III.  0.3910    IV.  0.3909

The average loss on the two check pieces is 5.7 milligrams, and the
average result of the four assay pieces is 0.3909. Add the average loss
to the average result, and there is got the corrected result, 0.3966.
And if 0.5 gram of alloy contain 0.3966 of silver, 1000 will contain
793.2 of silver, and this is the degree of fineness.

A correction for the loss in cupellation is always made in this way
when rich alloys are being assayed; and in the case of rich ores it may
be done after the manner of the first of the above illustrations. There
is another method of working which relies more on experiment. This is to
smelt the cupel as described further on (p. 114), and to again cupel the
resulting button of lead. The button of silver got in this second
cupellation is added to that first obtained. It will sometimes, but not
often, happen that the two buttons together will slightly exceed in
weight the silver which was actually present. This is because of the
retention in the buttons of a small quantity of lead. It has been stated
that the proportion of lead thus retained may be as much as 1% of the
silver present; this, however, can only be under exceptional conditions.
A determination of the actual silver in the buttons got in the series of
cupellations quoted on pages 102, 103, gave an average percentage of
99.85, so that even with the larger buttons the effect of the retained
lead would be only to increase the weight by about 1 milligram. In the
method of working with checks, the retained lead has no disturbing
influence.

~The proportion of lead required~ for the cupellation of any particular
alloy requires consideration. With too much lead the time occupied in
the process is increased, and so is the loss of silver; on the other
hand, too little lead is of greater disadvantage than too much. From 8
to 16 parts of lead are required for each part of silver alloy, or, if
gold is present, about twice as much as this must be used. For the
cupellation of 1 gram of a silver copper alloy containing different
percentages of copper, the following quantities of lead should be
used:--

  Percentage of Copper
      in Alloy.          Lead Required.

           5                 6 grams
          10                 8   "
          20                10   "
          30                12   "
          40                14   "
      50-100             16-18   "

The alloy, in not too large pieces, is wrapped in the required weight of
lead foil and charged into the cupel at once; or the lead may be put in
first, and, when the cupellation has fairly started, the alloy may be
added wrapped in tissue paper; or a portion of the lead may be first
started and the alloy wrapped in the remaining lead and subsequently
added. The cupellation of large quantities of alloy or of alloys which
contain tin, antimony, iron, or any substance which produces a scoria,
or corrodes the cupel, must be preceded by a scorification. The
advantages of this are that the slag is poorer in precious metal than
that found on a cupel and is more easily collected and cleaned; that
larger quantities of metal can be treated, and that, even if the
substance is in part infusible, or produces at the start a clinkery mass
or scoria, the oxide of lead gradually accumulates, fluxes the solid
matters, and produces a good final result; but if the oxide of lead by
itself is not sufficient for the purpose, borax or some other flux can
be easily added.

If the button of silver got is very small its weight may be estimated
from its size; but it must be remembered that the weight varies as the
cube of the diameter. If one button has twice the diameter of another it
is eight times as heavy and so on. Scales specially constructed for
measuring silver and gold buttons may be purchased; but it is much
better to make the measurement with the help of a microscope provided
with an eyepiece micrometer.

If the length of the long diameter of a silver button be taken the
following table will give the corresponding weight in milligrams:--

  ------------+-----------++------------+---------
   Diameter.  | Weight.   || Diameter.  | Weight.
  ------------------------------------------------
   0.04  inch |   3.6     || 0.015 inch |  0.19
   0.035   "  |   2.4     || 0.014  "   |  0.15
   0.03    "  |   1.5     || 0.013  "   |  0.12
   0.025   "  |   0.9     || 0.012  "   |  0.097
   0.02    "  |   0.45    || 0.011  "   |  0.075
   0.019   "  |   0.4     || 0.010  "   |  0.056
   0.018   "  |   0.33    || 0.008  "   |  0.028
   0.017   "  |   0.27    || 0.006  "   |  0.012
   0.016   "  |   0.23    || 0.004  "   |  0.004
  ------------+-----------++------------+---------

The weight of a corresponding button of gold is got by multiplying by
2.25. These figures are based on those given by Plattner, and apply only
to buttons of such shape as those left after cupellation. A sphere of
silver 0.01 inch in diameter would weigh 0.09 milligram, and a similar
sphere of gold weighs 0.167 milligram.

It is safer, however, to compare with a micrometer the diameter of the
button whose weight has to be determined with that of a standard button
of nearly equal size whose weight is known. The weights of the two
buttons are proportional to the cubes of their diameters. This plan of
working is described more fully in Appendix B., page 440.

~Calculation of the Results.~--After deducting for the silver added, and
correcting for the cupellation loss, the calculation is made in the
usual way; reporting as so many parts per thousand in the case of rich
alloys and as so many ounces and pennyweights, or better as ounces and
decimals of an ounce, in the case of poor alloys and ores.

In this last case, however, it is less fatiguing to refer to a set of
tables which give, either directly or by means of simple addition, the
produce corresponding to any weight obtained from certain given weights
of the substance. The following table gives the produce in ounces and
decimals of an ounce per ton of 2240 pounds:--

  ------------+----------------------------------------------------------
              |                    Weight of Ore taken.
   Weight of  |----------+----------+-----------+-----------+------------
   Metal got. | 3 grams. | 5 grams. | 20 grams. | 50 grams. | 100 grams.
  ------------+----------+----------+-----------+-----------+------------
     0.0001   |     1.09 |     0.65 |     0.16  |    0.06   |     0.03
     0.0002   |     2.18 |     1.31 |     0.33  |    0.13   |     0.06
     0.0003   |     3.27 |     1.96 |     0.49  |    0.20   |     0.10
     0.0004   |     4.36 |     2.61 |     0.65  |    0.26   |     0.13
     0.0005   |     5.44 |     3.27 |     0.82  |    0.33   |     0.16
     0.0006   |     6.53 |     3.92 |     0.98  |    0.39   |     0.19
     0.0007   |     7.62 |     4.57 |     1.14  |    0.46   |     0.23
     0.0008   |     8.71 |     5.23 |     1.31  |    0.52   |     0.26
     0.0009   |     9.80 |     5.88 |     1.47  |    0.59   |     0.29
     0.001    |    10.89 |     6.53 |     1.63  |    0.65   |     0.33
     0.002    |    21.78 |    13.07 |     3.27  |    1.31   |     0.65
     0.003    |    32.67 |    19.60 |     4.90  |    1.96   |     0.98
     0.004    |    43.56 |    26.13 |     6.53  |    2.61   |     1.31
     0.005    |    54.44 |    32.67 |     8.17  |    3.27   |     1.63
     0.006    |    65.33 |    39.20 |     9.80  |    3.92   |     1.96
     0.007    |    76.22 |    45.73 |    11.43  |    4.57   |     2.29
     0.008    |    87.11 |    52.27 |    13.07  |    5.23   |     2.61
     0.009    |    98.00 |    58.80 |    14.70  |    5.88   |     2.94
     0.01     |   108.89 |    65.33 |    16.33  |    6.53   |     3.27
     0.02     |   217.78 |   130.67 |    32.67  |   13.07   |     6.53
     0.03     |   326.67 |   196.00 |    49.00  |   19.60   |     9.80
     0.04     |   435.56 |   261.33 |    65.33  |   26.13   |    13.07
     0.05     |   544.44 |   326.67 |    81.67  |   32.67   |    16.33
     0.06     |   653.33 |   392.00 |    98.00  |   39.20   |    19.60
     0.07     |   762.22 |   457.33 |   114.33  |   45.73   |    22.87
     0.08     |   871.11 |   522.67 |   130.67  |   52.27   |    26.13
     0.09     |   980.00 |   588.00 |   147.00  |   58.80   |    29.40
     0.1      |  1088.89 |   653.33 |   163.33  |   65.33   |    32.67
     0.2      |  2177.78 |  1306.67 |   326.67  |  130.67   |    65.33
     0.3      |  3266.67 |  1960.00 |   490.00  |  196.00   |    98.00
     0.4      |  4355.56 |  2613.33 |   653.33  |  261.33   |   130.67
     0.5      |  5444.44 |  3266.67 |   816.67  |  326.67   |   163.33
     0.6      |  6533.33 |  3920.00 |   980.00  |  392.00   |   196.00
     0.7      |  7622.22 |  4573.33 |  1143.33  |  457.33   |   228.67
     0.8      |  8711.11 |  5226.67 |  1306.67  |  522.67   |   261.33
     0.9      |  9800.00 |  5880.00 |  1470.00  |  588.00   |   294.00
     1.0      | 10888.89 |  6533.33 |  1633.33  |  653.33   |   326.67
  ------------+----------+----------+-----------+-----------+------------

When, as in this table, the fraction of an ounce is expressed by two
places of decimals, it may be reduced to pennyweights (dwts.) by
dividing by 5. For example, 0.40 of an ounce is 8 dwts. The fraction of
a dwt. similarly expressed may be converted into grains with sufficient
exactness by dividing by 4. Thus, 1.63 ozs. equal 1 oz. 12.60 dwts., or
1 oz. 12 dwts. 15 grains. In England it is usual to report in ounces and
decimals of an ounce.

The way to use the table is best shown by an example. Suppose a button
of silver weighing 0.0435 gram was obtained from 20 grams of ore. Look
down the 20-gram column of the table, and select the values
corresponding to each figure of the weight, thus:--

  0.04   = 65.33 ozs. to the ton
  0.003  =  4.90         "
  0.0005 =  0.82         "
  ------   -----
  0.0435 = 71.05         "

Add these together. The produce is 71.05 ozs., or 71 ozs. 1 dwt. to the
ton.

Or, suppose an ore is known to contain 1.24 per cent. of silver. Look
down the 100-gram column, select the values, and add them together as
before.

  1.0  = 326.67 ozs. per ton
  0.2  =  65.33      "
  0.04 =  13.07      "
  ----   ------
  1.24 = 405.07      "

This gives 405 ozs. 1 dwt. 10 grains to the ton.

The calculation becomes more complicated when, as is frequently the
case, the ore contains metallic particles. These show themselves by
refusing to pass through the sieve when the ore is powdered. When they
are present, a large portion, or if feasible the whole, of the sample is
powdered and sifted. The weights of the sifted portion and of the
"metallics," or prills, are taken; the sum of these weights gives that
of the whole of the sample taken. It is very important that nothing be
lost during the operation of powdering.

Each portion has to be assayed separately. It is usual to assay a
portion of the sifted sample, say, 20 or 50 grams, and to add to the
produce of this its share of the "metallics." This way of calculating,
which is more convenient than correct, is illustrated by the following
example:--

  Weight of whole sample       400 grams
  Made up of sifted portions   399   "
       "     "Metallics"         1   "
                               ---
                               400   "

Twenty grams of the sifted portion, when assayed, gave 0.1050 gram of
silver. The whole of the "metallics" scorified and cupelled gave 0.842
gram of silver. Since the 20 grams assayed was 1-20th of the whole,
1-20th part of the 0.842 gram of silver (from the metallics) must be
added to its produce. We thus get 0.1471 gram (0.1050 + 0.0421).

Referring to the 20 gram column, we get--

  0.1    = 163.33
  0.04   =  65.33
  0.007  =  11.43
  0.0001 =   0.16
  ------   ------
  0.1471 = 240.25 ounces per ton.

A more legitimate method of calculation is as follows:--Calculate
separately the produce of each fraction as if they were from different
ores. Multiply each produce (best stated in per cents.) by the weight of
the corresponding fraction. Add together the products, and divide by the
weight of the whole sample. Taking the same example for illustration, we
have:--

  _Metallics._--Weight 1 gram.
         1 gram of it yielded 0.842 grams of silver.
     .'. Produce = 84.2 per cent.
         Produce multiplied by the weight is still ~84.2~.
  _Sifted Portion._--Weight 399 grams.
         20 grams of it yielded 0.105 gram of silver.
     .'. Produce = 0.525 per cent.
         Produce multiplied by weight (0.525 × 399) is ~209.475~.

Add together; and divide by 400, the weight of the whole sample--

             84.2
            209.475
            -------
       400) 293.675 (0.7342


0.7342 is the total produce of the ore in per cents.

Referring to the 100-gram column in the table we find 239.84 ounces to
the ton as the produce.

  0.7    = 228.67
  0.03   =   9.80
  0.004  =   1.31
  0.0002 =   0.06
           ------
           239.84

Comparing this with the result calculated by the first method--viz.,
240.26, we see that that was 0.38 oz., or between 7 and 8 dwts. too
high.

With ores containing "metallics" it is of great importance to powder the
whole of the selected sample without loss during the process; and of
even greater importance to well mix the sifted portion, of which the
last portions to come through the sieve are apt to be more than
ordinarily rich through the grinding down of some portions of the
metallic prills.

~Remarks on Cupellation.~--Cupellation is at once the neatest and the
most important of the dry methods of assaying. Its purpose is to remove
easily oxidisable metals, such as lead and copper, from silver and gold,
which are oxidisable with difficulty. Metals of the first class are
often spoken of as _base_, and gold and silver as _noble_ metals.

When lead is exposed to the action of air at a temperature a little
above redness, it combines with the oxygen of the air to form litharge,
an oxide of lead, which at the temperature of its formation is a
_liquid_. Consequently, if the lead rests on a porous support, which
allows the fused litharge to drain away as fast as it is formed, a fresh
surface of the lead will be continually exposed to the action of the
air, and the operation goes on until the whole of the lead has been
removed. Silver or gold exposed to similar treatment does not oxidise,
but retains its metallic condition; so that an alloy of lead and silver
similarly treated would yield its lead as oxide, which would sink into
the support, while the silver would remain as a button of metal.

The porous support, which is called _a cupel_(fig. 5), should absorb the
slag (oxide of lead, etc.) just as a sponge absorbs water, but must be
sufficiently fine-grained to be impervious to the molten metal. At first
sight it appears difficult to filter, as it were, a fluid slag from a
fluid metal; but an ordinary filter-paper damped with oil will allow
oils to run through and yet retain the water; but damped with water it
will allow water to run through and retain oils. Similarly, fused slags
damp and filter through a cupel, but the molten metal not damping it
withdraws itself into a button, which is retained. Although, of course,
if the cupel is very coarse-grained the metal may sink into the hollows.

Copper, antimony, tin, and most other metals, form powdery oxides, which
are not of themselves easily fusible, and it is necessary when these are
present to add some solvent or flux to render the oxide sufficiently
fluid. Fortunately, oxide of lead is sufficient for the purpose; hence,
mixed oxides of copper and lead, provided the lead is present in proper
proportion, form a fluid slag. In separating copper from silver or gold,
advantage is taken of this fact; for, although we cannot cupel an alloy
of copper and silver, it is easy to cupel an alloy of copper, silver and
lead. If, however, the lead is not present in sufficient quantity, the
whole of the copper will not be removed, and the button of silver, still
retaining copper, will be found embedded in a coating of black oxide of
copper. Copper oxidises less easily than lead does; and, consequently,
the alloy which is being cupelled becomes relatively richer in copper
as the operation proceeds. It is on this account that the ill-effects of
the copper make themselves felt at the close of the operation, and that
the oxide of copper is found accumulated around the button of silver.
Tin and antimony, on the other hand, are more easily oxidised; and the
tendency of their oxides to thicken the slag makes itself felt at the
commencement: if the button of alloy once frees itself from the ring or
crust of unfused oxide first formed, the cupellation proceeds quietly,
and leaves a clean button of silver in the centre. But in either case
the cupellation is imperfect, and should be repeated with a larger
proportion of lead. An unfused and, consequently, unabsorbed slag tends
to retain small buttons of alloy or metal, and thus cause serious loss.

There is a principle underlying many of the phenomena of dry silver
assaying which the student should endeavour to understand; and which
serves to emphasise and explain some facts which without an explanation
may present difficulties. If a button of melted lead be covered with a
layer of slag rich in oxide of lead, and a second metal be added, this
other metal distributes itself between the metal and slag in proportions
which depend mainly upon the ease with which it is oxidised, and to a
large extent upon the relative quantities of material present. Easily
oxidisable metals such as zinc, iron, antimony and tin, will go mainly
into the slag, and, if the proportion of the slag is large, very little
will go into the metal. On the other hand, with metals oxidisable with
difficulty, such as silver, gold, and platinum, the reverse holds true;
nearly the whole of the metals will go into the lead, and very little
into the slag. If, however, the slag be very rich, say in antimony, the
lead will contain antimony; and, on the other hand, if the lead be very
rich in silver, the slag will contain silver in appreciable quantity.
Copper, which is near lead in the facility with which it is oxidised,
will serve for the purpose of a detailed example. The results of actual
analyses of metal and slag formed in contact with each other are shown
in the following table:--

  ---------------------------+--------------------------
     Percentage Composition  | Percentage Composition
      of the Metal.          |   of the Slag.
  ----------+----------------+-----------+--------------
     Lead.  |   Copper.      |   Lead.   |   Copper.
      6.8   |    93.2        |    71.4   |    21.4
     20.0   |    80.0        |    78.0   |    17.0
     28.0   |    72.0        |    80.0   |    12.5
     32.0   |    68.0        |    86.0   |     6.7
     85.0   |    15.0        |    90.0   |     3.6
  ----------+----------------+-----------+--------------

It will be seen from this table that the slag is always much richer in
lead and poorer in copper than the metal with which it is in contact.
The ratio of lead to copper in these five samples is:--

               In the Metal.       In the Slag.
                1 : 14              1 : 0.3
                1 : 4               1 : 0.2
                1 : 2.5             1 : 0.16
                1 : 2               1 : 0.08
                1 : 0.16            1 : 0.04

Assuming these figures to be correct, the following statement is
approximately true. On oxidising an alloy of 10 grams of copper and 10
grams of lead, and pouring off the slag when 3 grams of lead have gone
into it, there will be a loss of (owing to the slag carrying it off)
about 0.2 gram of copper. On repeating the operation, the next 3 grams
of lead will carry with them about 0.5 gram of copper; and on again
repeating, 3 grams of lead will remove 0.8 gram of copper. Finally, the
last gram of lead will carry with it 0.3 gram of copper, and there will
be left a button of copper weighing 8.3 grams. The slag will have
carried off altogether 1.7 gram of copper, which is 17 per cent. of the
metal originally present.

With the more perfect exposure to the air, and quicker removal of the
slag, which results from heating on a cupel, the loss would be heavier.
Karsten got by actual experiment on cupelling copper and lead in equal
proportions, a loss of 21.25 per cent.

Going back to the example: if the slag were collected and fused with a
suitable reducing agent so as to convert, say, half of it into metal,
that half would contain nearly the whole of the copper (such a reduction
is called "cleaning the slag"). On reoxidising this metal, another
button of copper is formed which, added to the first, would reduce the
loss from 17 per cent. to, say, 7 or 8 per cent. And it is conceivable
that by a series of similar operations, almost the whole of the 10 grams
of copper originally taken might be recovered. In practice the problem
is (as far as the copper is concerned) not how to save, but how most
easily to remove it; and since the removal of this metal is quicker from
an alloy containing not too much lead, it is evident that two or three
operations with small quantities of lead will be more effectual than a
single treatment with a larger quantity. With those metals (tin,
antimony, &c.) which pass quickly into the slag, the contrary is true;
hence with these it is necessary to have enough lead present, so that
the slag formed at the outset shall contain enough oxide of lead to make
it fluid. As silver is so much less easily oxidised than copper, we
should reasonably expect that the proportion of silver carried off in
the oxide of lead would be considerably less than that of the copper
indicated in the above example. Indeed, there are one or two facts
which tend to encourage the hope that the operation may be conducted
without any loss. If a piece of pure silver foil is exposed on a cupel
to air at the usual temperature of cupellation, it undergoes very little
change; it does not even fuse; it loses nothing in weight, and does not
oxidise. In fact, even if oxide of silver were formed under these
conditions, it could not continue to exist, for it is decomposed into
silver and oxygen at a temperature considerably below redness. On the
other hand, oxide of silver is not reduced to metal by heat alone, when
mixed with an excess of oxide of lead; while metallic silver is
converted into oxide when heated with the higher oxides of lead, copper,
and some other metals. That silver, and even gold (which is more
difficult to oxidise than silver), may be carried off in the slag in
this way, is in agreement with general experience. If 10 grams of silver
are cupelled with 10 grams of lead, there will be a loss of about 50
milligrams of silver, which is in round numbers 1-30th of the
corresponding copper loss; with 10 grams of gold and 10 grams of lead,
the loss will be 4 or 5 milligrams, which is about 1-12th of the
corresponding silver loss.

~Determination of Silver in Assay Lead.~--Scorify 50 grams of the lead
with 0.5 gram of powdered quartz or glass at not too high a temperature.
When the eye has "closed in," pour; reject the slag, and cupel the
button of lead. Remove the cupel from the muffle immediately the
operation is finished. Weigh, and make a prominent note of the result in
the assay book, as so many milligrams of silver contained in 100 grams
of lead.

~Determination of Silver in Red Lead or Litharge.~--Fuse 100 grams of
the oxide with from 10 to 20 grams of borax; and in the case of litharge
with 2 grams or with red lead 4 grams of flour. Cupel the lead, and
weigh the button of silver. Note the result as in the last case.

~Determination of Silver in Argentiferous Lead.~--Be careful in taking
the sample, since with rich silver lead alloys the error from bad
sampling may amount to several parts per cent. Cupel two lots of 20
grams each, and weigh the buttons of silver. Add to these the estimated
cupel loss, and calculate the result. Or wrap each button of silver in
20 grams of assay lead, and re-cupel side by side with two fresh lots of
20 grams each of the alloy. Calculate the loss incurred, and add on to
the weight of the two fresh buttons got.

~Determination of Silver in Bullion.~--The remarks made under the last
heading as to the importance of correct sampling apply with equal force
here. Make a preliminary assay by cupelling 0.1 gram of the alloy with 1
gram of assay lead; calculate the percentage composition. Refer to the
table on page 105 to find what weight of lead is required for cupelling
1 gram of alloy.

Weigh out four lots of 1 gram each, and wrap them in the required
quantity of lead. Make two check pieces by weighing up two lots of fine
silver equal to that which you believe to be present in the assay
pieces; add copper to make up the weight to 1 gram, and wrap in the same
quantity of lead as was used for the assays.

[Illustration: FIG. 43.]

Prepare six cupels and charge them in the annexed order (fig. 43), and
cupel. Guard against spirting. Clean and weigh the buttons of silver.
Add the mean loss on the two check pieces to the mean weight of the four
assay pieces; this multiplied by 1000 will give the degree of fineness.

~Determination of Silver in Copper.~--The silver is best separated in
the wet way before cupelling, but if the proportion is not too small, it
can be found by cupellation. Weigh up 3 grams of the metal, wrap in 30
grams of sheet lead, and cupel; when the cupellation has proceeded for
fifteen minutes, add 20 grams more lead, and continue till finished.
Weigh the button of silver.

The cupellation loss will be five or six per cent. of the silver
present. Determine it by powdering the saturated portion of the cupel
and fusing in a large Cornish crucible with 30 grams each of soda and
borax, 10 grams of fluor spar, and 1-1/2 gram of charcoal. Cupel the
resulting button of lead, and add 10 grams more of lead towards the
close of the operation. Deduct the weight of silver contained in the
lead used from the weight of the two buttons, and calculate to ounces to
the ton.

In an experiment in which 0.1975 gram of silver was present, the weight
of the button from the first cupellation was 0.1867, and that of the
button from the second, after correcting for the lead added, was 0.0110
gram.

~Determination of Silver in Galena.~ _By Pot Assay._--Mix 20 grams of
the powdered ore with 30 grams of red lead, 20 grams of soda, and 5
grams of borax, as also with from 7 to 10 grams of nitre. Fuse and pour.
Clean the slag if the ore is rich. Cupel the buttons of lead. Make the
usual corrections and calculate in ounces to the ton.

_By Scorification._--Take 10 grams of the ore, 30 grams of lead, and
0.5 gram of borax. Scorify, clean the slag by adding anthracite after
the "eye" has closed in: cupel the button of lead. Weigh the button of
silver, make the necessary corrections, and calculate to ounces to the
ton.

The determination may also be made by cupelling the button of lead got
in the dry lead assay.

A sample of galena determined by the three methods gave the following
results:--

  By pot assay       7.18 ozs. per ton.
   " scorification   7.02      "
   " lead assay      6.72      "

~Determination of Silver in an Ore.~ _By Pot Assay._--Take 20 grams of
the powdered ore and mix with 30 grams of soda, 40 grams of red lead,
and 5 grams of borax, as also with from 2 to 3 grams of flour. Fuse:
pour. Clean the slag by fusing with 20 grams of red lead and two grams
of flour. Cupel the buttons of lead; weigh; make the necessary
corrections, and calculate to ounces to the ton.

_By Scorification._--Take 5 grams of the powdered ore, 50 grams of lead,
and 0.5 gram of "soda" or borax. Scorify. Clean the slag by fusing in a
crucible as in the pot assay. Cupel, &c.

  _Examples._--_By Pot Assay._--Ore taken 20 grams.
  Silver got                 0.2893 gram
  Silver from slag           0.0060   "
  Silver lost in cupellation 0.0100   "
                             ------
                             0.3053   "
  Deduct silver in red lead  0.0017   "
                             ------
  Silver in ore              0.3036   "  =  495.9 ozs. per ton.

       _By Scorification._--Ore taken, 3 grams.
  Silver got.                0.0425 gram
  Silver from slag           0.0022   "
  Silver lost in cupellation 0.0020   "
                             ------
                             0.0467   "
  Deduct silver in lead      0.0015   "
                             ------
  Silver in ore              0.0452   "  =  492.2 ozs. per ton.

~Determination of Silver in Silver Precipitate.~--This substance
contains, in addition to metallic silver and gold, sulphates of lead and
lime; oxides of zinc, copper, and iron; and more or less organic matter.
The sample as received is generally free from "water at 100° C."; and,
since it rapidly absorbs water, care should be taken in weighing it.

Since it contains combined water it is not suited for scorifying;
therefore the determination of silver and gold (fine metal) is made by
pot assay. Weigh up 5 grams of the precipitate, mix with 100 grams of
litharge and 1 gram of charcoal. Melt in a crucible at a moderate heat
and pour. Detach the slag, replace in the crucible, and, when fused, add
a mixture of 20 grams of litharge and 1 gram of charcoal. When the
fusion is again tranquil, pour; and cupel the two buttons of lead.

In a sample worked in this manner the mean of four determinations gave
0.6819 gram of "fine metal"; deducting 1 milligram for the silver
contained in the oxide of lead, and adding 8 milligrams for the
cupellation loss, there is got 0.6889 gram or 13.778 per cent. of silver
(and gold) in the sample.

~Determination of Silver in Burnt Ores.~ _By Pot Assay._--Roasted
cupriferous pyrites containing small quantities of gold and silver comes
under this heading. The following mixture will give a fluid slag which
is heavy and tough when cold:--

  Ore.   Borax.   Sand.   Litharge.   Charcoal.
  100      50      50       100           7

Mix; place in a large crucible; cover with salt; and melt down under
cover. When fused drop in an iron rod for a few minutes, and about a
couple of minutes after its withdrawal, pour the charge quickly into a
large conical mould. The button of lead should weigh about 50 grams.
Cupel and weigh the silver. The litharge may be replaced by red lead, in
which case another gram of charcoal powder must be added.

In our experience the results obtained by this method are about 20 per
cent. less than the actual content of the ore. The results of two
assays, after deducting for the silver in the litharge used, were 3.9
and 4.1 milligrams; and a third assay, in which 5.4 milligrams of silver
had been added, gave 9.2, which, after deducting the added silver,
leaves 3.8 milligrams. The average of the three results is 3.9
milligrams from the 100 grams of ore.

Two lots of 100 grams of the same ore treated in the wet way gave 5.2
and 5.0 milligrams of silver. Burnt ores from Spanish pyrites carry
about 0.005 per cent. of silver.


WET METHODS.

Silver is got into solution from its ores by attacking with nitric acid,
but it is best, after dissolving, to cautiously add dilute hydrochloric
acid, and to carefully avoid excess. If the quantity of silver is very
small the solution is allowed to stand twenty-four hours, but,
otherwise, it is warmed and filtered as soon as it clears. Dry the
residue and concentrate the silver in a button of lead by pot method or
scorification, according to the amount of stony matter present. Cupel
the lead, and the resulting button will be free from all metals, except
perhaps gold. It may be weighed; or dissolved in nitric acid, and the
silver determined gravimetrically in the diluted and filtered solution.
It is better to weigh the metal and afterwards to determine the gold in
it, estimating the silver by difference. Silver alloys are dissolved in
dilute nitric acid (free from chlorides), diluted, and filtered. The
solution is then ready for gravimetric determination.

Sulphuretted hydrogen precipitates silver (like copper), completely,
even from fairly acid solutions.


GRAVIMETRIC DETERMINATION.

Add dilute hydrochloric acid in small excess to the hot dilute solution,
which must contain free nitric acid. Heat and stir until the solution
clears. Decant through a small filter, and wash with hot water,
acidulated at first with a little nitric acid if bismuth is suspected to
be present. Dry quickly, transfer as much as possible of the precipitate
to a watch-glass; burn and ignite the filter paper, treating the ash
first with two drops of nitric acid and then with one of hydrochloric,
and again dry. Add the rest of the silver chloride and heat slowly over
a Bunsen burner until it begins to fuse. Cool and weigh.

The precipitate is silver chloride (AgCl) and contains 75.27 per cent.
of silver. The moist precipitate is heavy and curdy; it is decomposed by
direct sunlight, becoming violet under its influence. When heated it is
yellowish; and, since it is volatile at a high temperature, it must not,
in drying, be heated above its fusing point. The fused chloride can be
removed from the crucible (to which it adheres strongly) by digesting
with dilute acid and zinc.

For the determination of silver in nearly pure bullion the following
process is used:--Weigh up 1.5054 gram of the alloy. With this amount of
alloy each 2 milligrams of silver chloride formed is equivalent to 1
degree of fineness, so that the weight of the silver chloride obtained
(stated in milligrams and divided by 2) will give the degree of
fineness. Transfer to a bottle (known as "bottles for the Indian mint
assay") and dissolve in 10 c.c. of dilute nitric acid, then make up with
water to 200 c.c. and add 3 c.c. of dilute hydrochloric acid. Allow to
stand a few minutes and then shake. Fill the bottle completely with
water, allow to settle, and syphon off the clear liquid; pour on more
water, shake gently to break up the lumps, and again fill the bottle
with water. Invert over the mouth of the bottle a porous Wedgwood
crucible, somewhat similar to those used in gold parting. Take firm hold
of the crucible and bottle, and invert promptly so that the silver
chloride may be collected in the crucible. Allow to stand a little while
for the precipitate to settle, and then carefully remove the crucible
under water.[14] Drain off most of the water and break up the silver
chloride with the help of a well-rounded glass rod. This greatly
facilitates the subsequent drying. Dry first on the water bath and then
on the iron plate. Remove the dried silver chloride, by inverting the
crucible, and weigh it.

As an example, 3 determinations of silver in a coin carried out in this
way gave:--

  (1) 1.8500 gram AgCl = 925.0 fineness.
  (2) 1.8498     "     = 924.9     "
  (3) 1.8502     "     = 925.1     "

~Determination of Silver in Burnt Ores.~--Take 100 grams of the ore and
place in a large beaker of 2-1/2 litres capacity, and cover with 375
c.c. of hydrochloric acid. Boil for half an hour until the oxides are
dissolved and the residue looks like sand and pyrites; then add 20 c.c.
of nitric acid, and boil till free from nitrous fumes. Dilute to 2
litres with water, and pass a current of sulphuretted hydrogen till the
iron is reduced, the copper and silver precipitated, and the liquor
smells of the gas. This takes about one hour and a half.

Filter off the precipitate (rejecting the solution) and wash with warm
water. Dry and transfer to an evaporating dish, adding the ashes of the
filter paper. Heat gently with a Bunsen burner until the sulphur burns,
and then calcine until no more sulphurous oxide comes off. When cold add
30 c.c. of nitric acid, boil and dilute to 100 c.c. Add 1 c.c. of very
dilute hydrochloric acid (1 to 100),[15] stir well, and allow to stand
overnight. Decant on to a Swedish filter paper, dry and calcine.

Mix the ashes with 100 grams of litharge and 1 gram of charcoal, and
fuse in a small crucible. Detach the button of lead and cupel. Weigh and
make the usual corrections. As an example, 100 grams of ore treated in
this way gave 5.8 milligrams of silver; deducting 0.8 for the silver
added in the oxide of lead leaves 5 milligrams obtained from the ore.
Another experiment on 100 grams of the same ore to which 5 milligrams of
silver had been added gave 11.0 milligrams. Deduct 5.8 for the silver
added; this leaves 5.2 milligrams as the silver obtained from the ore.
These give, as a mean result, 0.0051 per cent., or 1.66 ounce per ton.

~Determination of Silver in Commercial Copper.~--For the method of doing
this, with an example and experiment, see under the heading of
_Examination of Commercial Copper_.


VOLUMETRIC METHODS.

There are two of these, one adapted for the determination of silver in
alloys of approximately known composition, and the other of more general
application. The first of these, generally known as "Gay-Lussac's"
method is, as regards its working, perfect in principle; but it requires
a practically constant quantity of silver, that is, one which varies by
a few milligrams only in each determination. It is a confirmatory method
rather than a determinative one. The other is known as "Volhard's," and
resembles in principle and method an ordinary volumetric process.

~Gay-Lussac's method~ is based on the precipitation of silver from a
nitric acid solution by a solution of sodium chloride. The point at
which the whole of the silver is precipitated being recognised by the
standard solution ceasing to give a precipitate. The process depends for
its success upon, (1) the ease which silver chloride separates out from
the solution leaving it clear after shaking, and, (2), the cloudiness
produced by the reaction of very small quantities of silver nitrate and
sodium chloride. In working, a quantity of the sodium chloride solution
equal to 1 gram of silver is added at once to the assay; and, when the
solution has been rendered clear by shaking, the residual silver (which
should not exceed a few milligrams) is estimated with the help of a
weaker solution of sodium chloride. The success in working evidently
depends upon the accuracy with which the first addition of the salt
solution is made. On this account the standard solution is run in from a
special pipette capable of delivering a practically invariable volume of
solution. It is not so important that this shall deliver exactly 100
c.c. as that in two consecutive deliveries the volume shall not differ
by more than 0.05 c.c. The dilute salt solution is one-tenth of the
strength of that first run in, and 1 c.c. of it is equivalent to 1
milligram of silver. Ordinarily it is run in 1 c.c. at a time (and an
ordinary burette may be used for this purpose), shaking between each
addition until it ceases to give a precipitate. If many such additions
have to be made the operation not only becomes tedious, but the
solution also ceases to clear after shaking, so that it becomes
impossible to determine the finishing point.

If the assay contains less than one gram of silver the first addition of
the dilute salt solution of course produces no precipitate. Five
milligrams of silver in solution (5 c.c.) is then added, and the assay
proceeded with in the usual way; 5 milligrams of silver being deducted
from the amount found.

There is required for the assay a _standard solution of sodium
chloride_, which is prepared by dissolving 5.4162 grams of the salt
(made by neutralizing carbonate of soda with hydrochloric acid) in water
and diluting to one litre. 100 c.c. of this is equivalent to 1 gram of
silver.

The weaker solution of salt is made by diluting 100 c.c. of the stronger
one to one litre. One c.c. of this will equal 1 milligram of silver, or
0.1 c.c. of the stronger solution.

A _standard solution of silver_ equivalent to the dilute salt solution
is made by dissolving 1 gram of fine silver in 10 c.c. of dilute nitric
acid, and diluting with water to one litre.

[Illustration: FIG. 44.]

The solution of salt is standardised as follows:--Weigh up 1.003 gram of
fine silver and dissolve in 25 c.c. of dilute nitric acid in a bottle
provided with a well-fitting flat-headed stopper. Heat on the water bath
to assist solution, resting the bottle in an inclined position. When
dissolved blow out the nitrous fumes with the help of a glass tube bent
at right angles. Run in from a stoppered pipette (as shown in fig. 44)
100 c.c. of the standard salt solution, and shake vigorously until the
solution clears. Fill an ordinary burette with the weaker standard salt
solution, and run 1 c.c. into the assay bottle, letting it run down the
side so that it forms a layer resting on the assay solution. If any
silver remains in solution a cloudy layer will be formed at the junction
where the two liquids meet. This is best observed against a black
background If a cloudiness is seen, shake, to clear the liquid, and run
in another c.c. of salt, and continue this until a cloudiness is no
longer visible. Deduct 1.5 c.c. from the amount of the weaker sodium
chloride solution run in. Divide the corrected reading by 10, and add to
the 100 c.c. This will give the volume of strong salt solution
equivalent to the silver taken.

If the first addition of the weaker salt solution causes no cloudiness
add 5 c.c. of the silver solution from an ordinary pipette, shake, and
then run in the weaker salt solution, working as before. These 5
milligrams of silver added must be allowed for before calculating. As
an example:--1.0100 gram of fine silver was taken for standardising a
solution and 4 c.c. of the weaker salt solution were run in. Deducting
1.5 and dividing by 10 gives 0.25 c.c. to be added to the 100 c.c.

        100.25 : 1.0100 :: 100 : _x_
                                 _x_ = 1.0075

which is the standard of the salt solution.

The method of working an assay may be gathered from the following
example:--In the determination of silver in some buttons left after
cupellation, it was assumed that these would contain 99.5 per cent. of
silver. For the assay it was necessary to take a quantity that should
contain a little more than 1.0075 grams of silver; then

        99.5 : 100 :: 1.0075 : _x_
                               _x_ = 1.0125

To ensure a slight excess, there was taken 1.0150 gram of the buttons,
which was treated in exactly the same way as for the standardising. The
quantity of the weaker salt solution required was 7 c.c.; deducting 1.5
c.c., and dividing by 10, gives 100.55 c.c. of strong salt solution,
which is equivalent to 1.0130 gram of silver. This being obtained from
1.015 gram of alloy, is equal to 99.8 per cent., or 998.0 fine.

~The Effect of Temperature.~--The standardising and the assay must be
done at the same time, since a difference of 5° C. makes a difference of
0.1 c.c. in measuring the 100 c.c. of strong solution of salt. It is
always best to prepare a standard with each batch of assays.

~SULPHOCYANATE METHOD.~--Volhard's process is based upon the
precipitation of silver in nitric acid solutions with potassium
sulphocyanate, the finishing point being the development of a
reddish-brown colour, produced by the action of the excess of
sulphocyanate upon ferric sulphate. The white sulphocyanate settles
readily, leaving the liquor clear; and a persistent brown coloration in
the liquid indicates the finish. The assay must be carried out in the
cold; and water free from chlorides[16] must be used.

_The standard sulphocyanate of potassium_ solution is made by dissolving
4-1/2 or 5 grams of the salt (KCyS) in water, and diluting to 1 litre.
100 c.c. are about equivalent to 0.5 gram of silver.

_The standard silver nitrate solution_ is made by dissolving 5 grams of
fine silver in 50 c.c. of dilute nitric acid, boiling off nitrous fumes,
and diluting to 1 litre.

The _indicator_ is a saturated solution of iron alum, or a solution of
ferric sulphate of equivalent strength made by titrating acid ferrous
sulphate with potassium permanganate. Use 2 c.c. for each assay.

The sulphocyanate solution is standardised by placing 50 c.c. of the
silver nitrate solution in a flask with 20 c.c. of dilute nitric acid,
diluting to 100 c.c. with water, and running in the sulphocyanate until
the greater part of the silver is precipitated; then adding 2 c.c. of
the ferric indicator, and continuing the titration until a reddish-brown
colour is developed, and remains permanent after shaking continuously.
The assay is similarly performed, the silver being used in the state of
a nitric acid solution.

The effect of variations in the conditions of the assay may be seen from
the following experiments, in which 20 c.c. of standard silver nitrate
were used:--

~Effect of Varying Temperature~:--

  Temperature            10° C.      30° C.      70° C.      100° C.
  Sulphocyanate reqd.   19.6 c.c.   19.3 c.c.   19.0 c.c.   18.6 c.c.

~Effect of Varying Nitric Acid~:--Varying nitric acid has no effect,
except that with a fairly acid solution the finishing point is somewhat
sharper.

  Nitric acid added      5 c.c.      10 c.c.     20 c.c.     50 c.c.
  Sulphocyanate reqd.   19.6 c.c.   19.5 c.c.   19.6 c.c.   19.6 c.c.

~Effect of Varying Bulk~:--

  Bulk                   50 c.c.    100 c.c.    200 c.c.    300 c.c.
  Sulphocyanate reqd.   19.5 c.c.   19.6 c.c.   19.6 c.c.   19.7 c.c.

~Effect of Varying Ammonic Nitrate~:--

  Ammonic nitrate        0 gram      1 gram      5 grams    10 grams
  Sulphocyanate reqd.   19.6 c.c.   19.6 c.c.   19.7 c.c.   19.9 c.c.

~Effect of Varying Silver~:--

  Silver added         1 c.c.   10 c.c.    20 c.c.    50 c.c.   100 c.c.
  Sulphocyanate reqd.  1.0 c.c.  9.70 c.c. 19.6 c.c.  49.4 c.c.  99.0 c.c.

This method is valuable for determining silver in salts, alloys, and
solutions, where no more than an ordinary degree of accuracy is
demanded. It is easy, and applicable under most of the usual conditions.
Its greatest disadvantage is the brown coloration produced by the
sulphocyanate when the assay is nearly, but not quite, finished; and the
slowness with which this is removed on shaking up with the precipitate.
This is worse with large quantities of precipitate, and if about 1 gram
of silver is present, it gives an indefiniteness to the finish which
lowers the precision of the process to about 1 in 500; this is useless
for the assays of bullion. One writer states that this inconvenience is
due to portions of liquid being entangled in the precipitate, but it
appears much more likely to be due to the action of the precipitate
itself. In attempting to apply the process to the assay of bullion by
working it on the principle of a Gay-Lussac assay, it was found that a
very considerable excess of silver was required to complete the
reaction. In these experiments 100 c.c. of "sulphocyanate" (very
accurately measured) was run into the solution containing the weighed
portion of bullion (fine silver) and, after shaking the solution, was
filtered. In the filtrate the remaining silver, if there should be any,
was determined by the ordinary titration, but with "sulphocyanate" of
one-tenth the strength. This final titration was quite satisfactory. The
amount of silver precipitated by the first 100 c.c., however, varied
with the quantity of silver present as in the following series.[17]

  Silver present.    Silver precipitated.
  1.1342 gram.         1.1322 gram.
  1.1375   "           1.1335   "
  1.1405   "           1.1351   "
  1.1484   "           1.1379   "

These, of course, preclude a method of the kind aimed at, and at the
same time emphasise the importance of uniformity of work in the ordinary
process. In the determination of chlorides in sea-water, Dittmar used a
combined method: precipitating the bulk of the silver as chloride, and
after filtering, determining the small excess of silver by
sulphocyanate. This modification answers admirably when applied to the
assay of bullion. In the ordinary Gay-Lussac method, the precipitation
of the bulk of the silver by the 100 c.c. of salt solution leaves
nothing to be desired, either as to ease in working or accuracy of
result; the silver precipitate settles quickly, and leaves a clear
liquor admirably fitted for the determination of the few milligrams of
silver remaining in solution. But the method of determining this
residual silver by adding successive small quantities of salt so long as
they continue to give a precipitate is unsatisfactory, and, judged on
its own merits apart from the rest of the process, could hardly escape
condemnation. It is clumsy in practice, for the continued adding of
small portions of salt solution is laborious and becomes impossible with
more than a few milligrams of silver in solution. The proposed
modification is simple; having precipitated the silver with the 100 c.c.
of salt solution, as described under Gay-Lussac's method (page 120),
shake till the liquor clears, and filter into a flask, washing with a
little distilled water. Add 2 c.c. of "ferric indicator" to the filtrate
and titrate with a standard "sulphocyanate solution" made by diluting
the ordinary standard solution to such an extent that 100 c.c. after
diluting shall be equivalent to 0.1 gram of silver.[18] Calculate the
weight of silver found by "sulphocyanate" and add it to the weight which
100 c.c. of the salt solution will precipitate.

An advantage of this modification is that an excess of 15 milligrams may
be determined as easily and exactly as 5. In standardising the salt
solution, then, weigh up, say 1.0150 gram of pure silver, dissolve and
titrate. Suppose 13.5 c.c. of "sulphocyanate" required; then these are
equivalent to .0135 gram of silver, (100 c.c. = .1); the silver
precipitated by the salt is 1.0150-.0135--_i.e._, 1.0015 gram, which is
the standard.

~Application of the Method to Assays for Arsenic.~--If silver nitrate be
added to a neutral solution of an arsenate of one of the alkali metals,
silver arsenate (Ag_{3}AsO_{4}), is thrown down as a dark-red
precipitate. If, after adding excess of silver nitrate to insure a
complete precipitation, the arsenate of silver be filtered off, the
weight of the arsenic could be estimated from the weight of silver
arsenate formed. But this may be done much more conveniently by
dissolving the precipitate in nitric acid, and titrating with
sulphocyanate; the silver found will be to the arsenic present as 324
(108×3) is to 75.

The mineral is best treated by the method given in the third paragraph
on page 382; but the solution, after being acidified with nitric acid,
should be made exactly neutral with ammonia. A small excess of silver
nitrate should then be added, and since acid is liberated in the
reaction, the liquor must again be neutralised.[19] The precipitate must
then be filtered off, and washed with distilled water. Then dissolve it
in the paper by slowly running over it 20 c.c. of dilute nitric acid.
Wash the filter with distilled water, collecting with the filtrate in a
small flask. Add 2 c.c. of "ferric indicator" and titrate.

If the sulphocyanate solution be made up with 11 or 12 grams of the
potassium salt to the litre, and be then standardised and diluted, so
that for 100 c.c. it shall equal 1.08 gram of silver, (see p. 38), then
it will also equal .25 gram of arsenic (As). Except for ores rich in
arsenic, it will be better to work with a solution one half this
strength. The standard as calculated from an experiment with pure silver
should be checked by another using pure resublimed white arsenic,
As_{2}O_{3}, which contains 75.75 % of the metal. The quantity of white
arsenic taken, .1 or .2 gram, should contain about as much arsenic as
will be present in the assays. It is converted into sodium arsenate by
evaporating to a small bulk with nitric acid and neutralising with soda.
The precipitation and titration of the silver arsenate should be exactly
as in the assays.

The difficulty of the method is in the neutralising; which has to be
very carefully done since silver arsenate is soluble in even faintly
acid solutions; one drop of nitric acid in 100 c.c. of water is enough
to produce an absolutely worthless result; and an excess of acid much
less than this is still very prejudicial. The addition of a little
sodium acetate to the solution after the final neutralising has a good
effect.

~Arsenic in Mispickel.~--Weigh up .250 gram of the finely-powdered ore,
and place in a Berlin crucible about 1-1/4 or 1-1/2 inch in diameter.
Treat with 10 or 12 drops, one drop at a time, of strong nitric acid,
warm very gently, but avoid much heating. Put on a thin layer of nitre,
and rather more than half fill the crucible with a mixture of equal
parts of soda and nitre. Heat quickly in the blow-pipe flame, and when
the mass is fused and effervescing, withdraw and allow to cool. Boil out
with water, filter and wash. Insert a piece of litmus paper and
cautiously neutralise with nitric acid, using ammonia to neutralise any
accidental excess of the acid. Add a gram or so of ammonium nitrate and
silver nitrate in excess, neutralise again with ammonia and add two or
three grams of sodium acetate. Filter off the precipitate, wash and
titrate. In the fusion care should be taken to avoid much effervescence
(an excess of the soda mitigates this) and the operation should be
stopped as soon as the whole has entered into fusion.


COLORIMETRIC DETERMINATION.

There is, properly speaking, no colorimetric method, but the following,
which is sometimes used, is based on similar principles. It is useful
for the determination of small quantities of silver in substances which
yield clear solutions with nitric acid.

Dissolve a weighed quantity of the substance in nitric acid, and dilute
to a definite bulk. Divide into two equal parts. To one, add a drop or
two of dilute hydrochloric acid, stir and filter. To the other, add a
similar amount of dilute acid, and then to the filtered portion run in
from a burette standard silver nitrate (1 c.c. = 0.5 milligram silver)
until the solutions are equally turbid. Calculate in the usual way.


GOLD.

Gold occurs in nature chiefly as metal. It always contains more or less
silver, and, in alluvial sands, &c., may be associated with platinum and
iridium.

Gold is insoluble in hydrochloric or nitric acid, but is dissolved by
aqua regia or by solutions of iodine, bromine, or chlorine. It is taken
up by mercury, forming an amalgam, from which the mercury may be driven
off by heat.

When gold occurs in particles of any size, it is readily detected by its
appearance, but when finely disseminated through a large quantity of
rock, it is separated and detected by the amalgamation assay--described
below--or by a process of washing somewhat similar to vanning, or by the
following test:--Powder and, if necessary, roast 50 to 100 grams of the
ore, put on it three or four crystals of iodine and enough alcohol to
cover it; allow to stand for half an hour; a piece of filter paper
moistened with the liquid and burnt leaves an ash with a distinctly
purple tint if any gold is present. It is better, however, to filter off
the solution, evaporate, and ignite. Then, either take up with mercury,
and ignite the amalgam so as to get a speck of the metallic gold; or
treat with a few drops of aqua regia, and test the solution with
stannous chloride: a purple coloration indicates gold.

~AMALGAMATION ASSAY.~--This does not attempt to give the total produce
of gold, but rather the quantity which can be extracted on a large
scale; therefore it should imitate as closely as possible the process
adopted in the mine or district for extracting the metal.

Take 2 lbs of the ore in powder and roast; make into a stiff paste with
hot water and rub up for an hour or so with a little mercury. Wash off
the sand carefully, and collect the amalgam. Drive off the mercury by
heat, and weigh the residual gold. It is best to cupel it with lead
before weighing.

In an experiment on a lot of ore which contained 0.189 gram of gold,
0.179 gram was obtained by the above process, equal to about 94-1/2 per
cent. recovered. With ores generally, the yield may be from 80 to 90 per
cent. of the actual gold present.


DRY ASSAY.

The dry assay of gold ores resembles in its main particulars the dry
assay for silver by the crucible method; and for much that is of
importance in its discussion the student is referred to what is written
under Silver on pp. 90-113.

~Size of Assay Charges.~--Gold ores rarely contain more than a few
ounces, often only a few pennyweights of gold to the ton; consequently,
the button of gold obtainable from such quantities of ore as may be
conveniently worked by assaying methods is often so small as to require
more than ordinary care in its manipulation. One milligram of gold forms
a button of about the size of one of the full-stops on this page, and
compared with a million similar particles of quartz (about four ounces),
represents a produce of a quarter of an ounce to the ton: a proportion
such as the assayer is frequently called on to determine. It is evident,
therefore, that a charge of half an ounce or less of the ore, such as is
usual with silver ores, would demand of the worker both skill and care
in the handling of the minute quantity of gold to be obtained from it.
Fortunately the work is simple and precise, so that in practised hands
and with only a 5-gram charge the assay of a 5-dwt. ore is practicable;
with so small a charge, however, the result is barely perceptible on a
sensitive balance: the button of gold should be measured under a
microscope. It follows, therefore, that larger charges of say 50, 100,
or even 200 grams, have an advantage in that they lessen the strain on
the worker's attention, and, except in the case of the poorest mineral,
bring the button of gold within the scope of the balance. On the other
hand, the inconvenience of the larger charges lies in the amount of
fluxes and consequent size of the crucibles required to flux them.

~Sampling.~--A further consideration in favour of the larger charges is
the matter of sampling. In preparing his ore, the student should ask
himself what reasonable expectation he has that the portion he puts in
the furnace will be of average richness. The larger charges are likely
to be nearer than the smaller ones to the average of the parcel of ore
from which they are taken. In explanation of this, let us suppose a
large heap of 5-dwt. ore, in sand of the coarseness of full-stops, and
containing all its gold in particles of 1 milligram, as uniformly
distributed as care and labour in the mixing can accomplish. Such a heap
could not possibly occur in practice, but it will serve for purposes of
illustration. Now, one ton of the sand, however taken, would contain
appreciably the same quantity of gold as any other ton. For a ton would
contain about 8000 particles of gold; and even if two separate tons
differed by as much as 100 particles (which they are just likely to do),
this would mean only a difference of 1 or 2 grains to the ton. On the
other hand, two portions of 14 lbs., which should contain on the average
50 particles of gold, are likely enough to differ by 10 particles, and
this, calculated on a ton, means a difference of 1 dwt. It is easy to
see that something like this should be true; for on calculating the
14-lb. lot up to a ton, the deviation from the average, whatever it may
be, is multiplied by 160; whereas, if the ton were made up by adding
14-lb. lot to 14-lb. lot, up to the full tale, then a large proportion
of the errors (some being in excess and some in defect) would neutralise
each other. An average which is practically true when dealing with
thousands, and perhaps sufficiently exact with hundreds, would be merely
misleading when applied to tens and units. Reasonable safety in
sampling, then, is dependent largely on the number of particles of gold
in the charge taken, and the risk of an abnormal result is less, the
larger the charge taken.

By doubling the charge, however, we merely double the number of
particles. Powdering finely is much more effective; for, since the
weight of a particle varies as the cube of the diameter, halving the
diameter of the particles increases their number eight-fold. If, now, we
modify our illustration by assuming the particles to have only one-sixth
the diameter of a full-stop (which would represent a powder of a
fineness not unusual in ores prepared for assaying), we should multiply
the number of particles by 200 (6 × 6 × 6 = 216). We should then
reasonably expect a 14-lb. parcel of the powder to give as safe a sample
as a ton of the sand would give; and portions of a size fit for crucible
work, say 50 or 100 grams, would be as safe as 10 or 20-lb. samples of
the coarser stuff. For example, 60 grams of such powder would contain,
for a 5-dwt. ore, about 100 particles; and in the majority of cases the
error due to sampling would be less than 10 or 12 grains to the ton, and
would only occasionally exceed a pennyweight. With richer ores the
actual deviation stated as so much to the ton of ore might be greater,
but it would represent a smaller proportion, stated in percentage of
the gold actually present, and would ultimately fall within the limits
of unavoidable error.

It will be seen that the size of the quartz particles has no direct
bearing on the argument; and, in fact, the coarseness of the quartz only
interferes by preventing the uniform mixing of the sand and by binding
together several particles of gold; in this last case, particles so
united must, of course, count as one larger particle. Now, there are
some natural ores in which the gold particles are all very small; with
these fine powdering and mixing yields a product from which a sample may
be safely taken. Then, again, in "tailings," before or after treatment
with cyanide, we have a similar material, inasmuch as the coarser gold
has been removed by previous amalgamation. With these, it is not unusual
to take the portion for assay without any further powdering, since they
are poor in gold, and have already been stamped and passed through a
sieve of say thirty holes to the inch (linear).

But there are other ores, in lump showing no visible gold, which contain
the gold in all possible degrees of fineness, from say prills of a
milligram or so down to a most impalpable powder. The treatment of these
cannot be so simple and straightforward. Suppose a parcel of 1000 grams
(say 2 lbs.) of such ore in fine powder, containing on an average 1
particle of 1 milligram (the presence or absence of which makes a
difference of .6 dwt. on the ton), 10 others of about .5 milligram (each
representing .3 dwt.), and 100 others, which are too coarse to pass
through an 80 sieve, and having an average weight of .1 milligram (each
.06 dwt.), and that the rest of the gold, equivalent altogether to 2
ounces to the ton, is so finely divided that a charge of 50 grams may be
taken without any considerable risk of its interfering with the
sampling. Then in a 50-gram charge there would be one chance in twenty
of getting the milligram particle, in which case the result would be
12.35 dwts. too high; on the other hand, if it were not present the
result would on this account be .65 dwt. too low. Of the ten
.5-milligram particles, it is as likely as not that one will be present,
and its presence or absence would cause an error of 3.3 dwts., more or
less. Of the 100 particles of .1 milligram, there would probably be from
3 to 7, instead of 5, the proper number; this would mean a variation of
2.6 dwts. from the true proportion. So that the probable result would
range about 5 dwts. more or less than the 2-1/2 ozs., which is the true
produce, and there are possibilities of astounding results. It is true
that the majority of the results would be well within these limits, and
now and again the heart of the student would be gladdened by a beautiful
concordance in duplicate assays; nevertheless, there can be no
reasonable expectation of a good assay, and to work in this way, on a
50-gram charge, would be to court failure. The coarse gold must ruin the
assay.

The difficulty may be met by concentrating the whole of the coarse gold
in a small fraction of the ore, by sifting and making a separate assay
of this fraction. A portion of the ore, of about 1000 grams, is ground
to a very fine powder and passed through an 80 sieve, re-grinding when
necessary, until only 20 or 30 grams is left of the coarser powder. This
is mixed with fluxes and carried through as a separate assay. The sifted
portion is _thoroughly mixed_, and a portion of it, say 30 or 50 grams,
taken for assay. The weights of the two portions must be known, and care
must be taken that nothing is lost in the powdering. The method of
calculating the mean result from the two assays is shown on page 109. In
this way of working there is no advantage in continuing the grinding
until the coarser fraction is reduced to a gram or so--rather the
contrary; and rubbing on until all the gold is sent through the sieve is
to be distinctly avoided. The student must bear in mind that what he is
aiming at is the exclusion of all coarse gold from the portion of ore of
which he is going to take only a fraction.

The question of the smaller sampling of gold ores has been dwelt on at
considerable length, as befits its importance, in order that the student
may be impressed with a sense of its true meaning. Sampling is not a
mystery, nor does the art lie in any subtle manner of division. It is,
of course, absolutely necessary that the stuff to be sampled shall be
well mixed, and the fractions taken, so that each part of the little
heap shall contribute its share to the sample. Moreover, it must be
remembered that tossing about is a poor sort of mixing, and that
everything tending to separate the large from the small, the light from
the heavy, or the soft from the hard (as happens in sifting), must be
avoided, or, if unavoidable, must be remedied by subsequent mixing.

With a well-taken sample, we may rely on a great majority of our results
falling within normal limits of error; but nothing can be more certain
than that, in a moderately large experience we shall get, now and again,
deviations much more considerable. These erratic assays can only be met
by the method of working duplicates, which call attention to the fault
by discordant results. Such faulty assays should be repeated in
duplicate, so that we may rest the decision on three out of four
determinations.

The likelihood of two very faulty assays being concordant is remote;
but with very important work, as in selling parcels of ore, even this
risk should be avoided, as concordance in these cases is demanded in the
reports of two or more assayers. The following actual reports on a
disputed assay will illustrate this: (a) 5 ozs. 1 dwt.; (b) 5 ozs.
10 dwts. 12 grains; (c) 5 ozs. 11 dwts.; (c) 5 ozs. 11 dwts. 12 grs.
The mean result of several assays, unless there be some fault in the
method, will be very fairly exact; and individual assays, with an
uncertainty of 1 in 20, may, by repetition, have this reduced to 1 in
100 or less.

~Assay Tons, etc.~--Having decided on taking a larger or smaller
portion, the exact quantity to be used will be either some round number
of grams, such as 50 or 100, easily calculable into percentage; or it
will be that known as the "Assay Ton" (see page 13) or some simple
multiple or fraction of it, which is easily calculable into ounces. The
reports, too, are at least as often made as ounces in the short ton of
2000 lbs., as on the more orthodox ton of 2240 lbs. Now the short ton is
equal to 29,166.6 troy ounces; and the corresponding "assay ton" is got
from it by replacing ounces by milligrams. The advantage of its use is
that if one assay ton of ore has been taken, the number of milligrams of
gold obtained is also the number of ounces of gold in a ton of the ore,
and there is absolutely no calculation. Even if half an assay ton has
been taken the only calculation needed is multiplying the milligrams by
two. On the other hand with a charge of two assay tons the milligrams
need halving. Where weights of this kind (_i.e._, assay tons) are not at
hand they may be easily extemporised out of buttons of tin or some
suitable metal, and it is better to do this than to array out the grams
and its fractions at each weighing. The sets of "assay tons," however,
are easily purchased. As stated on page 13, the assay ton for 2240 lbs.
is 32.6667 grams; and for the short ton, 29.1667 grams. If, however, the
round number of grams be used and the result brought by calculation to
the produce on 100 grams, the conversion to ounces to the ton may be
quickly effected by the help of the table on page 107. As this table
only deals with the ton of 2240 lbs., it is supplemented here by a
shortened one dealing only with _the produce of 100 grams_ and stating
the result in _ounces troy to the short ton of 2000 lbs_.

~Estimation of Small Quantities of Gold.~--_By the Balance._ In
estimating minute quantities of gold there are one or two points, of
importance to an assayer only in this assay, where they will often allow
one to avoid the working of inconveniently large charges. One of these
is known as "weighing by the method of vibrations."

            TABLE FOR CALCULATING OUNCES TO THE SHORT TON FROM THE
                    YIELD OF GOLD FROM 100 GRAMS OF ORE.
  ----------+---------+-----------+---------+-----------+---------
  Milligram.|Ounces to|Milligrams.|Ounces to|Milligrams.|Ounces to
            | the Ton.|           | the Ton.|           | the Ton.
  ----------+---------+-----------+---------+-----------+---------
     0.01   |  0.003  |    0.4    |  0.117  |    7.0    |  2.042
     0.02   |  0.006  |    0.5    |  0.145  |    8.0    |  2.333
     0.03   |  0.009  |    0.6    |  0.175  |    9.0    |  2.625
     0.04   |  0.012  |    0.7    |  0.204  |   10.0    |  2.916
     0.05   |  0.014  |    0.8    |  0.233  |   20.0    |  5.833
     0.06   |  0.017  |    0.9    |  0.262  |   30.0    |  8.750
     0.07   |  0.020  |    1.0    |  0.292  |   40.0    | 11.666
     0.08   |  0.023  |    2.0    |  0.583  |   50.0    | 14.583
     0.09   |  0.026  |    3.0    |  0.875  |   60.0    | 17.500
     0.10   |  0.029  |    4.0    |  1.167  |   70.0    | 20.416
     0.20   |  0.058  |    5.0    |  1.458  |   80.0    | 23.333
     0.30   |  0.087  |    6.0    |  1.750  |   90.0    | 26.250
  ----------+---------+-----------+---------+-----------+---------

Suppose a balance at rest in perfect equilibrium, with the
pointer exactly over the middle point of the scale. Let the scale be a
series of points at equal distances along a horizontal line; then, if a
small weight be placed on one pan, the pointer will deviate from its
vertical position and come to rest opposite some definite part of the
scale, which will depend upon the magnitude of the weight added. The law
determining this position is a very simple one; the deviation as
measured along the points of the scale varies directly as the weight
added. For example, with an ordinarily sensitive balance, such as is
used for general purposes, one milligram will move the pointer along,
say, three divisions of the scale; then two milligrams will move it six
divisions; half a milligram, one and a half divisions; and so on. Of
course, with a more sensitive balance the deviations will be greater.
Now the point at which the needle comes to rest is also the middle point
about which it vibrates when swinging. For example, if the needle swings
from the third to the seventh division on the right then [(7+3)/2] it
will come to rest on the fifth. In working by this method the following
conventions are useful: Always place the button to be weighed on the
left pan of the balance, the weights on the right; count the divisions
of the scale from the centre to right and left, marking the former + and
the latter -; thus -5 is the fifth division to the left. Then the
position of rest is half the algebraic sum of two readings. For example,
let the readings be 7 to the right and 3 to the left, then (+7-3)/2 =
+2. The mean division is the second division to the right. If the
student will place himself in front of a balance and repeat the
following observations and replace the figures here given by his own, he
will have no difficulty in grasping the method. First determine the
_bias_ of the balance; suppose the unloaded balance swings +1.25 and -1;
the bias then is (1.25-1)/2 = +.125 or one-eighth of a division to the
right. Now having put on the button to be weighed let the readings be
+7.5 and +9.25, and (7.5+9.25)/2 = +8.375. Then the effect of the button
has been to move the pointer from +.125 to +8.375, or 8.25 divisions to
the right; we should, therefore, add the weight equivalent of 8.25
divisions to the weights, whatever they may be on the right hand pan of
the balance; if the divisions were to the left (- divisions) we should
subtract. The value of 1 division is easily determined. Suppose the
button in the example were a 1 milligram weight, then we should have
found that 1 milligram = 8.25 divisions .'. 1 division = .121 milligram.
This method of working adds very considerably to the power of a balance
in distinguishing small quantities.

[Illustration: FIG. 44_a._]

_By the Microscope_.--The use of the microscope also is a real advantage
in estimating the weights of minute buttons of gold where there is no
undue risk in sampling, and where an error of say 1 in 20 on the
quantity of gold is tolerable. For ores with copper, lead, zinc, &c., as
well as for tailings rather poor in gold, this leaves a wide field of
usefulness. The method is described on page 440, but the description
needs supplementing for those who are not accustomed to the use of a
microscope. The eye-piece of a microscope (fig. 44_a_, A) unscrews at
_a_, showing a diaphragm at _b_, which will serve as a support for an
eye-piece micrometer. This last, B, is a scale engraved on glass, and
may be purchased of any optical instrument maker, though it may be
necessary to send the eye-piece to have it properly fitted. When resting
on the diaphragm it is in focus for the upper lens, so that on looking
through the microscope, the scale is clearly seen in whatever position
the instrument may be as regards the object being looked at. Suppose
this to be a small button of gold on a shallow, flat watch-glass, on the
stage of the microscope. Bring the button under the "objective" (_i.e._,
the nose of the microscope), which should be about a quarter of an inch
above the watch-glass; then looking through the instrument, raise the
tube until the button of gold, or at least some dust on the glass, comes
into focus. If the button is not in the field, rest the thumbs and index
fingers, using both hands, on the edge of the watch-glass, pressing
lightly but steadily, and give the glass a slow, short, sweeping motion;
the button will perhaps appear as an ill-defined blackness, because not
quite in focus. Bring this into the centre of the field. Raise or lower
the microscope until the button appears with sharp outlines. If the
scale does not cover the button, rotate the eye-piece; this will bring
the scale into a new position. Since the divisions over the button are
less distinct than the others, it is best to read the latter. Thus, in
fig. 44_b_, there are 36 divisions on one side of the button, and 35 on
the other, making altogether 71. The whole scale is 80, therefore the
diameter of the button is 9 divisions. The value of each division
obviously varies with the magnifying power employed. With most
microscopes there is a telescopic arrangement whereby the tube may be
lengthened; if this be done and the button again brought in focus, it
will be seen that, as measured on the scale, the button is much larger
than before. It is evident, therefore, the micrometer must always be
used in the same way. The method given in the appendix (page 440), for
finding the value of the scale when gold buttons are to be measured is
easy and satisfactory. When the button of gold is so small that there is
considerable risk of losing it in transferring to a watch-glass, it may
be measured on the cupel, but for this purpose it must be well
illuminated; this is best done by concentrating light on it with a lens,
or with what comes to the same thing, a clean flask filled with water.

[Illustration: FIG. 44_b._]

Most assayers, however, using a micrometer in this way, would like to
know its absolute value. To do this, a stage micrometer must be
purchased. This is like an ordinary microscope slide (fig. 44_a_, C),
and when looked at through a microscope it shows (fig. 44_c_) lines
ruled on the glass at distances of tenths and hundredths of a
millimetre, ten of each, so that the full scale is 1.1 mm. In the case
illustrated, 60 divisions of the scale in the eye-piece are just equal
to the 1.1 mm., therefore 1 division equals .0183 mm. A cube of this
diameter would contain (.0183×.0183×.0183) .0000061285 cubic mm. The
corresponding sphere is got by multiplying by .5236; this gives
.000003209 cb. mm. The weight of 1 cb. mm. of water is 1 milligram; and,
since gold is 19.2 times as heavy as water (sp. g. = 19.2), the contents
in cb. mm. must be multiplied by 19.2. This gives .0000616 milligram as
the weight of a sphere of gold measuring 1 division.

[Illustration: FIG. 44_c._]

If every result had to be calculated in this way the method would be
very laborious; but, having the figures for the first division, those of
the others may be calculated by multiplying by the cube of the
corresponding number. Thus, for the third division (3×3×3 = 27), the
content of the cube (.0000061285×27) is .0001655 cb. mm.; the content of
the sphere (.000003209×27) is .0000866 cb. mm.; and the corresponding
sphere of gold (.0000616×27) is .00166 milligram. With the help of a
table of cubes the whole calculation for 25 or 30 divisions may be made
in half an hour, and the results preserved in the form of a table will
simplify all future work.

~Assay Operations.~--The actual work of the assay resolves itself into
three operations:--(1) The fusion of the ore and concentration of the
"fine metal" (_i.e._, gold and silver) in a button of lead; (2) The
cupellation of the lead, whereby a button of fine metal is obtained; and
(3) the "parting" of the gold which separates it from the accompanying
silver. The following description takes the order as here given, but the
student, in learning the method, should first practise cupellation if he
has not already done so; next he should practise the separation of gold
from silver, taking known weights of fine gold (p. 63), varying from .5
or .3 gram down to quite minute quantities, and not resting satisfied
until a sensitive balance can barely distinguish between the weights of
gold taken and found. It may be noted here that if he has not a flatting
mill at his disposal, then for large buttons it is better to make an
alloy with eight or nine parts of silver to one of gold, and attack it
with acid without previous flattening, rather than accept the risk and
labour of beating out a less easily attacked alloy to the necessary
thinness with a hammer. It is only after a sense of security in gold
parting has been acquired, that the attack of an ore can be profitably
accomplished, and even then simple and easy ores should be first taken,
passing on to others more difficult, either because of a more complex
mineral composition or a difficulty in sampling.

~Concentration of the fine Metal in Lead.~--The best flux for quartz,
which makes up the earthy matter of most gold ores, is soda, and this is
best added as carbonate or bicarbonate. By theory,[20] 50 grams of
quartz will require 88.5 grams of the carbonate, or 140 grams of the
bicarbonate, to form sodium silicate, which is a glassy, easily-fusible
substance, making a good slag. If the bicarbonate is used, and heat is
applied gradually, steam and carbonic acid are given off at a
comparatively low temperature, and the carbonate is left; at a higher
temperature (about 800° C., or a cherry-red heat) the carbonate fuses
attacking the quartz, and giving off more carbonic acid; as the heat
increases, and the attack on the quartz (which of itself is infusible)
becomes complete, the whole mass settles down to a liquid sodium
silicate, which is sufficiently fluid to allow the gold and lead to
settle to the bottom. The fluid slag does to a certain extent dissolve
some of the crucible, but not seriously. In a perfect working of this
experiment, the first evolution of gases (steam and carbonic acid)
should be gentle, so as to run no risk of its blowing the fine powder
out of the crucible; and the heat at which the second evolution of
carbonic acid is produced should be maintained until the reaction is
completed, so that there may be little or no formation of gas in the
fused mass to cause an effervescence which may force some of the charge
over the edges of the crucible. Of course, in practice the ideal fusion
is not attained, but there is no difficulty in approaching it closely
enough to prevent the charge at any time rising above the level it
reached at first in the crucible, and this should be accomplished. It is
usual with quartzose ores to rely mainly on the action of carbonate of
soda, but not entirely. Litharge is also used; it forms, on fusion with
quartz, a silicate of lead, which is a yellow glass, easily fusible, and
more fluid in the furnace than silicate of soda is. By theory, 50 grams
of quartz would require 186 grams of litharge.[21] The reaction takes
place without evolution of gas, and in its working the only point is to
so regulate the heat that the litharge shall not fuse and drain under
the unattacked quartz, leaving it as a pasty mass on the surface. Now,
if in making up a charge for 50 grams of ore, we took 100 grams of
bicarbonate of soda (equivalent to about 63 grams of the carbonate),
this being five-sevenths of 140 grams (which by itself would be
sufficient), leaves two-sevenths of the quartz to be fluxed by other
reagents: two-sevenths of 186 grams (say 52 grams) of litharge would
serve for this purpose. But if we used 10 grams of borax, which has a
fluxing action about equal to that of the litharge, then 40 grams of the
latter, or (making an allowance for the quartz being not quite pure) say
35 grams, will suffice. The fluxes, then, for the 50 grams of ore would
be: bicarbonate of soda 100 grams, litharge 35 grams, and borax 10
grams; we could decrease any of these, and proportionately increase
either or both of the others, and still rely on getting a fusible slag,
which is the whole of the function of a flux, considered simply as a
flux. It should be remembered, however, that the slag is a bi-silicate
or acid slag, and that its acid character is increased by increasing the
proportion of borax.

But in addition to the fluxes there is required about 30 or 40 grams of
lead to collect the silver and gold. This is best added as litharge (say
40 grams) and flour (4 grams), or charcoal powder (2 grams). See pages
93 and 94. The full charge, then, would be:

  Ore                    50 grams.
  Bicarbonate of soda   100  "
  Litharge               75  "
  Borax                  10  "
  Flour                   4  "

These should be mixed, placed in a suitable crucible (a G Battersea,
round, will do), and heated, at first at a red heat, but finally much
hotter, so as to get a fluid and clean slag. When the charge has been in
tranquil fusion for some little time, take it out and pour it into an
iron mould. When cold, detach the button of lead. The slag should be
glassy, all through alike, and easily separable from the metal. With
ordinary ores, this slag may be considered as free from gold. In an
experiment in which 90 milligrams of gold were added, the full amount
was obtained from the lead produced by the first fusion. But in certain
cases, more especially where large amounts of metallic oxides are
present, the slag is not so clean, and with these the slag should be
powdered, mixed with 40 grams of litharge and 4 of flour, and melted
again; it is an advantage to add a small prill of say 2 or 3 milligrams
of silver to the charge, as it insures a visible product in the
cupellation. Indeed, this last precaution is a good one to be taken
wherever there is reason to expect very small buttons. It has the
further advantage, that, if the quantity of silver necessary for
inquartation is known, the right quantity may be added here, so as to
save a subsequent operation.

~Ores containing Oxides of Iron.~--Of the metallic oxides likely to be
present in a slag, oxide of iron is the most important. Gold is
occasionally found in a matrix of this substance, and in the assay of
"concentrates" largely made up of pyrites, this oxide will be formed in
the preliminary calcination. Now, the lower oxide of iron (ferrous
oxide, FeO) is easy to deal with; fused borax will dissolve about its
own weight of it, and a silicate of soda (such as makes up the bulk of a
slag in a gold assay) will take up at least half as much. But the higher
oxide (ferric oxide, Fe_{2}O_{3}) is more refractory; even 6 parts of
borax yields a poor product, and slags with any considerable percentage
of it are not satisfactory. A student attempting to recover gold from
some hæmatite (in which there was about half an ounce of the metal),
found in the slag nearly a gram of gold, although in the first fusion
the slag appeared perfectly fluid. There is, however, no difficulty in
getting good slags, even with large quantities of iron. For example,
with 50 grams of ferric oxide, 10 of quartz, 30 of borax, 30 of
soda,[22] 50 of litharge, and 7 of flour, the result was quite
satisfactory. So, too, was 25 of quartz, 50 of soda, 50 of litharge, and
7 of flour. It is well, however, in such cases to have an ample
proportion of flux and to aim at a larger button of lead than usual by
increasing the proportion of flour or charcoal (see also page 91). A
charge used on the Randt for roasted "concentrates" (which we may
roughly speak of as quartz and ferric oxide), is one assay ton (about 30
grams) each of ore, soda, and borax, and one and a half assay ton of
litharge and 2 grams of charcoal. Whilst, for the same material, from
which most of the gold has been extracted by "chloridising," 2.5 tons
each of ore, borax, and soda, 4 of litharge, and 4 grams of charcoal are
needed. This quantity requires a large crucible (I Battersea, round). In
this the proportion of silicate of soda and borax counted together is to
the oxide of iron as 4 to 1, on the supposition that the quartz and
oxide of iron of the ore are in about equal quantities; but, in the
larger charge especially, much oxide of lead would also remain as a
flux.

~Ores containing Sulphides.~--In assaying ores containing a large
proportion of pyrites or mispickel, or both, the best plan is to take a
portion and calcine so as to convert it into a product of the kind just
considered. The weighed portion of ore should be placed in a clean
crucible and be heated to incipient redness: with pyrites the first
effect is to drive off about half the sulphur as vapour which burns as
flame over the ore. At this stage care should be taken that there is no
great increase of temperature, otherwise there may be more or less
fusion, which would spoil the operation. When the sulphur flame ceases
the solid sulphide of iron burns with visible incandescence and the
charge should now be stirred with a flattened iron rod so as to expose
fresh portions to the air. The top of the furnace must be open, so that
air may have free access to the crucible. When stirring is no longer
followed by visible burning the heat may be raised to full redness. The
crucible is then lifted out (the stirrer still resting in it) and if the
charge gives off no odour of burning sulphur it is shaken out into an
iron mortar and mixed with the fluxes, taking care to clean the stirrer
in the mixture. The charge is then replaced in the crucible in which the
roasting was done and fused in the furnace. The resulting button of lead
is cupelled for fine metal. Ores rich in sulphides requiring this
treatment are frequently "concentrates." For their assay take 1 assay
ton (30 grams), calcine and mix with an equal weight of soda and of
borax (30 grams each), and half as much again of litharge (1.5 tons or
45 grams), and with 2 grams of charcoal or 5 grams of flour.

Where the sulphides are present in smaller proportion (10 per cent. or
less), they may be taken as serving the purpose of flour or charcoal
(see page 95); the sulphur and iron are oxidised at the expense of the
litharge with a consequent separation of lead as metal. If the
proportion of sulphides is not sufficient to give a large enough button
of lead, some charcoal or flour should be added. On the other hand, if
they are in small excess and give a button of lead somewhat sulphury,
_i.e._, hard and brittle, it may be remedied by the judicious addition
of nitre; this last reagent, however, should not be used in large
quantity. A plan much used to prevent sulphury buttons is to insert an
iron rod or a nail in the charge in the crucible; the iron takes the
sulphur forming sulphide of iron which in moderate quantity does not
form a separate layer of matte but dissolves in the slag. A slag formed
of 50 grams of quartz, 100 soda, and some borax, may take up in this way
some 10 or 12 grams of sulphide of iron. If, however, an ore gives a
layer of matte or speise, it is best to repeat the assay by the method
of calcining before fusion.

~Cyanide Charges, etc.~--In assaying the "tailings" which are to be
treated in a cyaniding plant the following charge is used:

  Tailings   3   assay tons or 100 grams.
  Litharge   4.5      "        150   "
  Soda       4.5      "        150   "
  Borax       .75     "         25   "

The sand is assayed without any further crushing and the assay is made
in duplicate.

The residues after treatment with cyanide, differing from the tailings
merely in being poorer in gold because of the extraction by the solution
of cyanide, are run down with the same fluxes in the same relative
proportions. But four charges of 2.5 assay tons (say 75 grams) are
worked, and two of the resulting buttons are scorified together and then
cupelled, etc., so as to give duplicate assays on charges of 5 assay
tons. This is one of the cases in which it is desirable to add a small
portion of silver before cupelling.

In assaying the "cyanide liquors" for gold, 2 assay tons of the liquor
are measured out (58.3 c.c. for the ton of 2000 lbs., 65.3 c.c. for the
other) and are evaporated to dryness in a lead dish weighing about 35
grams. Such a dish is easily extemporised out of a piece of lead foil,
if the ordinary vessel is not at hand; but care must be taken that the
lead is free from gold. The dish with the dried residue is then
scorified and the resulting button of lead is cupelled.

[Illustration: FIG. 44c.]

In some cases the fusion of the ore may be replaced by a treatment with
solution of cyanide of potassium and the gold recovered from the
solution in the way just described. For this purpose the ore should be
in not too fine powder, otherwise there will be great difficulty in
filtering; a sand which will pass a 30 sieve and having no large
proportion of very fine stuff will do. Not less than 200 grams should be
taken; and as an extraction apparatus a bell jar capable of holding half
as much again may be used. Such a jar may be extemporised by cutting off
the bottom of a bottle by leading a crack around it with a red hot
poker; or a lamp chimney will serve the purpose. The smaller mouth of
the jar is closed by a perforated cork provided with a clipped tube
after the manner of a burette (see fig. 44c). In the jar, just over
the cork, put a plug of loose asbestos or glass wool, or a piece of
sponge to act as a filter; a layer of broken glass, coarse at the bottom
and fine at the top, will serve the same purpose. On this, place the
charge of ore to be extracted. Prepare a solution of cyanide of
potassium in water, with 5 or 10 grams of the salt to the litre. It may
be that the whole point of the assay depends on the solution being of a
definite strength; as, for example, where the relative efficiency of
solutions of different strengths is being determined, when it will be
best to estimate the quantity of cyanide of potassium in the dilute
solution by the method given at the end of this article (page 160). Pour
the cyanide solution on to the ore, letting the first portions to come
through run into the beaker, but as soon as the ore is thoroughly wetted
close the clip and allow to stand for several hours. Then, opening the
clip, run through more cyanide solution and then water, so as to wash
the gold-carrying liquor thoroughly into the beaker. It is no matter if
the liquor is a little bit turbid; transfer it to a lead dish,
evaporate, scorify, and cupel in the usual fashion.

The assay of gold-zinc slimes, which is the precipitate formed by zinc
acting on cyanide solutions of gold, may be made by wrapping 2 or 3
grams in 40 grams of sheet lead and scorifying, cupelling, &c. The
amount of impurity in the stuff varies greatly; it is usually calcined
and mixed thoroughly with soda 40 per cent., borax 30 per cent., and
sand 10 per cent., and melted in graphite pots. The buttons of bullion
obtained are afterwards remelted with borax and run into bars, the
fineness of which varies from 600 to 830 thousandths. The bars are
sampled by chipping off diagonally opposite corners: or better, by
drilling, the drillings being freed from pieces of steel with the help
of a magnet.

~Cupellation.~[23]--The cupellation of lead for gold differs very little
from that of lead carrying silver. When the gold is accompanied by a
larger proportion of silver, and both have to be determined, the
cupellation must be conducted exactly as in a silver assay, the usual
precautions being taken to moderate the temperature so as to lessen the
cupellation loss and to promote a slow and undisturbed solidification in
order to avoid spirting. If, however, the gold predominates the finish
should be effected at a higher heat, as the melting-point of gold is
100° higher than that of silver. The bad effect of a higher temperature
in increasing the cupellation loss need hardly be considered in the case
of such small buttons of gold as are obtained in assaying gold ores, as
any loss there may be is hardly appreciable by the balance. With larger
quantities of gold, however (as in assaying gold bullion), this loss
becomes important; and it is therefore necessary to very carefully
regulate the temperature of the muffle so as to minimise the loss.

~The cupels~ are made of well-burnt bone-ash, of the fineness of coarse
wheat flour, moistened with one-twelfth its weight of water and
compressed into shape in suitable moulds. The moulds sold for this
purpose are often of unsuitable shape. Since lead has a specific gravity
of over 11, a cup to hold from 15 to 25 grams of molten lead need not
have a capacity of more than about 2 c.c. A hollow about 1 inch across
and 1/4 inch deep is sufficient; and the body of the cupel to absorb
this weight of lead should itself weigh from 20 to 25 grams. The button
of lead in a gold assay may be twice as heavy as this. For these larger
buttons a hollow 1-1/3 inch across and 1/3 inch deep will be sufficient.
If these larger cupels are not at hand the larger buttons will have to
be reduced in size by a scorification before cupelling. In some cases
this preliminary scorification is advantageous or even necessary: this
may be because the lead is hard and impure, or it may be that a very
small button of gold is expected. In the latter case it is best to
scorify the lead down to something less than 1 gram, and to perform the
cupellation on a specially prepared small fine cupel. These small cupels
are best made by grinding the unsaturated portion of a used cupel to a
fine powder, and compressing the dry powder into a small Berlin crucible
or scorifier; the face should be made quite smooth by pressure from a
pestle. On such cupels a small speck of gold (less than .01 milligram)
will be left in a good shape and easily visible; but the cupel must be
withdrawn from the muffle as soon as the cupellation is finished to make
sure of always getting the button in good condition. In places, such as
Mints, where large numbers of bullion assays are regularly made a
special form of cupel is used so that not less than six dozen assays may
all be cupelled at the same time in a muffle of ordinary size. These
cupels are square blocks, a little less than 2 inches across, and a
little more than three quarters of an inch deep. Each block carries four
hollows of about .7 inch across and .3 inch deep. A muffle, on a floor
space of 6 inches by 12, would take 3 of these blocks abreast and 6
deep, and thus provide the means for 72 assays.[24]

Cupels made with wet bone-ash should be slowly dried; and if in the
muffle they can be slowly brought to an orange-red heat it is all the
better. Under no circumstances must the lead be placed on the cupel
before the latter has been so thoroughly heated that it can no longer
give off steam or gas of any kind. For this gas bubbling through the
molten metal spatters it, thus spoiling one assay and throwing doubt on
all the rest. Again, the risk of freezing at the start is much greater
with a cupel which has not been properly heated.

The best plan is to do all the cupellations in batches. After the muffle
has cooled down for the withdrawal of the last batch, and the old cupels
have been taken out, the new cupels for the next batch should be put in
their place. The furnace should then be stoked and made ready for the
next cupellations; by the time the furnace is ready the cupels will be
ready also. There should be no unnecessary handling of the cupels once
they have been placed in the muffle.

~The cupellation temperature for gold~ is an orange-red heat or perhaps
a little hotter. Beginners, who are apt to overheat their furnace,
should avoid a heat which can properly be called yellow. Dr. T.K.
Rose[25] has determined the temperature of a muffle during the
cupellation of gold-silver alloys at the Royal Mint. In one muffle the
temperature ranged from 1065° to 1095° C.; the lower temperature was of
course in the front of the muffle. In another it ranged from 1022° to
1062°, and here the muffle appeared to the eye "decidedly cooler than
usual." The alloy left after cupelling was made up of 1 part of gold to
2-1/2 parts of silver, and was fused at 952°; hence the usual
temperature of cupellation was, say, 120° or 130° above the
melting-point of the residual metal. To obtain some real knowledge as to
the meaning of these figures, the student should prepare pointed pieces
of the following metals: silver, which melts at 945°; gold, which melts
at 1035°; and an alloy, half silver, half gold, which melts at 990°.
These should be placed on clean cupels in a muffle almost entirely
closed; the temperature should be very slowly raised, and the appearance
of the muffle when each metal begins to melt should be carefully noted.
The cupelling temperature in Dr. Rose's experiment was as much above the
melting-point of gold as this is above that of the silver-gold alloy.
The _finish of the cupellation_ of gold or gold-silver alloys is
practically the same as with pure silver; there is the same thinning out
of the litharge into a luminous film which becomes iridescent before the
brightening. But the danger of spirting decreases as the proportion of
gold becomes greater, and disappears when the gold is much over 30 per
cent. Nevertheless it is well to let such buttons become solid
undisturbed and protected from draughts in the body of the muffle. This
means closing the muffle and allowing the furnace to cool down somewhat
before withdrawing the cupels. Buttons solidified in this way are more
malleable than when they are withdrawn promptly on the finish of the
cupellation. This is important with large buttons, as in a bullion
assay. On the other hand, very small buttons, especially such as have to
be measured rather than weighed, should be withdrawn as soon as the
luminous film has disappeared. For when this is done the button can be
loosened from the cupel by merely touching it with the point of a pin,
and is then safely and easily transferred to a watch glass by touching
it with the head of a pin which has been moistened. It adheres to this,
and if the pin is not too wet comes off at once on touching the glass,
or in any case will do so on gentle warming.

Molten gold, with little or no silver, has a peculiar colour which is
easy to recognise; it is more globular than a button of silver of the
same size would be, and it shows less adhesion to the cupel. Just after
becoming solid it glows beautifully, and this is so marked that it is a
valuable help in finding the position of a button when it is more than
ordinarily minute.

If the button left from cupellation is yellow it is at least half gold,
and a rough guess as to the proportion of gold may be made from its
yellowness; the rest of the metal is generally silver. The presence of
platinum or one of the platinum group of metals makes the surface of the
button dull and crystalline. The native alloy of osmium and iridium does
not alloy with gold, however, but falls to the bottom of the molten
metal. It shows itself in the subsequent parting as a black spot or
streak on the under surface.

The buttons are removed from the cupel with a pair of pliers and then
brushed to remove adherent litharge and bone-ash. Some assayers advise
cleaning by dipping in warm dilute hydrochloric acid followed by washing
in water and drying. The button is next weighed. When the quantity of
silver obtained is not required to be known the weighing may sometimes
be omitted. The next operation in either case is parting either with or
without a previous inquartation.

_The loss of gold in cupellation_ is by no means always inconsiderable.
In three cupellations of 1 gram of gold with 20 grams of lead made
purposely at a very high temperature the cupel absorbed 6.04, 6.20, and
6.45 milligrams of gold. Hence at a high temperature there may easily be
a loss of more than half a per cent. of the gold. In ten cupellations
with the same quantities of gold and lead, but at an ordinary
temperature, the gold recovered from the cupels varied from 1.37 to 1.92
milligrams, and gave an average of 1.59 milligrams. In round numbers the
cupellation loss of pure gold is .15 per cent.

But if the gold be alloyed with _silver_ the loss is diminished, as is
shown by the following experiments. Gold, .3 gram, was cupelled with 10
grams of lead and varying amounts of silver, and the cupels were assayed
for gold with the following results:

  Silver in the alloy  .3 gram         .6 gram           .9 gram
  Gold in the cupel    .47 milligram   .32 milligram   .17 milligram

These, calculated on the .3 gram of gold, give the loss as .157, .107
and .057 per cent. respectively. The effect of _copper_, on the other
hand, is to increase the cupellation loss, which, silver being absent,
may from this cause rise to .3 per cent., even when the temperature is
not excessive.

In the ordinary assay of gold-copper alloys a constant weight of the
alloy is always taken; hence as the weight of copper in a cupel charge
increases, the weight of gold decreases. The silver, on the other hand,
is always very nearly two and a half times as much as the gold, whatever
its quantity may be. But the cupellation loss is smaller with less gold
and greater with more copper, and it so happens in these assays that
these two opposites nearly neutralise one another. Mr. W.F. Lowe[26]
found the gold recoverable from the cupels on which 20 grains of gold
bullion had been treated varied only between .014 and .015 grain (_i.e._
from .07 to .075 per cent. of the bullion treated), although the quality
of the bullion varied from 9 to 22 carat.[27] But in the poorest bullion
there was only 7.5 grains of pure gold, while in the richest there were
18.3 grains; yet each lost on the cupel the same weight of gold, viz.,
.014 grain. When reckoned in percentages of the actual gold present the
losses are .187 per cent. and .076 per cent. respectively. The heavier
percentage loss is mainly due to the increased quantity of copper.

As with silver so with gold the predominant cause of the cupellation
loss is the solution of the metal in the molten litharge which passes
into the cupel. Three lots of 1 gram of gold cupelled each with 20 grams
of lead repeatedly, so as to make 13 cupellations in all, lost in actual
weight 35.72 milligrams. The gold recovered from the cupels amounted
altogether to 34.56 milligrams. This shows that, compared with the
absorption by the cupel, the other causes of loss are inconsiderable.

The loss of gold by _volatilisation_ is, however, a real one. The dust
from the flues of assay furnaces has been tested on several occasions
and found to contain gold, though in small quantity. Thus Mr. Lowe found
.073 per cent. of silver and .00033 per cent. of gold in such a
material. The lead volatilised from a gold bullion assay would need to
be ten times as rich as this to account for a loss of gold equal to the
hundredth part of a milligram. Dr. Rose, in the paper already quoted,
believes that on a .5 gram charge of standard bullion the loss from
volatilisation is not less than .025 nor more than .05 milligram of
gold.

By way of conclusion it may be said that the cupellation loss of gold is
about .07 per cent., and that it is largely met or even over corrected
by a compensating error due to silver retained in the gold after
parting.

~Inquartation.~--The method of separating the gold from the silver in
gold-silver alloys by boiling with nitric acid does not act equally well
in all cases. An alloy half silver half gold, rolled to thin sheet and
boiled for half an hour with nitric acid, may still retain more than
two-thirds of its silver. An alloy of 1 part gold and 1.7 parts of
silver gives up practically the whole of its silver under similar
treatment. The gold is left in a coherent, though easily broken, sheet
retaining the shape of the original alloy. The gold thus left is quite
spongy and porous, so that the acid can penetrate into its innermost
portions. But if the silver is in large excess in the alloy, the
removal of the silver is less complete, and the residual gold, instead
of holding together in a form easy to manipulate, falls to a powder
which requires care and time in its treatment. The older assayers,
therefore, added silver to their gold in such proportion that the alloy
for parting should be one quarter gold to three quarters silver. This
operation they called _inquartation_.

The modern practice is to aim at getting an alloy with 2-1/2 parts of
silver and 1 part of gold. In gold bullion assays this proportion should
be obtained with fair exactness. And in the parting of such gold buttons
as are obtained in assaying ores it is well to aim at this proportion,
though absolute precision is not a matter of importance.

If the button left on cupelling the lead from an assay of an ore appears
white, it is best to assume that it already contains at least a
sufficiency of silver, in the absence of any knowledge to the contrary.
This will be true in almost all cases. But if, on parting, it does not
lose at least two-thirds of its weight, this indicates that the
assumption was not justified; and also what quantity of silver must be
added to the button before again attempting to part. Generally the fault
will be in the other direction; the silver will be in excess and the
gold will break up and demand very careful treatment.

If, however, such a button is yellow, then, from its weight and depth of
colour, a rough estimate can be made of how much gold is contained in
it. Silver must be added to make the total weight 3-1/2 times as much as
that of the gold supposed to be present. Thus, if the button weighs 10
milligrams and is supposed to contain 8 milligrams of gold, then 8
multiplied by 3-1/2 is 28; the button must, in such case, be made up to
28 milligrams by adding 18 milligrams of silver. In judging of the
quality of the gold button, no ordinary error will very seriously affect
the result. If, in the example just given, the quantity of gold present
was really 7 or even 9 milligrams of gold, the resulting alloy would
still have been suitable for such partings. In fact, in routine assays,
where the quantity as well as the quality of the gold is known within
fair limits, it is often the custom to add the silver for inquartation
to the lead during the first cupellation.

But in the assay of rich gold alloys such approximate work will not do.
If the composition is not already known with a fair degree of accuracy
_preliminary assays_ must be made. Weigh up two lots of 100 milligrams
of the alloy and wrap each in 3 grams of lead. To one add 300 milligrams
of silver. Cupel both. The button containing the added silver must be
flattened and boiled with 15 c.c. of nitric acid; and the resulting
gold must be washed, dried, ignited and weighed. This, in milligrams,
gives directly the percentage of gold. The weight of the other button
gives the percentage of gold and silver; the difference between the two
gives the percentage of silver. The rest will, perhaps, be copper.

The composition of the alloy being known, or having been determined as
just described, the calculation of how much silver must be added is
fairly simple. The following is an example. Suppose the bullion contains
92 per cent. of gold, 1 per cent. of silver and 7 per cent. of copper,
and that .5 gram of it is to be taken for an assay. The .5 gram, then,
will contain

  Gold        .460 gram
  Silver      .005  "
  Copper      .035  "

But the total silver required is .46 gram × 2.5. This equals 1.15.
Allowing for the .005 gram of silver already present, 1.145 gram of
silver must be added.

The silver is incorporated with the gold, and at the same time the
copper is eliminated, by cupelling with sheet lead. How much sheet lead
must be used will depend partly on how much bullion is taken, partly on
how much copper it contains. Four grams of lead will do for a .5 gram
charge; and for a .3 gram charge, 3 grams may be used. But with 20 per
cent. of copper these amounts should be doubled; with 40 per cent. of
copper they should be trebled; and with over 60 per cent. of copper four
times as much lead should be used. For small buttons of gold as little
lead as may be relied on to start cupelling may be taken; the lead may
conveniently be in the form of little cups made by folding lead foil on
a piece of glass rod. With a large number of bullion assays
systematically worked and checked a simple plan would be to always use
the quantity of lead required by the alloy containing most copper which
turns up for assay. This weight, cut out of lead foil, would be kept in
stock folded into little bags ready to receive the bullion and silver.

The silver used for inquartation must, of course, be free from gold and
is best prepared by the assayer who is to use it (see p. 66). It should
not be in long strips or angular pieces likely to perforate the lead in
which it is folded. When wrapped in the lead it should be in the middle
and should make as compact a parcel as possible.

Each little parcel, as completed, should be placed on a tray in its
properly numbered compartment. Its position here should correspond to
that it will occupy in the muffle and eventually in the cupel tray. The
cupellation must be made with all the requisite precautions. A good
smooth malleable button is needed for the next operation, which is known
as flatting.

[Illustration: FIG. 45.]

~Flatting.~--Small buttons, such as are got in assaying most gold ores,
are placed on a polished steel anvil and flattened by one or two blows
with a hammer. The flattened discs are heated to dull redness on a clean
cupel and are then ready for parting. Somewhat larger buttons may be
similarly treated, but they should be annealed (_i.e._ heated to redness
and allowed to cool) during the flattening. The silver-gold alloy left
from the cupellation is soft and bends like lead; but after hammering or
rolling it becomes harder, gets a spring in it like a piece of
mainspring and cracks or splits somewhat easily. There should be no
cracks or stripping or even roughness on the flattened metal, since such
defects may cause the loss of small particles either during the
flattening or in the subsequent treatment with acid. The softness of the
metal is restored by heating. In bullion assays the flatting of the
buttons requires care and practice for its skilful working. The strips
of alloy for parting should be of uniform thickness and condition so
that the action of the acid shall be equal in all cases. The button is
taken from the cupel, cleaned and placed on the anvil: it is then struck
a heavy blow which widens it to about 3/4 inch in diameter; this blow is
followed by two others, one a little in front, the other behind, which
lengthen the disc and give a very blunt roof-like slope to its upper
face. It should then be annealed. This may be done by putting it in a
just red-hot scorifier heated in a muffle: it very soon attains the
right heat and may then be transferred to a cold scorifier; the hot
scorifier should be put back into the muffle. The softened disc is then
taken to the rolls (Fig. 45). The rolls are loosened until the disc can
be pressed between them. Looking through the interval between them the
rolls should appear exactly parallel; if they are not, one adjusting
screw should be loosened and the other tightened until parallelism is
obtained. The rolls are now turned and the disc should be drawn through
without any great effort. Beginners are apt to err by trying to do too
much with one turn of the handle. It is easy to stop whilst the rolls
are only just gripping the metal and then to bring the disc back by
reversing the action. If the disc was originally level and the rolls
are parallel, the metal will appear as a strip which has been merely
lengthened. If the rolls are tighter on one side the strip will be
bowed; the tighter side will correspond with the outer curve of the
crescent. A mistake of this kind may be amended by passing the strip
through the rolls the other way, so as to reverse the irregularity and
so straighten the strip. The screw on the looser side should then be
tightened until parallelism is obtained; after which more care should be
taken to tighten the two screws equally. The rolling should be stopped
when the strip is 3 or 4 inches long and of the thickness of an ordinary
visiting card. The strip should be annealed during the rolling and again
at the finish.

~Parting.~--The thin sheet of metal is dropped into hot dilute nitric
acid and boiled for five or six minutes after the brisk action of the
acid on the metal has ceased. At this stage nearly all the silver has
gone into solution as nitrate of silver and the acid is charged with
this salt. This acid is poured off and the residual metal is again
boiled for from 20 to 30 minutes with a second lot of stronger acid.
This leaves the gold almost pure, though it may still retain from .05 to
.1 per cent. of silver. Treatment with the first acid only would
probably leave three or four times as much.

The _nitric acid_ used should be free from hydrochloric, sulphuric,
iodic and telluric acids. In testing it for the first of these add
nitrate of silver and dilute with distilled water; there should be no
turbidity. In testing for the others evaporate three lots in dishes over
a water-bath. Test one for sulphates by adding water and barium
chloride. Test another for iodates by taking up with a little water,
adding a few drops of starch paste and then dilute sulphurous acid
solution a little at a time; there should be no blue colour. Test the
third for tellurium by heating with 1 c.c. of strong sulphuric acid
until dense fumes come off; allow to cool considerably; a piece of tin
foil added to the warm acid develops a fine purple colour if only a
trace of tellurium is present.

The presence of lower oxides of nitrogen, which impart a brown colour to
the acid, is objectionable; they, however, are removed by boiling the
diluted acid before using it for parting. It is usual to keep a stock of
the acid suitably diluted to the two strengths required for the parting.
These are known as the parting acids. The _first parting acid_ is the
weaker and is used in the first attack on the metal. The specific
gravity generally recommended for it is about 1.2. It may be prepared
either by diluting the strong acid with about its own volume of
distilled water, or by suitably diluting the second parting acid which
has been already used in an assay; the small proportion of silver this
contains is not harmful for this purpose. The _second parting acid_ has
a specific gravity of about 1.3, and may be made by diluting the strong
acid with half its volume of distilled water.

_Parting in Flasks._--Flasks are most convenient for the larger
partings, as in bullion assays; and should always be used for this
purpose unless some of the special parting apparatus, like that used in
Mints, is available. Many assayers use flasks, though of a smaller size,
for the ordinary partings in assaying gold ores. The flasks are either
bulbs with long necks (Fig. 46) which ought to be heated on rose burners
of special construction; or they are small flat-bottomed conical flasks
which may be conveniently heated on a hot-plate and are, in this
respect, much easier to deal with in general work. The following
instructions apply to the parting of an alloy containing a few decigrams
of gold together with the proper proportion of silver.

[Illustration: FIG. 46.]

The strip from the rolls, after being softened by annealing, is folded
on itself on a glass rod into a roll or cornet. It should be so plastic
that it will retain the shape thus given it and not spring open on
removing the pressure of the fingers. About 50 c.c. of the first parting
acid are placed in a 6-ounce conical flask and heated to boiling; the
flask is then withdrawn, and tilted a little to one side, whilst the
cornet is cautiously dropped into it; there will be a sudden issue of
hot vapours and a prompt withdrawal of the hand is advisable. The flask
is replaced on the hot plate and the acid is kept boiling for 10 or 15
minutes. The flask is then withdrawn and the acid diluted with about an
equal volume of distilled water. If the flask has a thick glass band
around its neck, a little way down,[28] care must be taken to use hot
water, for any sudden chill will certainly crack the flask where it is
thus thickened. The liquor is carefully decanted into a clean beaker and
is then thrown into a jar marked "waste silver." About 40 c.c. of the
second parting acid, heated to boiling, is then poured into the flask,
which is then replaced on the hot plate. The boiling is continued for 15
or 20 minutes or even longer. At this stage bumping has to be specially
guarded against; after a little experience it is easy to see when this
is imminent and the flask should be withdrawn to a cooler part of the
plate; it is better to prolong the heating at a temperature below
boiling than to run the risk of disaster. Some of the older writers,
however, are rather insistent on vigorous boiling with large bubbles.
The addition of a small ball of well-burnt clay of about the size of a
pea has been recommended, as it lessens the tendency to irregular and
dangerous boiling. At the end of the treatment with the second acid the
flask is withdrawn from the plate and the acid is diluted with an equal
volume of distilled water. The liquor is carefully decanted into a
beaker, and then poured into a jar or Winchester marked "acid waste"; it
serves for making the first parting acid. The flask is then washed twice
with hot distilled water; the washings must be carefully decanted from
the gold. The flask is then filled with water. A parting cup (size B) is
then placed over its mouth, like a thimble on the tip of a finger. This
cup is of unglazed porous earthenware of such texture that it absorbs
the last few drops of water left on drying; and with a surface to which
the gold does not adhere even on ignition. The gold should fall out
cleanly and completely on merely inverting the cup over the pan of the
balance. The flask and cup are then inverted so that the flask stands
mouth down in the cup; a little of the water from the flask flows into
the cup, but only a little. The gold falls steadily through the water
into the cup. When time has been allowed for even the finest of the gold
to have settled into the cup, the flask is removed. This is easiest done
under water. The cup, with the flask still resting in it, is dipped
under water in a basin; as soon as the neck of the flask is immersed the
crucible can safely be drawn away from under it and then lifted out of
the water. The flask should not be taken away first, for the rush of
water from it may easily sweep the gold out of the cup. The water in the
cup is then drained off and the cup is dried at not too high a
temperature; for if the last drop or two of water should boil there is
danger of spattering the gold out of the crucible. When it is dry, the
cup is heated on a pipe-clay triangle over a Bunsen burner, or on a slab
of asbestos in a muffle, to a dull-red heat. This brings the gold to
"colour"; that is, the loose tender dark coloured gold becomes bright
yellow and coherent; and is in a state fit to be transferred to the
balance and weighed. All unnecessary transferences must be avoided. As
soon as the cup is cool it may be inverted over the pan of the balance,
when the gold will fall out cleanly or, at the worst, a gentle tap with
the finger will be sufficient to detach it.

_Parting in test-tubes_, or in the smaller conical flasks, is used in
the assay of gold ores of ordinary richness. The work is exactly like
that just described in all its main features. Generally speaking much
less acid will be used; for example, in test-tubes and for small
buttons, 3 or 4 c.c. of each acid is quite enough. Again, the action
need not be so prolonged; 10 or 15 minutes in each acid is sufficient.
So, too, the heating may be less; it is very convenient to support the
test-tubes in a water-bath, or merely to rest them in a beaker of
boiling water; and there is no serious objection to doing this. A
smaller parting cup should be used; the A size is suitable. The button,
on the other hand, should be beaten thinner than is needed for the
larger partings. If the silver should be in excess and the gold becomes
much broken up, ample time should be given for subsidence from the
test-tube or flask into the parting cup.

_Parting in glazed crucibles or dishes._--This method of working has the
advantage that there is no transference of the gold until it is placed
on the pan of the balance. On the other hand, in the boiling more care
is required in adjusting the temperature. The following instructions
apply to the treatment of very small buttons, to which the method is
more particularly applicable; but very little modification is needed for
the treatment of larger buttons. The smallest sized Berlin crucibles
answer admirably. They should be cleaned by treatment with hot and
strong sulphuric acid, followed by washing in distilled water; the
comfort and ease of working mainly depends on the thoroughness of this
cleaning. The crucible, one-third full with the first parting acid, is
heated on the hot plate until the acid is almost boiling. The flattened
and annealed button is dropped into it and the heating continued with,
at most, gentle boiling for a few minutes. The crucible is then filled
with distilled water, which cools it enough for easy handling; and when
the gold has settled the liquor is poured off along a glass rod into a
clean beaker. Any greasiness of the crucible makes itself felt here and
is very objectionable. The crucible is then one-third filled with the
second parting acid and the heating resumed, care being taken not to
raise the temperature too high; this should be continued much longer
than before, say for five or ten minutes or even longer according to the
size of the button. Distilled water is again added and, when it is
drained off, the washing with distilled water is twice repeated. It will
not be possible to drain off the last drop of water; but if the gold is
coherent, the crucible can be so inclined that this drop drains away
from the gold, in which case the drying can be done rapidly; the boiling
of the water will do no harm. But when the gold is much broken up, it
will collect in the middle of this drop and the drying must be done
gently; best by putting the crucible in a warm place. When dry, the
crucible is heated till the gold changes colour, but the heat must be
kept well below redness. When cold, the gold is transferred directly to
the pan of the balance. With minute specks of gold which will require
measuring, it is best to put a small piece of lead foil (say .1 gram) in
the crucible over the gold, and then heat the crucible to above redness
over a blowpipe. Whilst the lead is oxidising it is easily swept round
in a bath of molten litharge by merely tilting the crucible. In this way
any separated specks of gold can be taken up with certainty. When the
worker is satisfied that the lead has had ample opportunity for taking
up the gold, the lead must be kept in one place and the heat slowly
lowered. By this means the button becomes supported in comparatively
pure litharge and when solid can be picked out quite easily with a pair
of pliers and in a very clean condition. The lead button is then
cupelled on a very fine cupel, as already described. The method of
working last described destroys the crucible. If the gold is not quite
so small this may be avoided. A small piece of lead foil should be
hammered out until it is perfectly flexible. It is then shaped into a
tray and the gold is transferred to it. The lead is then folded over,
with the help of two pins; and cupelled.

If the crucible shows a black stain on heating it is because some silver
remains through bad washing. It shows poor work and the assay should be
repeated.

_The silver retained in the gold after parting_ is, in bullion assays,
an important matter; it is roughly equal to the loss of gold due to
absorption by the cupel. Mr. Lowe working on .5 oz. of gold, obtained by
parting in assaying bullion, found it to contain .123 per cent. of
silver. Dr. Rose in some special assay pieces found by a less direct
method of assaying, from .06 to .09 per cent. of silver. The proportion
of silver retained varies in a marked way with the proportion of gold to
silver in the alloy before parting. It is generally stated that the
retained silver is least when this proportion is 1 to 2-1/2, and more or
less silver than this leads to a less pure gold after parting.

_Platinum_ in an alloy being parted is dissolved along with the silver
either altogether or in part. It imparts a straw yellow colour to the
parting acid. _Palladium_ gives an orange colour to the acid.

_The loss of gold by solution in the acid during parting_ is small, but
easily demonstrable. On a 500-milligram charge of bullion it may amount
to from .05 to .15 milligram; _i.e._ from .01 to .03 per cent. It is due
to gold actually dissolved and not merely held in suspension.

~Assaying with checks. Surcharge.~--It will be seen from what has been
stated that the errors in gold parting are of two kinds: viz. (1) a loss
of gold on the cupel and to a less extent by solution in the acid, and
(2) an apparent gain of gold due to the retention of silver in the
parted material. Both errors are small, and as they are of an opposite
character they tend to neutralise each other. Hence they are altogether
without effect on the accuracy of the assays of ores when the total gold
is reckoned in milligrams. And even with the larger amounts present in
bullion assays their influence is so small that an uncorrected result is
still fairly accurate; the resultant error would not be more than one
part in two or three thousand.

It is customary to report the purity of bullion, or its fineness as it
is called, in parts per thousand of bullion. The sum of the errors of an
assay, which is called the _surcharge_, is reported in the same way.
Thus a surcharge of + .3 means that the gold as weighed was .3 part per
1000 more than the gold actually present. But a surcharge - .3 means
that on the whole there was a loss of .3 part per 1000 in the assay.

Speaking roughly the retained silver will vary with the weight of gold
present; if one alloy contains twice as much gold as another the
retained silver will be about twice as much also. On the other hand, as
already explained, the cupellation loss on the poorer alloy is as much
as, or even more than, with the richer one, because of the copper, &c.
present. With rich gold alloys the silver more than compensates for the
loss and the surcharge is positive; but with poorer alloys the loss is
greater and the surcharge is negative.

In Mints and places where bullion assays must be made with the highest
attainable accuracy, the surcharge is determined by experiment, and the
proper correction is made in the reports on the bullion. This is done by
making assays of gold of the highest degree of purity alongside of those
of the bullion whose quality has to be determined. These "checks" are so
made that they do not differ from the actual assays in any material
point. Thus, being of the same quality and weight and undergoing exactly
the same treatment, they may reasonably be expected to have the same
surcharge as the assays they imitate. Suppose the bullion being assayed
varies only a little, up or down, from 900 gold and 100 copper in the
thousand, and that .5 gram of it is used in each assay. A quantity of
gold differing only a little from .450 gram would be very exactly
weighed and placed with .050 gram of copper in the same weight of lead
as is being used in the other assays. It would be cupelled, parted, &c.,
as nearly as possible under the same conditions as the actual assays.
Suppose the pure gold weighed .45016 gram and the parted gold weighed
.45025 gram, the gain in weight, .00009 gram, would be deducted from the
actual assays. A surcharge correction is never applied except to bullion
of the same quality as that represented by the "check assay" it was
calculated from.

It is evident that unless the gold is of the highest degree of purity
these check assays will introduce an error almost equal to that which it
is designed to remedy. Moreover, to work the checks to the greatest
advantage, a very systematic and uniform method of working must be
adopted.

~Parting in special apparatus.~--One plan for obtaining greater
uniformity is to stamp each cornet with a number for purposes of
identification, and to treat several, including one or more check assays
in the same acid contained in a beaker; all the assays under these
conditions evidently receive precisely the same acid treatment. Such a
plan can of course only be adopted where there is no risk of the gold
breaking up during the parting. An improvement on this is to have a
porcelain basin[29] about 8-1/2 inches in diameter and with a capacity
of about 1-1/2 litres. It is provided with a porcelain cover with 30
numbered holes through which tubes dip into the acid. The cover is
removable. The tubes are like test-tubes and are supported by the cover;
their bottoms are perforated with holes or slits. The acid is placed in
the basin and boiled over a flat burner; it enters the tubes through the
slits. The cornets are placed each in its proper tube. When the boiling
is finished, the cover with the tubes is lifted and at the same time the
acid drains back into the basin. A dip into a basin of distilled water
washes at one operation all 30 assays. The cover is then put on a basin
containing the stronger parting acid which is already boiling. This
boiling is continued for half an hour. The cover with the 30 cornets is
then lifted out from the acid and dipped two or three times in distilled
water to wash off the last traces of acid. To transfer the cornets from
the tubes to the porous cups the whole of the tube must be dipped under
the water; otherwise the operation is exactly as when working with
test-tubes.

A still simpler method of working is to use small platinum cups[30]
provided with fine slits which admit the acid but retain the gold. A
number of these, say 60, are supported on a platinum tray. The parting
acids are boiled in platinum dishes under a hood; and the 60 cornets
(each in its proper cup) are placed in the acid all at once: the tray
carrying the cups is provided with a handle suitable for this purpose.
After a proper boiling the tray is lifted out of the weaker acid into
the stronger one, where it undergoes the second boiling. It is next
dipped several times in distilled water and lastly, after a gentle
drying, it is raised to an annealing temperature which must not be too
high for fear of the gold sticking to the platinum. After cooling, the
cornets are transferred from the platinum cups directly to the pan of
the balance. Here all 60 cornets have exactly the same treatment and the
"checks" may be compared with great exactness with the other assays
accompanying them. There is, too, a great saving of labour.[31]

~Silver, &c., in gold bullion.~--The base metals are generally
determined by cupelling .5 gram of the alloy with 5 grams of lead. The
loss in cupellation having been allowed for by any of the usual methods
(see p. 104) the gold and silver contents are given. By deducting the
gold the proportion of silver is obtained. The silver is generally
determined by difference in this way. If it is desired to dissolve out
the copper, silver, &c., and to determine them in the wet way, the gold
must first be alloyed with a sufficiency of some other metal to render
it amenable to the attack by acid. Cadmium is the metal generally
recommended, and the alloy is made by melting together a weighed portion
of the gold with five or six times its weight of cadmium in a Berlin
crucible and under a thin layer of potassium cyanide.

~Lead with gold or silver.~--Large quantities of lead carrying gold and
silver are sold to refiners in bars weighing about 100 lbs. each. The
assay of these alloys presents no special difficulties, but the sampling
of them is a question which may be profitably discussed.[32]

A molten metal may be conceived to have all the physical states observed
in ordinary liquids, although these cannot be actually seen owing to its
opaqueness. There is no doubt that _pure_ lead at a temperature only a
little above its melting-point can contain a large proportion of gold in
such a manner that it may in a figurative way be spoken of as a clear
solution. Any small portion withdrawn from the molten metal would afford
a perfect sample. The same would be true of any pure alloy of lead and
silver in which the silver does not exceed the proportion of 2-1/2 per
cent.[33] On the other hand, if the molten metal contains much more than
.5 per cent. of zinc, more than .1 per cent. of copper, or a larger
quantity of silver, it may be likened to a turbid liquor. The
resemblance holds good so far that if the molten lead be further heated,
whereby its solvent power on the added metal is increased, the turbidity
will disappear, or at least be considerably diminished. A portion taken
at random from such a molten metal may, or may not, give a good sample.
The suspended insoluble matter will tend to concentrate itself in the
upper or lower parts of the liquid according to whether it is heavier or
lighter than it; and this separation may occur with extreme slowness or
with fair rapidity. However, it is generally agreed that in the case of
such alloys as occur in practice, samples taken in this way are quite
satisfactory and are the best obtainable. The precautions insisted on
are that the lead shall be made as hot as practicable; that it shall be
stirred up at the time of taking the sample; and that the portion
withdrawn shall be taken out with a ladle at least as hot as the molten
metal. The further precaution that if any dross be on the surface of the
metal it shall be skimmed off and separately sampled and assayed is
almost too obvious to require mention. An alternative and, perhaps,
better way of taking the sample is to withdraw portions at equal
intervals from the stream of metal whilst the pot is being emptied;
equal weights taken from these portions and mixed (by melting or in some
other way) give a fair sample of the whole. In addition, separate assays
of each portion will show to what extent the metal lacks uniformity in
composition For example, samples taken at the beginning, middle, and end
of a run gave the following results in ozs. of silver per ton: 475, 472,
466, showing an average result of 471 ozs. Fifteen fractions taken at
regular intervals during the same pouring ranged from 475 ozs. to 464
ozs.: the average result was 469.8 ozs. The same lead cast into bars and
sampled by sawing gave an average of 470 ozs.[34] In another case[35]
samples drawn at the beginning, middle, and end of a run gave 1345 ozs.,
1335 ozs. and 1331 ozs. The mean result in such cases is always a
reasonably safe one, but evidently where the metal varies a good deal it
is safer to take more than three dips.

Imagine such lead run into moulds and allowed to become solid as bars;
the difference between bar and bar would not be greater than that
between corresponding dip samples. But in each bar the distribution of
the silver and gold is very seriously affected during solidification.
Chips taken from the same bar of auriferous lead may show in one place
23 ozs. of gold to the ton, in another 39 ozs.; similarly with silver
they may vary as much as from 900 ozs. to 1500 ozs. to the ton.

This rearrangement of the constituents of a bar takes place whilst the
lead is partly solid, partly liquid. The most useful conception of such
half-solidified metal is that of a felted spongy mass of skeleton
crystals of comparatively pure lead saturated with a still fluid
enriched alloy. If the solidification of an ingot of impure tin be
watched it will be evident that the frosted appearance of the surface is
due to the withdrawal of the fluid portion from a mat of crystals of
purer tin which have been for some time solid and a contraction of the
mass. The shrinking of the last part to become solid is further shown by
the collapse of the surface of the ingot where weakest; that is, a
furrow is formed on the flat surface. In other cases of fused metal
there is expansion instead of contraction in this final stage of the
solidification, and the enriched alloy then causes the upper face of the
ingot to bulge outwards. There are other causes effecting the
redistribution of the metals through the ingot. There can be no general
rule of wide application showing which part of a bar is richest and
which poorest in the precious metals. This will depend on the quantities
of gold or silver, on the quantities and kinds of other metals present
and on the manner of casting. The student is advised to consult Mr.
Claudet's paper which has been already referred to.

The best method of sampling such bars is to melt them all down and to
take a dip sample of the molten metal in one or other of the methods
already described. According to Mr. Claudet this should be done in all
cases where the gold exceeds one or two ounces or where the silver
exceeds 200 ozs. to the ton. If during the melting down some dross has
formed this must be skimmed off, weighed and separately sampled and
assayed. The clean lead also must be weighed, sampled and assayed. The
mean result must be calculated. Thus 14 tons 5 cwts. of clean lead
assaying 32 ozs. to the ton will contain 456 ozs. of silver; 15 cwt.
dross assaying 20 ozs. to the ton will contain 15 ozs. of silver. The 15
tons of lead and dross will contain 471 ozs. of silver or 31.4 ozs. per
ton.

Of the methods of sampling which avoid melting the bars, that known as
sawing is the only one which is thoroughly satisfactory. In it the bars
are brought to a circular saw having fine teeth and are sawn across
either completely or halfway through; in this way a quantity of lead
sawdust is obtained (say 1 lb. or so from a bar) which represents
exactly the average of the bar along the particular cross section taken
and approximately that of the whole bar. A bar of lead, which by dip
assay gave 334 ozs. to the ton, gave on three transverse sections 333
ozs., 335 ozs. and 331 ozs. The variation may be greater than this, but
with a large number of bars, where each bar is cut across in as far as
possible a different place, these variations tend to neutralise each
other and a good sample is obtained. Two or three cwt. of sawdust may be
obtained in this way; this is thoroughly mixed and reduced by
quartering in the usual way or by a mechanical sampler. A sample of 2 or
3 lbs. is sent to the assayer. This being contaminated with the oil used
in lubricating the saw is freed from it by washing with carbon
bisulphide, ether or benzene and dried. Then, after mixing, 100 to 200
grams of it are carefully weighed and placed in a hot crucible, the heat
of which should be sufficient to melt all the lead. The molten lead
should not be overheated and should show no loss due to the melting. The
removal of the oil may have decreased the weight by perhaps one half per
cent. If the lead gives dross on heating it may be melted under 10 or 20
grams of potassium cyanide, which prevents the formation of dross.
Samples are sometimes taken with a drill, gouge or chisel, though no
method of this kind is quite satisfactory. One plan adopted is to use a
punch which, when driven into the bar, gives a core or rod of metal
about half as long as the bar is thick and about one-eighth of an inch
across. With five bars side by side it is customary to drive in the
punch at one end on the first bar, and at the opposite end on the last
one, and on the others in intermediate positions in such a manner that
all the holes will be along a diagonal of the rectangle enclosing the
bars. The bars are then turned over and similar portions punched out
through the bottoms of the bars and along the other diagonal. Or one set
of five may be sampled along the top and the next set along the bottom
of the bars.

Silver and gold present in bars of copper are subject to the same
irregularity of distribution as in lead. The sampling of such bars is
guided by the same principles.[36]


CYANIDES.

The cyanides ought perhaps to be considered along with chlorides,
bromides and iodides in Chapter XV. But they are treated here because
they owe their importance to their use in the extraction of gold and
because their determination has become a part of the ordinary work of an
assayer of gold ores.

Formerly, the cyanide most easily obtained in commerce was potassium
cyanide; and it was generally sold in cakes which might contain as
little as 40 per cent. or as much as 95 per cent. of the pure salt. It
became customary to express the quality of a sample of commercial
cyanide by saying it contained so much per cent. of potassium cyanide.
The commercial product now made by improved methods of manufacture is
actually sodium cyanide, but is called "potassium cyanide" (probably
with the words "double salt" on the label); it contains cyanide
equivalent to something over 100 per cent. of potassium cyanide in
addition to a large proportion of sodium carbonate and other impurities.
What is wanted in most cases is merely a soluble cyanide, and it is a
matter of indifference whether the base be sodium or potassium. But
since 49 parts of sodium cyanide (NaCN = 49) are equivalent to 65 parts
of potassium cyanide (KCN = 65) it is evident that a pure sample of
sodium cyanide would contain cyanide equivalent to little less than 133
per cent. of potassium cyanide. Therefore a sample of cyanide reported
on in this way may be rich in cyanide, and yet have much impurity.

The commonest impurity in commercial cyanide is carbonate of sodium or
potassium. This may be tested for by dissolving, say, 2 grams in a
little water and adding barium chloride. There may be formed a white
precipitate of barium carbonate, which if filtered off, washed and
treated with acid, will dissolve with effervescence. Cyanate may be
tested for in the solution from which the barium carbonate has been
filtered by adding a little soda and boiling; if cyanates are present
they decompose, giving off ammonia (which may be tested for in the
steam) and yielding a further precipitate of barium carbonate.[37] If
the soda alone gave a further precipitate of barium carbonate, this may,
perhaps, be due to the presence of bicarbonates. Alkaline sulphides may
be present in small quantity in commercial cyanide. Their presence is
shown at once when the sample is being tested for its strength in
cyanide, inasmuch as the first few drops of silver nitrate solution
produce at once a darkening of the liquor. A special test for sulphide
may be made by adding a drop or two of solution of acetate of lead to
four or five c.c. of soda solution and adding this to a clear solution
of the suspected cyanide. This will cause a black precipitate or colour,
if any sulphide is present.

The cyanides of the heavier metals combine with the alkaline cyanides to
form double cyanides. Some of these, ferrocyanide and ferricyanide of
potassium for example, have such characteristic properties that the fact
that they are cyanides may be overlooked. Others, such as potassium zinc
cyanide (K_{2}ZnCy_{4}), have much less distinctiveness: they behave
more or less as a mixture of two cyanides and are, moreover, so easily
decomposed that it may be doubted if they can exist in dilute alkaline
solutions. In reporting the cyanide strength of a cyanide liquor as
equivalent to so much per cent. of potassium cyanide, there is a
question as to whether the cyanide present in the form of any of these
double cyanides should be taken into account. It must be remembered that
the object of the assay is not to learn how much of the cyanide exists
in the solution as actual potassium cyanide; reporting the strength in
terms of this salt is a mere matter of convenience; what is really
desired is to know how much of the cyanide present in the liquor is
"free" or "available" for the purposes of dissolving gold. Every one is
agreed as to the exclusion of such cyanides as the following: potassium
ferrocyanide (K_{4}FeCy_{6}), potassium ferricyanide (K_{3}FeCy_{6}),
potassium silver cyanide (KAgCy_{2}), and potassium aurocyanide
(KAuCy_{2}); and the double cyanides with copper or nickel. But with
cyanide liquors containing zinc the position is less satisfactory. One
method of assay gives a lower proportion of cyanide when this metal is
present; and the loss of available cyanide thus reported depends, though
in a fitful and uncertain way, upon the quantity of zinc present. The
other method of assay reports as full a strength in cyanide as if no
zinc were present. Unfortunately, using both methods and accepting the
difference in the results as a measure of the quantity of zinc present,
or at any rate of the zinc present as cyanide, is not satisfactory. It
appears best to use the method which ignores the zinc; and to determine
the amount of zinc by a special assay of the liquor for this metal.

The cyanide present as hydrogen cyanide or prussic acid (HCy) is
practically useless as a gold solvent. Hence any report on the strength
of a cyanide liquor which assigned to this the same value as its
equivalent of alkaline cyanide would be misleading. On the other hand,
it is "available cyanide" inasmuch as a proper addition of sodium
hydrate[38] would restore its value. The question of the presence or
absence of free prussic acid is involved in the larger one as to whether
the cyanide solution has the right degree of alkalinity. The assay for
"cyanide" should include the hydrogen cyanide with the rest.

A rough test of the power of a cyanide liquor for dissolving gold may be
made by floating a gold leaf on its surface and noting the time required
for its solution. This test might, perhaps, be improved by taking, say,
20 c.c. of the liquor and adding three or four gold leaves so that the
gold shall always be in considerable excess. The liquor should not be
diluted as this will affect the result. It should be allowed to stand
for a definite time, say at least two or three hours, or better, that
corresponding to the time the liquor is left in contact with the ore in
actual practice. The liquor should then be filtered off and, with the
washings, be evaporated in a lead dish as in the assay of cyanide
liquors for gold (p. 141). The gold obtained on cupelling, less any gold
and silver originally present in the liquor, would be the measure of the
gold dissolving power.


THE ASSAY FOR CYANIDE BY TITRATION WITH SILVER NITRATE.

The determination of the quantity of a cyanide is made by finding how
much silver nitrate is required to convert the whole of the cyanide into
potassium silver cyanide[39] or one of the allied compounds. It will be
seen from the equation that 170 parts by weight of silver nitrate are
required for 130 parts by weight of potassium cyanide. As already
explained it is customary to report the cyanide-strength in terms of
potassium cyanide, even when only the sodium salt is present. One gram
of potassium cyanide will require 1.3076 gram of silver nitrate. _The
standard solution of silver nitrate_ is made by dissolving 13.076 grams
of silver nitrate in distilled water and diluting to 1 litre; 100 c.c.
of such a solution are equivalent to 1 gram of potassium cyanide.[40]

The titration is performed in the usual way, running the standard
solution of silver nitrate into a solution containing a known weight or
volume of the material containing the cyanide. The _finishing point_ is
determined in one of two ways, both of which are largely used. In the
first place, as long as there remains any free cyanide in the solution
the silver nitrate will combine with it forming the double cyanide and
yielding a clear solution; but as soon as all the free cyanide is used
up the silver nitrate will react with the double cyanide[41] forming
silver cyanide, which separates as a white precipitate and renders the
solution turbid. But, in the second place, if potassium iodide is
present in the solution the excess of silver nitrate will react with
it,[42] rather than with the double cyanide; and silver iodide will
separate as a yellowish turbidity which is easily recognised.

In working with pure solutions, the two finishing points give the same
results; and this is true even when there is much difference in the
degree of dilution. The finishing point with the iodide, however, has
an advantage in precision. Moreover, it is but little affected by
variations in alkalinity, which render the other finishing point quite
useless. The great difference between the two is shown when zinc is
present in the solution. In this case, when working without the iodide,
the first appearance of a turbidity is less distinct; the turbidity
increases on standing and as a finishing point is unsatisfactory. It can
be determined with precision only by very systematic working and after
some experience. The turbidity is due to the separation of an insoluble
zinc compound. A most important point (to which reference has already
been made) is that less silver nitrate is required to give this
turbidity and, consequently, a lower strength in cyanide is reported. On
the other hand, as much silver nitrate is required to give the yellow
turbidity due to silver iodide as would be required if no zinc were
present.

Unfortunately the difference in the two titrations does not depend
merely on the quantity of zinc present; as it is also influenced by the
extent of dilution, the degree of alkalinity of the solution, and the
quantity of cyanide present. In an experiment with .055 gram of zinc
sulphate and .1 gram of potassium cyanide the difference in the two
finishing points was only .1 c.c.; whereas with .4 gram of potassium
cyanide, the other conditions being the same, the difference was 1.5
c.c. of standard silver nitrate. On the assumption that all the zinc was
present as potassium zinc cyanide (K_{2}ZnCy_{4}) the difference should
have been 5 c.c. in each case. Again, repeating the experiment with .4
gram of potassium cyanide, but with .11 gram of crystallised zinc
sulphate, the difference was 6.5 c.c.: that is, merely doubling the
quantity of zinc increased the difference by more than four times. Hence
it would appear better to use the method with the iodide and make a
separate assay for the zinc. But since the student may be called on to
use the other method, he is advised to practice it also.

~The assay without iodide.~--The standard solution of silver nitrate is
placed in a small burette divided into tenths of a c.c. Ten c.c. of the
cyanide solution to be assayed is transferred to a small flask and
diluted with water to about 70 c.c. The silver solution is then run in
from the burette (with constant shaking of the flask), a little at a
time but somewhat rapidly, until a permanent turbidity appears. Since 1
c.c. of the silver nitrate solution corresponds to .01 gram of potassium
cyanide, it also corresponds to .1 per cent. of this salt counted on the
10 c.c. of cyanide solution taken. The titration should be performed in
a fairly good uniform light. The learner should practice on a fairly
pure solution of potassium cyanide at first, and this may conveniently
have a strength of about 1 per cent. For practice with solutions
containing zinc make a solution containing 1.1 gram of crystallised zinc
sulphate in 100 c.c. and slowly add measured quantities of from 1 to 5
c.c. of this to the 10 c.c. of cyanide liquor before diluting for the
titration.

If a cyanide solution blackens on the addition of the silver nitrate it
contains sulphide. In this case, shake up a considerable bulk of the
liquor with a few grams of lead carbonate, allow to settle and make the
assay on 10 c.c. of the clear liquor.

If the cyanide liquor be suspected to contain free prussic acid, take 10
c.c. for the assay as usual; but, before titrating, add .1 or .2 gram of
sodium carbonate. On no condition must caustic soda or ammonia be added.
The difference between the results, with and without the addition of
carbonate of soda, is supposed to measure the quantity of free prussic
acid. If this has to be reported it is best done as "prussic acid
equivalent to ... per cent. of potassium cyanide." Suppose, for example,
the difference in the two titrations equals 1 c.c. of standard silver
nitrate; the prussic acid found would be equivalent to .1 per cent. of
potassium cyanide.

~The assay with iodide.~--The standard solution of silver nitrate is
placed in a burette divided into tenths of a c.c. Take 10 c.c. of the
cyanide liquor, which should previously have been treated with white
lead for the removal of sulphides if these happened to be present.
Transfer to a small flask, add 3 or 4 drops of a solution of potassium
iodide and 2 or 3 c.c. of a solution of sodium hydrate; dilute to 60 or
70 c.c. with water. If much zinc is present the soda may be increased to
20 or 30 c.c. with advantage. The standard solution should be run in
somewhat rapidly, but a little at a time, so that the precipitate at
first formed shall be small and have only a momentary existence. The
titration is continued until there is a permanent yellowish turbidity.
The most satisfactory and exact finish is got by ignoring any faint
suspicion of a turbidity and accepting the unmistakable turbidity which
the next drop of silver nitrate is sure to produce. This finishing point
gives results which are exactly proportional to the quantity of cyanide
present; and it can be recognised with more than ordinary precision even
in solutions which are not otherwise perfectly clear.

Each c.c. of the standard silver nitrate solution corresponds to .01
gram of potassium cyanide; and if 10 c.c. of the liquor are taken for
assay this corresponds to .1 per cent. or 2 lbs. to the short ton or
2.24 lbs. to the long ton. As already explained the result should be
reported as "cyanide equivalent to so much per cent. of potassium
cyanide."

The following experimental results were obtained with a solution of
potassium cyanide made up to contain about 1.2 per cent. of the salt.

~Effect of varying cyanide.~--The bulk before titration was in each case
60 c.c.; 2 c.c. of soda and 3 drops of potassium iodide were used in
each case.

  Cyanide added   40 c.c.   30 c.c.    20 c.c.   10 c.c.   5 c.c.   1 c.c.
  Silver required 47.0 c.c. 35.25 c.c. 23.5 c.c. 11.7 c.c. 5.8 c.c. 1.15 c.c.

Accepting the result for 40 c.c. as correct, the others are in very
satisfactory agreement.

~Effect of varying dilution.~--The conditions were those of the 40 c.c.
experiment in the last series; but varying amounts of water were used in
diluting.

  Water added        none       100 c.c.   200 c.c.   400 c.c.
  Silver required    47.0 c.c.  47.0 c.c.  47.0 c.c.  47.05 c.c.

Very considerable dilution therefore has no effect.

~Effect of varying soda.~--The conditions were those of the 40 c.c.
experiment in the first series, except that varying amounts of soda
solution were used.

  Soda added        none        10 c.c.   30 c.c.
  Silver required   46.95 c.c.  47.0 c.c. 47.0 c.c.

This alkali therefore has no prejudicial effect.

~Effect of ammonia.~--Soda causes turbidity in some cyanide liquors;
with these it should be replaced by 2 or 3 c.c. of dilute ammonia with a
gram or so of ammonium chloride. The following experiments with dilute
ammonia show that larger quantities of this reagent must be avoided.

  Ammonia added      none      10 c.c.     30 c.c.    60 c.c.
  Silver required  46.95 c.c.  47.15 c.c.  47.7 c.c.  49.5 c.c.

~Effect of sodium bicarbonate.~--In this experiment 1 gram of
bicarbonate of soda was used instead of the soda or ammonia of the other
experiments. The silver nitrate required was only 46.45 c.c. instead of
the 47.0 c.c. which is the normal result. This is probably due to the
liberation of prussic acid and shows the importance of having the
solution alkaline.

~Effect of zinc.~--In each experiment 40 c.c. of the cyanide solution
and .5 gram of zinc sulphate crystals were used and the bulk was made up
to 100 c.c. before titrating.

  Soda added        1 c.c.     5 c.c.    10 c.c.    25 c.c.
  Silver required  47.1 c.c.  47.0 c.c.  46.9 c.c.  46.9 c.c.

The work was easier with the more alkaline solutions. The titration in
the presence of zinc is comparatively easy, but, in learning it, it is
well to have a burette with cyanide so that if a titration be overdone
it can be brought back by the addition of 1 or 2 c.c. more cyanide and
the finish repeated; a quarter of an hour's work in this way will ensure
confidence in the method.

~Effect of other substances.~--It was found that an alkaline cyanate,
sulphocyanate, ferrocyanide, nitrite, borate, silicate or carbonate has
no effect. The ferricyanide had a small influence and, as might be
expected, hyposulphite is fatal to the assay. The addition of salts of
lead and cadmium was without effect. On the other hand, nickel produces
its full effect; and the quantity of nickel added can be calculated with
accuracy from the extent of its interference with the titration.

~Assay of commercial cyanide of potassium.~--Break off 20 or 30 grams of
the cyanide in clean fresh pieces, weigh accurately to the nearest
centigram. Dissolve in water containing a little sodium hydroxide;
transfer to a 2-litre flask: dilute to 2 litres; add a few grams of
white lead; shake up and allow to settle. Run 50 c.c. of the clear
liquor from a burette into an 8 oz. flask; add 2 or 3 c.c. of soda
solution and 3 drops of potassium iodide. Titrate with the standard
solution of silver nitrate. The percentage may be calculated by
multiplying the number of c.c. used by 40 (50 c.c. is one fortieth of
the 2 litres) and dividing by the weight of commercial cyanide
originally taken.

~Alkalinity of commercial potassium cyanide and of cyanide
solutions.~--Hydrocyanic acid like carbonic acid has no action on
methyl-orange;[43] hence the alkaline cyanides may be titrated with
"normal acid" as easily as the carbonates or hydrates. 100 c.c. of
normal acid will neutralise 6.5 grams of pure potassium cyanide.[44] A
solution of commercial cyanide prepared as for the assay last described,
but best without the addition of white lead, may be used for the test.
Take 50 c.c. of it; tint faintly yellow with methyl-orange and titrate
with normal acid till the liquor acquires a permanent reddish tint. In
the case of the purer samples of cyanide the quantity of acid used will
correspond exactly with that required to neutralise the actual quantity
of cyanide present as determined by the assay with nitrate of silver.
The less pure samples will show an excess of alkalinity because of the
presence of sodium carbonate or of potassium carbonate.

In comparing the alkalinity and cyanide strength of a solution the
simplest plan is to take 65 c.c. of the solution and titrate with
normal acid; for in this case each c.c. of normal acid corresponds to .1
per cent. of potassium cyanide. In systematic assays of this kind, the
alkalinity would no doubt be generally in excess of that required by the
cyanide present: there would be no inconvenience in recording such
excess in terms of potassium cyanide.

~Determination of the acidity of an ore.~--Most ores have the power of
destroying more or less of the alkalinity of a cyanide solution and in a
proportionate degree of damaging its efficiency. An assay is needed to
determine how much lime or soda must be added for each ton of ore in
order to counteract this. Whether this acidity should be reported in
terms of the lime or of the soda required to neutralise it will depend
on which of these reagents is to be used in the actual practice. Again,
if the ore is washed with water before treating with cyanide on the
large scale, then the assay should be made of the acidity of the ore
after a similar washing.

The _standard solutions of acid and alkali_ used for this determination
may be one-fifth normal. 200 c.c. of the normal solution should be
diluted to 1 litre in each case, 1 c.c. of the resulting solutions would
be equivalent to 8 milligrams of soda (NaHO) or 5.6 milligrams of lime,
CaO. It must be remembered this refers to the pure bases in each case.
Suppose it is desired to report as so many lbs. of lime to the short ton
(2000 lbs.) of ore. Since 1 c.c. of the standard solution is equivalent
to 5.6 milligrams of lime, if we take 2000 times this weight of ore
(_i.e._ 11,200 milligrams or 11.2 grams) for the assay, each c.c. of
standard solution will be equivalent to 1 lb. of lime to the short
ton.[45]

~Total acidity.~--Weigh out 11.2 grams of the ore, place them in a
four-inch evaporating dish and measure on to it from a burette 10 or 20
c.c. of the standard solution of soda. Stir the soda solution into the
ore and allow to stand for 15 or 20 minutes with occasional stirring.
Stir up with 30 or 40 c.c. of water, float a piece of litmus paper on
the liquid and titrate with the standard solution of acid. If the ore is
strictly neutral the quantity of "acid" required to redden the litmus
will be the same as the quantity of "soda" originally used. If the ore
is acid, less acid will be used. For example, if 10 c.c. of soda were
used and only 7 c.c. of acid were required, the ore will have done the
work of the remaining 3 c.c. of acid. And the ton of ore will require 3
lbs. of lime to neutralise its acidity.

~Acidity after washing.~--Take 11.2 grams of the ore; wash thoroughly
with water and immediately treat the residue, without drying, exactly as
just described.

~Examination of cyanide solutions for metals, &c.~--Take a measured
quantity of the solution, say 20 c.c.[46] and evaporate in a small dish
with, say, half a c.c. of strong sulphuric acid. Evaporate at first, on
a water-bath in a well ventilated place, but finish off with a naked
Bunsen flame, using a high temperature at the end in order to completely
decompose the more refractory double cyanides. Allow to cool; moisten
with strong hydrochloric acid; warm with a little water and test for the
metals in the solution by the ordinary methods. Since the quantities of
the metals likely to be present may be given in milligrams the work must
be carefully performed. It may be worth while to determine the
proportions of lime and magnesia as well as those of the metals proper.

Or the 20 c.c. of cyanide liquor may be evaporated with 5 c.c. of strong
nitric acid to dryness and gently ignited and the residue taken up with
2 or 3 c.c. of strong hydrochloric acid.

Copper, iron, and zinc can be rapidly determined in such a solution, as
follows. Dilute with water to 10 or 15 c.c., add an excess of ammonia,
and filter. The precipitate will contain the iron as ferric hydrate;
dissolve it in a little hot dilute sulphuric acid: reduce with
sulphuretted hydrogen; boil off the excess of gas, cool and titrate with
standard potassium permanganate (p. 236). Determine the copper in the
filtrate colorimetrically (p. 203); but avoid further dilution. Then add
dilute hydrochloric acid, so as to have an excess of 4 or 5 c.c. after
neutralising the ammonia; add some clean strips of lead foil, and boil
until the solution has for some time become colourless. Titrate with
standard potassium ferrocyanide (p. 263) without further dilution, and
bearing in mind that at most only one or two c.c. will be required.

~Examination of an ore for "cyanicides."~--Place 100 grams of the ore
with 200 c.c. of a cyanide solution of known strength (say .1 or .2 per
cent.) in a bottle and agitate for a definite time, such as one or two
days. Filter off some of the liquor and assay for cyanide, using say 20
c.c. Calculate how much cyanide has been destroyed in the operation.
Evaporate 20 c.c. with sulphuric or nitric acid and examine for metal.
Test another portion for sulphides, &c.

The student who has mastered the methods of assaying can greatly improve
himself by working out such problems as the above.


PLATINUM.

Platinum occurs in nature in alluvial deposits associated with gold and
some rare metals, generally in fine metallic grains, and, occasionally,
in nuggets. It is a grey metal with a high specific gravity, 21.5 when
pure and about 18.0 in native specimens. It is fusible only at the
highest temperature, and is not acted on by acids.

It is dissolved by warm aqua regia, forming a solution of "platinic
chloride," H_{2}PtCl_{6}. This substance on evaporation remains as a
brownish red deliquescent mass; on drying at 300° C. it is converted
into platinous chloride, PtCl_{2}, and becomes insoluble, and at a
higher temperature it is converted into platinum. All platinum compounds
yield the metal in this way. Platinic chloride combines with other
chlorides to form double salts, of which the ammonic and potassic
platino-chlorides are the most important.

Platinum alone is not soluble in nitric acid; but when alloyed with
other metals which dissolve in this acid it too is dissolved; so that in
gold parting, for example, if platinum was present, some, or perhaps the
whole of it would go into solution with the silver. Such alloys,
however, when treated with hot sulphuric acid leave the platinum in the
residue with the gold.

Platinum is detected when in the metallic state by its physical
characters and insolubility in acids. In alloys it may be found by
dissolving them in nitric acid or in aqua regia, evaporating with
hydrochloric acid, and treating the filtrate with ammonic chloride and
alcohol. A heavy yellow precipitate marks its presence.

The assay of bullion, or of an alloy containing platinum, may be made as
follows: Take 0.2 gram of the alloy and an equal weight of fine silver,
cupel with sheet lead, and weigh. The loss in weight, after deducting
that of the silver added, gives the weight of the base metals, copper,
lead, &c. Flatten the button and part by boiling with strong sulphuric
acid for several minutes. _When cold_, wash, anneal, and weigh. The
weight is that of the platinum and gold. The silver may be got by
difference. Re-cupel the metal thus got with 12 or 15 times its weight
of silver, flatten and part the gold with nitric acid in the usual way
(see under _Gold_), and the platinum will dissolve. The gold may contain
an alloy of osmium and iridium; if so, it should be weighed and treated
with aqua regia. The osmiridium will remain as an insoluble residue,
which can be separated and weighed. Its weight deducted from that
previously ascertained will give the weight of the gold.

When the platinum only is required, the alloy must be dissolved by
prolonged treatment with aqua regia, the solution evaporated to dryness,
and the residue extracted with water. The solution thus obtained is
treated with ammonic chloride in large excess and with some alcohol. A
sparingly soluble[47] yellow ammonic platinum chloride is thrown down,
mixed, perhaps, with the corresponding salts of other metals of the
platinum group. Gold will be in solution. The solution is allowed to
stand for some time, and then the precipitate is filtered off, washed
with alcohol, dried, and transferred (wrapped in the filter paper) to a
weighed crucible. It is ignited, gently at first, as there is danger of
volatilising some of the platinum chloride, and afterwards intensely.
With large quantities of platinum the ignition should be performed in an
atmosphere of hydrogen. Cool and weigh as metallic platinum.


IRIDIUM

Occurs in nature alloyed with osmium as osmiridium or iridosmine, which
is "rather abundant in the auriferous beach sands of Northern
California" (Dana). It occurs in bright metallic scales, which do not
alloy with lead, and are insoluble in aqua regia. Iridium also occurs in
most platinum ores, and forms as much as two per cent. of some
commercial platinum. In chemical properties it resembles platinum, but
the ammonic irido-chloride has a dark red colour, and on ignition leaves
metallic iridium, which does not dissolve in aqua regia diluted with
four or five times its volume of water and heated to a temperature of
40° or 50° C.

The other metals of the platinum group are Palladium, Rhodium, Osmium,
and Ruthenium. They differ from gold, platinum, and iridium by the
insolubility of their sulphides in a solution of sodium sulphide.
Palladium is distinguished by the insolubility of its iodide; and Osmium
by the volatility of its oxide on boiling with nitric acid.


MERCURY.

Mercury occurs native and, occasionally, alloyed with gold or silver in
natural amalgams; but its chief ore is the sulphide, cinnabar. It is
comparatively rare, being mined for only in a few districts. It is
chiefly used in the extraction of gold and silver from their ores
(amalgamation); for silvering mirrors, &c.

Mercury forms two series of salts, mercurous and mercuric, but for the
purposes of the assayer the most important property is the ease with
which it can be reduced to the metallic state from either of these.
Mercury itself is soluble in nitric acid, forming, when the acid is hot
and strong, mercuric nitrate. Cinnabar is soluble only in aqua regia.
Mercurous salts are generally insoluble, and may be converted into
mercuric salts by prolonged boiling with oxidising agents (nitric acid
or aqua regia). The salts of mercury are volatile, and, if heated with a
reducing agent or some body capable of fixing the acid, metallic mercury
is given off, which may be condensed and collected.

Mercury is separated from its solutions by zinc or copper, or it may be
thrown down by stannous chloride, which, when in excess, gives a grey
powder of metallic mercury, or, if dilute, a white crystalline
precipitate of mercurous chloride. Nitric acid solutions of mercury
yield the metal on electrolysis; and, if the pole on which the metal
comes down be made of gold or copper, or is coated with these, the
separated mercury will adhere thereto. It may then be washed and
weighed.

The best tests for mercury next to obtaining globules of the metal are:
(1) a black precipitate with sulphuretted hydrogen from acid solutions,
which is insoluble in nitric acid; and (2) a white precipitate with
stannous chloride.


DRY METHOD.

[Illustration: FIG. 47.]

Weigh up 5 grams, if the ore is rich, or 10 grams, if a poorer mineral.
Take a piece of combustion tube from 18 inches to 2 feet long, closed at
one end, and place in it some powdered magnesite, so as to fill it to a
depth of 2 or 3 inches, and on that a layer of an equal quantity of
powdered lime (not slaked). Mix the weighed sample of ore in a mortar
with 10 grams of finely powdered lime and transfer to the tube; rinse
out the mortar with a little more lime, and add the rinsings. Cover with
a layer of six or seven inches more lime and a loosely fitting plug of
asbestos. Draw out the tube before the blowpipe to the shape shown in
fig. 47, avoiding the formation of a ridge or hollow at the bend which
might collect the mercury. Tap gently, holding the tube nearly
horizontal, so as to allow sufficient space above the mixture for the
passage of the gases and vapours which are formed. Place the tube in a
"tube furnace," and, when in position, place a small beaker of water so
that it shall just close the opening of the tube. The point of the tube
should not more than touch the surface of the water. Bring the tube
gradually to a red heat, commencing by heating the lime just behind the
asbestos plug, and travelling slowly backwards. When the portion of the
tube containing the ore has been heated to redness for some time the
heat is carried back to the end of the tube. The magnesite readily gives
up carbonic acid, which fills the tube and sweeps the mercury vapour
before it. Some of the mercury will have dropped into the beaker, and
some will remain as drops adhering to the upper part of the neck. Whilst
the tube is still hot cut off the neck of the tube just in front of the
asbestos plug (a drop of water from the wash bottle will do this), and
wash the mercury from the neck into the beaker. The mercury easily
collects into a globule, which must be transferred, after decanting off
the bulk of the water, to a weighed Berlin crucible. The water is
removed from the crucible, first by the help of filter paper, and then
by exposing in a desiccator over sulphuric acid, where it should be left
until its weight remains constant. It should not be warmed.

_Example_:--5 grams of an ore treated in this way gave 4.265 grams of
mercury, equivalent to 85.3 per cent. Pure cinnabar contains 86.2 per
cent.


WET METHODS.

_Solution._--Since solutions of chloride of mercury cannot be boiled
without risk of loss,[48] nitric acid solutions should be used wherever
possible. No mercury-containing minerals are insoluble in acids; but
cinnabar requires aqua regia for solution. In dissolving this mineral
nitric acid should be used, with just as much hydrochloric acid as will
suffice to take it up.

To separate the mercury, pass sulphuretted hydrogen in considerable
excess through the somewhat dilute solution. The precipitate should be
black, although it comes down at first very light coloured. It is
filtered, washed, and transferred back to the beaker, and then digested
with warm ammonic sulphide. The residue, filtered, washed, and boiled
with dilute nitric acid, will, in the absence of much lead, be pure
mercuric sulphide. If much lead is present, a portion may be
precipitated as sulphate, but can be removed by washing with ammonic
acetate. To get the mercury into solution, cover with nitric acid and a
few drops of hydrochloric, and warm till solution is effected. Dilute
with water to 50 or 100 c.c.


GRAVIMETRIC DETERMINATION.

This may be made by _electrolysis_. The same apparatus as is used for
the electrolytic copper assay may be employed, but instead of a cylinder
of platinum one cut out of sheet copper should be taken, or the platinum
one may be coated with an evenly deposited layer of copper. Fix the
spiral and weighed copper cylinder in position, couple up the battery,
_and when this has been done_ put the nitric acid solution of the
mercury in its place.[49] The student had better refer to the
description of the _Electrolytic Copper Assay_.

The mercury comes down readily, and the precipitation is complete in a
few hours: it is better to leave it overnight to make sure of complete
reduction. Disconnect the apparatus, and wash the cylinder, first with
cold water, then with alcohol. Dry by placing in the water oven for two
or three minutes. Cool and weigh: the increase in weight gives the
amount of metallic mercury.

It must be remembered that copper will precipitate mercury without the
aid of the battery; but in this case copper will go into solution with a
consequent loss in the weight of the cylinder: this must be avoided by
connecting the battery before immersing the electrodes in the assay
solution. The electrolysed solution should be treated with an excess of
ammonia, when a blue coloration will indicate copper, in which case the
electrolysis is unsatisfactory. With a little care this need not happen.
Gold cylinders may preferably be used instead of copper; but on platinum
the deposit of mercury is grey and non-adherent, so that it cannot be
washed and weighed.


VOLUMETRIC METHODS.

Several methods have been devised: for the details of these the student
is referred to Sutton's "Handbook of Volumetric Analysis."


QUESTIONS.

1. The specific gravity of mercury is 13.596. What volume would 8 grams
occupy?

2. If 3.169 grams of cinnabar gave 2.718 grams of mercury, what would be
the percentage of the metal in the ore?

3. Pour solution of mercuric chloride on mercury and explain what
happens.

4. On dissolving 0.3 gram of mercury in hot nitric acid, and passing
sulphuretted hydrogen in excess through the diluted solution, what
weight of precipitate will be got?

FOOTNOTES:

[9] Lead may be granulated by heating it to a little above the melting
point, pouring it into a closed wooden box, and rapidly agitating it as
it solidifies.

[10] A rod of iron placed in the crucible with the assays will decompose
any regulus that may be formed.

[11] With buttons poor in silver the lowering of the temperature at this
stage is not a matter of importance.

[12] 100 grams of the lead, or of its oxide, will contain from 1.5 to
2.5 milligrams.

[13] Still the precautions of having cupels well made from bone ash in
fine powder, and of working the cupellation at as low a temperature as
possible are very proper ones, provided they are not carried to an
absurd excess.

[14] Be careful to remove the crucible before taking the bottle out of
the basin of water; if this is not done the chloride may be washed out
of it.

[15] 1 c.c. of this dilute acid will precipitate 8 or 9 milligrams of
silver.

[16] Chlorides interfere not merely by removing silver as insoluble
silver chloride, but also by making it difficult to get a good finishing
point, owing to the silver chloride removing the colour from the
reddened solution.

[17] These results were obtained when using ammonium sulphocyanate, and
cannot be explained by the presence of such impurities as chlorides, &c.

[18] Multiply the _standard_ by 1000, and dilute 100 c.c. of the
standard solution to the resulting number of c.c. Thus, with a solution
of a standard .495, dilute 100 c.c. to 495 c.c., using, of course,
distilled water.

[19] HNa_{2}AsO_{4} + 3AgNO_{3} = Ag_{3}AsO_{4} + HNO_{3} + 2NaNO_{3}.

[20] SiO_{2} + Na_{2}CO_{3} = CO_{2} + Na_{2}SiO_{3}
     SiO_{2} + 2NaHCO_{3} = 2CO_{2} + Na_{2}SiO_{3} + H_{2}O.

[21] PbO + SiO_{2} = PbSiO_{3}

[22] Here and elsewhere in this article when a flux is spoken of as soda
the bicarbonate is meant.

[23] See the description of the process commencing on p. 98 and the
explanatory remarks on p. 110.

[24] Percy, _Metallurgy of Silver and Gold_, p. 258.

[25] "Limits of Accuracy attained in Gold-bullion Assay," _Trans. Chem.
Soc._, 1893.

[26] "Assaying and Hall-marking at the Chester Assay Office." W.F. Lowe.
_Journ. Soc. Chem. Industry_, Sept. 1889.

[27] Fine or pure gold is 24 carat. Nine carat gold therefore contains 9
parts of gold in 24 of the alloy; eighteen carat gold contains 18 parts
of gold in 24; and so on.

[28] The mouth of the flask must not have a rim around it.

[29] See "Assaying and Hall-marking at the Chester Assay Office," by
W.F. Lowe. _Journ. Soc. Chem. Industry_, Sept. 1889.

[30] Percy, _Metallurgy of Silver and Gold_, p. 263.

[31] See also "The Assaying of Gold Bullion," by C. Whitehead and T.
Ulke. _Eng. and Mining Journal_, New York, Feb. 12, 1898.

[32] Consult Percy's _Metallurgy of Silver and Gold_, p. 172; A.C.
Claudet, _Trans. Inst. Mining and Metallurgy_, vol. vi. p. 29; G.M.
Roberts _Trans. Amer. Inst. Mining Engineers_, Buffalo Meeting, 1898; J.
and H.S. Pattinson, _Journ. Soc. Chem. Industry_, vol. xi. p. 321.

[33] Heycock and Neville, _Journ. Chem. Soc._, 1892, p. 907.

[34] G.M. Roberts.

[35] A.C. Claudet.

[36] "The Sampling of Argentiferous and Auriferous Copper," by A.R.
Ledoux. _Journ. Canadian Mining Institute_, 1899.

[37] NaCNO + BaCl_{2} + NaHO + H_{2}O = NH_{3} + BaCO_{3} + 2 NaCl.

[38] HCy + NaHO = NaCy + H_{2}O.

[39] 2KCN + AgNO_{3} = KAg(CN)_{2} + KNO_{3}.

[40] If it be desired to make a solution so that 100 c.c. shall be
equivalent to 1 gram of sodium cyanide, then 18.085 grams of silver
nitrate should be taken for each litre.

[41] AgNO_{3} + KAgCy_{2} = 2 AgCy + KNO_{3}.

[42] AgNO_{3} + KI = AgI + KNO_{3}.

[43] See pp. 322, 323, and 324 for a description of the methods for
measuring the quantity of acid or alkali.

[44] KCN + HCl = KCl + HCN

[45] Taking 16.0 grams of ore, each c.c. = 1 lb. of soda to the short
ton. The corresponding figures for the long ton are 12.544 grams for
lime and 17.92 grams for soda.

[46] In which case each .01 gram of metal found equals 1 lb to the short
ton of solution.

[47] 100 c.c. of water dissolves 0.66 gram of the salt; it is almost
insoluble in alcohol or in solutions of ammonic chloride.

[48] According to Personne mercuric chloride is not volatilised from
boiling solutions when alkaline chlorides are present.

[49] The solution should contain about 0.25 gram of mercury, and a large
excess of nitric acid must be avoided.



CHAPTER X.

COPPER--LEAD--THALLIUM--BISMUTH--ANTIMONY.


COPPER.


Copper occurs native in large quantities, especially in the Lake
Superior district; in this state it is generally pure. More frequently
it is found in combination. The ores of copper may be classed as oxides
and sulphides. The most abundant oxidised ores are the carbonates,
malachite and chessylite; the silicates, as also the red and black
oxides, occur less abundantly. All these yield their copper in solution
on boiling with hydrochloric acid.

The sulphides are more abundant. Copper pyrites (or yellow ore),
erubescite (or purple ore), and chalcocite (or grey ore) are the most
important. Iron pyrites generally carries copper and is frequently
associated with the above-mentioned minerals. These are all attacked by
nitric acid. They nearly all contain a small quantity of organic matter,
and frequently considerable quantities of lead, zinc, silver, gold,
arsenic, bismuth, &c.

The copper ores are often concentrated on the mine before being sent
into the market, either by smelting, when the product is a regulus or
matte, or by a wet method of extraction, yielding cement copper or
precipitate. A regulus is a sulphide of copper and iron, carrying from
30 to 40 per cent. of copper. A precipitate, which is generally in the
form of powder, consists mainly of metallic copper. Either regulus or
precipitate may be readily dissolved in nitric acid.

Copper forms two classes of salts, cuprous and cupric. The former are
pale coloured and of little importance to the assayer. They are easily
and completely converted into cupric by oxidising agents. Cupric
compounds are generally green or blue, and are soluble in ammonia,
forming deep blue solutions.


DRY ASSAY.

That, for copper, next after those for gold and silver, holds a more
important position than any other dry assay. The sale of copper ores has
been regulated almost solely in the past by assays made on the Cornish
method. It is not pretended that this method gives the actual content of
copper, but it gives the purchaser an idea of the quantity and quality
of the metal that can be got by smelting. The process is itself one of
smelting on a small scale. As might be expected, however, the assay
produce and the smelting produce are not the same, there being a smaller
loss of copper in the smelting. The method has worked very well, but
when applied to the purchase of low class ores (from which the whole of
the copper is extracted by wet methods) it is unsatisfactory. The
following table, which embodies the results of several years' experience
with copper assays, shows the loss of copper on ores of varying produce.
The figures in the fourth column show how rapidly the proportion of
copper lost increases as the percentage of copper in the ore falls below
30 per cent. For material with more than 30 per cent. the proportion
lost is in inverse proportion to the copper present.


LOSS OF COPPER.

  -----------------+------------+-----------+------------------
   Copper present. | Dry Assay. |  Margin.  |   Loss on 100
                   |            |           | Parts of Copper.
  -----------------+------------+-----------+------------------
      Per cent.    |  Per cent. | Per cent. |
        100        |    98      |    2.0    |        2.0
         95        |    92-1/2  |    2.5    |        2.6
         90        |    87-3/8  |    2.6    |        2.9
         85        |    82-3/8  |    2.6    |        3.0
         80        |    77-3/8  |    2.6    |        3.2
         75        |    72-3/8  |    2.6    |        3.5
         70        |    67-1/2  |    2.5    |        3.6
         65        |    62-1/2  |    2.5    |        3.8
         60        |    57-5/8  |    2.4    |        4.0
         55        |    52-3/4  |    2.3    |        4.2
         50        |    47-3/4  |    2.2    |        4.4
         45        |    43      |    2.0    |        4.5
         40        |    38-1/8  |    1.8    |        4.6
         35        |    33-1/4  |    1.7    |        4.8
         30        |    28-1/2  |    1.50   |        5.0
         25        |    23-1/2  |    1.50   |        6.0
         20        |    18-1/2  |    1.56   |        7.8
         18        |    16-1/2  |    1.53   |        8.5
         16        |    14-1/2  |    1.48   |        9.3
         14        |    12-5/8  |    1.40   |       10.0
         12        |    10-5/8  |    1.37   |       11.4
         10        |     8-3/4  |    1.28   |       12.8
          8        |     6-7/8  |    1.14   |       14.3
          6        |    5       |    1.05   |       17.5
          5        |    4       |    1.00   |       20.0
          4        |    3       |    1.00   |       25.0
          3.75     |    2-3/4   |    0.97   |       26.0
          3.50     |    2-9/16  |    0.94   |       27.0
          3.25     |    2-5/16  |    0.91   |       28.0
          3.00     |    2-1/8   |    0.87   |       29.0
          2.75     |    1-15/16 |    0.82   |       30.0
          2.50     |    1-3/4   |    0.77   |       31.0
          2.25     |    1-1/2   |    0.72   |       32.0
          2.00     |    1-5/16  |    0.66   |       33.0
  -----------------+------------+-----------+------------------

The wet assay being known, the dry assay can be calculated with the help
of the above table by deducting the amount in the column headed "margin"
opposite the corresponding percentage. For example, if the wet assay
gives a produce of 17.12 per cent., there should be deducted 1.5; the
dry assay would then be 15.62, or, since the fractions are always
expressed in eighths, 15-5/8. With impure ores, containing from 25 to 50
per cent. of copper, the differences may be perhaps 1/4 greater.

Wet methods are gradually replacing the dry assay, and it is probable
that in the future they will supersede it; for stock-taking, and the
various determinations required in smelting works and on mines, they are
generally adopted, because they give the actual copper contents, and
since it is obvious that a knowledge of this is more valuable to the
miner and smelter. Moreover, the working of the dry method has been
monopolised by a small ring of assayers, with the double result of
exciting outside jealousy and, worse still, of retarding the development
and improvement of the process.

The principal stages of the dry assay are: (1) the concentration of the
copper in a regulus; (2) the separation of the sulphur by calcining; (3)
the reduction of the copper by fusion; and (4) the refining of the metal
obtained.

The whole of these operations are not necessary with all copper
material. Ores are worked through all the stages; with mattes, the
preliminary fusion for regulus is omitted; precipitates are simply fused
for coarse copper, and refined; and blister or bar coppers are refined,
or, if very pure, subjected merely to washing.

The quantity of ore generally taken is 400 grains, and is known as "a
full trial"; but for rich material, containing more than 50 per cent. of
copper, "a half trial," or 200 grains, is used.

~Fusion for Regulus.~--The ore (either with or without a previous
imperfect roasting to get rid of any excess of sulphur) is mixed with
borax, glass, lime, and fluor spar; and, in some cases, with nitre, or
iron pyrites, according to the quality of the ore. The mixture is placed
in a large Cornish crucible, and heated as uniformly as possible in the
wind furnace, gradually raising the temperature so as to melt down the
charge in from 15 to 20 minutes. The crucible is removed and its
contents poured into an iron mould. When the slag is solid, it is taken
up with tweezers and quenched in water. The regulus is easily detached
from the slag. It should be convex above and easily broken, have a
reddish brown colour, and contain from 40 to 60 per cent. of copper. A
regulus with more than this is "too fine," and with less "too coarse." A
regulus which is too fine is round, compact, hard, and of a dark bluish
grey on the freshly broken surface. A coarse regulus is flat and coarse
grained, and more nearly resembles sulphide of iron in fracture and
colour.

If an assay yields a regulus "too coarse," a fresh determination is made
with more nitre added, or the roasting is carried further. With low
class ores a somewhat coarse regulus is an advantage. If, on the other
hand, the regulus is too fine, less nitre or less roasting is the
remedy. With grey copper ores and the oxidised ores, iron pyrites is
added.

~Calcining the Regulus.~--It is powdered in an iron mortar and
transferred to a small Cornish crucible, or (if the roasting is to be
done in the muffle) to a roasting dish or scorifier. The calcining is
carried out at a dull red heat, which is gradually increased. The charge
requires constant stirring at first to prevent clotting, but towards the
end it becomes sandy and requires less attention. If the temperature
during calcination has been too low sulphates are formed, which are
again reduced to sulphides in the subsequent fusion. To prevent this the
roasted regulus is recalcined at a higher temperature, after being
rubbed up with a little anthracite. The roasted substance must not smell
of burning sulphur when hot. It is practically a mixture of the oxides
of copper and iron.

~Fusion for Coarse Copper.~--The calcined regulus is mixed with a flux
consisting of borax and carbonate of soda, with more or less tartar
according to its weight. Some "assayers" use both tartar and nitre, the
former of course being in excess. The charge is returned to the crucible
in which it was calcined, and is melted down at a high temperature, and,
as soon as tranquil, poured. When solid it is quenched and the button of
metal separated.

The slag is black and glassy. The small quantity of copper which it
retains is recovered by a subsequent "cleaning," together with the slags
from the next operation.

The button of "coarse copper" obtained must be free from a coating of
regulus. It will vary somewhat in appearance according to the nature and
quantity of the impurities.

~Refining the Coarse Copper.~--The same crucible is put back in the
furnace, deep down and under the crevice between the two bricks. When it
has attained the temperature of the furnace the coarse copper is dropped
into it and the furnace closed. The copper will melt almost at once with
a dull surface, which after a time clears, showing an "eye." Some
refining flux is then shot in from the scoop (fig. 48), and, when the
assay is again fluid, it is poured. When cold the button of metal is
separated.

[Illustration: FIG. 48.]

The button of "fine" copper is flat or pitted on its upper surface, and
is coated with a thin orange film; it must have the appearance of good
copper. If it is covered with a red or purple film, it is overdone or
"burnt." If, on the other hand, it has a rough, dull appearance, it is
not sufficiently refined. Assays that have been "burnt" are rejected.
Those not sufficiently fine are treated as "coarse copper," and again
put through the refining operation.

~Cleaning the Slags.~--These are roughly powdered and re-fused with
tartar, etc., as in the fusion for coarse copper. The button of metal
got is separated (if big enough refined) and weighed.

The details of the process are slightly varied by different assayers:
the following will be good practice for the student.

~Determination of Copper in Copper Pyrites.~--Powder, dry, and weigh up
20 grams of the ore. Mix with 20 grams each of powdered lime and fluor,
15 grams each of powdered glass and borax, and 5 or 10 grams of nitre.
Transfer to a large Cornish crucible and fuse under a loose cover at a
high temperature for from 15 to 20 minutes. When fluid and tranquil pour
into a mould. When the slag has solidified, but whilst still hot, quench
by dipping two or three times in cold water. Avoid leaving it in the
water so long that it does not dry after removal. When cold separate the
button, or perhaps buttons, of regulus by crumbling the slag between the
fingers. See that the slag is free from regulus. It should be light
coloured when cold and very fluid when hot. Reject the slag.

Powder the regulus in a mortar and transfer to a small crucible.
Calcine, with occasional stirring, until no odour of sulphurous oxide
can be detected. Shake back into the mortar, rub up with about 1 gram of
powdered anthracite, and re-calcine for 10 minutes longer.

Mix the calcined regulus with 10 grams of tartar, 20 grams of soda, and
3 grams of borax; and replace in the crucible used for calcining. Fuse
at a bright red heat for 10 or 15 minutes. Pour, when tranquil.

As soon as solid, quench in water, separate the button of copper, and
save the slag.

To refine the copper a very hot fire is wanted, and the fuel should not
be too low down in the furnace. Place the crucible well down in the fire
and in the middle of the furnace. The same crucible is used, or, if a
new one is taken, it must be glazed with a little borax. When the
crucible is at a good red heat, above the fusing point of copper, drop
the button of copper into it, and close the furnace. Watch through the
crevice, and, as soon as the button has melted and appears clear showing
an eye, shoot in 10 grams of refining flux, close the furnace, and, in a
few minutes, pour; then separate the button of copper. Add the slag to
that from the coarse copper fusion, and powder. Mix with 5 grams of
tartar, 0.5 gram of powdered charcoal, and 2 grams of soda. Fuse in the
same crucible, and, when tranquil, pour; quench, and pick out the prills
of metal.

If the copper thus got from the slags is coarse looking and large in
amount, it must be refined; but, if small in quantity, it may be taken
as four-fifths copper. The combined results multiplied by five give the
percentage of copper.

The refining flux is made by mixing 3 parts (by measure) of powdered
nitre, 2-1\2 of tartar, and 1 of salt. Put in a large crucible, and stir
with a red-hot iron until action has ceased. This operation should be
carried out in a well-ventilated spot.

For pure ores in which the copper is present, either as metal or oxide,
and free from sulphur, arsenic, &c., the concentration of the copper in
a regulus may be omitted, and the metal obtained in a pure state by a
single fusion.[50] It is necessary to get a fluid neutral slag with the
addition of as small an amount of flux as possible. The fusion should be
made at a high temperature, so as not to occupy more than from 20 to 25
minutes. Thirty grams of ore is taken for a charge, mixed with 20 grams
of cream of tartar, and 10 grams each of dried borax and soda. If the
gangue of the ore is basic, carrying much oxide of iron or lime, silica
is added, in quantity not exceeding 10 grams. If, on the other hand, the
gangue is mainly quartz, oxide of iron up to 7 grams must be added.

_Example._--Twenty grams of copper pyrites, known to contain 27.6 per
cent. of copper, gave by the method first described 5.22 grams of
copper, equalling 26-1/8 per cent. Another sample of 20 grams of the
same ore, calcined, fused with 40 grams of nitre, and washed to ensure
the removal of arsenic and sulphur, and treated according to the second
method, gave a button weighing 5.27 grams, equalling 26-3/8 per cent.
The ore contained a considerable quantity of lead. Lead renders the
assay more difficult, since after calcination it remains as lead
sulphate, and in the fusion for coarse copper reappears as a regulus on
the button.

~The Estimation of Moisture.~--The Cornish dry assayer very seldom makes
a moisture determination. He dries the samples by placing the papers
containing them on the iron plate of the furnace.

It is well known that by buying the copper contents of pyrites by
Cornish assay, burning off the sulphur, and converting the copper into
precipitate, a large excess is obtained.


NOTES ON THE VALUATION OF COPPER ORES.

Closely bound up with the practice of dry copper assaying is that of
valuing a parcel of copper ore. The methods by which the valuation is
made have been described by Mr. Westmoreland,[51] and are briefly as
follows:--The produce of the parcel is settled by two assayers, one
acting for the buyer, the other for the seller; with the help, in case
of non-agreement, of a third, or referee, whose decision is final. The
dry assayers who do this are in most cases helped, and sometimes,
perhaps, controlled, by wet assays made for one or both of the parties
in the transaction.

In the case of "ticketing," the parcels are purchased by the smelters by
tender, and the value of any particular parcel is calculated from the
average price paid, as follows:--The "standard," or absolute value of
each ton of fine copper in the ore, is the price the smelters have paid
for it, plus the returning charges or cost of smelting the quantity of
ore in which it is contained. The value of any particular parcel of ore
is that of the quantity of fine copper it contains, calculated on this
standard, minus the returning charges. The ton consists of 21 cwts.,
and it is assumed that the "settled" produce is the actual yield of the
ore.

If at a ticketing in Cornwall 985 tons of ore containing 63.3 tons of
fine copper (by dry assay) brought £2591 12s., the standard would be
£83 15s. This is calculated as follows:--The returning charge is fixed
at 55s. per ton of ore. This on 985 tons will amount to £2708 15s.
Add this to the actual price paid, and there is got £5300 as the value
of the fine copper present. The weight of copper in these 985 tons being
63.3 tons, the standard is £5300/63.3, or £83 15s. (nearly).

The value of a parcel of 150 tons of a 6 per cent. ore on the same
standard would be arrived at as follows:--The 150 tons at 6 per cent.
would contain 9 tons (150×6/100) of fine copper. This, at £83 15s.
per ton, would give £753 15s. From this must be deducted the returning
charges on 150 tons of ore at 55s. per ton, or £412 10s. This leaves
£341 5s. as the value of the parcel.

At Swansea the returning charge is less than in Cornwall, and varies
with the quality of the ore. This appears equitable, since in smelting
there are some costs which are dependent simply on the number of tons
treated, and others which increase with the richness. The returning
charge then is made up of two parts, one fixed at so much (12s. 2d.)
per ton of ore treated, and the other so much (3s. 9d.) per unit of
metal in the ore. In this way the returning charge on a ton of ore of
8-3/4 produce would be (12s. 2d.)+(8-3/4×(3s. 9d.)), or £2 5s.

If, for example, Chili bars, containing 96 per cent. of copper, bring
£50 per ton, the standard is £71 9s. 4d. It is got at in this way.
The returning charge on a 96 per cent. ore is (12s. 2d.)+(96×(3s.
9d.)), or £18 12s. 2d. This added to £50 gives £68 12s. 2d.,
and this multiplied by 100 and divided by 96 (100 tons of the bars will
contain 96 tons of fine copper) will give £71 9s. 4d.

The price of 100 tons of pyrites, containing 2-1/4 per cent. of copper
by dry assay, would be got on this standard as follows:--The parcel of
ore would contain 2-1/4 tons of copper. This multiplied by the standard
gives £160 16s. 0d. From this must be deducted the returning charge,
which for 1 ton of ore of this produce would be (12s. 2d.) + (2-1/4×(3s.
9d.)) or £1 0s. 7d., and on the 100 tons is £102 18s. 4d. This would
leave £57 17s. 10d. as the price of the parcel, or 11s. 7d. per ton.
This would be on the standard returning charge of 45s. (for 8-3/4 per
cent. ore); if a smaller returning charge was agreed on, say 38s., the
difference in this case, 7s., would be added to the price per ton.


WET METHODS.

The solubility of the ores of copper in acid has already been described,
but certain furnace products, such as slags, are best opened up by
fusion with fusion mixture and a little nitre.

The method of dissolving varies with the nature of the ore. With 5 grams
of pyrites, a single evaporation with 20 c.c. of nitric acid will give a
residue completely soluble in 30 c.c. of hydrochloric acid. If the ore
carries oxide of iron or similar bodies, these are first dissolved up by
boiling with 20 c.c. of hydrochloric acid, and the residue attacked by
an addition of 5 c.c. of nitric. When silicates decomposable by acid are
present, the solution is evaporated to dryness to render the silica
insoluble; the residue extracted with 30 c.c. of hydrochloric acid, and
diluted with water to 150 c.c. It is advisable to have the copper in
solution as chloride. To separate the copper, heat the solution nearly
to boiling (best in a pint flask), and pass a rapid current of
sulphuretted hydrogen for four or five minutes until the precipitate
settles readily and the liquid smells of the gas. When iron is present
it will be reduced to the ferrous state before the copper sulphide
begins to separate. The copper appears as a brown coloration or black
precipitate according to the quantity present. Filter through a coarse
filter, wash with hot water containing sulphuretted hydrogen, if
necessary. Wash the precipitate back into the flask, boil with 10 c.c.
of nitric acid, add soda till alkaline, and pass sulphuretted hydrogen
again. Warm and filter, wash and redissolve in nitric acid, neutralise
with ammonia, add ammonic carbonate, boil and filter. The copper freed
from impurities will be in the solution. Acidulate and reprecipitate
with sulphuretted hydrogen. When the nature of the impurities will allow
it, this process may be shortened to first filtering off the gangue,
then precipitating with sulphuretted hydrogen and washing the
precipitate on the filter first with water and then with ammonium
sulphide.

Having separated the copper as sulphide, its weight is determined as
follows. Dry and transfer to a weighed porcelain crucible, mix with a
little pure sulphur, and ignite at a red heat for 5 or 10 minutes in a
current of hydrogen. Allow to cool while the hydrogen is still passing.
Weigh. The subsulphide of copper thus obtained contains 79.85 per cent.
of copper; it is a greyish-black crystalline mass, which loses no weight
on ignition if air is excluded.

Copper may be separated from its solutions by means of sodium
hyposulphite. The solution is freed from hydrochloric and nitric acids
by evaporation with sulphuric acid; diluted to about a quarter of a
litre; heated nearly to boiling; and treated with a hot solution of
sodium hyposulphite (added a little at a time) until the precipitate
settles and leaves the solution free from colour. The solution contains
suspended sulphur. The precipitate is easily washed, and under the
proper conditions the separation is complete, but the separation with
sulphuretted hydrogen is more satisfactory, since the conditions as to
acidity, &c., need not be so exact.

Zinc or iron is sometimes used for separating copper from its solutions,
but they are not to be recommended.


ELECTROLYTIC ASSAY.

The separation of copper by means of a current of electricity is largely
made use of, and forms the basis of the most satisfactory method for the
determination of this metal. If the wire closing an electric circuit be
broken, and the two ends immersed in a beaker of acidulated water or
solution of any salt, the electricity will pass through the liquid,
bringing about some remarkable changes. Hydrogen and the metals will be
liberated around that part of the wire connected with the zinc end of
the battery, and oxygen, chlorine, and the acid radicals will be set
free around the other. Different metals are deposited in this way with
varying degrees of ease, and whether or not any particular metal will be
deposited depends--(1) on the conditions of the solution as regards acid
and other substances present, and (2) on the _intensity_ of the current
of electricity used. For analytical purposes the metal should be
deposited not only free from the other metals present, but also as a
firm coherent film, which may afterwards be manipulated without fear of
loss. This is, in the case of copper and many other metals, effected by
a simple control of the conditions. It is necessary that the electrodes,
or wires which bring the electricity into the solution, should be made
of a material to which the deposited metal will adhere, and which will
not be attacked by substances originally present or set free in the
solution. They are generally made of platinum. There are various
arrangements of apparatus used for this purpose, but the following plan
and method of working is simple and effective, and has been in daily use
with very satisfactory results for the last five or six years.

The battery used is made up of two Daniell cells, coupled up for
intensity as shown in fig. 49--that is, with the copper of one connected
with the zinc of the other. For eight or ten assays daily the quart size
should be used, but for four or five two pint cells will be sufficient.

[Illustration: FIG. 49.]

The outer pot of each cell is made of sheet copper, and must be clean
and free from solder on the inside. It is provided near the top with a
perforated copper shelf in the shape of a ring, into which the inner or
porous cell loosely fits. It is charged with a saturated solution of
copper sulphate, and crystals of this salt must be added, and always
kept in excess. When the battery is at work copper is being deposited on
the inner surface of this pot.

The inner or porous pot contains the zinc rod, and is charged with a
dilute acid, made by diluting one volume of sulphuric acid up to ten
with water. The object of the porous pot is to prevent the mixing of the
acid and copper sulphate solutions, without interrupting the flow of
electricity. The copper sulphate solution will last for months, but the
acid must be emptied out and recharged daily.

The zinc rods must be well amalgamated by rubbing with mercury under
dilute acid until they show a uniformly bright surface. They should not
produce a brisk effervescence when placed in the acid in the porous pot
before coupling up.

The battery when working is apt to become dirty from the "creeping" of
the copper and zinc sulphate solution. It must be kept away from the
working bench, and is best kept in a box on the floor.

[Illustration: FIG. 50.]

The connection of the battery with, and the fixing of, the electrodes
may be made by any suitable arrangement, but the following is a very
convenient plan. The wire from the zinc is connected by means of a
binding screw with a piece of stout copper wire, which, at a distance
sufficiently great to allow of easy coupling with the battery, is led
along the back of a piece of hard wood. This is fixed horizontally
about one foot above the working bench. The general arrangement is shown
in fig. 50, in which, however, for the sake of economy of space, the
battery is placed on the working bench instead of on the floor. The
piece of wood is one inch square and three or four feet long. It is
perforated from front to back at distances of six inches by a number of
small holes, in which are inserted screws like that shown in fig. 51.
These are known as "terminals," and may be obtained of any electrician.
The head of each screw is soldered to the wire mentioned above as
running along the back and as being connected with the zinc end of the
battery. These terminals serve to fix the electrodes on which the copper
is to be deposited. The wire from the copper end of the battery is
similarly connected by a connecting screw (fig. 52) with another wire (H
in fig. 53), which runs along the top of the rod and has soldered to it,
at distances of six inches, cylindrical spirals of copper wire. These
should project from the rod at points about half-way between the
terminals already described. They may be made by wrapping copper wire
around a black-lead pencil for a length of about three inches.

[Illustration: FIG. 51.]

[Illustration: FIG. 52.]

[Illustration: FIG. 53.]

The rod is perforated from top to bottom with a series of small holes,
one in advance of each terminal but as near it as possible. Into these
short pieces of glass tube are inserted to ensure insulation. These
receive the other electrodes, which are connected with the wire leading
to the copper end of the battery, through the spirals, with the help of
a binding screw. The figure will make this clear. (Fig. 53.)

[Illustration: FIG. 54]

~The electrodes~ consist of a platinum spiral and cylinder. The spiral
should have the shape shown in A, fig. 54. When in work it is passed
through one of the holes fitted with glass tubes and connected with the
copper end of the battery. The thickness of the wire of which it is made
is unimportant, provided it is stout enough to keep its form and does
not easily bend. The spiral will weigh about 8 grams. The cylinder (C,
fig. 54) will weigh about 12 grams. It should have the shape shown in
the figure. In working it is clamped to one of the terminals, and on it
the copper is deposited. A cylinder will serve for the deposition of
from 1 to 1.5 gram of copper. It is made by rivetting a square piece of
foil on to a stiff piece of wire, and then bending into shape over a
glass tube or piece of rounded wood. Each cylinder carries a distinctive
number, and is marked by impressing Roman numerals on the foil with the
blade of a knife. The weight of each is carefully taken and recorded.
They lose slightly in weight when in use, but the loss is uniform, and
averages half a milligram per month when in daily use. The cylinders are
cleaned from deposited copper by dissolving off with nitric acid and
washing with water; and from grease by igniting.

The ~beakers~, to contain the solution of copper to be electrolysed, are
ordinary tall beakers of about 200 c.c. capacity, and are marked off at
100 c.c. and 150 c.c. They are supported on movable stands, consisting
of wooden blocks about six inches high and three inches across. The bar
of wood which carries the connecting wires and electrodes is permanently
fixed over the working bench, at such a height that, with the beakers
resting on these blocks, the electrodes shall be in position for
working.

To fix the electrodes to the rod, remove the stand and beaker and pass
the long limb of the spiral up through one of the glass tubes. Connect
it with the free end of the copper spiral by means of a connecting screw
(fig. 52), and then draw out and bend the copper spiral so that the
platinum one may hang freely. Screw the wire of the cylinder to the
terminal, and, if necessary, bend it so that the cylinder itself may be
brought to encircle the rod of the spiral in the manner shown in fig.
53.

The ~general method of working~ is as follows:--The quantity of ore to
be taken for an assay varies with the richness of the ore, as is shown
in the following table:--

  Percentage of Copper     Quantity of Ore
      in the Ore.           to be taken.

       1 to   5               5 grams
       5 to  10               3   "
      10 to  30               2   "
      30 to  50               1.5 "
      50 to 100               1   "

The weighed quantity of ore is dissolved by evaporating with nitric acid
and taking up with hydrochloric, as already described. Any coloured
residue which may be left is generally organic matter: it is filtered
off, calcined, and any copper it contains is estimated colorimetrically.
Nearly always, however, the residue is white and sandy. The copper is
separated from the solution as sulphide by means of a rapid current of
sulphuretted hydrogen. The liquid is decanted off through a filter, the
precipitate washed once with hot water and then rinsed back into the
flask (the filter paper being opened out) with a jet of water from a
wash bottle. Fifteen c.c. of nitric acid are added to the contents of
the flask, which are then briskly boiled until the bulk is reduced to
less than 10 c.c. The boiling down is carried out in a cupboard free
from cold draughts, so as to prevent the condensation of acid and steam
in the neck of the flask. Twenty c.c. of water are next added, and the
solution is warmed, and filtered into one of the beakers for
electrolysis. The filtrate and washings are diluted with water to the
100 c.c. mark, and the solution is then ready for the battery. It must
not contain more than 10 per cent. by volume of nitric acid.

The number and weight of the platinum cylinder having been recorded,
both electrodes are fixed in position and the wooden block removed from
under them. The beaker containing the copper solution is then brought up
into its place with one hand, and the block replaced with the other so
as to support it. All the assays having been got into position, the
connecting wires are joined to the battery. If everything is right
bubbles of oxygen at once stream off from the spiral, and the cylinder
becomes tarnished by a deposit of copper. If the oxygen comes off but no
copper is deposited, it is because the assay solution contains too much
nitric acid. If no action whatever takes place, it is because the
current is not passing. In this case examine the connections to see that
they are clean and secure, and the connecting wires to see that they are
not touching each other.

The action is allowed to go on for sixteen or seventeen hours, so that
it is best to let the current act overnight. In the morning the
solutions will appear colourless, and a slow stream of oxygen will still
be coming off from the spiral.

A wash-bottle with cold distilled water and two beakers, one with
distilled water and the other with alcohol, are got ready. The block is
then removed, the spiral loosened and lowered with the beaker. The
cylinder is next detached and washed with a stream of water from the
wash-bottle, the washings being added to the original solution. The
current from the battery is not stopped until all the cylinders are
washed. After being dipped in the beaker of water and once or twice in
that with the alcohol, it is dried in the water-oven for about three
minutes, and then weighed. The increase in weight is due to deposited
copper. This should be salmon-red in colour, satin-like or crystalline
in appearance, and in an even coherent deposit, not removed by rubbing.
It is permanent in air when dry, but sulphuretted hydrogen quickly
tarnishes it, producing coloured films. With ores containing even very
small proportions of bismuth, the deposited copper has a dark grey
colour, and when much of this metal is present the copper is coated with
a grey shaggy deposit.

It still remains to determine any copper left undeposited in the
solution. This does not generally exceed four or five milligrams, and
is estimated colorimetrically. Thirty c.c. of dilute ammonia (one of
strong ammonia mixed with one of water) are added to the electrolysed
solution, which is then diluted up to the 150 c.c. mark with water. It
is mixed, using the spiral as stirrer, and, after standing a few minutes
to allow the precipitate to settle, 100 c.c. of it are filtered off
through a dry filter for the colorimetric determination. Since only
two-thirds of the solution are taken for this, the quantity of copper
found must be increased by one-half to get the quantity actually
present.

[Illustration: FIG. 55.]

The ~colorimetric determination~ may be made in the manner described
under that head, but where a number of assays are being carried out it
is more convenient to have a series of standard phials containing known
amounts of copper in ammoniacal solution. By comparing the measured
volume of the assay solution with these, the amount of copper present is
determined at a glance. These standard bottles, however, can only be
economically used where a large number of assays are being made daily.

A convenient plan is to get a quantity of white glass four-ounce phials,
like that in fig. 55, and to label them so that they shall contain 100
c.c. when filled up to the bottom of the labels. The labels should be
rendered permanent by coating with wax, and be marked with numbers
indicating the milligrams of copper present. The bottles are stopped
with new clean corks, and contain, in addition to the specified quantity
of copper, 6 c.c. of nitric acid and 10 c.c. of strong ammonia, with
sufficient water to make up the bulk to 100 c.c. The copper is best
added by running in the requisite amount of a standard solution of
copper, each c.c. of which contains 0.001 gram of the metal.

The standard bottles should be refilled once every three or four months,
since their colorimetric value becomes slowly less on keeping. The
following determinations of a set which had been in use for three months
will illustrate this. The figures indicate milligrams of copper in 100
c.c.: the first row gives the nominal and the second row the actual
colorimetric value of the standards. The difference between the two
shows the deterioration.

  1  2  3  4    6    8   10  12  14
  1  2  3  3.7  5.5  7.5  9  11  13

The amount of copper in the assay is got by increasing that found
colorimetrically by one-half and adding to that found on the platinum
cylinder. The percentage is calculated in the usual way. The following
examples will illustrate this, as well as the method of recording the
work in the laboratory book:--

  ---------------------------------------------
  Cylinder I. + Cu                       9.5410
  Cylinder I.                            9.5170
                                         ------
                                         0.0240
  By colour 100 c.c. = 0.0015}
                       0.0007}           0.0022
                       ------            ------
                       0.0022            0.0262
  IX. Sample.    Took 5 grams.
                                Copper = 0.52%
  ---------------------------------------------
  Cylinder VI. + Cu                     10.5705
  Cylinder VI.                          10.0437
                                        -------
                                         0.5268
  By colour, 100 c.c. = 0.0070}
                        0.0035}          0.0105
                        ------           ------
                        0.0105           0.5373
  Matte, No. 1070. Took 1.5 gram.
                               Copper = 35.82%
  ---------------------------------------------
  Cylinder XIII. + Cu                   12.0352
  Cylinder XIII.                        11.0405
                                        -------
                                         0.9947
  By colour 100 c.c. = 0.0005}
                       0.0002}           0.0007
                       ------            ------
                       0.0007            0.9954
  X. Sample, Cake copper. Took 1.0053 gram.
                               Copper = 99.00%
  ---------------------------------------------

In the electrolytic assay of metals, alloys, precipitates, and other
bodies rich in copper, the preliminary separation of the copper by
sulphuretted hydrogen is unnecessary. It is sufficient to dissolve the
weighed sample in 10 c.c. of nitric acid, boil off nitrous fumes, dilute
to 100 c.c. with water, and then electrolyse.

~General Considerations.~--In the preliminary work with the copper
sulphide there is a small loss owing to its imperfect removal in washing
the filter paper, and another small loss in dissolving in nitric acid
owing to the retention of particles in the fused globules of sulphur. To
determine its amount the filter-papers and sulphur were collected from
forty assays, and the copper in them determined. The average amount of
copper in each assay was 0.175 gram; that left on the filter paper was
0.00067 gram; and that retained by the sulphur 0.00003 gram; thus
showing an average loss from both sources of 0.00070 gram. The
determinations from another lot of forty-two similar assays gave on an
average

  Copper left on filter paper      0.00070 gram
  Copper retained by sulphur.      0.00004  "

The loss from these sources is trifling, and need only be considered
when great accuracy is required.

The deposition of the copper under the conditions given is satisfactory,
but, as already stated, if the solution contain more than 10 per cent.
of nitric acid it is not thrown down at all; or if a stronger current is
used, say that from three Bunsen cells, it will be precipitated in an
arborescent brittle form, ill adapted for weighing. It may be noted here
that increasing the size of the cells does not necessarily increase the
intensity of the current.

In two determinations on pure electrotype copper the following results
were obtained:--

  Copper Taken.       Copper Found.
   0.8988 gram         0.8985 gram
   0.8305  "           0.8303  "

The presence of salts of ammonia, &c., somewhat retards the deposition,
but has no other ill effect.

The organic matter generally present in copper ores interferes more
especially in the colorimetric determination of the residual copper. It
can be detected on dissolving the ore as a light black residue insoluble
in nitric acid. It is filtered off at once, or, if only present in small
amount, it is carried on in the ordinary process of the assay and
separated in the last filtration before electrolysis.

The following experiments were made to test the effect of the presence
of salts of foreign metals in the solution during the precipitation of
copper by electrolysis:--

  --------------+----------------------------------------+---------------
  Copper Taken. |           Other Metal Added.           | Copper Found.
  --------------+----------------------------------------+---------------
   0.1000 gram  | 0.1000 gram of silver                  |    0.1800
   0.1050   "   | 0.1000   "       "                     |    0.2000
   0.1030   "   | 0.1000   "     mercury                 |    0.2010
   0.1037   "   | 0.1000   "       "                     |    0.2015
   0.1020   "   | 0.1000   "     lead                    |    0.1020
   0.1030   "   | 0.1000   "       "                     |    0.1028
   0.1010   "   | 0.1000   "     arsenic                 |    0.1010
   0.1007   "   | 0.1000   "       "                     |    0.1022
   0.1030   "   | 0.1000   "     antimony                |    0.1050
   0.1034   "   | 0.1000   "       "                     |    0.1057
   0.0990   "   | 0.1200   "     tin                     |    0.0990
   0.1014   "   | 0.1000   "       "                     |    0.1015
   0.1000   "   | 0.1000   "     bismuth                 |    0.1662
   0.1040   "   | 0.1000   "     cadmium                 |    0.1052
   0.1009   "   | 0.1300   "    zinc                     |    0.1017
   0.1014   "   | 0.1000   "    nickel                   |    0.1007
   0.1079   "   | 0.1200   "    iron                     |    0.1089
   0.1054   "   | 0.1000   "    chromium (Cr_{2}O_{3})   |    0.1035
   0.1034   "   | 0.1000   "        "    (K_{2}CrO_{4})  |    0.1010
   0.1075   "   | 0.1000   "    aluminium                |    0.1078
   0.1010   "   | 0.1000   "    manganese                |    0.0980
  --------------+----------------------------------------+---------------

It will be seen from these that mercury, silver, and bismuth are the
only metals which are precipitable[52] along with the copper under the
conditions of the assay. Mercury, which if present would interfere, is
separated because of the insolubility of its sulphide in nitric acid.

Bismuth is precipitated only after the main portion of the copper is
thrown down. It renders the copper obviously unsuitable for weighing. It
darkens, or forms a greyish coating on, the copper; and this darkening
is a delicate test for bismuth. In assaying ores containing about three
and a half per cent. of copper, and known to contain bismuth in
quantities scarcely detectable in ordinary analysis, the metal deposited
was distinctly greyish in colour, and would not be mistaken for pure
copper. Ten grams of this impure copper were collected and analysed,
with the following results:--

  Copper       99.46 per cent.
  Bismuth      00.30    "
  Iron         00.14    "
  Arsenic      00.10    "
              ------
              100.00

The quantity of copper got in each assay was 0.175 gram, and
consequently the bismuth averaged 0.00053 gram.

To separate the bismuth in such a case the deposit is dissolved off by
warming it in the original solution. The bismuth is precipitated by the
addition of ammonic carbonate, and the solution, after filtering and
acidifying with nitric acid, is re-electrolysed.

~Determination of Copper in Commercial Copper.~--Take from 1 to 1.5
gram, weigh carefully, and transfer to a beaker; add 20 c.c. of water
and 10 c.c. of nitric acid; cover with a clock glass, and allow to
dissolve with moderate action; boil off nitrous fumes, dilute to 100
c.c., and electrolyse. The cylinder must be carefully weighed, and the
electrolysis allowed to proceed for 24 hours. The weight found will be
that of the copper and silver. The silver in it must be determined[53]
and deducted.

~Determination of Copper in Brass, German Silver, or Bronze.~--Treat in
the same manner as commercial copper. If nickel is present, the few
milligrams of copper remaining in the electrolysed solution should be
separated with sulphuretted hydrogen, the precipitated sulphide
dissolved in nitric acid, and determined colorimetrically.


VOLUMETRIC PROCESSES.

There are two of these in use, one based on the decolorising effect of
potassic cyanide upon an ammoniacal copper solution, and the other upon
the measurement of the quantity of iodine liberated from potassic iodide
by the copper salt. The cyanide process is the more generally used, and
when carefully worked, "on certain understood and orthodox conditions,"
yields good results; but probably there is no method of assaying where a
slight deviation from these conditions so surely leads to error. An
operator has no difficulty in getting concordant results with duplicate
assays; yet different assayers, working, without bias, on the same
material, get results uniformly higher or lower; a difference evidently
due to variations in the mode of working. Where a large number of
results are wanted quickly it is a very convenient method. The iodide
process is very satisfactory when worked under the proper conditions.


CYANIDE METHOD.

The process is based upon the facts--(1) that when ammonia is added in
excess to a solution containing cupric salts, ammoniacal copper
compounds are formed which give to the solution a deep blue colour; and
(2) that when potassic cyanide is added in sufficient quantity to such a
solution the colour is removed, double cyanides of copper and potassium
or ammonium being formed.[54] In the explanation generally given the
formation of cuprous cyanide is supposed[55]; but in practice it is
found that one part of copper requires rather more than four parts of
cyanide, which agrees with the former, rather than the latter,
explanation.

Reliance on the accuracy of the process cannot rest upon the
supposition that the cyanide required for decoloration is proportional
to the copper present, for varying quantities of ammonia salts, ammonia
and water, and differences of temperature have an important effect. The
results are concordant and exact only when the cyanide is standardised
under the same conditions as it is used. It is best to have the assay
solution and that used for standardising as nearly as possible alike,
and to titrate the two solutions side by side. This demands an
approximate knowledge of the quantity of copper contained in the ore and
a separation of the bulk of the impurities.

For the titration there is required a standard solution of potassium
cyanide made by dissolving 42 grams of the salt, known to dealers as
Potassium Cyanide (Gold), in water and diluting to one litre: 100 c.c.
of this will be about equivalent to one gram of copper. For poor ores
the solution may conveniently be made half this strength.

The solution of the ore and the separation of the copper as sulphide are
effected in the same ways as have been already described for
electrolysis. Similarly, too, the sulphide is attacked with 15 c.c. of
nitric acid and the assay boiled down to 10 c.c. Add 20 c.c. of water
and warm, filter into a pint flask, wash well with water, and dilute to
about 150 c.c.; add 30 c.c. of dilute ammonia, and cool.

Prepare a standard by dissolving a quantity of electrotype copper
(judged to be about the same as that contained in the assay) in 20 c.c.
of water and 10 c.c. of nitric acid, boil off the nitrous fumes, and
dilute to 150 c.c.: add 30 c.c. of dilute ammonia and cool.

Fill a burette with the standard cyanide solution. The burette with
syphon arrangement, figured on page 52, is used. A number of titrations
can be carried on at the same time provided the quantity of copper
present in each is about the same. This is regulated in weighing up the
ore. The flasks must of course be marked, and should be arranged in
series on a bench in front of a good light and at such a height that the
liquid can be looked through without stooping. Supposing about 50 c.c.
of cyanide will be required, 30 c.c. should be run into each, and each
addition be recorded as soon as made; then run 15 c.c. into each. The
solutions will now probably show marked differences of tint: add 1 c.c.
of cyanide to the lighter ones and more to the darker, so as to bring
the colours to about the same depth of tint. They should all be of
nearly equal tint just before finishing. At the end add half a c.c. at a
time until the colours are completely discharged. A piece of damp filter
paper held between the light and the flask assists in judging the
colour when nearly finished. Overdone assays show a straw yellow colour
which deepens on standing.

The following will illustrate the notes recorded of five such assays and
one standard:--

  (1) 30 c.c. 15 c.c. 5 c.c. 2 c.c. 1 c.c. 1/2 c.c. -- c.c. = 53-1/2 c.c.
  (2) 30  "   15  "   1  "   1  "   1  "   1/2  "   --  "   = 48-1/2  "
  (3) 30  "   15  "   3  "   1  "   1  "   1/2  "   --  "   = 50-1/2  "
  (4) 30  "   15  "   5  "   2  "   1  "   1/2  "  1/2  "   = 54      "
  (5) 30  "   15  "   2  "   1  "   1  "   1/2  "   --  "   = 49-1/2  "
  (6) 30  "   15  "   2  "   1  "   1  "   1/2  "  1/2  "   = 50  standard

Three grams of ore were taken, and the standard contained 0.480 gram of
copper.

In this series the difference of half a c.c. means about 0.15 per cent.
on the ore; with a little practice it is easy to estimate whether the
whole or half of the last addition should be counted.

To get satisfactory results, the manner of finishing once adopted must
be adhered to.

The following experiments show the effect of variation in the conditions
of the assay:--Use _a solution of copper nitrate_, made by dissolving 10
grams of copper in 50 c.c. of water and 35 c.c. of nitric acid, and
diluting to a litre. 100 c.c. = 1 gram of copper.

~Effect of Varying Temperature.~--In these experiments 20 c.c. of copper
nitrate were used, with 10 c.c. of nitric acid, 30 c.c. of dilute
ammonia, and water to 200 c.c. The results were--

  Temperature         15°        30°        70°       100°
  Cyanide required  21.5 c.c.  20.8 c.c.  19.7 c.c.  18.8 c.c.

The temperature is that of the solution _before_ titrating. These show
the importance of always cooling before titrating, and of titrating the
assay and standard at the same temperature.

~Effect of Varying Bulk.~--The quantities of copper, acid, and ammonia
were the same as in the last-mentioned experiments. The results were:--

  Bulk             100.0 c.c.  200.0 c.c.  300.0 c.c.  400.0 c.c.
  Cyanide required  23.3  "     21.7  "     21.4  "     21.4  "

These show that large variations in bulk must be avoided.

~Effect of Varying Ammonia.~--The quantities of copper and acid were the
same as in the series of experiments last noticed. The bulk was 200 c.c.
The results were:--

  Dilute ammonia    20.0 c.c.  30.0 c.c.  50.0 c.c.  100.0 c.c.
  Cyanide required  20.9  "    21.7  "    22.3  "     24.6  "

~Effect of Varying Acid.~--The quantities of copper and water were the
same as in the last-noticed set of experiments: 30 c.c. of dilute
ammonia were used.

  Nitric acid        5.0 c.c.  10.0 c.c.  15.0 c.c.
  Cyanide required  21.6  "    21.7  "    21.5 "

On adding nitric acid to the solution it combines with a portion of the
ammonia to form ammonic nitrate; it will be seen from the last series of
experiments that the lessening of the amount of free ammonia will
decrease the quantity of cyanide required; but, on the other hand, the
ammonic nitrate which is at the same time formed will increase the
amount required; under the conditions of the assay these two effects
neutralise each other, and such differences in the quantity of acid as
are likely to occur are unimportant.

~Effect of Varying Ammonic Salts.~--The quantities of copper, water, and
ammonia were the same as in the last mentioned set of experiments, but
no nitric acid was used.

  Ammonic nitrate added  1 gram     5 grams   10 grams   20 grams
  Cyanide required      21.2 c.c.  22.1 c.c.  23.1 c.c.  24.1 c.c.

These show that combined ammonia seriously affects the titration, and
that the principle sometimes recommended of neutralising the acid with
ammonia, and then adding a constant quantity of ammonia, is not a good
one, because there is then an interference both by the ammonia and by
the variable quantity of ammonic salts.

The same quantity of combined ammonia has the same effect, whether it is
present as sulphate, nitrate, chloride, or acetate, as the following
experiments show. Four lots of 20 c.c. of "copper nitrate" were taken,
and 20 c.c. of dilute ammonia added to each. These were carefully
neutralised with the respective acids, rendered alkaline with 30 c.c.
more of ammonia, cooled, diluted to bulk, and titrated. The results
were:--

  With sulphuric acid     22.5 c.c. of cyanide
    "  nitric acid        22.6  "        "
    "  hydrochloric acid  22.6  "        "
    "  acetic acid        22.5  "        "

~Effect of Foreign Salts.~--Sulphates, nitrates and chlorides of sodium
or potassium have no action, whilst the hydrates, carbonates,
bicarbonates, sulphites, and nitrites have an important effect. The
interference of ammonic salts has already been shown.

Salts of silver, zinc, and nickel react with cyanide just as copper
does, and consequently interfere. Ferrous salts are sure to be absent,
and ferric salts yield ferric hydrate with the ammonia, which is not
acted on by the cyanide, but, owing to its bulkiness, it settles slowly;
this lengthens the time required for titration, and so modifies the
manner of working. _An assay should not be worked with ferric hydrate
present, unless the standard contains about the same amount of it._ On
mines it is often inconvenient to separate the copper by means of
sulphuretted hydrogen; hence it is customary to titrate without
previous separation. In this case, instead of standardising the cyanide
with electrotype copper, a standard ore should be used. This should be
an ore (of the same kind as those being assayed) in which the copper has
been carefully determined.

~Effect of Varying Copper.~--In these experiments 10 c.c. of nitric
acid, 30 c.c. of ammonia, and water to 200 c.c. were used.

  Copper nitrate present 1.0 c.c. 10.0 c.c. 20.0 c.c. 50.0 c.c. 100.0 c.c.
  Cyanide required       0.7  "   11.2  "   21.7  "   54.5 "    108.1  "

These results show that under the conditions laid down the various
causes of disturbance nearly neutralise one another, and the results
within a fair range are practically proportional.

~Determination of Copper in Copper Pyrites.~--Weigh up 2 grams of the
dried and powdered ore, and place in an evaporating dish about four
inches in diameter. Cover with 20 c.c. of nitric acid and put on a hot
plate. Evaporate to dryness without further handling. Allow to cool and
take up with 30 c.c. of hydrochloric acid, boil, dilute, and transfer to
a pint flask, filtering if necessary. Make up the bulk with the washings
to about 150 c.c. Precipitate with sulphuretted hydrogen, filter, and
wash back the precipitate into the flask. Add 15 c.c. of nitric acid,
and boil down rapidly to 10 c.c. Dilute, add 30 c.c. of dilute ammonia,
make up to 150 c.c., and cool. For the standard, weigh up 0.5 gram of
copper, more or less, according to the quantity judged to be present in
the assay. Dissolve in 20 c.c. of dilute nitric acid, boil off nitrous
fumes, add 30 c.c. of dilute ammonia, make up to the same bulk as that
of the assay, and cool. Titrate the two solutions side by side and as
nearly as possible in the same manner.

Since the assay solution is often turbid from the presence of small
quantities of lead and of iron from incomplete washing, and since this
slight precipitate is very slow in settling, the standard can hardly be
compared strictly with the assay. This can be counteracted by
precipitating in both solutions a mixture of ferric and aluminic
hydrates, which settles readily and leaves the supernatant liquor clear.
To effect this, boil the nitric acid solutions with 30 c.c. of a
solution containing 15 grams each of alum and ferrous sulphate to the
litre. In an actual determination 2 grams of the ore were taken and
compared with 0.5 gram of copper. The assay required 57.7 c.c. of
cyanide and the standard 52.5 c.c.

        52.5 : 0.5 :: 57.7 : 0.5495

This on 2 grams of ore = 27.47 per cent.; the same sample by
electrolysis gave 27.60 per cent. of copper.

~Determination without Previous Separation.~--Dissolve up 2 grams as
before, but, instead of passing sulphuretted hydrogen, add 30 c.c. of
dilute ammonia, shake well, and cool. Prepare a standard by dissolving
0.5 gram of copper in 1 c.c. of nitric acid, add 0.6 gram of iron in the
form of ferric chloride and 20 c.c. of hydrochloric acid, dilute to
about 150 c.c., add 30 c.c. of dilute ammonia, and cool. Titrate the two
solutions side by side. In a determination on the sample last used, 58
c.c. were required for the assay and 53 c.c. for the standard, which
indicates 27.3 per cent. of copper.

This method of working is somewhat rough.


IODIDE METHOD.

This is based upon the fact that when potassic iodide in excess is added
to a strong solution of a cupric salt in a faintly acid solution,
cuprous iodide is formed and an equivalent of iodine liberated.[56] The
iodine is measured by titrating with a solution of sodium
hyposulphite,[57] using starch paste as indicator. The iodine is soluble
in the excess of potassium iodide, forming a deep brown solution; the
hyposulphite is added until this brown colour is almost removed. Starch
paste is then added, and strikes with the remaining iodine a dirty blue
colour. The addition of the "hypo" is continued until the blue colour is
discharged. The end reaction is sharp; a drop is sufficient to complete
it.

As regards the titration, the process leaves little to be desired; the
quantity of "hypo" required is strictly proportional to the copper
present, and ordinary variations in the conditions of working are
without effect. The presence of salts of bismuth masks the end reaction
because of the strong colour imparted to the solution by the iodide of
bismuth. Under certain conditions there is a return of the blue colour
in the assay solution after the finishing point has apparently been
reached, which is a heavy tax on the patience and confidence of the
operator. This is specially apt to occur when sodium acetate is present,
although it may also be due to excessive dilution.

~The standard "hypo" solution~ is made by dissolving 39.18 grams of the
crystallised salt (Na_{2}S_{2}O_{3}.5H_{2}O) in water and diluting to
one litre. One hundred c.c. will equal one gram of copper.

The starch solution is made by mixing 1 gram of starch into a thin paste
with cold water, pouring it into 200 c.c. of boiling water, and
continuing the boiling for a minute or so. The solution must be cold
before use, and about 2 c.c. is used for each assay. It should not be
added until the bulk of the iodine has been reduced.

To standardise the "hypo," weigh up 0.3 or 0.4 gram of pure copper,
dissolve in 5 c.c. of dilute nitric acid, boil off nitrous fumes, and
dilute with an equal bulk of cold water. Add "soda" solution until a
permanent precipitate is obtained, and then 1 c.c. of acetic acid. This
should yield a clear solution. Fill an ordinary burette with the "hypo."
Add 3 grams of potassium iodide crystals to the copper solution, and,
when these are dissolved, dilute to 100 c.c. with water. Run in the
"hypo" solution rather quickly until the brown colour is nearly
discharged--_i.e._, to within 3 or 4 c.c. of the finish. Add 2 c.c. of
the starch solution, and continue the addition of the "hypo" a few drops
at a time until the tint suddenly changes to a cream colour. The blue
colour must not return on standing three or four minutes. Calculate the
standard in the usual way.

In assaying ores, the copper is dissolved and separated with
sulphuretted hydrogen as in the other processes, but the sulphide should
be washed more completely to ensure the absence of iron salts.

The following experiments show the effect of variation in the conditions
of the assay. Use a solution of copper sulphate containing 39.38 grams
of copper sulphate crystals (CuSO_{4}.5H_{2}O) in the litre. 100 c.c.
equal 1.00 gram of copper.

~Effect of Varying Temperature.~--The assay after the addition of the
potassic iodide must be kept cold, else iodine may be volatilised.

~Effect of Varying Potassium Iodide.~--In various descriptions of the
process the amount of iodide required is variously stated at from "a few
crystals" to as much as 10 grams. The proportion required by theory for
1 gram of copper is a little over 5 grams: an excess, however, is
required to keep the liberated iodine in solution. On economic grounds
this excess should not be extravagant; if the student uses 10 parts of
the iodide for each part of copper in the assay he will have sufficient.
In the experiments there were used 20 c.c. of the copper sulphate, with
varying amounts of potassic iodide, and the following results were
got:--

  Potassic iodide added    1.5 gram      3 grams       5 grams
  "Hypo" required          20.0 c.c.     20.0 c.c.     20.0 c.c.

In these the iodide was added direct to the solution containing the
copper, which was afterwards diluted to 100 c.c. and titrated. In
another series the iodide was added after the dilution to 100 c.c., and
the results were:--

  Potassic iodide added   1.5 gram   3 grams    5 grams    10 grams
  "Hypo" required         20.0 c.c.  20.1 c.c.  20.0 c.c.  20.0 c.c.

~Effect of Varying Bulk.~--In these experiments, 20 c.c. of copper
sulphate were taken, 3 grams of potassic iodide added, and also water to
the required bulk.

  Bulk               20.0 c.c.    100.0 c.c.   200.0 c.c.    500.0 c.c.
  "Hypo" required    20.0  "       20.0  "      20.0  "       19.9  "

In the last of these experiments the colour was discharged at 18 c.c.,
but gradually returned until 19.9 c.c. had been run in. It will be seen
that considerable variation in bulk does not interfere.

~Effect of Acetic Acid.~--These experiments were like the last series
mentioned, but the bulk was 100 c.c., and varying amounts of acetic acid
were added.

  Acetic acid added       0 c.c.  1.0 c.c.  5.0 c.c. 10.0 c.c. 20.0 c.c.
  "Hypo" required      20.0  "   20.1  "   20.1  "   20.0  "   20.2  "

Acetic acid, then, does not interfere to any serious extent.

~Effect of Varying Sodium Acetate.~--These experiments were like those
last mentioned, but without acetic acid, and with varying amounts of
sodium acetate.

  Sodium acetate added   0 gram     1 gram     2 grams    5 grams   10 grams
  "Hypo" required       20.0 c.c.  20.0 c.c.  20.2 c.c.  19.3 c.c.  18.2 c.c.

In the 5 grams experiment, when the finishing point had been apparently
reached the colour slowly returned; but as the results generally on
titrating were not satisfactory a repetition of the experiment was made
with the addition of 5 c.c. of acetic acid, which gave an equally bad
result.

~Effect of Foreign Salts.~--The conditions of these experiments were the
same as before. The salts were added and dissolved before the addition
of the potassium iodide. Using 5 grams (or in the case of the acids, 5
c.c.), the results were as follows:--

                               dilute       Acetic
  Salt added          --      H_{2}SO_{4}   acid       NaAc         NaCl
  "Hypo" required  20.0 c.c.  20.0 c.c.     20.1 c.c.  19.3 c.c.    20.1 c.c.

  Salt added       KNO_{3}    Na_{2}SO_{4}  AmCl       Am_{2}SO_{4}
  "Hypo" required  20.2 c.c.  18.7 c.c.     20.0 c.c.  19.9 c.c.

The low result with the sulphate of soda was evidently due to the
formation of a sparingly soluble double salt, which removed copper from
the solution; on adding a little acetic acid the full amount of "hypo"
was required. The effect of the presence of certain metals is important.
The method of determining it was to add the substance to the solution
containing the copper, and partly precipitate with soda solution; then
treating with 1 c.c. of acetic acid, adding the iodide, and proceeding
as before.

      Substance Added.                   "Hypo" Required,
            -                                20. c.c
  0.050 gram arsenic as As_{2}O_{5}          20.0 "
  0.050   "  antimony as SbCl_{5}            19.8 "
  0.050   "  lead as Pb(NO_{3})_{2}          20.1 "

A similar experiment with 0.050 gram of bismuth nitrate could not be
determined, the end-reaction being masked. Bismuth iodide is soluble in
potassic iodide, forming a brown solution, the colour of which is very
similar to that produced by iodine; and although it does not strike a
blue colour with starch, "hypo" has an action on it.

A similar experiment with 0.050 gram of iron as ferric chloride required
22.3 c.c. of "hypo," and the colour returned on standing. This shows
that ferric acetate liberates iodine under the conditions of the assay.
Trying to counteract this, by adding to a similar solution 0.5 gram of
phosphate of soda dissolved in a little water, 19.7 c.c. of "hypo" were
required instead of 20.0, but the assay showed signs of returning
colour.

In standardising, the same result was obtained, whether the copper was
present as nitrate or sulphate before neutralising.

~Effect of Varying Copper.~--With the same conditions as before, but
with varying amounts of copper and a proportionally increasing quantity
of iodide, the results were:--

  Copper present   1.0 c.c. 10.0 c.c. 20.0 c.c. 50.0 c.c. 100.0 c.c.
  "Hypo" required  1.0  "   10.0  "   20.0  "   50.0  "   100.0  "

showing the results to be exactly proportional.[58]

~Determination of Copper in Copper Pyrites.~--Take 2 grams of the dried
and powdered ore and treat in a porcelain dish with 20 c.c. of nitric
acid, and evaporate to dryness. Take up with 30 c.c. of hydrochloric
acid, dilute, and transfer to a pint flask; make up with water to 200
c.c., warm, and pass sulphuretted hydrogen to excess. Filter, and wash
the precipitate with water acidified with sulphuric acid. Wash the
precipitate back into the flask, and dissolve with 15 c.c. of nitric
acid. Evaporate almost to dryness; add 20 c.c. of water, and boil till
free from nitrous fumes; filter off the sulphur and gangue; neutralise
with soda, avoiding excess; add 1 or 2 c.c. of acetic acid, and shake
till clear. Add 5 grams of potassium iodide, dilute to 100 c.c., and
titrate. The following is an example:--

  0.5 gram of copper required    50.5 c.c. "hypo."
  The assay required             55.6  "     "

which is equal to 27.5 per cent. of copper.


COLORIMETRIC PROCESS.

This is based on the blue coloration of ammoniacal copper solutions. The
quantity of copper in 100 c.c. of the assay solution should not be more
than 15 milligrams, or less than half a milligram. It is not so delicate
as most other colorimetric methods, but nevertheless is a very useful
one.

The manner of working is the same as that described under iron.

~Standard Copper Solution.~--Weigh up 0.5 gram of electrotype copper,
dissolve in 10 c.c. of nitric acid, boil off nitrous fumes, and dilute
to 1 litre. 1 c.c. = 0.5 milligram.

In nearly all cases it will be necessary to separate the copper with
sulphuretted hydrogen from a solution of about 5 grams of the material
to be assayed. The filter paper containing the sulphide (and, probably,
much sulphur) is dried and burnt. The ashes are dissolved in 5 c.c. of
dilute nitric acid, 10 c.c. of dilute ammonia added, and the solution
filtered through a coarse filter into a Nessler tube, washing the paper
with a little dilute ammonia.

The estimation of the colour and calculation of the result are made in
the way described on page 44.

The effect of varying conditions on the assay may be seen from the
following experiments.

~Effect of Varying Temperature.~--The effect of increased temperature is
to slightly decrease the colour, but this can only be observed when a
fair quantity of copper is present.

   1.0 c.c. at 15° showed the colour of 1.0 c.c. at 70°
   2.5          "     "    "          " 2.7      "
   5.0          "     "    "          " 5.0      "
  10.0          "     "    "          " 9.0      "

~Effect of Varying Ammonia.~--The solution must, of course, contain free
ammonia; about 5 c.c. of dilute ammonia in 50 c.c. bulk is the quantity
to be used in the experiments. A larger quantity affects the results,
giving lower readings and altering the tint. With small quantities of
ammonia the colour approaches a violet; with larger, a sky-blue.

   2.5 c.c. with 25 c.c. of strong ammonia read 2.2 c.c.
   5.0       "       "        "      "      "   4.0  "
  10.0       "       "        "      "      "   8.0  "

~Effect of Ammonic Salts.~--The following table shows the results after
addition of ammonic salts:--

  --------------+-----------------+------------------+------------------
   C.c. Present.|  With 10 grams  |   With 10 grams  |   With 10 grams
                | Ammonic Nitrate.| Ammonic Chloride.| Ammonic Sulphate.
  --------------+-----------------+------------------+------------------
        2.5     |      2.5        |        2.5       |        2.0
        5.0     |      5.0        |        5.3       |        4.3
       10.0     |     10.0        |       10.0       |        8.5
  --------------+-----------------+------------------+------------------

These show that sulphates should be avoided, and either nitrate or
chloride solutions be used in the standard as well as in the assay.

~Determination of Copper in a Manganese Ore.~--Treat 3 grams of the ore
with 20 c.c. of hydrochloric acid, and evaporate to dryness. Take up
with 10 c.c. of hydrochloric acid; dilute to about 200 c.c., and pass
sulphuretted hydrogen until the solution smells of the gas; filter,
burn, take up with 5 c.c. of dilute nitric acid, add 10 c.c. of dilute
ammonia, and filter into the Nessler tube, and make up with the washings
to 50 c.c. Into the "standard" tube put 5 c.c. of dilute nitric acid and
10 c.c. of dilute ammonia. Make up to nearly 50 c.c. with water, and run
in the standard copper until the colours are equal. In a determination 4
c.c. (= 2.0 milligrams of copper) were required; this in 3 grams of ore
= 0.07 per cent.

~Determination of Copper in "Black Tin."~--Weigh up 3 grams of the dried
ore, boil with 10 c.c. of hydrochloric acid, and afterwards add 1 c.c.
of nitric; boil off nitrous fumes, evaporate to about 5 c.c., dilute to
50 c.c., add 20 c.c. of dilute ammonia; stir, and filter. If much iron
is present, dissolve the precipitate of ferric hydrate in acid, and
reprecipitate with ammonia. Mix the two filtrates, and dilute to 100
c.c. Take 50 c.c. for the test. A sample of 3 grams of an ore treated in
this way required 5.2 c.c. of standard copper to produce equality of
tint. This gives 0.35 per cent.

~Determination of Copper in Tin.~--Weigh up 1 gram of the sample,
transfer to an evaporating dish, and cover with 30 c.c. of aqua regia.
Warm until the metal has dissolved, then evaporate almost to dryness.
Take up with a few c.c. of hydrochloric acid and again evaporate.

Dissolve the residue in 10 c.c. of dilute hydrochloric acid and transfer
to a 100 c.c. flask. Add 10 c.c. of dilute ammonia and make up with
water to the containing mark.

Filter off 50 c.c. of the solution into a Nessler glass and determine
the copper in it colorimetrically.


EXAMINATION OF COMMERCIAL COPPER.

Very pure copper can be obtained in commerce, owing to the demand for
metal of "high conductivity" for electrical purposes, which practically
means for metal free from impurities.

Much of the metal sold contains as much as one per cent. of foreign
substances, of which arsenic is the most important. The other elements
to be looked for are bismuth, lead, antimony, silver, gold, iron,
nickel, cobalt, sulphur, and oxygen. In "blister copper" (which is the
unrefined metal), aluminium, silicon, and phosphorus may be met with.

~Oxygen.~--All commercial copper carries oxygen; most of it is present
as cuprous oxide, which is dissolved by molten copper. The estimation of
oxygen is often made "by difference." The copper and the other
impurities being determined, the rest is assumed to be oxygen. Probably
this is nearly correct, but the whole of the oxygen should not be
ascribed to cuprous oxide; for any arsenic the metal contained would be
present as cuprous arsenite, since arsenide of copper and cuprous oxide
could not exist together at the temperature of fusion without
interacting. In the report of the analysis, it is best to state the
proportion of oxygen thus:--

  Oxygen  ---- per cent. by difference.

There is a method of determination by fusing 5 or 10 grams in a brasqued
crucible, and counting the loss as oxygen; and another method for the
determination of cuprous oxide based on the reaction of this substance
with nitrate of silver.[59] About 2 grams of silver nitrate, dissolved
in 100 c.c. of water, is allowed to act upon 1 gram of the copper in the
cold. The precipitate is filtered off, washed thoroughly with water, and
the basic salt dissolved and determined colorimetrically.

One part of copper found represents 1.68 part of cuprous oxide, or 0.19
part of oxygen. Copper generally carries from 0.1 to 0.2 per cent. of
oxygen.

~Silver~ is found in most samples, but occurs in variable proportions;
when it amounts to 30 ounces per ton it has a commercial value. To
determine its amount, dissolve 10 grams of the copper in 35 c.c. of
nitric acid and 50 c.c. of water, boil off nitrous fumes, and dilute to
about 100 c.c. One or two c.c. of dilute hydrochloric acid (one to 100
of water) are added, stirred in, and the precipitate allowed to settle
for twenty-four hours. Filter through a double Swedish paper, dry, burn,
and cupel the ashes with one gram of sheet lead.

Ten grams of a sample of copper gave in this way 4.7 milligrams of
silver. Ten grams of the same copper, to which 24 milligrams of silver
had been added gave 28.2 milligrams.

~Gold.~--To determine it, dissolve 10, 20, or 50 grams of the sample in
35, 70, or 175 c.c. of nitric acid and an equal volume of water, boil
till free from nitrous fumes, and dilute to double its volume. Allow to
stand for some time, decant on to a filter, dry, burn, and cupel the
ashes with 1 gram of sheet lead. If silver is present, owing to traces
of chlorides in the re-agents used, "parting" will be necessary. (See
_Gold_.)

Working in this way on 20 grams of copper, to which 1.8 milligram of
gold had been added, a button weighing 2.0 milligrams was obtained.

~Antimony~ is not a frequent impurity of copper: it can be detected in
quantities over 0.1 per cent. by a white residue of Sb_{2}O_{4},
insoluble in nitric acid. With material containing only small quantities
of antimony the white oxide does not show itself for some time, but on
long-continued boiling it separates as a fine powder. It is best (when
looking for it) to evaporate the nitric acid solution to the
crystallising point, to add a little fresh nitric acid and water, and
then to filter off the precipitate. After weighing it should be examined
for arsenic and bismuth.

~Lead.~--Refined coppers are often free from lead, anything more than
traces being seldom found; in coarse coppers it is sometimes present in
considerable quantities.

Its presence may be detected in the estimation of the copper
electrolytically, the platinum spiral becoming coated with a brown or
black deposit of lead dioxide. The depth of colour varies with the lead
present, and obviously could be made the basis of an approximate
estimation. The colour shows itself within an hour or so, but is best
observed when all the copper has been deposited.

Electrolysing a solution of one gram of pure copper, to which 0.5
milligram of lead had been added, the deposit was dark brown; in a
similar solution with 1 milligram of lead it was much darker, and with 2
milligrams it was black. Under the conditions of the assay the dioxide
cannot be weighed, as it partly dissolves on breaking the current. When
lead has been found, its quantity may be estimated by evaporating to
dryness the nitric acid solution to which an excess of sulphuric acid
has been added, taking up with water, and filtering off and weighing the
lead sulphate.

The separation of traces of lead as chromate is a fairly good one.
Dissolve 5 grams of the copper in 17 c.c. of nitric acid and an equal
volume of water; boil off nitrous fumes, neutralise with soda, and
afterwards acidulate with acetic acid; and dilute to a litre. Add 20
grams of sodium acetate, warm, and precipitate the lead with a dilute
solution of potassium chromate. Copper chromate (yellow) may be at the
same time thrown down, but it is readily soluble on diluting. Filter off
the precipitate; wash it into a beaker and pass sulphuretted hydrogen;
oxidise the sulphide and weigh as lead sulphate. Treated in this way 5
grams of copper yielded sulphate of lead equal to 2.0 milligrams of
lead. Five grams of the same sample to which 10 milligrams of lead were
added gave 11.4 milligrams.

~Nickel and Cobalt.~--Nickel is always present in larger or smaller
quantities in commercial copper, and, perhaps, has an influence on the
properties of the metal. It is determined as follows:--Dissolve 10 grams
of the copper in 35 c.c. of nitric acid and an equal bulk of water, boil
off nitrous fumes and neutralise with soda, add 2 grams of carbonate of
soda dissolved in water, boil, and filter. Acidify the filtrate with 2
or 3 c.c. of dilute nitric acid and dilute to 1 or 1-1/2 litres. Pass
sulphuretted hydrogen through the cold solution till the copper is all
down and the liquid smells of the gas. Filter and evaporate the filtrate
to a small bulk, and determine the nickel by electrolysing the solution
rendered ammoniacal, or by precipitating as sulphide and weighing as
sulphate. (See under _Nickel_.) The precipitate, after weighing, should
be tested for cobalt. If present it is separated with potassium nitrite
as described under _Cobalt_. Ten grams of copper gave 6.0 milligrams of
nickel; and another lot of 10 grams of the same copper, to which 10.0
milligrams of nickel had been added, gave 17.2 milligrams.

~Sulphur.~--The amount of sulphur in refined copper is very small,
seldom exceeding 0.005 per cent. In coarse copper, as might be expected,
it is found in larger quantities.

In determining it, it is first converted into sulphuric acid, and then
precipitated and weighed as barium sulphate. The precipitation cannot be
effected from a nitric acid solution. Ten grams of copper are dissolved
in nitric acid, as for the other determinations, and then boiled with
excess of hydrochloric acid till the nitric acid is completely removed.
There is then added a few drops of a dilute solution of baric chloride,
and the solution is allowed to stand for some hours. The baric sulphate
is filtered off and weighed.

The necessity for precipitating from a hydrochloric acid solution is
seen from the following determinations. In each experiment 10 grams of
copper was used, and a known weight of sulphur, in the form of copper
sulphate, added.

  --------------+---------------------------+---------------------
  Sulphur added.|     Sulphur found in      |  Sulphur found in
                |Hydrochloric Acid Solution.|Nitric Acid Solution.
  --------------+---------------------------+---------------------
   5 milligrams |        8 milligrams       |    0.03 milligrams
  10     "      |       11      "           |    0.03      "
  15     "      |       17      "           |   12.00      "
  --------------+---------------------------+---------------------

~Bismuth.~--Nearly all samples of copper contain bismuth, but only in
small quantities. It is best determined colorimetrically as described
under _Bismuth_. The method of concentrating and preparing the solution
for colorimetric assay is as follows. Dissolve 10 grams of copper in
nitric acid, as in the other determinations; neutralise with soda; add 1
or 1.5 grams of bicarbonate of soda and boil for ten minutes; filter,
dissolve the precipitate in hot dilute sulphuric acid; add sulphurous
acid and potassium iodide in excess, and boil till free from iodine.
Filter and dilute to 500 c.c. Take 50 c.c. of the yellow solution for
the determination. A few c.c. of a dilute solution of sulphurous acid (1
in 100) will prevent the liberation of iodine. The following experiments
test the method of separation. Ten grams of copper were treated as above
and precipitated with 1.5 gram of "soda;" the precipitate contained 0.6
milligram of bismuth (= 0.006 per cent.). The filtrate treated with
another 1.5 gram of "soda" gave a precipitate which was free from
bismuth. To the filtrate from this was added 1.0 milligram of bismuth,
and another fraction was precipitated with 1.5 gram of "soda." In this
precipitate was found 1.0 milligram of bismuth. To the filtrate another
milligram of bismuth was added and the separation with "soda" repeated.
The bismuth was separated from this precipitate with ammonic carbonate
before determination, and 0.9 milligram was found.

~Arsenic.~--The proportion of arsenic in copper varies from 0.01 to 0.75
per cent. whilst in coarse copper it may amount to 2 or even 3 per cent.
To determine it, dissolve 5, 10, or 20 grams of the copper (according to
the amount of arsenic present) in 18 c.c., 35 c.c., or 70 c.c. of nitric
acid, and an equal volume of water. Boil off the nitrous fumes, dilute
to 100 c.c. and neutralise with soda; add 1.5 or 2 grams of carbonate of
soda dissolved in a little water, and boil. Filter (washing is
unnecessary) and dissolve back into the flask with a little dilute
hydrochloric acid; add 30 c.c. of dilute ammonia and 25 c.c. of
"magnesia mixture," and allow to stand overnight. The whole of the
arsenic is precipitated as ammonic-magnesic arsenate in one hour, but it
is advisable to leave it longer. The precipitate may be dried and
weighed, or, better, titrated with uranium acetate. (See _Arsenic_.) To
test this method of separation 10 grams of pure copper were taken and
0.200 gram of arsenic dissolved with it. The arsenic was determined by
titration with uranium acetate, and 0.200 gram was found. Two other
similar experiments with 0.080 and 0.010 gram of arsenic added, gave
0.079 and 0.012 gram respectively.

Antimony or bismuth may be present without interfering with the
titration. With 0.100 gram of antimony and 0.100 gram of arsenic, 0.100
gram of arsenic was found; and in another case, with 0.100 gram of
bismuth and 0.060 gram of arsenic, 0.060 gram was found. In these
experiments the antimony and bismuth were present in the assay solution
when titrated. For a gravimetric determination they would require to be
removed before precipitating with "magnesia mixture."

Phosphorus, if present, counts as arsenic in the proportion of 1 to 2.4;
but, except in the case of coarse coppers, it is always absent.

Iron, if present, interferes by forming a white flocculent precipitate
of ferric arsenate after the addition of the sodium acetate and
preliminary to the titration. Each milligram of iron abstracts, in this
way, 1.3 milligrams of arsenic.

~Iron.~--Refined coppers carry traces of iron, varying from 0.001 to
0.01 per cent. It is best determined during the arsenic estimation. The
precipitate of the ammonic-magnesic arsenate will contain the whole of
the iron as ferric hydrate. On dissolving in hydrochloric acid,
neutralising with ammonia, adding 5 c.c. of sodic acetate, diluting, and
boiling, it reappears as a white precipitate of ferric arsenate. It is
filtered off (the arsenic being estimated in the filtrate), dissolved in
warm hydrochloric acid, and determined colorimetrically as described
under _Iron_. A series of experiments testing the separation is there
given.

~Phosphorus.~--Refined coppers do not carry phosphorus, although it may
be present in "coarse copper" up to 1 per cent. or more. In such samples
the following method is adopted for the estimation of both phosphorus
and arsenic. Dissolve 10 grams of copper and 0.1, 0.2, or 0.3 gram of
iron wire (according to the amount of arsenic and phosphorus present) in
35 c.c. of nitric acid and an equal volume of water. Add soda till the
free acid is nearly neutralised. Next add a strong solution of sodium
acetate, until the solution ceases to darken on further addition, then
dilute with water to half a litre. The solution is best contained in a
large beaker; it is next heated to the boiling point, and at once
removed and allowed to settle. If the precipitate is light coloured it
is evidence that sufficient iron has not been added, or, if it is green,
from basic copper salts, it shows that the solution was not sufficiently
acid. In either case start afresh. Filter off the precipitate and wash
with hot water containing a little sodium acetate, dissolve it off the
filter with hot dilute hydrochloric acid, add ammonia in excess, and
pass sulphuretted hydrogen for five minutes. Warm at about 70° C. for a
quarter of an hour. Filter. The clear yellow filtrate contains the
arsenic and phosphorus. Add dilute sulphuric acid in excess; filter off
the yellow precipitate of sulphide of arsenic, dissolve it in nitric
acid, and titrate with uranium acetate, as described under _Arsenic_.

The filtrate from the sulphide of arsenic is rendered alkaline with
ammonia and "magnesia mixture" added. The solution is stirred, and
allowed to stand overnight. The precipitate of ammonic-magnesic
phosphate is filtered off, dissolved, and titrated with uranium acetate,
using the same standard solution as is used in the arsenic assay: 0.5
gram of arsenic equals 0.207 gram of phosphorus.

~Copper.~--The method of determining this has been described under
_Electrolytic Assay_.

In the method of concentration by fractional precipitation with sodic
carbonate (which is adopted in most of these determinations) the
precipitate will contain all the bismuth, iron, and alumina; the arsenic
and phosphorus as cupric arsenate and phosphate; and the greater part of
the lead, antimony, and silver. The nickel and cobalt, and the sulphur
as sulphuric acid, will remain in solution with the greater part of the
copper.


PRACTICAL EXERCISES.

1. According to a wet assay 2 grams of a certain ore contained 0.3650
gram of copper. What would you expect the dry assay produce to be?

2. A standard solution is made by dissolving 25 grams of potassic
cyanide and diluting to a litre. Assuming the salt to be 98 per cent.
real cyanide, what would 100 c.c. of the solution be equivalent to in
grams of copper?

3. How would you make a solution of "hypo" of such strength that 100
c.c. shall equal 0.633 gram of copper?

4. What weight of ore, containing 17.0 per cent. of copper, would you
take in order to get about 0.5 gram of copper in solution for
electrolysis?

5. The solution of copper in nitric acid is effected by the following
reaction:--

3Cu + 8HNO_{3} = 3Cu(NO_{3})_{2} + 4H_{2}O + 2NO.

What volume of nitric acid will be required to dissolve 1 gram of
copper?


LEAD.

The chief ore of lead is galena, a sulphide of lead, common in most
mining districts, and frequently associated with blende and
copper-pyrites. It always carries more or less silver; so that in the
assay of the ore a silver determination is always necessary. Carbonate
(cerussite), sulphate (anglesite), and phosphate (pyromorphite) of lead
also occur as ores, but in much smaller quantities.

Lead ores are easily concentrated (owing to their high specific gravity,
&c.) by mechanical operations, so that the mineral matter sent to the
smelter is comparatively pure.

Lead is readily soluble in dilute nitric acid. The addition of sulphuric
acid to this solution throws down heavy, white, and insoluble lead
sulphate.

Galena is soluble in hot hydrochloric acid, sulphuretted hydrogen being
evolved; but the action is retarded by the separation of the sparingly
soluble lead chloride. If a rod of zinc is placed in this solution,
metallic lead is precipitated on it as a spongy mass, the lead chloride
being decomposed as fast as it is formed. The opening up of the ore is
thus easily effected, the sulphur going off as sulphuretted hydrogen,
and the lead remaining in a form easily soluble in dilute nitric acid.
Galena itself is readily attacked by nitric acid, part of the lead going
into solution, and the rest remaining as insoluble lead sulphate. The
sulphate is due to the oxidation of the sulphur by nitric acid; its
amount will vary with the quantity and concentration of the acid used.
Sulphate of lead is soluble in solutions of ammonium or sodium acetate;
or it may be converted into carbonate by boiling with carbonate of soda.
The carbonate, after washing off the sulphate of soda, dissolves easily
in nitric acid. The precipitation of lead from acid solutions with
sulphuric acid, and the solubility of the precipitate in ammonium
acetate, distinguishes it from all other metals. The addition of
potassium chromate to the acetate solution reprecipitates the lead as a
yellow chromate.


DRY ASSAY.

The dry assay of lead is largely used, but it is only applicable to rich
or concentrated ores, and even with these only gives approximate
results. Both lead and lead sulphide are sensibly volatile at a
moderately-high temperature; hence it is necessary to obtain a slag
which is easily fusible. As a reducing agent iron is almost always used,
and this is added either in the form of an iron rod, or the crucible
itself is made of this metal. The flux used is carbonate of soda.

When a clay crucible is used, the method of working is as
follows:--Weigh up 25 grams of the dry and powdered ore, mix with an
equal weight of "soda" and 2 grams of tartar; place in a crucible (E.
Battersea round), and then insert a piece of iron rod about half an inch
in diameter, and of such a length that it will just allow the crucible
to be covered. The rod should be pushed down so as to touch the bottom
of the crucible, and the mixture should be covered with a sprinkling of
borax. Place in a furnace heated to, but not above, redness, and cover
the crucible. In about twenty minutes the charge will be fused: the
fusion is complete when bubbles of gas are no longer being evolved; and
then, but not till then, the iron is withdrawn, any adhering buttons of
lead being washed off by dipping the rod a few times in the slag. Cover
the crucible, leave it for a minute or two, and then pour. Detach the
slag, when cold, by hammering. The weight of the button multiplied by 4
gives the percentage. The commoner errors of students in working the
process are too high a temperature and too quick a withdrawal.

A sample of ore treated in this manner gave on duplicate assay 17.5 and
17.6 grams of lead, equalling 70.0 and 70.4 per cent. respectively. By
wet assay the sample gave 73.3 per cent. Using an iron crucible, the
results will be 1 per cent. or so higher. The crucible must be made of
wrought iron; and, if it has been previously used, should be cleaned by
heating to dull redness and scraping the scale off with a stirrer. Take
30 grams of the ore, mix with 30 grams of "soda" and 3 grams of tartar;
put the mixture in the crucible, and cover with a sprinkling of borax;
heat for about twenty minutes at not too high a temperature, and then
scrape down the slag adhering to the side with a stirrer. Leave in the
furnace till action has ceased. Before pouring, tap the pot gently, and
then tilt it so as to make the slag wash over the part of the crucible
along which the charge is to be poured. Pour; and, when cold, clean and
weigh the button of metal. A crucible may be used from ten to twenty
times.

These assays are for ores containing the lead chiefly as sulphide. For
oxidised ores, charcoal or tartar is employed as the reducing agent. The
student may practise on red lead as follows:--Take 30 grams of red lead;
mix with 10 grams each of borax and "soda" and about 1.5 gram of
powdered charcoal; place in a small clay crucible with a cover (C.
Battersea round), fuse at a gentle heat, and pour when action ceases.
This assay will only take a few minutes.

Where lead is present as phosphate (as in the case of pyromorphite), or
mixed with phosphates (as sometimes happens), carbonate of soda is a
suitable flux; but the phosphate of soda which is formed makes a thick
tenacious slag, which is very apt to be carried out of the pot by the
escaping gas. A wide-mouthed clay pot is taken and a little fluor spar
added. For the assay of pyromorphite the following charge may be
used:--Ore, 20 grams; "soda," 25 grams; tartar, 7 grams; and fluor spar,
5 grams; and 2 grams of borax as a cover. This will melt down in about
ten minutes, and should be poured as soon as tranquil.


WET ASSAY.

In the case of galena the best method of getting the lead into solution
is to treat with hydrochloric acid and zinc. Put 1 gram of the ore in an
evaporating dish 4 inches across, and cover with 10 c.c. of dilute
hydrochloric acid. Heat till the evolution of sulphuretted hydrogen
becomes sluggish, and then drop in a piece of zinc rod. If the solution
effervesces too strongly, dilute it. Continue the heating until the
sulphide is seen to be all dissolved; when the lead is all precipitated,
pour off the liquid and wash twice with cold water. Peel off the
precipitated lead with the help of a glass rod, and then clean the zinc.
Cover the lead with 20 c.c. of water and 5 c.c. of dilute nitric acid,
and heat gently till dissolved; all the lead will be in solution, and,
when filtered off from the gangue, will be ready for a gravimetric
determination. For volumetric work this filtering is unnecessary.

The chief objection to this method is that commercial zinc carries
considerable quantities of lead. Although this can be determined and
allowed for, the correction required is in most cases too large to be
satisfactory. The following method is applicable in all cases, but is
more troublesome:--Treat 1 gram of the ore with 10 c.c. of dilute nitric
acid in an evaporating dish covered with a clock-glass, and evaporate
till nearly dry. Take up with 50 c.c. of water, and add 10 c.c. of
dilute sulphuric acid. Filter. The residue contains the lead as
sulphate, together with the insoluble matter of the ore and globules of
sulphur. Warm with a solution of ammonium acetate, and filter. The lead
will be in the filtrate, and is recovered in a state fit for direct
gravimetric estimation by the addition of dilute sulphuric acid. If the
volumetric method is to be used, the lead sulphate should be dissolved
out with a solution of sodium acetate instead of with the ammonium salt
solution.


GRAVIMETRIC DETERMINATION.

The lead is separated and precipitated as sulphate, as already
described. The solution must be allowed to stand, and the clear liquid
be decanted through a filter. Transfer the precipitate, and wash with
very dilute sulphuric acid (1 or 2 c.c. in 100 c.c. of water). The acid
must be completely removed with one or two washes with cold water, and
then with alcohol. The volume of liquid required for washing is small,
as the precipitate is dense and easily cleaned; but the washing must be
carefully done, since if any acid remains it will, on drying, char the
paper, and render the subsequent work troublesome. Dry, transfer to a
watch-glass, and burn the filter paper, collecting its ash in a weighed
porcelain crucible. The filter paper must be freed as much as possible
from the lead sulphate before burning, and the ash treated with a drop
or two of nitric and sulphuric acids. Transfer the lead sulphate to the
crucible; ignite gently, keeping the temperature below redness; cool,
and weigh. The precipitate will contain 73.6 per cent. of lead oxide or
68.3 per cent. of lead.

~Determination of Lead in Commercial Zinc.~--Take 10 grams of zinc, and
treat (without heating) with 60 c.c. of dilute hydrochloric acid. When
the zinc is nearly all dissolved, decant off the clear liquid, and
dissolve the residue in 2 c.c. of dilute nitric acid. Evaporate till
most of the acid is removed; dilute to 20 or 30 c.c. with water, and add
10 c.c. of dilute sulphuric acid. Filter off, and weigh the lead
sulphate. Ten grams treated in this way gave--0.1610 gram of lead
sulphate, equivalent to 1.10 per cent. of lead.


VOLUMETRIC METHOD.

This is based upon the reaction between chromate of potash and soluble
lead salts in neutral solutions, whereby an insoluble yellow chromate of
lead is produced.[60] An excess of the chromate is required to complete
the reaction, so that the point at which an indicator shows the presence
of undecomposed chromate cannot be satisfactorily taken as the finish.
Therefore an excess of the standard chromate must be run in, and such
excess determined.


Chromate of lead is not precipitated from strong nitric acid solutions,
and only incompletely from dilute ones. Acids generally are detrimental
to the precipitation, and must be neutralised before titrating. If the
lead is present as sulphate in sodic acetate solution, it is well to
render it distinctly alkaline with ammonia.

Lead chromate precipitated in the cold is a lemon-yellow, light
precipitate, very difficult to filter: on heating to 40° C. the colour
becomes orange; at 60° C. it assumes a deeper hue, and becomes
flocculent; and at a boiling temperature it still further darkens and
settles readily. These changes in colour are not due to any chemical
change, as will be seen by testing the filtrate for chromium or lead:
this is an advantage to the assay, since it is only at the higher
temperature that the precipitate can be easily filtered. The lead is not
completely precipitated, but the amount remaining in solution is only 2
or 3 milligrams, which is just sufficient to give a dark coloration with
sulphuretted hydrogen.

_The standard chromate of potash solution_ is made by dissolving 7.13
grams of bichromate of potash and 2.0 grams of caustic soda in water,
and diluting to 1 litre; or 9.40 grams of the neutral chromate
(K_{2}CrO_{4}) may be dissolved and diluted to 1 litre: 100 c.c. will be
equivalent to 1.000 gram of lead.

_Standard Lead Solution._--16 grams of nitrate of lead (Pb(NO_{3})_{2})
are dissolved in water and diluted to 1 litre; 100 c.c. will contain
1.000 gram of lead.

_Acetate of Soda Solution._--250 grams of the crystallised salt
(NaAc.3H_{2}O) are dissolved, and diluted to 1 litre. Use 40 c.c. for
each assay.

In the titration the assay solution should measure 150 to 200 c.c., and
should be boiling or nearly so. It is best contained in a pint flask,
and the standard chromate solution used with an ordinary burette. Run in
the chromate solution in a steady stream until the whole of the lead has
been precipitated. The amount required for this may be calculated: for
example, 1 gram of an 80 per cent. ore would require 80 c.c. A little of
the assay may be filtered off, and if it does not show a yellow colour
in the filtrate run in 2 c.c. more of the standard solution and continue
this addition till a colour is shown. After this run in another c.c. to
ensure an excess, dilute to 250 c.c., and heat to boiling; allow to
settle for three or four minutes, filter off 50 c.c. into a Nessler
glass, and determine the excess of chromate colorimetrically. The excess
found in the 50 c.c. must, of course, be multiplied by five, and then be
deducted from the quantity of chromate originally run into the assay
solution. The quantity to be deducted should not exceed 3 c.c. Where a
number of determinations are made the colorimetric estimation is
facilitated by using a series of standard phials similar to those
described under the _Electrolytic Copper Assay_. The determination is
rendered sharper and less liable to error by the addition of a few drops
of acetic acid to convert the chromate into bichromate. The same
chromate solution must be used in this determination as was used in the
precipitation.

In standardising the chromate solution, the standard lead nitrate
solution is used. A quantity containing about as much lead as the assay
is supposed to contain is measured off, rendered alkaline with dilute
ammonia, and then neutralised with acetic acid, using a small piece of
litmus paper dropped into the solution as indicator. Then dilute, boil,
and titrate. When the lead in the assay has been separated as sulphate
and dissolved in sodic acetate, less chromate is apparently required,
and in this case it will be necessary to precipitate the lead in the
standard with an equivalent of sodic sulphate and redissolve in sodic
acetate just as in the assay. In these solutions (although there is
considerable chromate in excess) a further addition of 5 or 6 c.c. of
the chromate solution will cause a further precipitate. The following
experiments show the effect of variation in the conditions of the
assay:--

~Effect of Varying Temperature.~--Twenty c.c. of lead nitrate solution
and 10 grams of sodium acetate were used; diluted to 200 c.c., heated to
the desired temperature, and titrated. The results were:--

  Temperature             15°        30°        50°        100°
  "Chromate" required  19.8 c.c.  19.5 c.c.  19.3 c.c.   19.2 c.c.

The first two of these filtered badly, the precipitate coming through
the filter; the last was very satisfactory in the working.

~Effect of Varying Bulk.~--Using 20 c.c. of lead nitrate, and 10 grams
of sodium acetate as before, diluting to the required bulk, heating to
boiling, and titrating, the results were:--

  Bulk                 100.0 c.c. 200.0 c.c.  500.0 c.c. 1000.0 c.c.
  "Chromate" required   19.6  "    19.3  "     19.4  "     19.4  "

~Effect of Varying Acetic Acid.~--Since the experiments are carried out
in the presence of sodic acetate, acetic acid is the only acid whose
effect need be considered. Working as before, but with 200 c.c. bulk and
varying amounts of the acid, the results were:--

  Acid present           --        10.0 c.c.   20.0 c.c.   40.0 c.c.
  "Chromate" required  19.7 c.c.   19.1  "     18.5  "     17.3  "

These experiments show that only slight quantities of acid are
admissible.

~Effect of Varying Sodium Acetate.~--With the same conditions as before,
but with varying weights of sodium acetate, the results were:--

  Sodium acetate present   --      5 grams  10 grams  25 grams  50 grams
  "Chromate" required   19.7 c.c. 19.6 c.c. 19.6 c.c. 18.8 c.c. 17.8 c.c.

These experiments show that excessive quantities of sodium acetate must
be avoided. Ammonium acetate interferes to a greater extent, and if
both acetic acid and this salt are present, each exerts its disturbing
influence. With 10 grams of ammonium acetate, 19.4 c.c. of the chromate
solution were required instead of 19.7 c.c. in the absence of this salt;
with 10 grams of the acetate and 10 c.c. of acetic acid, only 18.6 c.c.
were required.

~Effect of Foreign Salts.~--As already stated, sulphates interfere.
Twenty c.c. of the lead nitrate solution were taken, precipitated with
sulphate of soda, and the precipitate dissolved in 10 grams of sodium
acetate and titrated as before. Duplicate experiments required 18.6 c.c.
and 18.7 c.c. of the chromate solution. A similar experiment with 40
c.c. of lead nitrate required 37.4 c.c. of chromate. If the sulphate had
not been present, the results would have been 19.7 c.c. and 39.4 c.c.
respectively.

~Effect of Varying Lead.~--In these experiments the conditions were as
before, but with varying amounts of lead.

  Lead nitrate solution present  10.0 c.c. 20.0 c.c. 50.0 c.c. 100.0 c.c.
  Chromate solution required.     9.4  "   19.7  "   48.8  "    98.2  "

~Determination of Lead in Galena.~--Weigh up 1 gram of the powdered and
dried ore, and boil in an evaporating dish with 10 c.c. of dilute
hydrochloric acid. When the action becomes sluggish, dilute with an
equal bulk of water, and add a weighed piece of zinc rod about 1 inch
long and quarter-inch across. Keep up a moderate action by warming till
the ore is seen to be completely attacked and the lead precipitated.
Decant off the solution, wash once, strip off the lead, wash and weigh
the remaining zinc. Dissolve the lead in 5 c.c. of dilute nitric acid,
and 5 c.c. of water with the aid of heat. Dilute and transfer to a pint
flask; add a slight excess of dilute ammonia, and render faintly acid
with acetic acid. Dilute to 150 c.c., heat to boiling, and run in the
standard chromate in slight excess, noting the amount required, and make
up to 250 c.c. with water. Boil the solution, allow to settle for a
minute or so, filter off 50 c.c., and determine the excess of chromate
colorimetrically. As an example, 1 gram of an impure galena was
precipitated with 75 c.c. of standard chromate (100 c.c. = 1.020 gram
lead). The excess found in 50 c.c. was 0.3 c.c., which, multiplied by 5,
gives 1.5 c.c. as the excess in the whole solution. The remaining 73.5
c.c. of "chromate" required by the assay, are equivalent to 0.7497 gram
of lead. The zinc used up weighed 1.5 grams, and contained 0.0165 gram
of lead. Thus we get--

      Lead in the assay    0.7497 gram
      Lead from the zinc   0.0165  "
                           ------
  .'. Lead in the galena   0.7332  "

  Equivalent to 73.3 per cent.

Another sample, in which the galena was accompanied with a large
quantity of pyrites, gave the following results:--Three grams were
treated with 30 c.c. of dilute hydrochloric acid and a rod of zinc. The
zinc and lead were carefully transferred to another vessel, the zinc
cleaned, and the lead (dissolved in 5 c.c. of dilute nitric acid and 20
c.c. of water) treated as before.

  14.5 c.c. of the chromate were required = 0.1479 gram lead
  Lead in 2 grams of zinc                 = 0.0220     "
                                            ------
  .'. Lead in 3 grams of the ore          = 0.1259     "

  Equivalent to 4.20 per cent.

The same ore gave by separation of the lead with sulphuretted hydrogen,
and conversion into sulphate, 4.16 per cent.

With fairly pure ores, free from sulphate, the assay may be made more
quickly as follows: Dissolve 1 gram of the finely powdered ore by
boiling gently with 40 c.c. of dilute hydrochloric acid for 15 minutes;
cool; add a few drops of permanganate; neutralise with ammonia, add
acetic acid and a little sodium acetate. Titrate with standard chromate.


COLORIMETRIC PROCESS.

This is based upon the brown coloration produced in very dilute
solutions of lead by the action of a solution of sulphuretted hydrogen.
The quantity of lead in the 50 c.c. of the assay solution must not much
exceed 0.5 milligram, nor be less than 0.01. The sulphuretted hydrogen
is used in the form of a solution, and is not bubbled through the assay.
The principle of working is the same as previously described.

_Standard Lead Solution._--Each c.c. of this should contain 0.1
milligram of lead. It is made by diluting 10 c.c. of the solution of
lead nitrate, described under the volumetric process, to 1 litre.

_Sulphuretted hydrogen water_ is made by passing a current of the washed
gas into water till the latter is saturated.

Five c.c. of the sulphuretted hydrogen water are put into a Nessler
tube, the measured portion of the assay solution added, and the whole
diluted with water to the 50 c.c. mark. Into the standard Nessler tube
the same amount of the sulphuretted hydrogen water is put, and diluted
to nearly 50 c.c. The standard lead solution is then run in till the
tints are equal. The assay solution must not contain much free acid, and
if the conditions will allow it, may with advantage be rendered alkaline
with ammonia. The chief cause of disturbance is the precipitation of
lead sulphide forming a black turbid solution instead of a brown clear
one. This may be caused by using hot solutions or an excess of acid.
Other metals precipitable by sulphuretted hydrogen must be absent as
well as strong oxidising agents.

~Effect of Varying Temperature.~--The effect of increased temperature is
to change the colour from brown to black, and to render the estimation
difficult.

  1 c.c. at 15° C. showed the colour of 0.5 c.c. at 60° C.
  2  "      "        "          "      1.5   "   at 60° C.
  3  "      "        "          "      5.0   "   at 50° C.

~Effect of Varying Time.~--The colour becomes lighter on standing: 2
c.c. on standing 10, 20, and 40 minutes became equal in colour to 1.7
c.c.

~Effect of Acids and Ammonia.~--Two c.c. of the solution with 2 c.c. of
dilute hydrochloric acid became cloudy and equivalent to about 2.5 c.c.;
and a similar result was got with 2 c.c. of dilute sulphuric acid. With
2 c.c. of dilute ammonia the solution became somewhat darker, or equal
to 2.3 c.c.; but gave a very clear solution easy to compare.

~Determination of Lead in Commercial Zinc.~--Dissolve 0.1 gram of the
metal in 1 c.c. of dilute nitric acid evaporates till a solid separates
out, dilute to 100 c.c. with water, and take 20 c.c. for assay. A sample
treated in this way required 2.4 c.c.; this multiplied by 5 gives 12.0
c.c., equivalent to 1.2 milligram of lead, or 1.2 per cent. By
gravimetric assay the sample gave 1.10 per cent.


PRACTICAL EXERCISES.

1. Thirty grams of galena gave on dry assay 21 grams of lead; and this,
on cupellation, gave 15 milligrams of silver. Calculate the results in
per cents. of lead and in ounces of silver to the ton of ore.

2. How many ounces of silver to the ton would be contained in the lead
got from this ore if the loss in smelting is equal to that of the assay?

3. Having given you a sample of white lead freed from oil by washing
with ether, how would you proceed to determine the percentage of lead in
it?


THALLIUM.

Thallium is a rare metal, found in small quantities in some varieties of
iron and copper pyrites, and in some lithia micas. It resembles lead in
appearance. Its compounds resemble the salts of the alkalies in some
respects; and, in others, those of the heavy metals.

It is detected by the green colour which its salts impart to the flame.
This, when examined with the spectroscope, shows only one bright green
line.

It is separated and estimated by dissolving in aqua regia; converting
into sulphate by evaporation with sulphuric acid; separating the second
group of metals with sulphuretted hydrogen in the acid solution, boiling
off the excess of the gas; nearly neutralising with carbonate of soda;
and precipitating the thallium with an excess of potassic iodide. On
allowing the liquid to stand for some time a bright yellow precipitate
of thallous iodide separates out. This is collected on a weighed filter;
washed with cold water, finishing off with alcohol; dried at 100° C.,
and weighed. The precipitate is thallous iodide TlI, and contains 61.6
per cent. of thallium.


BISMUTH.

Bismuth is nearly always found in nature in the metallic state; but
occasionally it is met with as sulphide in bismuthine and as carbonate
in bismutite. It is also found in some comparatively rare minerals, such
as tetradymite, combined with tellurium, and associated with gold. In
minute quantities it is widely distributed: it is a common constituent
of most copper ores; hence it finds its way into refined copper, which
is seldom free from it. It is occasionally met with in silver in
sufficient quantity to interfere with the working qualities of that
metal.

Bismuth compounds are used in medicine and in the manufacture of alloys.
Bismuth possesses many useful properties. It has considerable commercial
value, and sells at a high price.

The metal is brittle, breaks with a highly crystalline fracture, and has
a characteristic reddish-yellow colour. It is almost insoluble in
hydrochloric, but readily dissolves in nitric, acid; and gives, if the
acid is in excess, a clear solution. Bismuth salts have a strong
tendency to separate out as insoluble basic compounds; this is more
especially true of the chloride which, on diluting with a large volume
of water, becomes milky; the whole of the bismuth separating out. The
nitrate, carbonate, and hydrate yield the oxide (Bi_{2}O_{3}) on
ignition. This oxide closely resembles litharge. It combines with
silica, forming fluid slags; and at a red heat is liquid enough to be
absorbed by a cupel; in fact, bismuth may take the place of lead in
cupellation. The metal itself is easily fusible, and may be separated
from its ores by liquation.

The assay of bismuth by wet methods presents little difficulty, and is
fairly accurate. The price of the metal is such that only methods which
yield good results should be adopted; and, since bismuth is volatile at
the temperature of the furnace, and is found mixed with ores not easy to
flux, as also with metals which are not easily separated by the dry
method, the dry assay can only be considered as having a qualitative
value.


DRY ASSAY.

~By Liquation.~--This is adapted to ores containing the bismuth as
metal. Take 20 grams of the powdered ore and place in a crucible with a
perforated bottom, put this crucible into another of about the same size
and lute the joint. Lute on a cover, place in the furnace and heat to
redness. The bismuth melts readily and drains into the lower crucible
from which, when cold, it is taken and weighed.

~By Fusion.~--For fairly pure ores the process is as follows:--Take 20
grams of the ore and mix with 20 grams of fusion mixture, 10 grams of
salt and 5 or 10 grams of potassium cyanide; place in a crucible, cover,
and fuse at a moderate temperature for about fifteen minutes; pour; when
cold detach the metal and weigh.

For coppery ores in which the metals are present as sulphides use the
fluxes just given with 2 grams of charcoal (instead of the cyanide) and
a little sulphur.

For coppery ores in which the metals are present as oxides, mix 20 grams
of the ore with 10 grams of fusion mixture, 4 grams of salt, 4 grams of
sulphur and 2 grams of charcoal; and fuse.

A considerable percentage of bismuth is lost in these assays; it is
stated as being nearly 8 per cent. of the metal present.


WET METHODS.

~Detection.~--Bismuth is detected by dissolving the substance in nitric
or hydrochloric acid and precipitating the diluted solution with
sulphuretted hydrogen. The precipitated sulphides, after digesting with
soda and washing, are dissolved in nitric acid and the solution boiled
with ammonium carbonate. The precipitate is washed and then warmed with
dilute sulphuric acid. The solution will contain the bismuth. Add a
solution of potassium iodide in excess, and boil; a yellow or dark brown
solution proves that bismuth is present. Another good test for small
quantities of bismuth is to add tartaric acid to the solution to be
tested, and then to make it alkaline with potash. Add a few c.c. of
Schneider's liquid,[61] and heat. A brownish-black colour is produced by
as little as one part of bismuth in 200,000 of solution. The test is not
applicable in the presence of mercury, copper, or manganese.

Compounds of bismuth fused with cyanide of potassium in a Berlin
crucible readily give a globule of bismuth which is recognised by its
appearance and fracture.

~Solution and Separation.~--The solution of bismuth compounds presents
no difficulty. They are soluble in nitric acid or aqua regia, and,
provided the solution is sufficiently acid, they remain dissolved. In
separating it from other metals the solution is made up to about 100
c.c. and treated with a current of sulphuretted hydrogen. The bismuth
comes down in a tolerably strong acid solution. The sulphide is decanted
on to a filter and washed. It is next digested with ammonic sulphide;
or, better (especially when other metals are present), dissolved in
nitric acid, and treated with an excess of ammonia and a current of
sulphuretted hydrogen. The precipitate is filtered off and evaporated to
dryness with nitric acid. It is taken up with a few drops of sulphuric
acid and a little water; and warmed and filtered, if necessary. The
filtrate is nearly neutralised with ammonia; ammonium carbonate added in
slight excess; and the liquid heated to boiling and filtered. The
bismuth will be contained in the precipitate with perhaps traces of
lead, antimony, tin, or sometimes iron from incomplete separation or
washing. When only traces of a precipitate are got it must be tested.
The bismuth precipitate is readily soluble in dilute nitric acid.


GRAVIMETRIC DETERMINATION.

The bismuth having been separated and dissolved in nitric acid[62] is
precipitated (after dilution) by the addition of carbonate of ammonium
in slight excess, and boiling. The precipitate is filtered off, washed
with hot water, dried, ignited, and weighed. The ignition should be
performed carefully at not above a low red heat. The oxide which is
formed has, at this temperature, a dark yellow or brown colour, and
becomes yellow on cooling. It is bismuthic oxide (Bi_{2}O_{3}) and
contains 89.65 per cent. of bismuth. Fusion with potassium cyanide at a
temperature just sufficient to melt the salt reduces it to the metal
which falls to the bottom and runs into a globule. The button of metal
may be weighed, but it often sticks tenaciously to the bottom of the
crucible. The precipitation with ammonic carbonate must not be made in a
sulphate or chloride solution; since basic compounds would then be
thrown down, and the result on weighing would either be too low (because
of the volatilisation of the chloride), or too high (because of the
retention of sulphuric acid).

Bismuth compounds in a nitric acid solution are readily decomposed by
the electric current, but the deposited bismuth is not coherent. It
comes down in shaggy tufts which are difficult to wash and easy to
oxidise.


VOLUMETRIC ASSAY.

There are two methods which have been proposed; one based on the
precipitation as chromate and the estimation of the chromic acid; and
the other on the precipitation as oxalate and subsequent titration with
permanganate of potash. These offer little advantage over the easy
gravimetric determination.


COLORIMETRIC METHOD.

Bismuth iodide dissolves in excess of potassium iodide, forming a
yellow-coloured solution, indistinguishable in colour from that given by
iodine. The colour, however, is not removed by boiling or by sulphurous
acid. Since none of the commoner metals give such a colour, and free
iodine is easily separated by boiling, this method is specially suited
for small determinations of bismuth.

It requires a _solution of bismuth_, made by dissolving 0.1 gram of
bismuth in a drop or so of nitric acid, evaporating with a little
sulphuric acid and diluting with water to 1 litre. 1 c.c. will contain
0.1 milligram of bismuth. And a _solution of sulphurous acid_, made by
diluting 10 c.c. of the commercial acid to 1 litre with water.

The determination is made in the usual way: 50 c.c. of the prepared
solution, which should not carry more than 0.75 milligram nor less than
0.01 milligram of bismuth, are placed in a Nessler tube and the colour
compared with that observed in a similar tube containing water and
potassium iodide on adding the standard solution of bismuth.

The assay solution is prepared by separating the bismuth with
sulphuretted hydrogen, boiling the precipitate with nitric acid, and
evaporating with sulphuric acid. Take up with water, add 10 or 20 c.c.
of solution of potassium iodide, boil off any iodine liberated, dilute,
filter, and make up to 100 c.c. According to the depth of colour take
10, 20, or 50 c.c. and transfer to the Nessler tube. Add a few c.c. of
the solution of sulphurous acid. Into the other Nessler tube put as much
potassium iodide solution as is contained in the assay tube, with
sulphurous acid and water to within a few c.c. of the bulk. Then add the
standard bismuth solution till the tints are equal.

The student must be careful not to confuse the colour of the bismuth
iodide with that of free iodine. If the yellow colour is removed by
boiling and returns on standing it is due altogether to iodine; if it is
lessened by the addition of a few drops of the dilute sulphurous acid,
it is in part due to it. Hence the necessity of having a little free
sulphurous acid in each tube. A strong solution must not be used, since
it liberates iodine from potassium iodide.

The following experiments illustrate the effect of variation in the
conditions of the assay:--

~Effect of Varying Temperature.~--At a higher temperature the colour is
somewhat lessened.

  1.0 c.c. at 15° C. showed the colour of 0.8 c.c. at 70° C.
  2.5  "       "        "          "      2.0          "
  5.0  "       "        "          "      5.0          "

~Effect of Free Acid.~--

  2.5 c.c. with 5 c.c. of nitric acid equalled 2.5 c.c.
  5.0  "        "       sulphuric acid    "    5.0  "

Hydrochloric acid almost completely removes the colour, which, however,
is restored by the addition of a few crystals of potassium iodide.

~Effect of Alkalies.~--Ammonia, soda, or potash destroys the colour, but
it is restored on acidifying with nitric or sulphuric acid.

~Effect of Ammonic Salts.~--The following table shows the results after
addition of ammonic salts:--

  -------------+----------------+-----------------+-----------------
  C.c. present.|With 10 grams   |With 10 grams    |With 10 grams
               |Ammonic Nitrate.|Ammonic Sulphate.|Ammonic Chloride.
  -------------+----------------+-----------------+-----------------
    1.0 c.c.   |    0.9 c.c.    |    1.1 c.c.     |       --
    2.5  "     |    2.5  "      |    2.7  "       |       --
    5.0  "     |    5.0  "      |    5.5  "       |       --
  -------------+----------------+-----------------+-----------------

Ammonic chloride, like hydrochloric acid, removes the colour, which may
be restored on the addition of more potassium iodide. Nitrates and
sulphates do not thus interfere.

~Effect of Foreign Salts.~--Sodic hyposulphite almost completely removes
the colour. Copper salts liberate iodine; but when this has been removed
by boiling and the cuprous iodide has been filtered off there is no
further interference. Dilute solutions of lead salts give no colour.


PRACTICAL EXERCISES.

1. A fusible alloy is made up of 8 parts of bismuth, 5 of lead, and 3 of
tin. What weight of oxide of bismuth, Bi_{2}O_{3}, would you get on the
analysis of 1 gram of it?

2. What weight of bismuth can be got from 2 grams of the subnitrate
BiONO_{3}.H_{2}O?

3. How would you detect and separate arsenic, lead, and copper in a
sample of bismuth?


ANTIMONY.

Antimony occurs in the native state, but is rare; its common ore is
antimonite, the sulphide (Sb_{2}S_{3}). Jamesonite and other sulphides
of lead and antimony are frequently met with. Sulphide of antimony is
also a constituent of fahlerz and of many silver ores.

Antimonite occurs generally in fibrous masses, has a lead-like metallic
lustre, is easily cut with a knife, and melts in the flame of a candle.

Antimony itself has a very crystalline fracture, is brittle, and has a
bluish-white colour. It is used in the preparation of alloys with lead
and tin for the manufacture of type-metal. It is readily fusible, and
imparts hardness and the property of taking a sharp cast to its alloys.
It is practically insoluble in hydrochloric acid. On boiling with strong
nitric acid it is converted into antimonic oxide (Sb_{2}O_{5}), which is
a powder almost insoluble in this acid or in water, but which may be got
into solution with difficulty by the prolonged action of hydrochloric
and tartaric acids. Antimonic oxide is converted on ignition into the
tetroxide (Sb_{2}O_{4}) with loss of oxygen. Antimony forms two series
of salts, antimonious and antimonic; and advantage is taken of this in
its determination volumetrically. Either sulphide of antimony yields
antimonious chloride on boiling with hydrochloric acid, sulphuretted
hydrogen being given off; and, in the case of antimonic sulphide,
sulphur is deposited. Antimonious is converted into antimonic chloride
by treatment with permanganate of potash in an acid solution. Antimonic
chloride and potassium iodide react, forming antimonious chloride and
free iodine. This latter may be got rid of by boiling. Sulphide of
antimony is separated from the ore by liquation; this regulus is met
with in commerce as "crude antimony."


DRY ASSAY.

An approximate determination of the amount of sulphide of antimony in an
ore may be made by fusing and liquating in a luted double crucible in
the manner described under bismuth. This is unsatisfactory. The
determination of metallic antimony in an ore is made either by fusion
with potassium cyanide or by fusion with iron, as in the galena assay.
Both methods yield poor results; and, where iron is used, it must be
added in quantity only sufficient for desulphurising; this amounts to
about 40 per cent. in pure ores. If the iron is in excess it alloys with
the reduced antimony. If, on the other hand, it is insufficient, the
metal will contain sulphur; or sulphide of antimony will be lost in the
slag.

The following note, for which we are indebted to Mr. Bedford McNeill,
A.R.S.M., gives a description of the method adopted in the commercial
valuation of a parcel of antimony ore:--

The antimony smelter, when he wishes to determine the value of any
parcel of ore--usually the sulphide--that may be offered for sale,
practically has recourse to the smelting operation. That is, a quantity
of 2 or 3 cwts. taken by his sampler having been obtained, he treats it
under the immediate supervision of the foreman smelter as if it formed
part of the ore in process of daily reduction at his works. He thus
determines by actual trial the output which it may fairly be anticipated
will be yielded by the bulk, and upon the result of this trial or assay,
and the knowledge gained of the actual behaviour of the ore under
treatment, he bases his tender, knowing that, should he secure the
parcel, he may confidently expect a similar return.

Briefly, the process consists of the three ordinary operations of--

  (a) Singling or removing most of the antimony from the ore;
  (b) Doubling;
  (c) Refining or "starring."

But in the assay sufficient information is generally given by the first
two of these.

A new pot having been taken and made hot in the furnace, 40 or 45 lbs.
of the ore is weighed in (the mineral from the necessities of sampling
not exceeding walnut size); 1 to 3 lbs. of salt cake is now added to
render the separation of the resulting sulphide of iron more easy, as
also to assist in the fusion of the gangue; 20 to 25 lbs. of tin-plate
scrap, beaten more or less into ball shape, is weighed, placed on the
top of the ore and salt cake, and the whole brought to a state of
fusion. The foreman from time to time takes notice of the behaviour of
the ore under the working conditions. Ores that manifest a tendency to
"boil" or "froth " require the admixture of other more sluggish mineral
in order to render their reduction economically practicable.

After 1-1/4 to 1-1/2 hours (the time depending mainly on the
temperature), the contents of the crucible are usually in a state of
tranquil fusion. The pot is now lifted from the fire, and its contents
transferred to a conical iron mould, the empty pot being immediately put
back into the fire, and the latter "mended" with sufficient coke for
another run. The conical mould (when dealing with a "strange" ore, and
the possibility of insufficient iron being present to satisfy the
sulphur contents) is wiped inside with clay previous to pouring in the
molten charge. Otherwise the mould itself will be attacked, and the
contents after solidifying will require to be chiselled out piecemeal.

A further 40 lbs. of the ore is now charged into the crucible with iron
as above; but before this second charge is ready to be drawn an
inspection of the first may suggest the addition of either 3 or 5 lbs.
more iron, or 5 or 10 lbs. more ore.

It is a good fault rather to aim at an excess of iron as tending to
clean the ore from antimony, any of the latter that (from an
insufficiency of iron) may be left in the slag from the first process
being irretrievably lost; whereas, if the iron be in excess, that which
is combined with the crude antimony resulting from the first process is
easily got rid of by adding 3 to 5 lbs. or so of ore in the second
process.

This latter, as practised for the determination of the value of a parcel
of ore, consists in selecting two of the best quality singles, resulting
from perhaps four or five trials as above, and running them down with a
few pounds of salt cake, or a mixture of salt cake with American potash,
and (as is generally necessary) a small addition of ore.

Upon the final result (confirmed perhaps on another pair of singles,
and, judging from the total weight or output of the metal as calculated
from the ore used in "singling," plus any added in the "doubling," the
crystalline fracture and face of the metal, its colour, etc.) the price
to be offered for the parcel of ore is fixed.


WET METHODS.

~Detection.~--The antimony, if any, being got into solution by treating
the ore with hydrochloric acid or aqua regia may be detected by
evaporating with hydrochloric acid, diluting, and filtering into the
cover of a platinum crucible or (better) a platinum dish. A small lump
of zinc is then added, and, if antimony is present, _the dish_ will in a
minute or so be stained black with a deposit of metallic antimony. This
stain is removed by nitric, but not by hydrochloric, acid. The reaction
is delicate and characteristic; arsenic under like conditions is evolved
as arseniuretted hydrogen, and tin is deposited as metal _on the zinc_.

~Solution.~--Ores, &c., containing antimony are best opened up by
boiling with hydrochloric acid or aqua regia; treatment with nitric
acid should be avoided wherever possible, since it forms antimonic acid,
which is subsequently dissolved only with difficulty. Salts of antimony
in solution have a tendency to form insoluble basic salts; so that care
must be exercised in diluting. Compounds such as antimonite which are
soluble in hydrochloric should be dissolved at once in that acid.

~Separation.~--To the solution add potash in excess and a little free
sulphur, and pass a current of sulphuretted hydrogen for some minutes;
allow to digest for an hour or so on a hot plate; filter; and wash the
residue. Acidulate the filtrate with hydrochloric acid: the precipitate
will contain the antimony (as Sb_{2}S_{5}), and possibly arsenic or tin.
The precipitate is transferred to a beaker and boiled with hydrochloric
acid; the solution is filtered off and diluted. Add a few crystals of
tartaric acid, and pass a current of sulphuretted hydrogen for some
time. The first flocculent precipitate will become denser, and render
the filtering more easy. Transfer the precipitate (after washing free
from chlorides) to a Berlin dish, and treat cautiously with fuming
nitric acid. The action of this acid on the sulphide is very violent.
Evaporate and ignite, transfer to a silver dish, and fuse with four or
five times its weight of caustic soda, cool and extract with a little
water, then add an equal volume of alcohol, and allow to stand
overnight. Filter, wash with dilute alcohol. (The filtrate will contain
the tin.) The residue contains the antimony as antimonate of soda, and
is dissolved off the filter with hot dilute hydrochloric, with the help
of a little tartaric, acid. The filtrate is now ready for the
gravimetric determination.


GRAVIMETRIC ASSAY.

Pass a current of sulphuretted hydrogen through the solution containing
the antimony to which a little tartaric acid has been previously added.
Pass the gas till the precipitate becomes dense, and the antimony is all
down. The solution must not be too strongly acid. Filter off the
precipitate, wash with hot water, dry in the water oven, transfer to a
weighed porcelain dish, and cautiously treat with fuming nitric acid.
Continue the action on the water bath till the sulphur and antimony are
completely oxidised. Evaporate; ignite, gently at first, then strongly
over the blast; cool, and weigh. The residue is a white infusible
powder, and consists of antimony tetroxide, Sb_{2}O_{4}, containing
78.94 per cent. of the metal.

~Determination of Antimony as Bigallate.~--What appears to be a very
good method has been worked out by M.A. Guyard, and is described in
Crookes' _Select Methods_, p. 398.

The antimony must be in solution as antimonious chloride, and must not
be accompanied by an excess of hydrochloric acid. To ensure these
conditions, the solution is treated with potassium iodide until no more
iodine is evolved, and is then evaporated to remove the excess of
hydrochloric acid. To the concentrated, and nearly neutral, solution a
freshly-prepared solution of gallic acid is added in slight excess. A
bulky white precipitate is formed that settles rapidly. The solution is
diluted with hot water and washed by decantation. Then the precipitate
is collected on a weighed double filter, washed once or twice with hot
water, and dried at 100° C. The dried substance is antimony bigallate,
and contains 40.85 per cent. of antimony. It should be completely
soluble in ammonium sulphide. The solution in which the antimony is
precipitated need not be quite free from other metals.


VOLUMETRIC METHOD.

This is based on the reduction of antimonic chloride (SbCl_{5}) to
antimonious (SbCl_{3}) by the action of potassium iodide in strong
hydrochloric acid solution.[63] Iodine is at the same time liberated,
and the amount of antimony reduced is got at by titrating with sodium
hyposulphite, which measures the iodine set free.

The standard solution of sodium hyposulphite is made by dissolving 41.32
grams of the salt (Na_{2}S_{2}O_{3}.5H_{2}O) in water, and diluting to 1
litre. One hundred c.c. will be equivalent to about 1 gram of antimony.

It is standardised with the help of a solution of antimony made as
follows:--Weigh up 5 grams of powdered antimony, transfer to a flask,
and cover with 50 c.c. of hydrochloric acid; boil, and add nitric acid
(5 or 10 drops at a time) until the metal is dissolved. Allow the action
of the nitric acid to cease before adding more. Boil down to a small
bulk, add 250 c.c. of hydrochloric acid, and dilute to nearly 1 litre.
Warm until any precipitate which has formed is redissolved; allow to
cool slowly, and run in from a pipette a weak solution of permanganate
until a faint brown colour is produced. Dilute to exactly 1 litre; 100
c.c. contain 0.5 gram of antimony as antimonic chloride.

In standardising, take 50 c.c. of the antimony solution, and transfer to
a flask; add 2 grams of potassium iodide crystals, and when dissolved,
after standing a few minutes, run in the solution of "hypo" from an
ordinary burette until the greater part of the iodine has been reduced.
Add a few drops of starch solution, and continue the addition of the
"hypo" until the muddy-green colour changes to a clear brownish-yellow.
The solution must be shaken after each addition of the "hypo."

In determining antimony in ore, weigh up 0.5 to 1 gram, and dissolve in
hydrochloric acid with, if necessary, the help of chlorate of potash.
The antimony is separated as sulphide, redissolved in hydrochloric acid,
and oxidised with a crystal of chlorate of potash. Chlorine is boiled
off, and the solution diluted with an equal bulk of water. To the clear
cold solution potassium iodide is added, and after a few minutes the
liberated iodine is titrated with "hypo," as already described. The
method only yields satisfactory results when the standard and assay are
carried out alike.

FOOTNOTES:

[50] "Modern American Methods of Copper Smelting" (Dr. Peters).

[51] "Journal of the Society of Chemical Industry," vol. v. No. 2.

[52] Lead when present is precipitated on the _spiral_ in the form of a
dark powder of dioxide (PbO_{2}). Manganese is also thrown down on the
spiral as dioxide (MnO_{2}), the solution at the same time becomes
violet from the formation of permanganic acid.

[53] See the method given under _Examination of Commercial Copper_.

[54] CuSO_{4} + 4KCy = 2KCy.CuCy_{2} + K_{2}SO_{4}.

[55] 2CuSO_{4} + 3KCy + Am_{2}O = Cu_{2}Cy_{2} + Am_{2}SO_(4) +
K_{2}SO_{4} + KCyO.

[56] 2CuSO_{4} + 4KI = Cn_{2}I_{2} + 2I + 2K_{2}SO_{4}.

[57] 2Na_{2}S_{2}O_{3} + 2I = 2NaI + Na_{2}S_{4}O_{6}.

[58] For further information, see Appendix B., and a paper by J.W.
Westmoreland, _Journal of the Society of Chemical Industry_, vol. v.
p. 48.

[59] 3Cu_{2}O + 6AgNO_{3} + 3H_{2}O = 2Cu_{2}H_{3}O_{3}NO_{3} +
2Cu(NO_{3})_{2} + 6Ag. (Insoluble basic salt.)

[60] K_{2}CrO_{4} + Pb(NO_{3})_{2} = PbCrO_{4} + 2KNO_{3}

[61] Made by dissolving 12 grams of tartaric acid and 4 grams of
stannous chloride in water, and adding potash solution till it is
alkaline. The solution should remain clear on heating to 60° or 70° C.

[62] It must be remembered that arsenate of bismuth is completely
insoluble in this acid.

[63] SbCl_{5} + 2KI = I_{2} + SbCl_{3} + 2KCl.



CHAPTER XI.

IRON--NICKEL--COBALT--ZINC--CADMIUM.


IRON.

Iron rusts or oxidises very readily, and, consequently, is rarely found
in the metallic state in nature; such native iron as is found being
generally of meteoric origin or imbedded in basalt and other igneous
rocks. It chiefly occurs as oxide, as in magnetite, hæmatite, and in the
brown iron ores and ochres. Chalybite, which is carbonate of iron, is an
ore of great importance. Iron is found combined with sulphur in
pyrrhotine and pyrites, and together with arsenic in mispickel. It is a
common constituent of most rocks, imparting to them a green, black, or
brown colour; and is present, either as an essential part or as an
impurity, in most substances.

The chemistry of iron is somewhat complicated by the existence of two
oxides, each of which gives rise to a well-marked series of compounds.
Those derived from the lower oxide, known as ferrous salts, are
generally pale and greenish. Ferric salts are derived from the higher
oxide, and are generally red, brown, or yellow. The existence of these
two well-marked families of salts renders the assay of iron
comparatively easy, for the quantity of iron present in a solution can
be readily measured by the amount of oxidising or reducing agent
required to convert it from the one state into the other--that is, from
ferrous to ferric, or from ferric to ferrous, as the case may be.

In the red and brown iron ores and ochres ferric iron is present; in
chalybite the iron is in the ferrous state; and in magnetite it is
present in both forms. Traces of iron in the ferrous state may be found
(even in the presence of much ferric iron) by either of the following
tests:--

     1. Ferricyanide of potassium gives a blue precipitate or green
     coloration; with ferric salts a brown colour only is produced.

     2. A solution of permanganate of potassium is decolorised by a
     ferrous salt, but not by a ferric one.

Traces of ferric iron can be detected (even in the presence of much
ferrous iron) by the following tests:--

     (1) By the brown or yellow colour of the solution, especially
     when hot.

     (2) By giving a pink or red coloration with sulphocyanide of
     potassium.

Substances containing oxide of iron yield the whole of the iron as metal
when fused at a high temperature with charcoal and suitable fluxes. The
metal, however, will contain varying proportions of carbon and other
impurities, and its weight can only afford a rough knowledge of the
proportion of the metal in the ore. There are two or three methods of
dry assay for iron, but they are not only inexact, but more troublesome
than the wet methods, and need not be further considered. Chalybite and
the hydrated oxides dissolve very readily in hydrochloric acid; hæmatite
and magnetite dissolve with rather more difficulty. Iron itself, when
soft, is easily soluble in dilute hydrochloric, or sulphuric, acid.
Pyrites, mispickel, &c., are insoluble in hydrochloric acid, but they
are readily attacked by nitric acid. Certain minerals, such as chrome
iron ore, titaniferous iron ore, and some silicates containing iron,
remain in the residue insoluble in acids. Some of these yield their iron
when attacked with strong sulphuric acid, or when fused with the acid
sulphate of potash. Generally, however, it is better in such stubborn
cases to fuse with carbonate of soda, and then attack the "melt" with
hydrochloric acid.

When nitric acid, or the fusion method, has been used, the metal will be
in solution in the ferric state, no matter in what condition it existed
in the ore. But with dilute hydrochloric or sulphuric acid it will
retain its former degree of oxidation. Hydrochloric acid, for example,
with chalybite (ferrous carbonate) will give a solution of _ferrous_
chloride; with hæmatite (ferric oxide) it will yield _ferric_ chloride;
and with magnetite (ferrous and ferric oxides) a mixture of ferrous and
ferric chlorides. Metallic iron yields solutions of _ferrous_ salts. It
is convenient to speak of the iron in a ferrous salt as ferrous iron,
and when in the ferric state as ferric iron. Frequently it is required
to determine how much of the iron exists in an ore in each condition. In
such cases it is necessary to keep off the air whilst dissolving; the
operation should, therefore, be performed in an atmosphere of carbonic
acid.

~Separation.~--The separation of the iron from the other substances is
as follows:--Silica is removed by evaporating the acid solution, and
taking up with acid, as described under _Silica_; the whole of the iron
will be in solution. The metals of Groups I. and II. are removed by
passing sulphuretted hydrogen, and at the same time the iron will be
reduced to the ferrous state. The solution should be filtered into a 16
oz. flask, boiled to get rid of the gas, and treated (whilst boiling)
with a few drops of nitric acid, in order to convert the whole of the
iron into the ferric state. When this condition is arrived at, an
additional drop of nitric acid causes no dark coloration. The boiling
must be continued to remove nitrous fumes. Next add caustic soda
solution until the colour of the solution changes from yellow to red.
The solution must be free from a precipitate; if the soda be
incautiously added a permanent precipitate will be formed, in which case
it must be redissolved with hydrochloric acid, and soda again, but more
cautiously, added. After cooling, a solution of sodium acetate is added
until the colour of the solution is no longer darkened. The solution,
diluted to two-thirds of the flaskful with water, is heated to boiling.
Long-continued boiling must be avoided. The precipitate is filtered
quickly through a large filter, and washed with hot water containing a
little acetate of soda.

The precipitate will contain all the iron and may also contain alumina,
chromium, titanium, as well as phosphoric, and, perhaps, arsenic
acids.[64]

Dissolve the precipitate off the filter with dilute sulphuric acid,
avoiding excess, add tartaric acid and then ammonia in excess. Pass
sulphuretted hydrogen, warm, and allow the precipitate to settle. Filter
and wash with water containing a little ammonic sulphide.


GRAVIMETRIC METHOD.

Dissolve the precipitate in dilute hydrochloric acid; peroxidise with a
few drops of nitric acid and boil, dilute to about 200 c.c., add ammonia
(with constant stirring) till the liquid smells of it, and heat to
boiling. Wash as much as possible by decantation with hot water.
Transfer to the filter, and wash till the filtrate gives no indication
of soluble salts coming through. The filtrate must be colourless and
clear. The wet precipitate is very bulky, of a dark-brown colour and
readily soluble in dilute acids, but insoluble in ammonia and dilute
alkalies. When thrown down from a solution containing other metals it is
very apt to carry portions of these with it, even when they are by
themselves very soluble in ammoniacal solutions. It must be dried and
ignited, the filter paper being burnt separately and its ash added. When
further ignition ceases to cause a loss of weight, the residue is ferric
oxide (Fe_{2}O_{3}), which contains 70 per cent. of iron. The weight of
iron therefore can be calculated by multiplying the weight of oxide
obtained by 0.7.

The presence of ammonic chloride causes loss of iron during the
ignition, and organic matter causes an apparent loss by reducing the
iron to a lower state of oxidation. When the iron in the solution much
exceeds 0.2 gram the volumetric determination is generally adopted, as
the bulkiness of the precipitate of ferric hydrate makes the gravimetric
method very inconvenient.


VOLUMETRIC METHODS.

As already explained these are based on the measurement of the volume of
a reagent required to bring the whole of the iron from the ferrous to
the ferric state (oxidation), or from the ferric to the ferrous
(reduction). Ferrous compounds are converted into ferric by the action
of an oxidising agent in the presence of an acid. Either permanganate or
bichromate of potash is generally used for this purpose.[65]

Ferric compounds are reduced to ferrous by the action of:--

  (1)  Stannous chloride;
  (2)  Sulphuretted hydrogen;
  (3)  Sodium sulphite; or
  (4)  Zinc.[66]

The processes, then, may be divided into two kinds, one based on
oxidation and the other on reduction. In each case the titration must be
preceded by an exact preparation of the solution to be assayed in order
that the iron may be in the right state of oxidation.


~PERMANGANATE AND BICHROMATE METHODS.~

These consist of three operations:--

  (1) Solution of the ore;
  (2) Reduction of the iron to the ferrous state; and
  (3) Titration.

~Solution.~--The only point to be noticed concerning the first operation
(in addition to those already mentioned) is that nitric acid must be
absent. If nitric acid has been used, evaporate to dryness, of course
without previous dilution; add hydrochloric or sulphuric acid, and boil
for five or ten minutes. Dilute with water to about 100 c.c., and warm
until solution is complete.

The reduction is performed by either of the following methods:--

1. _With Stannous Chloride._--Fill a burette with a solution of stannous
chloride,[67] and cautiously run the liquid into the hot assay solution
(in which the iron is present as _chloride_) until the colour is
discharged. A large excess of the stannous chloride must be avoided.
Then add 5 c.c. of a 2-1/2 per cent. solution of mercuric chloride, this
will cause a white precipitate (or a grey one if too large an excess of
the stannous chloride has been added). Boil till the solution clears,
cool, dilute, and titrate.

2. _With Sulphuretted Hydrogen._--Cool the solution and pass through it
a current of washed sulphuretted hydrogen till the liquid smells
strongly of the gas after withdrawal and shaking. A white precipitate of
sulphur will be formed, this will not interfere with the subsequent
titration provided it is precipitated in the cold. If, however, the
precipitate is coloured (showing the presence of the second group
metals), or if the precipitation has been carried out in a hot solution,
it should be filtered off. Boil the solution until the sulphuretted
hydrogen is driven off; this may be tested by holding a strip of filter
paper dipped in lead acetate solution in the steam issuing from the
flask. The presence of sulphuretted hydrogen should be looked for rather
than its absence. It is well to continue the boiling for a few minutes
after the gas has been driven off. Cool and titrate.

3. _With Sodium Sulphite._--Add ammonia (a few drops at a time) until
the precipitate first formed redissolves with difficulty. If a permanent
precipitate is formed, redissolve with a few drops of acid. To the warm
solution add from 2 to 3 grams of sodium sulphite crystals. The solution
will become strongly coloured, but the colour will fade away on standing
for a few minutes in a warm place. When the colour is quite removed, add
20 c.c. of dilute sulphuric acid, and boil until the steam is quite free
from the odour of sulphurous acid. Cool and titrate.

4. _With Zinc._--Add about 10 grams of granulated zinc; if the hydrogen
comes off violently add water; if, on the other hand, the action is very
slow, add sufficient dilute sulphuric acid to keep up a brisk
effervescence. The reduction is hastened by warming, and is complete
when the solution is quite colourless and a drop of the liquid tested
with sulphocyanate of potassium gives no reaction for ferric iron.
Filter through "glass wool" or quick filtering paper. The zinc should be
still giving off gas rapidly, indicating a freely acid solution; if not,
acid must be added. Wash with water rendered acid. Cool and titrate.

With regard to the relative advantages of the different methods they may
be roughly summed up as follows:--The stannous chloride method has the
advantage of immediately reducing the ferric iron whether in hot or cold
solution and under varied conditions in regard to acidity, but has the
disadvantage of similarly reducing salts of copper and antimony, which,
in a subsequent titration, count as iron. Moreover, there is no
convenient method of eliminating any large excess of the reagent that
may have been used; and, consequently, it either leaves too much to the
judgment of the operator, or entails as much care as a titration.
Students generally get good results by this method.

The sulphuretted hydrogen method also has the advantage of quick
reduction under varying conditions, and the further one of adding
nothing objectionable to the solution; in fact it removes certain
impurities. The disadvantages are the necessity for boiling off the
excess of the gas, and of filtering off the precipitated sulphur,
although this last is not necessary if precipitated cold. The tendency
with students is to get high results. The sodium sulphite method has the
advantages of being clean and neat, and of requiring no nitration. On
the other hand it requires practice in obtaining the best conditions for
complete reduction; and, as with sulphuretted hydrogen, there is the
necessity for boiling off the gas, while there is no simple and delicate
test for the residual sulphurous acid. In addition, if an excess of
sodium sulphite has been used and enough acid not subsequently added,
the excess will count as iron. Students generally get low results by
this method.

The advantages of the zinc method are, that it is easily worked and that
the excess of zinc is readily removed by simply filtering. The
disadvantages are the slowness[68] with which the last portions of
ferric iron are reduced, the danger of loss by effervescence, the
precipitation of basic salts, and, perhaps, of iron, and the loading of
the solution with salts of zinc, which in the titration with bichromate
have a prejudicial effect. The tendency in the hands of students is to
get variable results, sometimes low and sometimes high.

Generally speaking, the sulphuretted hydrogen and sodium sulphite
methods are to be preferred. Carefully worked each method will yield
good results.

The titration may be done with a standard solution of (1) permanganate
of potash, or (2) bichromate of potash.

1. _With Permanganate of Potash._--Prepare a standard solution by
dissolving 2.82 grams of the salt and diluting to one litre. The
strength of this should be 100 c.c. = 0.5 gram of iron, but it varies
slightly, and should be determined (and afterwards checked every two or
three weeks) by weighing up 0.2 gram of iron wire, dissolving in 10 c.c.
of dilute sulphuric acid, diluting to about 100 c.c., and titrating.

The standard solution must be put in a burette with a glass stopcock, as
it attacks india-rubber. The assay should be contained in a pint flask,
and be cooled before titrating. The standard solution must be run in
until a pinkish tinge permeates the whole solution; this must be taken
as the finishing point. When certain interfering bodies are present this
colour quickly fades, but the fading must be ignored. With pure
solutions the colour is fairly permanent, and a single drop of the
potassium permanganate solution is sufficient to determine the finishing
point.

2. _With Bichromate of Potash._--Prepare a standard solution by
dissolving 4.39 grams of the powdered and dried salt in water, and
diluting to 1 litre. This solution is permanent, its strength is
determined by dissolving 0.2 gram of iron wire in 10 c.c. of dilute
sulphuric acid, diluting to about a quarter of a litre, and titrating.

Also prepare a test solution by dissolving 0.1 gram of ferricyanide of
potassium in 100 c.c. of water. This solution does not keep well and
must be freshly prepared.

An ordinary burette is used. The assay is best contained in a glazed
earthenware dish, and may be titrated hot or cold. To determine the
finishing point, place a series of drops of the ferricyanide solution on
a dry white glazed plate. The drops should be of about the same size and
be placed in lines at fairly equal distances. The bichromate is run in,
in a steady stream, the assay solution being continuously stirred until
the reaction is sensibly slackened. Then bring a drop of the assay with
the stirrer in contact with one of the test drops on the plate. The
standard can be safely run in 1 c.c. at a time, so long as the test drop
shows signs of a precipitate. When only a coloration is produced run in
cautiously a few drops at a time so long as two drops of the assay gives
with the test a colour which is even faintly greener than two drops of
the assay solution placed alongside. The finishing point is decided and
practically permanent, although it demands a little practice to
recognise it. The titration with permanganate of potassium has the
advantage of a more distinct finishing point and easier mode of working;
its application, however, is somewhat limited by the disturbing effects
of hydrochloric acid. The bichromate method has the advantage of a
standard solution which does not alter in strength, and the further one
of being but little affected by altering conditions of assay.
Hydrochloric acid has practically no effect on it. Both methods give
accurate results and are good examples of volumetric methods.

The following results illustrate the extent to which the methods may be
relied on; and the influence which the various conditions of experiment
have on the assay.

Solutions of ferrous sulphate and of ferrous chloride were made
containing 0.5 gram of iron in each 100 c.c., thus corresponding to the
standard solutions of permanganate and bichromate of potassium. These
last were prepared in the way already described. The solution of ferrous
sulphate was made by dissolving 5.01 grams of iron wire in 100 c.c. of
dilute sulphuric acid and diluting to 1 litre. A similar solution may be
made by dissolving 24.82 grams of pure ferrous sulphate crystals in
water, adding 100 c.c. of dilute sulphuric acid, and diluting to 1
litre.

~Rate of Oxidation by Exposure to Air.~--This is an important
consideration, and if the rate were at all rapid would have a serious
influence on the manner of working, since exclusion of air in the
various operations would be troublesome. 20 c.c. of the solution of
ferrous sulphate were taken in each experiment, acidified with 10 c.c.
of dilute sulphuric acid, and diluted to 100 c.c. The solution was
exposed, cold, in an open beaker for varying lengths of time, and
titrated with permanganate of potassium.

  Time exposed    1 hour  1 day   2 days  3 days
  c.c. required    19.2    19.1    19.0    19.0

These results show that the atmospheric oxidation in cold solutions is
unimportant. With boiling solutions the results are somewhat different;
a solution which at the outset required 20 c.c. of permanganate of
potassium, after boiling for an hour in an open beaker (without any
precautions to prevent oxidation), water being added from time to time
to replace that lost by evaporation, required 19.2 c.c. If the solution
be evaporated to dryness the oxidising power of concentrated sulphuric
acid comes into play, so that very little ferrous iron will be left. A
solution evaporated in this way required only 2.2 c.c. of permanganate
of potassium.

~Effect of Varying Temperature.~--In these experiments the bulk was in
each case 100 c.c., and 10 c.c. of dilute sulphuric acid were present.
The permanganate required by

    1 c.c. of ferrous sulphate was, at 15° 1.0 c.c., and at 70° 1.1 c.c.
   10  "            "            "         9.7           "      9.8   "
  100  "            "            "        97.7           "     96.8   "

The lower result with the 100 c.c. may be due to oxidation from
exposure.

~Effect of Varying Bulk.~--The following experiments show that
considerable variations in bulk have no practical effect. In each case
20 c.c. of ferrous sulphate solution and 10 c.c. of dilute acid were
used.

  Bulk of assay           30 c.c.   100 c.c.   500 c.c.   1000 c.c.
  Permanganate required   20.0 "     20.0 "     20.2 "      20.5 "

~Effect of Free Sulphuric Acid.~--Free acid is necessary for these
assays; if there is an insufficiency, the assay solution, instead of
immediately decolorising the permanganate, assumes a brown colour. The
addition of 10 c.c. of dilute sulphuric acid suffices to meet
requirements and keep the assay clear throughout. The following
experiments show that a considerable excess of acid may be used without
in the least affecting the results. In each case 20 c.c. of ferrous
sulphate were used.

Dilute sulphuric
  acid       1.0 c.c.   5.0 c.c.  10.0 c.c.  20.0 c.c.  50.0 c.c.  100.0 c.c.
Permanganate
  required   19.3  "    19.3  "    19.3  "    19.3  "    19.3   "    19.3  "

~Effect of Foreign Salts.~--When the assay has been reduced with zinc
varying quantities of salts of this metal pass into solution, the amount
depending on the quantity of acid and iron present. Salts of sodium or
ammonium may similarly be introduced. It is essential to know by
experiment that these salts do not exert any effect on the titration.
The following series of experiments show that as much as 50 grams of
zinc sulphate may be present without interfering.

  Zinc sulphate present    0 gram       1 gram      10 grams    50 grams
  Permanganate required   19.3 c.c.    19.3 c.c.    19.3 c.c.   19.3 c.c.

Magnesium, sodium, and ammonium salts, are equally without effect.

  Ammonic sulphate present     0 gram      1 gram    10 grams
  Permanganate required       19.3 c.c.   19.2 c.c.  19.3 c.c.

Sodic sulphate present       0 gram      1 gram    10 grams
Permanganate required       19.3 c.c.   19.3 c.c.  19.3 c.c.

Magnesic sulphate present    0 gram      1 gram    10 grams
Permanganate required       19.3 c.c.   19.3 c.c.  19.3 c.c.

~Effect of Varying Amounts of Iron.~--It is important to know within
what limits the quantity of iron in an assay may safely vary from that
used in standardising. In the following experiments the conditions as to
bulk, acidity, and mode of working were the same as before:--

  Ferrous sulphate solution taken  1 c.c. 10 c.c. 20 c.c. 50 c.c. 100 c.c.
  Permanganate required            1.0 "   9.7 "  19.6 "  48.9 "   97.7 "

The ferrous sulphate solution is here a little weaker than that of the
permanganate of potassium, but the results show that the permanganate
required is proportional to the iron present.

~Titrations in Hydrochloric Solutions.~--These are less satisfactory
than those in sulphuric solutions, since an excess of hydrochloric acid
decomposes permanganate of potassium, evolving chlorine, and since the
finishing point is indicated, not by the persistence of the pink colour
of the permanganate, but by a brown coloration probably due to
perchloride of manganese. Nevertheless, if the solution contains only
from 5 to 10 per cent. of free hydrochloric acid (sp. g. 1.16) the
results are the same as those obtained in a sulphuric acid solution.
Equal weights (0.1 gram) of the same iron wire required exactly the same
quantity of the permanganate of potassium solution (20 c.c.) whether the
iron was dissolved in dilute sulphuric or dilute hydrochloric acid. The
following series of experiments are on the same plan as those given
above with sulphuric acid solutions. A solution of ferrous chloride was
made by dissolving 5.01 grams of iron wire in 50 c.c. of dilute
hydrochloric acid and diluting to 1 litre. The dilute hydrochloric acid
was made by mixing equal volumes of the acid (sp. g. 1.16) and water.

~Rate of Atmospheric Oxidation.~--20 c.c. of the ferrous chloride
solution were acidified with 10 c.c. of the dilute hydrochloric acid and
diluted to 100 c.c. This solution was exposed cold in open beakers.

  Time exposed              --      1 hour    1 day     2 days    3 days
  Permanganate required  19.6 c.c. 19.6 c.c. 19.5 c.c. 19.4 c.c. 19.5 c.c.

Similar solutions boiled required, before boiling, 20 c.c.; after
boiling for one hour, replacing the water as it evaporated, 19.3 c.c.;
and after evaporation to a paste and redissolving, 17.0 c.c.

~Effect of Varying Temperature.~--Solutions similar to the last were
titrated and gave the following results:--

  Temperature            15°        30°        50°          70°
  Permanganate required  19.8 c.c.  19.6 c.c.  19.5 c.c.    19.4 c.c.

~Effect of Varying Bulk.~--As before, 20 c.c. of the iron solution, and
10 c.c. of the dilute acid were diluted to the required volumes and
titrated.

  Bulk                     30 c.c.    100 c.c.    500 c.c.    1000 c.c.
  Permanganate required    20.4 "      20.3 "      20.8 "       21.5 "

The variation due to difference in bulk here, although only equal to an
excess of 0.7 milligram of iron for each 100 c.c. of dilution, are about
three times as great as those observed in a sulphuric acid solution.

~Effect of Free Hydrochloric Acid.~--In these experiments 20 c.c. of the
ferrous chloride solution were used with varying quantities of acid, the
bulk of the assay in each case being 100 c.c.

  Dilute acid present      5 c.c.  10 c.c.  50 c.c.  100 c.c.
  Permanganate required   20.2 "   20.2 "   20.5 "    21.0 "

The last had a very indistinct finishing point, the brown coloration
being very evanescent. The effect of the acid is modified by the
presence of alkaline and other sulphates, but not by sulphuric acid.
Repeating the last experiment we got--

  Without further addition                 21.0 c.c.
  With 100 c.c. of dilute sulphuric acid   22.0  "
   "   10 grams ammonic sulphate           20.5  "
   "   10   "   sodic sulphate             20.0  "
   "   10   "   magnesium sulphate         20.4  "
   "   10   "   manganese sulphate         20.2  "

The results with these salts, in counteracting the interference of the
acid, however, were not a complete success, since the end-reactions were
all indistinct, with the exception, perhaps, of that with the manganese
sulphate.

~Effect of Varying Amounts of Iron.~--In these experiments the bulk of
the assay was 100 c.c., and 10 c.c. of acid were present.

  Ferrous chloride used   1 c.c.  10 c.c.  20 c.c.  50 c.c.  100 c.c.
  Permanganate required   1.1 "   10.3 "   20.3 "   50.4 "   100.1 "

In making himself familiar with the permanganate of potassium titration,
the student should practise by working out a series of experiments
similar to the above, varying his conditions one at a time so as to be
certain of the cause of any variation in his results. He may then
proceed to experiment on the various methods of reduction.

_A solution of ferric chloride_ is made by dissolving 5.01 grams of iron
wire in 50 c.c. of hydrochloric acid (sp. g. 1.16), and running from a
burette nitric acid diluted with an equal volume of water into the
boiling iron solution, until the liquid changes from a black to a
reddish-yellow. About 4.5 c.c. of the nitric acid will be required, and
the finishing point is marked by a brisk effervescence. The solution of
iron should be contained in an evaporating dish, and boiled briskly,
with constant stirring. There should be no excess of nitric acid. Boil
down to about half its bulk; then cool, and dilute to one litre with
water. Twenty c.c. of this solution diluted to 100 c.c. with water, and
acidified with 10 c.c. of dilute hydrochloric acid, should not
decolorise any of the permanganate of potassium solution; this shows the
absence of ferrous salts. And 20 c.c. of the same solution, boiled with
20 c.c. of the ferrous sulphate solution, should not decrease the
quantity of "permanganate" required for the titration of the ferrous
sulphate added. In a series of experiments on the various methods of
reduction, the following results were got. The modes of working were
those already described.

(1) _With Stannous Chloride._--Twenty c.c. of the ferric chloride
solution required, after reduction with stannous chloride, 20 c.c. of
"permanganate." Fifty c.c. of a solution of ferrous chloride, which
required on titration 49.8 c.c. of "permanganate," required for
re-titration (after subsequent reduction with stannous chloride) 50 c.c.
of the permanganate solution.

(2) _With Sulphuretted Hydrogen._--Two experiments with this gas, using
in each 20 c.c. of the ferric chloride solution, and 10 c.c. of
hydrochloric acid, required (after reduction) 20.2 c.c. and 20.1 c.c. of
"permanganate." Repeating the experiments by passing the gas through a
nearly boiling solution, but in other respects working in the same way,
21.3 c.c. and 21.6 c.c. of the permanganate solution were required. The
sulphur was not filtered off in any of these. In another experiment, in
which 50 c.c. of the ferrous sulphate solution were titrated with
"permanganate," 48 c.c. of the latter were required. The titrated
solution was next reduced with sulphuretted hydrogen, brought to the
same bulk as before, and again titrated; 47.9 c.c. of the permanganate
of potassium solution were required.

(3) _With Sodium Sulphite._--Twenty c.c. of the ferric chloride
solution, reduced with sodium sulphite, required 19.9 c.c. of
"permanganate." In one experiment 50 c.c. of the ferrous sulphate
solution were titrated with "permanganate"; 49.3 c.c. of the
last-mentioned solution were required. The titrated solution was reduced
with sodium sulphite, and again titrated; it required 49.2 c.c. of the
permanganate of potassium solution.

(4) _With Zinc._--Twenty c.c. of the ferric chloride solution, reduced
with zinc and titrated, required 20.8 c.c. of "permanganate." Fifty c.c.
of a solution of ferrous sulphate which required 49.7 c.c. of
"permanganate," required for re-titration, after reduction with zinc,
49.7 c.c.

The student should next practise the titration with bichromate, which is
more especially valuable in the estimation of hydrochloric acid
solutions. The following experiments are on the same plan as those
already given. In each experiment (except when otherwise stated) there
were present 20 c.c. of the ferrous chloride solution, and 10 c.c. of
dilute hydrochloric acid, and the bulk was 300 c.c.

~Effect of Varying Temperature.~--The quantities of the bichromate of
potassium solution required were as follows:--

  Temperature           15°        30°        70°        100°
  Bichromate required   20.2 c.c.  20.3 c.c.  20.3 c.c.  20.4 c.c.

~Effect of Varying Bulk.~--

  Bulk                  50 c.c.  100 c.c.  200 c.c.  500 c.c.  1000 c.c.
  Bichromate required   20.4 "    20.4 "    20.4 "    20.5 "     20.8 "

~Effect of Varying Acid.~--In these, variable quantities of dilute
hydrochloric acid were used.

  Acid present          10 c.c.  50 c.c.  100 c.c.
  Bichromate required   20.3 "   20.3 "    20.2 "

~Effect of Foreign Salts.~--The effect of the addition of 10 grams of
crystallized zinc sulphate was to decrease the quantity of "bichromate"
required from 20.3 c.c. to 20.1 c.c., but the colour produced with the
test-drop was very slight at 18.5 c.c., and with incautious work the
finishing point might have been taken anywhere between these extremes.
Zinc should not be used as a reducing agent preliminary to a
"bichromate" titration. Ten grams of ammonic sulphate had the effect of
rendering the finishing point faint for about 0.5 c.c. before the
titration was finished, but there was no doubt about the finishing point
when allowed to stand for a minute. The student should note that a
titration is not completed if a colour is developed on standing for five
or ten minutes. Ten grams of sodic sulphate had no effect; 20.3 c.c.
were required.

~Effect of Varying Iron.~--The results are proportional, as will be seen
from the following details:--

  Ferrous chloride
    present             1.0 c.c. 10.0 c.c. 20.0 c.c. 50.0 c.c. 100.0 c.c.
  Bichromate required   1.0  "   10.2  "   20.3  "   51.0  "   102.3  "

The student may now apply these titrations to actual assays of minerals.
The following examples will illustrate the mode of working and of
calculating the results:--

~Determination of Iron in Chalybite.~--Weigh up 1 gram of the dry
powdered ore, and dissolve in 10 c.c. of dilute sulphuric acid and an
equal volume of water with the aid of heat. Avoid evaporating to
dryness. Dilute and titrate. The result will give the percentage of iron
existing in the ore in the ferrous state. Some ferric iron may be
present. If it is wished to determine this also, add (in dissolving
another portion) 10 c.c. of dilute hydrochloric acid to the sulphuric
acid already ordered, and reduce the resulting solution before
titrating. By dissolving and titrating (without previous reduction) one
has a measure of the ferrous iron present; by dissolving, reducing, and
then titrating, one can measure the total iron; and as the iron exists
in only two conditions, the total iron, less the ferrous iron, is the
measure of the ferric iron.

~Determination of Iron in Brown or Red Ores or Magnetite.~--Weigh up 0.5
gram of the ore (powdered and dried at 100° C.), and dissolve in from 10
to 20 c.c. of strong hydrochloric acid, boiling until all is dissolved,
or until no coloured particles are left. Dilute, reduce, and titrate.

~Determination of Iron in Pyrites.~--Weigh up 1 gram of the dry powdered
ore, and place in a beaker. Cover with 10 c.c. of strong sulphuric acid,
mix well by shaking, and place on the hot plate without further handling
for an hour or so until the action has ceased. _Allow to cool_, and
dilute to 100 c.c. Warm until solution is complete. Reduce and titrate.

~Determination of Iron in Substances Insoluble in Acids.~--Weigh up 1
gram of the ore, mix with 5 or 6 grams of carbonate of soda and 0.5 gram
of nitre by rubbing in a small mortar, and transfer to a platinum
crucible. Clean out the mortar by rubbing up another gram or so of soda,
and add this to the contents of the crucible as a cover. Fuse till
tranquil. Cool. Extract with water. If the ore carries much silica,
evaporate to dryness with hydrochloric acid to separate it. Re-dissolve
in hydrochloric acid, and separate the iron by precipitating with
ammonia and filtering. If only a small quantity of silica is present,
the aqueous extract of the "melt" must be filtered, and the insoluble
residue washed and dissolved in dilute hydrochloric acid. Reduce and
titrate.

A convenient method of at once separating iron from a solution and
reducing it, is to add ammonia, pass sulphuretted hydrogen through it,
filter, and dissolve the precipitate in dilute sulphuric acid. The
solution, when boiled free from sulphuretted hydrogen, is ready for
titrating.


STANNOUS CHLORIDE PROCESS.

The colour imparted to hot hydrochloric acid solutions by a trace of a
ferric compound is so strong, and the reducing action of stannous
chloride is so rapid, that a method of titration is based upon the
quantity of a standard solution of stannous chloride required to
completely decolorise a solution containing ferric iron. This method is
more especially adapted for the assay of liquors containing much ferric
iron and of those oxidised ores which are completely soluble in
hydrochloric acid. It must be remembered, however, that it only measures
the ferric iron present, and when (as is generally the case) the total
iron is wanted, it is well to calcine the weighed portion of ore
previous to solution in order to get the whole of the iron into the
higher state of oxidation, since many ores which are generally supposed
to contain only ferric iron carry a considerable percentage of ferrous.

_The stannous chloride solution_ is made by dissolving 20 grams of the
commercial salt (SnCl_{2}.2H_{2}O) in 100 c.c. of water with the help of
20 c.c. of dilute hydrochloric acid, and diluting to a litre. The
solution may be slightly opalescent, but should show no signs of a
precipitate. The strength of this is about equivalent to 1 gram of iron
for each 100 c.c. of the solution, but it is apt to lessen on standing,
taking up oxygen from the air, forming stannic chloride. A larger
proportion of hydrochloric acid than is ordered above would remove the
opalescence, but at the same time increase this tendency to atmospheric
oxidation, as the following experiments show. The stannous chloride
solution (20 c.c.) was mixed with varying amounts of strong hydrochloric
acid (sp. g. 1.16), diluted to 100 c.c., and exposed in open beakers for
varying lengths of time; and the residual stannous chloride measured by
titration with permanganate. The quantities required were as follows:--

  Time Exposed.  50 per cent. Acid.  10 per cent. Acid.  1 per cent. Acid.
     1 hour          33.2 c.c.            34.4 c.c.          34.5 c.c.
     1 day            5.0  "              24.0  "            27.6  "
     2 days           3.0  "              14.5  "            21.3  "

These indicate very clearly the increased susceptibility to oxidation in
strongly acid solutions.

_A standard solution of ferric chloride_ is prepared in the same manner
as that described under the experiments on the methods of reduction; but
it should be of twice the strength, so that 100 c.c. may contain 1 gram
of iron. This solution is used for standardising the stannous chloride
when required; and must be carefully prepared; and tested for the
presence of nitric acid.

The titration is more limited in its application than either of the
oxidising processes because of the restrictions as to bulk, quality and
quantity of free acid present, and other conditions of the solution to
be assayed. The following experiments show the conditions necessary for
a successful titration.

~Effect of Varying Temperature.~--Twenty c.c. of ferric chloride
solution with 20 c.c. of strong hydrochloric acid, diluted to 50 c.c.,
gave the following results when titrated:--

  Temperature                 15°        30°        70°        100°
  Stannous chloride required  22.8 c.c.  22.0 c.c.  22.1 c.c.   22.0 c.c.

The finishing point, however, is more distinct the hotter the solution;
so that it is best in all cases to run the standard into the boiling
solution.

~Effect of Varying Bulk.~--Solutions containing the same quantity of
iron and acid as the last, but diluted to various bulks, and titrated
while boiling, gave the following results:--

  Bulk                          30 c.c.   100 c.c.   500 c.c.
  Stannous chloride required    21.5 "     21.7 "     24.3 "

~Effect of Varying Quantities of Hydrochloric Acid.~--In these
experiments the bulk before titration was 50 c.c. except in the last, in
which it was 70 c.c. With less than 5 c.c. of strong hydrochloric acid
the finishing point is indistinct and prolonged.

  Strong hydrochloric
    acid present         5 c.c.  10 c.c.  20 c.c.  30 c.c.  50 c.c.
  Stannous chloride
    required            21.1 "   21.1 "   21.2 "   21.8 "   22.2 "

~Effect of Free Sulphuric Acid.~--In these experiments 20 c.c. of
hydrochloric acid were present, and the bulk was 50 c.c.

  Strong sulphuric acid
    present                -- c.c.   3 c.c.   5 c.c.  10 c.c.
  Stannous chloride
    required              21.6 "    22.3 "   22.9 "   23.1 "

This interference of strong sulphuric acid may be completely
counteracted by somewhat modifying the mode of working. Another
experiment, like the last of this series, required 21.6 c.c.

~Effect of Foreign Salts.~--Experiments in which 10 grams of various
salts were added showed them to be without effect. The results were as
follows:--

  Salt present                    --        AmCl     Am_{2}SO_{4}  MgCl_{2}
  Stannous chloride required  21.6 c.c.  21.6 c.c.    21.6 c.c.   21.6 c.c.

  Salt present                 CaCl_{2}    FeCl_{2}    Al_{2}Cl_{6}
  Stannous chloride required   21.8 c.c.   21.6 c.c.     21.6 c.c.

~Effect of Varying Iron.~--Titrating a solution (with 20 c.c. of
hydrochloric acid) measuring 50 c.c., and kept boiling, the quantity of
stannous chloride solution required is practically proportional to the
iron present.

  Ferric chloride
    added             1 c.c.  10 c.c.  20 c.c.  50 c.c.  100 c.c.
  Stannous chloride
    required          1.1 "   10.5 "   20.6 "   51.4 "   102.6 "

The student, having practised some of the above experiments, may proceed
to the assay of an iron ore.

~Determination of Iron in Brown Iron Ore.~--Weigh up 1 gram of the dried
and powdered ore, calcine in the cover of a platinum crucible, and
dissolve up in an evaporating dish[69] with 20 c.c. of strong
hydrochloric acid. When solution is complete, dilute to 50 c.c. after
replacing any acid that may have been evaporated. Boil, and run in the
stannous chloride solution until the colour is faintly yellow; boil
again, and continue the addition of the stannous chloride solution,
stirring continuously until the solution appears colourless. Note the
quantity of the stannous chloride solution required. Suppose this to be
59 c.c. Take 60 c.c. of the standard ferric chloride solution, add 20
c.c. of hydrochloric acid, boil and titrate in the same way as before.
Suppose this to require 61 c.c. Then as 61 is equivalent to 60 of the
iron solution, 59 is equivalent to 58.13.[70] This gives the percentage.
It is not necessary to standardise the stannous chloride solution in
this way with each sample assayed, the ratio 61: 60 would serve for a
whole batch of samples; but the standardising should be repeated at
least once each day.


COLORIMETRIC METHOD.

This method is valuable for the determination of small quantities of
iron present as impurities in other metals or ores. It is based on the
red coloration developed by the action of potassic sulphocyanate on acid
solutions of ferric salts.

_Standard Ferric Chloride Solution._--Take 1 c.c. of the ferric chloride
solution used for standardising the stannous chloride solution, add 2
c.c. of dilute hydrochloric acid, and dilute to 1 litre with water. 1
c.c. = 0.01 milligram.

_Solution of Potassic Sulphocyanate._--Dissolve 60 grams of the salt in
water, and dilute to a litre. It should be colourless. Use 10 c.c. for
each test.

The quantity of the substance to be weighed for the assay should not
contain more than a milligram of iron; consequently, if the ore contain
more than 0.1 per cent. of that metal, less than a gram of it must be
taken.

The method is as follows:--Weigh up 1 gram of the substance and dissolve
in a suitable acid; dilute; and add permanganate of potash solution
until tinted. Boil for some time and dilute to 100 c.c. Take a couple of
Nessler tubes, holding over 100 c.c., but marked at 50 c.c.; label them
"1" and "2"; and into each put 10 c.c. of the potassic sulphocyanate
solution and 2 c.c. of dilute hydrochloric acid. The solutions should be
colourless. To "1" add 10 c.c. of the assay solution, and dilute to the
50 c.c. mark. To the other add water, but only to within 5 or 10 c.c. of
this mark. Now run in the standard ferric chloride solution from a small
burette, 1 c.c. at a time, stirring after each addition till the colour
is nearly equal to that of the assay (No. 1). At this stage bring the
solution to the same level by diluting, and make a further addition of
the standard ferric chloride solution till the colours correspond. The
amount of iron will be the same in each tube; that in the standard may
be known by reading off the volume from the burette and multiplying by
0.01 milligram.

If the 10 c.c. of the assay solution gave a colour requiring more than 5
or 6 c.c. of the standard ferric chloride solution, repeat the
determination, taking a smaller proportion.

The effect of varying conditions on the assay will be seen from the
following experiments:--

~Effect of Varying Temperature.~--The effect of increase of temperature
is to lessen the colour; in fact, by boiling, the colour can be entirely
removed. All assays are best carried out in the cold.

  1 c.c. at 15° would only show the colour of 0.75 c.c. at 45°
  2     "         "         "         "       1.75       "
  5     "         "         "         "       4.0        "

~Effect of Time.~--The effect of increase of time is to increase the
colour, as will be seen from the following experiments:--

  2 c.c. on standing 10 minutes became equal to 2.25 c.c.
  2       "          20    "      "      "      2.75  "
  2       "          40    "      "      "      3.00  "

~Effect of Free Acid.~--If no acid at all be present, the sulphocyanate
of potassium solution removes the colour it first produces, so that a
certain amount of acid is necessary to develop the colour. The use of a
large excess has a tendency to increase the colour produced.

5 c.c. nitric acid (sp. g. 1.4) read 3.7 c.c. instead of 2 c.c. with the
dilute acid.

5 c.c. sulphuric acid (sp. g. 1.32) read 2.2 c.c. instead of 2 c.c. with
the dilute acid.

5 c.c. hydrochloric acid (sp. g. 1.16) read 2.5 c.c. instead of 2 c.c.
with the dilute acid.


~Effect of Foreign Metals.~--Lead, mercury, cadmium, bismuth, arsenic,
tin, antimony, nickel, cobalt, manganese, aluminium, zinc, strontium,
barium, calcium, magnesium, sodium, or potassium, when separately
present in quantities of from 100 to 200 times the weight of iron
present, do not interfere if they have previously been brought to their
highest oxidised condition by boiling with nitric acid or by treating
with permanganate. Arsenic and phosphoric acids interfere unless an
excess of free hydrochloric or other acid is present. Oxalic acid (but
not tartaric acid) in minute quantities destroys the colour. Nitrous
acid strikes a red colour with the sulphocyanate of potassium;
consequently, when nitric acid has been used in excess, high results may
be obtained. Copper and some other metals interfere, so that in most
cases it is advisable to concentrate the iron before estimating it. A
blank experiment should always be made with the reagents used in order
to determine the iron, if any, introduced during the solution, &c., of
the substance assayed.

~Determination of Iron in Metallic Copper.~--This may be most
conveniently done during the estimation of the arsenic. The small
quantity of white flocculent precipitate which may be observed in the
acetic acid solution before titrating, contains the whole of the iron as
ferric arsenate. It should be filtered off, dissolved in 10 c.c. of
dilute hydrochloric acid, and diluted to 100 c.c.; 10 c.c. of this may
be taken for the estimation. For example: 10 grams of copper were taken,
and the iron estimated; 3.0 c.c. of standard ferric chloride solution
were used, equivalent to 0.03 milligram of iron; this multiplied by 10
(because only 1/10th of the sample was taken) gives 0.3 milligram as the
iron in 10 grams of copper. This equals 0.003 per cent.

In a series of experiments with this method working on 10-gram lots of
copper, to which known quantities of iron had been added, the following
were the results:--

  Iron present   0.015%   0.070%   0.100%   0.495%
  Iron found     0.015"   0.061"   0.087"   0.522"

When no arsenic is present in the copper, the iron can be separated by
fractionally precipitating with sodic carbonate, dissolving in ammonia,
and filtering off the ferric hydrate. Coppers generally carry more iron
the less arsenic they contain.

~Determination of Iron in Metallic Zinc.~--Dissolve 1 gram of zinc in 10
c.c. of dilute hydrochloric acid, adding a drop or two of nitric acid
towards the end to effect complete solution. Boil, dilute, and tint with
the permanganate of potassium solution; boil till colourless, and dilute
to 100 c.c. Take 10 c.c. for the determination. Make a blank experiment
by boiling 10 c.c. of dilute hydrochloric acid with a drop or two of
nitric acid; add a similar quantity of the permanganate of potassium
solution, boiling, &c., as before. The quantity of iron in zinc varies
from less than 0.005 to more than 2.0 per cent. When 1 gram is taken
and worked as above, each c.c. of ferric chloride solution required
indicates 0.01 per cent. of iron.

~Determination of Iron in Metallic Tin.~--Cover 1 gram of tin with 5
c.c. of hydrochloric acid, add 1 c.c. of nitric acid, and evaporate to
dryness. Take up with 2 c.c. of dilute hydrochloric acid, add 10 c.c. of
the potassic sulphocyanate solution, and make up to 50 c.c. Probably the
colour developed will be brown instead of red owing to the presence of
copper; in this case, add to the standard as much copper as the assay is
known to contain (which must have previously been determined; see
_Copper_); the titration is then carried out in the usual way.

Or the iron may be separated from the copper in the tin by the following
process:--Dissolve 5 grams of metal in 30 c.c. of hydrochloric acid and
5 c.c. of nitric acid, and evaporate to dryness. Take up with 5 c.c. of
dilute hydrochloric acid, add 10 grams of potash dissolved in 30 c.c. of
water, and warm till the tin is dissolved. Pass sulphuretted hydrogen,
boil, cool, and filter. The iron and copper will be in the precipitate.
They are separated in the ordinary manner.


PRACTICAL EXERCISES.

1. Calculate from the following determinations the percentages of
ferrous, ferric, and total iron in the sample of ore used.

1 gram of ore dissolved and titrated required 26.7 c.c. of bichromate of
potassium solution.

1 gram of ore dissolved, reduced, and titrated required 43.5 c.c. of
bichromate of potassium solution.

Standard = 1.014.

2. One gram of an ore contained 0.307 gram of ferrous iron and 0.655
gram of total iron. The iron existing as oxide, what are the percentages
of ferrous oxide (FeO) and ferric oxide (Fe_{2}O_{3}) in the ore?

3. One gram of brown iron ore dissolved in hydrochloric acid required
59.2 c.c. of stannous chloride (standard = 0.930). Another gram
dissolved in acid and titrated with "permanganate" required 8.2 c.c.
(standard = 0.4951). Calculate the percentages of ferrous, ferric, and
total iron.

4. Another gram of the same ore, roasted, dissolved and titrated with
stannous chloride, required 63.5 c.c. To what extent does this result
confirm the others?

5. Two grams of a metal were dissolved and diluted to 100 c.c. Five c.c.
were taken for a colorimetric determination, and required 4.5 c.c. of
the standard ferric chloride solution. What is the percentage of iron in
the metal?


NICKEL.

Nickel and cobalt are closely related in their chemical properties, and
may best be considered together. Nickel is the commoner of the two, and
is met with in commerce alloyed with copper and zinc as German silver;
as also in the coinage of the United States and on the Continent. It is
used for plating polished iron and steel goods, forming a coating little
liable to rust and taking a good polish. The ores of nickel are not very
common. Kupfernickel and chloanthite are arsenides of nickel with,
generally, more or less iron and cobalt. Noumeite and garnierite are
hydrated silicates of nickel and magnesia. The chief sources of nickel
are these silicates, which are found in large quantity in New Caledonia;
and a pyrites found in Norway, containing three or four per cent. of the
metal. In smaller quantities it is more widely distributed, being
frequently met with in copper ores; consequently, commercial copper is
rarely free from it.

Nickel is readily soluble in moderately concentrated nitric acid. Its
salts are mostly green, and soluble in excess of ammonia, forming blue
solutions; in these respects it resembles copper. The acid solutions,
however, are not precipitated by sulphuretted hydrogen, although in
alkaline solutions a black sulphide is formed which is insoluble in
dilute hydrochloric acid. If the sulphide is formed in a solution
containing much free ammonia, the precipitation is incomplete, some
sulphide remaining in the solution and colouring it dark brown. These
reactions serve to distinguish and separate nickel from other metals,
except cobalt. If the separated sulphide be heated in a borax bead, the
colour obtained will be a sherry brown in the outer flame, and grey or
colourless in the inner flame if nickel only is present. In the presence
of cobalt these colours are masked by the intense and characteristic
blue yielded in both flames by that metal.


DRY ASSAY.

The dry assay of nickel (cobalt being at the same time determined) is
based on the formation of a speise which will carry the cobalt, nickel,
copper, and some of the iron of the ore in combination with arsenic. A
speise of this kind, fused and exposed at a red heat to air, first loses
arsenide of iron by oxidation. It is only when the iron has been
oxidised that the arsenide of cobalt begins to be attacked; and when the
removal of the cobalt is complete, the nickel commences to pass into the
slag, the copper being left till last. The changes are rendered evident
by fusion in contact with borax. The process is as follows:--Weigh up 5
grams of the ore, and calcine thoroughly on a roasting dish in the
muffle. Rub up with some anthracite, and re-roast. Mix intimately with
from 3 to 5 grams of metallic arsenic, and heat in a small covered clay
crucible at dull redness in a muffle until no more fumes of arsenic come
off (about 15 minutes). Take out the crucible, and inject a mixture of
20 grams of carbonate of soda, 5 grams of flour, and 2 grams of fused
borax. Place in the wind furnace, and raise the temperature gradually
until the charge is in a state of tranquil fusion. Pour; when cold,
detach the button of speise, and weigh.

Weigh out carefully a portion of about 1 gram of it. Place a shallow
clay dish in the muffle, and heat it to bright redness; then add about
1.5 gram of borax glass wrapped in a piece of tissue paper; when this
has fused, drop the piece of speise into it. Close the muffle until the
speise has melted, which should be almost at once. The arsenide of iron
will oxidise first, and when this has ceased the surface of the button
brightens. Remove it from the muffle, and quench in water as soon as the
button has solidified. The borax should be coloured slightly blue.
Weigh: the loss is the arsenide of iron. Repeat the operation with the
weighed button on another dish, using rather less borax. Continue the
scorification until a film, green when cold, floating on the surface of
the button shows that the nickel is beginning to oxidise. Cool,
separate, and weigh the button as before. The loss is the arsenide of
cobalt.

If copper is absent, the speise is now arsenide of nickel.

The weight of nickel corresponding to the arsenide got is calculated by
multiplying by 0.607; and, similarly, the weight of the cobalt is
ascertained by multiplying the loss in the last scorification by
0.615.[71] It must be remembered that the nickel and cobalt so obtained
are derived from a fraction only of the speise yielded by the ore taken,
so that the results must be multiplied by the weight of the whole of the
speise, and divided by the weight of the fragment used in the
determination. As an example, suppose 5 grams of ore gave 3.3 grams of
speise, and 1.1 gram of this gave 0.8 gram of nickel arsenide. Then--

       0.8×0.607 = 0.4856 gram of nickel
  0.4856×3.3/1.1 = 1.456 gram of nickel

And this being obtained from 5 grams of ore is equivalent to 29.12 per
cent.

When copper is also present, weigh up accurately about 0.5 gram of gold,
and place it on the scorifier with the button of nickel and copper
arsenide, using borax as before. Scorify until the button shows the
bluish-green colour of a fused gold-copper alloy. Then cool, and weigh
the button of copper and gold. The increase in weight of the gold button
gives the copper as metal. The weight of the copper multiplied by 1.395
is the weight of the copper arsenide (Cu_{3}As) present. The difference
will be the nickel arsenide.

The student should enter the weighings in his book as follows:

  Ore taken  -- grams
  Speise got --   "

  Speise taken                            -- grams
  Arsenides of cobalt, nickel, and copper --  "
      "        nickel and copper          --  "
               Gold added                 --  "
               Gold and copper got        --  "
               Showing Cobalt             -- per cent.
               Nickel                     --    "
               Copper                     --    "


WET METHODS.

~Solution and Separation.~--Two or three grams of a rich ore, or 5 to 10
grams if poor, are taken for the assay. If much arsenic is present (as
is usually the case), the ore must be calcined before attacking with
acids. Transfer to a flask; and boil, first with hydrochloric acid until
the oxides are dissolved, and then with the help of nitric acid, until
nothing metalliferous is left. Dilute, nearly neutralise with soda, and
separate the iron as basic acetate,[72] as described in page 233.
Through the filtrate pass sulphuretted hydrogen till saturated. Allow to
settle (best overnight), filter, and wash. Transfer the precipitate to a
beaker, and dissolve in nitric acid. Dilute with water, pass
sulphuretted hydrogen, and filter off the precipitate, if any. Boil off
the gas, add ammonia until a precipitate is formed, and then acidify
somewhat strongly with acetic acid. Pass sulphuretted hydrogen in a slow
stream until any white precipitate of zinc sulphide, there may be,
begins to darken. Filter; to the filtrate add ammonia, and pass
sulphuretted hydrogen. The precipitate will contain the nickel and
cobalt as sulphides.

Where small quantities of nickel and cobalt are present, and an
approximate determination is sufficient, they can be concentrated as
follows:--Remove the copper, &c., by passing sulphuretted hydrogen
through the acid solution and filtering; add ammonia to the filtrate,
and again pass sulphuretted hydrogen; then heat nearly to boiling, and
filter. Dissolve the precipitate off the filter with dilute hydrochloric
acid; the residue will contain nearly all the nickel and cobalt as
sulphides.

~Separation of Nickel and Cobalt.~--Dissolve the sulphides separated as
above in nitric acid; render alkaline with a solution of potash, then
acidify with acetic acid; add a concentrated solution of _nitrite_ of
potash. The liquid after this addition must have an acid reaction. Allow
to stand for 24 hours in a warm place. Filter off the yellow precipitate
of nitrite of potash and cobalt, and wash with a 10 per cent. solution
of acetate of potash. The cobalt is determined in the precipitate in the
way described under _Cobalt_. The nickel is separated from the solution
by boiling with sodic hydrate, filtering, and dissolving the precipitate
in nitric acid. The solution will contain the nickel.


GRAVIMETRIC DETERMINATION.

The solution, which contains the nickel free from other metals, is
heated, and a solution of sodic hydrate added in slight excess. The
precipitate is filtered off, washed with boiling water, dried, ignited
at a red heat, and weighed when cold. The ignited substance is nickel
oxide (NiO), and contains 78.67 per cent. of nickel. The oxide is a
green powder, readily and completely soluble in hydrochloric acid, and
without action on litmus paper. It is very easily reduced by ignition in
hydrogen to metallic nickel.

[Illustration: FIG. 56.]

Nickel is also determined by electrolysis, as follows:--The nitric acid
solution is rendered strongly ammoniacal, and placed under the
electrolytic apparatus used for the copper assay. Three cells (fig. 56),
however, must be used, coupled up for intensity, that is, with the zinc
of one connected with the copper of the next. The electrolysis is
allowed to go on overnight, and in the morning the nickel will be
deposited as a bright and coherent film. A portion of the solution is
drawn off with a pipette; if it smells of ammonia, has no blue colour,
and gives no precipitate with ammonic sulphide, the separation is
complete. Wash the cylinder containing the deposited metal, first with
water and then with alcohol, as in the copper assay. Dry in the water
oven, and weigh. The increase in weight is metallic nickel.

As an example:--There was taken 1 gram of a nickel alloy used for
coinage. It was dissolved in 10 c.c. of nitric acid, and diluted to 100
c.c. with water. The copper was then precipitated by electrolysis. It
weighed 0.734 gram. The solution, after electrolysis, was treated with
sulphuretted hydrogen, and the remaining copper was thrown down as
sulphide, and estimated colorimetrically. This amounted to 3-1/2
milligrams. The filtrate was evaporated, treated with ammonia, warmed,
and filtered. The ferric hydrate was dissolved in dilute acid, and
reprecipitated, dried, ignited, and weighed. Its weight was 0.0310 gram.
The two filtrates were mixed, and reduced in bulk to about 50 c.c.; a
considerable excess of ammonia was added, and the nickel precipitated by
electrolysis. It weighed 0.2434 gram. These quantities are equivalent
to:

  Copper   73.75 per cent.
  Nickel   24.34    "
  Iron      2.17    "
          ------
          100.26


VOLUMETRIC DETERMINATION.

An alkaline solution of potassium cyanide, to which a little potassium
iodide has been added, can be assayed for its strength in cyanide by
titrating with a standard solution of silver nitrate. Nickel interferes
with this assay, doing the work of its equivalent of silver; and the
quantity of nickel present can be calculated from the amount of its
interference in the titration. A volumetric assay for nickel is based on
this. It has the disadvantage of all indirect titrations in that it
requires two standard solutions. On the other hand it gives good results
even under unfavourable conditions, and is applicable in the presence of
much zinc. Small quantities of cobalt will count as so much nickel, but
larger quantities make the assay unworkable. Some of the other
metals--lead for example--have no appreciable effect; but practically
the solution demands a preliminary treatment which would result in their
removal. Nevertheless it is a very satisfactory method and makes the
determination of nickel quick and comparatively easy in most cases.

_The standard solution of silver nitrate_ is made by dissolving 14.48
grams of recrystallised silver nitrate in distilled water and diluting
to 1 litre: 100 c.c. of this solution are equivalent to 0.25 gram of
nickel.[73]

_The standard solution of potassium cyanide_ should be made so as to be
exactly equal to the silver nitrate solution. This can be done as
follows: Weigh up 12 grams of good potassium cyanide (95 per cent.),
dissolve in water, add 50 c.c. of a 10 per cent. solution of sodium
hydrate and dilute to 1 litre. Fill one burette with this and another
with the solution of silver nitrate. Run 50 c.c. of the cyanide into a
flask; add a few drops of potassium iodide solution and titrate with the
standard silver nitrate until there is a distinct permanent yellowish
turbidity. The titration is more fully described under _Cyanide_, p.
165. The cyanide solution will be found rather stronger than the silver
nitrate; dilute it so as to get the two solutions of equal value. For
example, 51.3 c.c. of silver nitrate may have been required: then add
1.3 c.c. of water to each 50 c.c. of the cyanide solution remaining. If
the full 950 c.c. are available, then add to them 24.7 c.c. of water.
After mixing, take another 50 c.c. and titrate with the silver nitrate;
the two solutions should now be exactly equal. The cyanide solution,
being strongly alkaline with soda, keeps very well; but its strength
should be checked from time to time by titrating with silver nitrate;
should there be any slight inequality in the strengths of the two
solutions it is easily allowed for in the calculations.

~The titration.~--The solution, containing not much more than 0.1 gram
of nickel, and free from the interfering metals, must be cooled. It is
next neutralised and then made strongly alkaline with a solution of soda
(NaHO); an excess of 20 or 30 c.c. suffices. This will produce a
precipitate. The cyanide solution is now run in from a burette until the
solution clears, after which an excess of about 20 c.c. is added. It is
well to use some round number of c.c. to simplify the calculation. Add a
few drops of potassium iodide solution, and run in the standard solution
of silver nitrate from a burette. This should be done a little at a
time, though somewhat rapidly, and with constant shaking, till a
permanent yellow precipitate appears. If the addition of the cyanide did
not result in a perfectly clear solution, this is because something
besides nickel is present. The residue may be filtered off, though with
a little practice the finishing-point may be detected with certainty in
the presence of a small precipitate. If the student has the slightest
doubt about a finish he should run in another 5 c.c. of the cyanide and
again finish with silver nitrate. The second result will be the same as
the first. For example, if 40 c.c. of cyanide and 30 c.c. of silver
nitrate were required at the first titration, then the 45 c.c. of
cyanide in the second titration will require 35 c.c. of silver nitrate.
The difference between the quantities of the two solutions used in each
case will be 10 c.c. It is this difference in the readings of the two
burettes which measures the quantity of nickel present. Each c.c. of the
difference is equal to .0025 gram of nickel. But if the cyanide solution
is not exactly equal in strength to the silver nitrate, the quantity of
cyanide used should be calculated to its equivalent in silver nitrate
before making the subtraction.

The following experimental results illustrate the accuracy of the assay
and the effect upon it of varying conditions. A solution containing 1
gram of nickel sulphate (NiSO_{4}.6H_{2}O) in 100 c.c. was used. By a
separate assay the sulphate was found to contain 22.25 per cent. of
nickel. For the sake of simplicity the results of the experiments are
stated in weights of nickel in grams.

~Effect of varying excess of Cyanide Solution.~--In each experiment
there was 20 c.c. of the nickel solution, equal to .0445 gram of nickel.
There were also 10 c.c. of soda solution, 3 or 4 drops of potassium
iodide and sufficient water to bring the bulk to 100 c.c. before
titrating.

  Cyanide in excess  6 c.c.  4 c.c.  8 c.c.  12 c.c.  25 c.c.
  Nickel found       .0434   .0436   .0440   .0442    .0444

Although the difference between the highest and lowest of these results
is only 1 milligram, their meaning is quite obvious. The excess of
cyanide should not be less than 20 c.c.

~Effect of varying the quantity of Soda.~--There were two series of
experiments, one with 2 c.c. of nickel solution (= .0044 gram of
nickel), the other with 20 c.c. The conditions were as before, except
that the quantity of soda was varied.

  Soda added                5 c.c.  15 c.c.  30 c.c.
  Nickel found, 1st series  .0037   .0042    .0045
    "      "    2nd series  .0444   .0444    .0442

These show that the presence of much soda, though it has only a small
effect, is beneficial rather than otherwise. Ammonia has a bad effect,
if present in anything like the same quantities.

~Effect of varying the Nickel.~--In experiments with 10, 20, and 40 c.c.
of the nickel solution, the results were:--

  Nickel present  .0222  .0445  .0890
  Nickel found    .0220  .0442  .0884

~Effect of Zinc.~--In these experiments 20 c.c. of nickel solution (=
.0445 gram of nickel), 10 c.c. of soda, 6 drops of potassium iodide and
water to 100 c.c. were used. The excess of cyanide was purposely kept at
from 10 to 15 c.c., which is hardly sufficient.

  Zinc added     0     .25 gram.  .5 gram.
  Nickel found  .0442  .0440      .0407

On increasing the excess of cyanide to over 20 c.c. and doubling the
quantity of soda, the experiment with 0.5 gram of zinc gave 0.441 gram
of nickel. Hence the titration is satisfactory in the presence of zinc
provided that not fewer than 20 or 30 c.c. of soda are used, and that
the excess of cyanide is such that not fewer than 20 or 30 c.c. of
silver nitrate are required in the titration. Moreover, these
precautions should be taken whether zinc is present or not.

~Effect of other Metals.~--If metals of the first and second groups are
present they should be removed by passing sulphuretted hydrogen and
filtering. If _iron_ is present it must be removed, since ferrous salts
use up much cyanide, forming ferrocyanides, and ferric salts yield
ferric hydrate, which obscures the end reaction. Hence the sulphuretted
hydrogen must be boiled off and the iron removed as basic ferric acetate
by the method described on p. 233. If the precipitate is bulky it should
be dissolved in a little dilute acid, neutralised and again precipitated
as basic acetate. The nickel will be in the two filtrates. In the
absence of manganese and cobalt the titration may be made without
further separation.

_Manganese_ does not directly interfere, but the precipitated hydrate,
which rapidly darkens through atmospheric oxidation, obscures the end
reaction. It may be removed by passing sulphuretted hydrogen through the
filtrate from the acetate separation: sulphides of nickel, cobalt and
zinc will be precipitated, whilst manganese remains in solution: the
addition of more sodium acetate may assist the precipitation. The
precipitate must be filtered off and dissolved in nitric acid: the
solution should be evaporated to dryness. The filtrate may retain a
little nickel; if so, add ammonia till alkaline, then acidify with
acetic acid and again filter; any small precipitate obtained here should
be added to that first obtained.

It is only when _cobalt_ is present that any further separation is
required. Cobalt hydrate takes up oxygen from the air, and on adding
potassium cyanide some may refuse to dissolve; and the solution itself
acquires a brown colour, which becomes deeper on standing. At this stage
the cobalt is easily separated. The solution containing the nickel and
cobalt with no great excess of acid, is made alkaline by adding 20 c.c.
of soda exactly as in preparing for a titration. So, too, the solution
of cyanide is added so as to have an excess of 20 or 30 c.c.; the
solution may have a brown colour, but if it is not quite clear it _must_
be filtered. Then warm (boiling is not needed) and add from 50 to 100
c.c. of bromine water. This throws down all the nickel as black
peroxide in a condition easy to filter. Filter it off and wash with
water. The precipitate can be dissolved off the filter with the greatest
ease by a little warm sulphurous acid. The filtrate and washings, boiled
till free from sulphurous acid, yield the nickel as sulphate in a clean
condition.

~Determination of Nickel in Nickel Sulphate Crystals.~--Take 0.5 gram of
the salt, dissolve in 50 c.c. of water and add 25 c.c. of solution of
soda. Run in from a burette, say, 60 c.c. "cyanide." Add a few drops of
potassium iodide and titrate back with "silver nitrate." Suppose 15.5
c.c. of the latter is required. Then 15.5 c.c. subtracted from 60 c.c.
leaves 44.5 c.c., and since 100 c.c. = 0.25 gram of nickel, 44.5 c.c.
will equal 0.11125 gram of nickel. This in 0.5 gram of the salt equals
22.25 per cent.

~Determination of Nickel in German Silver.~--Weigh up 0.5 gram of the
alloy, and dissolve in a dish with 5 or 10 c.c. of dilute nitric acid.
Add 5 c.c. of dilute sulphuric acid and evaporate till all the nitric
acid is removed. Cool, take up with 50 c.c. of water, and when dissolved
pass sulphuretted hydrogen through the solution. Filter off the
precipitate and wash with water containing sulphuretted hydrogen and
dilute sulphuric acid. Boil down the filtrate and washings to get rid of
the excess of the gas; add some nitric acid and continue the boiling.
Cool, neutralise the excess of acid with soda, add 1 gram of sodium
acetate and boil. Filter off the precipitate which contains the iron.
The filtrate, cooled and rendered alkaline with soda, is ready for the
titration.


COBALT

Occurs less abundantly than nickel. Its chief ores are smaltite and
cobaltite, which are arsenides of cobalt, with more or less iron,
nickel, and copper. It also occurs as arseniate in erythrine, and as
oxide in asbolan or earthy cobalt, which is essentially a wad carrying
cobalt.

It is mainly used in the manufacture of smalts for imparting a blue
colour to glass and enamels. The oxide of cobalt forms coloured
compounds with many other metallic oxides. With oxide of zinc it forms
"Rinman's green"; with aluminia, a blue; with magnesia, a pink. This
property is taken advantage of in the detection of substances before the
blow-pipe.

The compounds of cobalt in most of their properties closely resemble
those of nickel, and the remarks as to solution and separation given for
the latter metal apply here. Solutions of cobalt are pink, whilst those
of nickel are green.

The detection of cobalt, even in very small quantity, is rendered easy
by the strong blue colour which it gives to the borax bead, both in the
oxidising and in the reducing flame. It is concentrated from the ore in
the same way as nickel, and should be separated from that metal by means
of potassic nitrite in the way described. The dry assay of cobalt has
been given under _Nickel_.


GRAVIMETRIC METHOD.

The yellow precipitate from the potassium nitrite, after being washed
with the acetate of potash, is washed with alcohol, dried, transferred
to a weighed porcelain crucible, and cautiously ignited with an excess
of strong sulphuric acid. The heat must not be sufficient to decompose
the sulphate of cobalt, which decomposition is indicated by a blackening
of the substance at the edges. The salt bears a low red heat without
breaking up. If blackening has occurred, moisten with sulphuric acid,
and ignite again. Cool and weigh. The substance is a mixture of the
sulphates of cobalt and potash (2CoSO_{4} + 3K_{2}SO_{4}), and contains
14.17 per cent. of cobalt.

Cobalt is also gravimetrically determined, like nickel, by electrolysis,
or by precipitation with sodic hydrate. In the latter case, the ignited
oxide will be somewhat uncertain in composition, owing to its containing
an excess of oxygen. Consequently, it is better to reduce it by igniting
at a red heat in a current of hydrogen and to weigh it as metallic
cobalt.


PRACTICAL EXERCISES.

1. In the dry assay of an ore containing cobalt, nickel, and copper, the
following results were obtained. Calculate the percentages. Ore taken, 5
grams. Speise formed, 0.99 gram. Speise taken. 0.99 gram. Arsenides of
cobalt, nickel, and copper got, 0.75 gram. Arsenide of nickel and copper
got, 0.54 gram. Gold added, 0.5 gram. Gold and copper got, 0.61 gram.

2. Calculate the percentage composition of the following compounds:
Co_{2}As, Ni_{2}As, and Cu_{2}As.

3. A sample of mispickel contains 7 per cent. cobalt. What weight of the
mixed sulphates of potash and cobalt will be obtained in a gravimetric
determination on 1 gram of the ore?

4. 0.3157 gram of metal was deposited by the electrolysis of a nickel
and cobalt solution. On dissolving in nitric acid and determining the
cobalt 0.2563 gram of potassium and cobalt sulphates were got. Find the
weights of cobalt and nickel present in the deposit.

5. What should be the percentage composition of pure cobaltite, its
formula being CoAsS?


ZINC.

Zinc occurs in nature most commonly as sulphide (blende); it also occurs
as carbonate (calamine) and silicate (smithsonite). Each of these is
sufficiently abundant to be a source of the metal.

The metal is known in commerce as "spelter" when in ingots, and as sheet
zinc when rolled. It is chiefly used in the form of alloys with copper,
which are known as brasses. It is also used in the form of a thin film,
to protect iron goods from rusting--galvanised iron.

Ores of zinc, more especially blende, are met with in most lead, copper,
gold, and silver mines, in larger or small quantities scattered through
the lodes. Those ores which generally come under the notice of the
assayer are fairly rich in zinc; but alloys and metallurgical products
contain it in very varying proportions.

Zinc itself is readily soluble in dilute acids; any residue which is
left after boiling with dilute hydrochloric or sulphuric acid consists
simply of the impurities of the metal; this is generally lead.

All zinc compounds are either soluble in, or are decomposed by, boiling
with acids, the zinc going into solution. Zinc forms only one series of
salts, and these are colourless. Their chief characteristic is
solubility in an alkaline solution, from which sulphuretted hydrogen
produces a white precipitate of zinc sulphide. Zinc is detected by
dissolving the substance in hydrochloric or nitric acid, boiling, and
adding sodic hydrate in excess, filtering, and adding ammonic sulphide
to the filtrate. The precipitate contains the zinc, which can be
dissolved out by boiling with dilute sulphuric acid, and detected by the
formation of a white precipitate on the addition of potassic
ferrocyanide.

The dry assay of zinc can only be made indirectly, and is
unsatisfactory. Zinc is volatile, and at the temperature of its
reduction is a gas. It is impracticable to condense the vapour so as to
weigh the metal, consequently its amount is determined by loss. The
following method gives approximate results: Take 10 grams of the dried
and powdered ore and roast, first at a low temperature and afterwards at
a higher one, with the help of carbonate of ammonia to decompose the
sulphates formed; cool and weigh. The metals will be present as oxides.
Mix with 2 grams of powdered charcoal and charge into a black-lead
crucible heated to whiteness, cover loosely, and leave in the furnace
for about a quarter of an hour. Uncover and calcine the residue, cool
and weigh. The loss in weight multiplied by 8.03 gives the percentage of
zinc in the ore.


WET METHODS.

Solution and separation may be effected as follows: Treat 1 or 3 grams
of the substance with 10 or 30 c.c. of hydrochloric acid or aqua regia;
evaporate to dryness; take up with 10 c.c. of hydrochloric acid and
dilute to 100 c.c.; heat nearly to boiling; saturate with sulphuretted
hydrogen; filter, and wash with water acidulated with hydrochloric acid.
Boil off the sulphuretted hydrogen and peroxidise with a few drops of
nitric acid. Cool; add caustic soda till nearly, but not quite,
neutralised, and separate the iron as basic acetate by the method
described under _Iron_. To the filtrate add ammonia till alkaline, and
pass sulphuretted hydrogen. Allow to settle and decant on to a filter.
Dissolve off the precipitate from the filter with hot dilute
hydrochloric acid. The solution will contain the zinc, together with any
manganese the ore contained, and, perhaps, traces of nickel and cobalt.
If the zinc is to be determined volumetrically, and manganese is
present, this latter is separated with carbonate of ammonia, as
described further on; but if a gravimetric method is used, and only
small quantities of manganese are present, it is better to proceed as if
it were absent, and to subsequently determine its amount, which should
be deducted.


GRAVIMETRIC DETERMINATION.

The solution containing the zinc is contained in an evaporating dish,
and freed from sulphuretted hydrogen by boiling, and, if necessary, from
an excess of acid by evaporation. The evaporating dish must be a large
one. Cautiously add sodium carbonate to the hot, moderately dilute
solution, until the liquid is distinctly alkaline, and boil. Allow the
precipitate to settle, decant on to a filter, and wash with hot water.
Dry, transfer to a porcelain crucible (cleaning the paper as much as
possible), add the ash, ignite, and weigh. The substance weighed is
oxide of zinc, which contains 80.26 per cent. of the metal. It is a
white powder, becoming yellow when heated. It must not show an alkaline
reaction when moistened. If it contains manganese this metal will be
present as sesquioxide (Mn_{2}O_{3}). Its amount can be determined by
dissolving in dilute acid and boiling with an excess of sodic hydrate.
The oxide of manganese will be precipitated, and can be ignited and
weighed. Its weight multiplied by 1.035 must be deducted from the weight
of oxide of zinc previously obtained. The results yielded by the
gravimetric determination are likely to be high, since the basic
carbonate of zinc frequently carries down with it more or less soda
which is difficult to wash off.


VOLUMETRIC DETERMINATION

This method is based on the facts that zinc salts in an acid solution
decompose potassium ferrocyanide, forming a white insoluble zinc
compound; and that an excess of the ferrocyanide can be detected by the
brown coloration it strikes with uranium acetate. The method resembles
in its working the bichromate iron assay. The standard solution of
potassium ferrocyanide is run into a hot hydrochloric acid solution of
the zinc until a drop of the latter brought in contact with a drop of
the indicator (uranium acetate) on a white plate strikes a brown colour.
The quantity of zinc in the solution must be approximately known; run in
a little less of the ferrocyanide than is expected will be necessary;
test a drop or two of the assay, and then run in, one or two c.c. at a
time, until the brown colour is obtained. Add 5 c.c. of a standard zinc
solution, equivalent in strength to the standard "ferrocyanide,"
re-titrate, and finish off cautiously. Of course 5 c.c. must be deducted
from the reading on the burette. The precipitate of zinc ferrocyanide
formed in the assay solution is white; but if traces of iron are
present, it becomes bluish. If the quantity of ferrocyanide required is
known within a few c.c., the finishing point is exactly determined in
the first titration without any addition of the standard zinc solution.
Unfortunately this titration serves simply to replace the gravimetric
determination, and does not, as many volumetric processes do, lessen the
necessity for a complete separation of any other metals which are
present. Most metals give precipitates with ferrocyanide of potassium in
acid solutions. If the conditions are held to, the titration is a fairly
good one, and differences in the results of an assay will be due to
error in the separation. Ferric hydrate precipitated in a fairly strong
solution of zinc will carry with it perceptible quantities of that
metal. Similarly, large quantities of copper precipitated as sulphide by
means of sulphuretted hydrogen will carry zinc with it, except under
certain nicely drawn conditions. When much copper is present it is best
separated in a nitric acid solution by electrolysis. The titration of
the zinc takes less time, and, with ordinary working, is more
trustworthy than the gravimetric method.

_The standard ferrocyanide solution_ is made by dissolving 43.2 grams of
potassium ferrocyanide (K_{4}FeCy_{6}.3H_{2}O) in water, and diluting to
a litre. One hundred c.c. are equal to 1 gram of zinc.

_The standard zinc solution_ is made by dissolving 10 grams of pure zinc
in 50 c.c. of hydrochloric acid and 100 or 200 c.c. of water, and
diluting to 1 litre, or by dissolving 44.15 grams of zinc sulphate
(ZnSO_{4}.7H_{2}O) in water with 30 c.c. of hydrochloric acid, and
diluting to 1 litre. One hundred c.c. will contain 1 gram of zinc.

_The uranium acetate solution_ is made by dissolving 0.2 gram of the
salt in 100 c.c. of water.

To standardise the "ferrocyanide" measure off 50 c.c. of the standard
zinc solution into a 10 oz. beaker, dilute to 100 c.c., and heat to
about 50° C. (not to boiling). Run in 47 or 48 c.c. of the
"ferrocyanide" solution from an ordinary burette, and finish off
cautiously. Fifty divided by the quantity of "ferrocyanide" solution
required gives the standard.

In assaying ores, &c., take such quantity as shall contain from 0.1 to 1
gram of zinc, separate the zinc as sulphide, as already directed.
Dissolve the sulphide off the filter with hot dilute hydrochloric acid,
which is best done by a stream from a wash bottle. Evaporate the
filtrate to a paste, add 5 c.c. of dilute hydrochloric acid, dilute to
100 c.c. or 150 c.c., heat to about 50° C., and titrate. Manganese, if
present, counts as so much zinc, and must be specially separated, since
it is not removed by the method already given. The following method will
effect its removal. To the hydrochloric acid solution of the zinc and
manganese add sodium acetate in large excess and pass sulphuretted
hydrogen freely. Allow to settle, filter off the zinc sulphide and wash
with sulphuretted hydrogen water. The precipitate, freed from manganese,
is then dissolved in hydrochloric acid and titrated.

The following experiments show the effect of variation in the conditions
of the assay:--

~Effect of Varying Temperature.~--Using 20 c.c. of the standard zinc
solution, 5 c.c. of dilute hydrochloric acid, and diluting to 100 c.c.

  Temperature               15° C.     30° C.     70° C.    100° C.
  "Ferrocyanide" required  20.6 c.c.  20.3 c.c.  20.3 c.c.  20.3 c.c.

The solution can be heated to boiling before titrating without
interfering with the result; but it is more convenient to work with the
solution at about 50° C. Cold solutions must not be used.

~Effect of Varying Bulk.~--These were all titrated at about 50° C., and
were like the last, but with varying bulk.

  Bulk                     25.0 c.c.  50.0 c.c.  100.0 c.c.  200.0 c.c.
  "Ferrocyanide" required  20.2  "    20.4  "     20.3  "     20.4  "

Any ordinary variation in bulk has no effect.

~Effect of Varying Hydrochloric Acid.~-- With 100 c.c. bulk and varying
dilute hydrochloric acid the results were:--

  Acid added        0.0 c.c.   1.0 c.c.   5.0 c.c.   10.0 c.c.   20.0 c.c.
  "Ferrocyanide"
    required       24.4  "    20.2  "    20.3  "     20.3  "     20.7  "

~Effect of Foreign Salts.~--The experiments were carried out under the
same conditions as the others. Five grams each of the following salts
were added:--

  Salt added         { Ammonic    Ammonic    Sodium      Sodium
                     { chloride.  sulphate.  chloride.   sulphate.
  "Ferrocyanide"
    required           20.3 c.c.  20.5 c.c.  20.6 c.c.   20.4 c.c.

  Salt added         { Potassium  Magnesium   Nil.
                     { Nitrate.   sulphate.
  "Ferrocyanide"
    required           20.2 c.c.  20.4 c.c.   20.4 c.c.

In a series of experiments in which foreign metals were present to the
extent of 0.050 gram in each, with 20 c.c. of zinc solution and 5 c.c.
of dilute hydrochloric acid, those in which copper sulphate, ferrous
sulphate, and ferric chloride were used, gave (as might be expected) so
strongly coloured precipitates that the end reaction could not be
recognised. The other results were:--

                                          "Ferrocyanide"
                                            required.
  With nothing added.                        20.3 c.c.
   "   0.050 gram lead (as chloride)         20.9  "
   "   0.050   "  manganese (as sulphate)    25.5  "
   "   0.050   "  cadmium (as sulphate)      23.5  "
   "   0.050   "  nickel (as sulphate)       26.2  "

~Effect of Varying Zinc.~--These were titrated under the usual
conditions, and gave the following results:--

    Zinc added     1.0 c.c.  10.0 c.c.  20.0 c.c.  50.0 c.c.  100.0 c.c.
  "Ferrocyanide"
    required       1.1  "    10.2  "    20.3  "    50.6  "    101.0  "

~Determination of Zinc in a Sample of Brass.~--Take the solution from
which the copper has been separated by electrolysis and pass
sulphuretted hydrogen until the remaining traces of copper and the lead
are precipitated, filter, boil the solution free from sulphuretted
hydrogen, put in a piece of litmus paper, and add sodic hydrate solution
in slight excess; add 10 c.c. of dilute hydrochloric acid (which should
render the solution acid and clear); warm, and titrate.

A sample of 0.5 gram of brass treated in this manner required 16.4 c.c.
of "ferrocyanide" (standard 100 c.c. = 0.9909 zinc), which equals 0.1625
gram of zinc or 32.5 per cent.

~Determination of Zinc in Blende.~--Dissolve 1 gram of the dried and
powdered sample in 25 c.c. of nitric acid with the help of two or three
grams of potassium chlorate dissolved in the acid. Evaporate to complete
dryness, taking care to avoid spirting. Add 7 grams of powdered ammonium
chloride, 15 c.c. of strong ammonia and 25 c.c. of boiling water; boil
for one minute and see that the residue is all softened. Filter through
a small filter, and wash thoroughly with small quantities of a hot one
per cent. solution of ammonium chloride. Add 25 c.c. of hydrochloric
acid to the filtrate. Place in the solution some clean lead foil, say 10
or 20 square inches. Boil gently until the solution has been colourless
for three or four minutes. Filter, wash with a little hot water; and
titrate with standard ferrocyanide.

~Determination of Zinc in Silver Precipitate.~--This precipitate
contains lead sulphate, silver, copper, iron, zinc, lime, &c. Weigh up 5
grams of the sample, and extract with 30 c.c. of dilute sulphuric acid
with the aid of heat. Separate the copper with sulphuretted hydrogen,
peroxidise the iron with a drop or two of nitric acid, and separate as
acetate. Render the filtrate ammoniacal, pass sulphuretted hydrogen;
warm, and filter. Dissolve the precipitated zinc sulphide in dilute
hydrochloric acid, evaporate, dilute, and titrate. Silver precipitates
carry about 2.5 per cent. of zinc.


GASOMETRIC METHOD.

Metallic zinc is readily soluble in dilute hydrochloric or sulphuric
acid, hydrogen being at the same time evolved.[74] The volume of the
hydrogen evolved is obviously a measure of the amount of zinc present in
the metallic state. The speed with which the reaction goes on (even in
the cold) and the insolubility of hydrogen renders this method of assay
a convenient one. It is especially applicable to the determination of
the proportion of zinc in zinc dust. The apparatus described in the
chapter on gasometric method is used. The method of working is as
follows: Fill the two burettes with cold water to a little above the
zero mark, place in the bottle about 0.25 gram of the substance to be
determined, and in the inner phial or test tube 5 c.c. of dilute
sulphuric acid; cork the apparatus tightly and allow to stand for a few
minutes; then bring the water to the same level in the two burettes by
running out through the clip at the bottom. Read off the level of the
liquid in the graduated burette. Turn the bottle over sufficiently to
spill the acid over the zinc, and then run water out of the apparatus so
as to keep the liquid in the two burettes at the same level, taking care
not to run it out more quickly than the hydrogen is being generated.
When the volume of gas ceases to increase, read off the level of the
liquid, deduct the reading which was started with; the difference gives
the volume of hydrogen evolved. At the same time read off the volume of
air in the "volume corrector," which must be fixed alongside the gas
burettes. Make the correction. For example: A piece of zinc weighing
0.2835 gram was found to give 99.9 c.c. of gas at a time when the
corrector read 104 c.c.[75] Then the corrected volume is

        104 : 100 :: 99.9 : _x_.
                            _x_ = 96.0 c.c.

100 c.c. of hydrogen at 0° C. and 760 mm. is equivalent to 0.2912 gram
of zinc; therefore the quantity of zinc found is

        100 : 96 :: 0.2912 : _x_.
                             _x_ = 0.2795 gram of zinc.

This being contained in 0.2835 gram of metal is equivalent to 98.5 per
cent.

As an example of a determination in which reducing the volume of
liberated hydrogen to 0° C. and 760 mm. is avoided, the following may be
taken:--

        0.2315 gram of pure zinc gave 82.1 c.c. of gas;
and the volume of air in the corrector was 103.6 c.c.

        0.2835 gram of the assay gave 99.9 c.c. of gas;
and the volume of air in the corrector was 104.0 c.c.;

        104 : 103.6 :: 99.9 : _x_.
                              _x_ = 99.5 c.c.

This is the volume of gas got in the assay if measured under the same
conditions as the standard,

        82.1 : 99.5 :: 0.2315 : _x_.
                                _x_ = 0.2806.

        Then 0.2835 : 0.2806 :: 100: _x_.
                                     _x_ = 98.9 per cent.

As these assays can be made quickly, it is well for the sake of greater
accuracy to make them in duplicate, and to take the mean of the
readings. One set of standardisings will do for any number of assays.
The student must carefully avoid unnecessary handling of the bottle in
which the zinc is dissolved.

~Colorimetric Method.~--Zinc salts being colourless, there is no
colorimetric determination.


EXAMINATION OF COMMERCIAL ZINC.

Take 20 grams of zinc, and dissolve them in dilute nitric acid; boil,
allow to settle; filter; wash, dry; ignite the precipitate, if any, and
weigh as oxide of tin. Examine this for arsenic.

~Lead.~--Add ammonia and carbonate of ammonia to the liquid, and boil,
filter off the precipitate, wash with hot water. Digest the precipitate
with dilute sulphuric acid; filter, wash, and weigh the sulphate of
lead.

~Iron.~--To the filtrate from the sulphate of lead add ammonia, and pass
sulphuretted hydrogen; digest, and filter. (Save the filtrate.) Dissolve
the precipitate in hydrochloric acid, oxidise with nitric acid, and
precipitate with ammonia. Wash, ignite, and weigh as ferric oxide.
Calculate to iron.

~Arsenic.~--To the filtrate from the sulphide of iron add hydrochloric
acid in slight excess; filter off, and wash the precipitate. Rinse it
back into the beaker, dissolve in nitric acid, filter from the sulphur,
and add ammonia, in excess, and magnesia mixture. Filter off the
ammonic-magnesic arsenate, and wash with dilute ammonia. Dry, ignite
with nitric acid, and weigh as magnesic pyrarsenate. Calculate to
arsenic, and add to that found with the tin.

~Copper.~--To the filtrate from the ammonia and ammonic carbonate add
sulphuric acid in small excess, and pass sulphuretted hydrogen. Allow to
settle, filter, and wash. Rinse the precipitate into a beaker, boil with
dilute sulphuric acid, and filter. (Save the filtrate.) Dry, burn the
paper with the precipitate, treat with a drop or two of nitric acid,
ignite, and weigh as copper oxide. Calculate to copper.

~Cadmium.~--To the filtrate from the sulphide of copper add ammonia, so
as to nearly neutralise the excess of acid, and pass sulphuretted
hydrogen. Collect and weigh the precipitate as cadmium sulphide, as
described under _Cadmium_.


PRACTICAL EXERCISES.

1. What weight of hydrogen will be evolved in dissolving 1 gram of zinc
in dilute sulphuric acid?

2. How many c.c. would this quantity of hydrogen measure at 0° C. and
760 m.m.? (1 litre weighs 0.0896 gram).

3. 0.23 gram of zinc are found to give 77.9 c.c. of hydrogen. In another
experiment under the same conditions 80.2 c.c. are got. What weight of
zinc was used for the second experiment?

4. A sample of blende is found to contain 55 per cent. of zinc. What
percentage of zinc sulphide did the sample contain?

5. How much metallic lead would be precipitated from a solution of lead
acetate by 1 gram of zinc?


CADMIUM.

Cadmium occurs in nature as cadmium sulphide in greenockite, CdS, which
is very rare. It is widely diffused in calamine, blende, and other zinc
ores, forming, in some cases, as much as 2 or 3 per cent. of the ore.
Oxide of cadmium forms the "brown blaze" of the zinc smelters.

Sulphide of cadmium is used as a pigment (cadmium yellow); and the metal
and some of its salts are useful reagents.

The salts of cadmium closely resemble those of zinc. The hydrate,
however, is insoluble in excess of potash, and the sulphide is insoluble
in dilute acids. It forms only one series of salts.

Cadmium is detected by giving with sulphuretted hydrogen in solutions,
not too strongly acid, a yellow precipitate, which is insoluble in
solutions of the alkalies, alkaline sulphides, or cyanide of potassium.

~Solution and Separation.~--Substances containing cadmium are soluble in
acids. The solution is evaporated to dryness (to render any silica that
may be present insoluble) and taken up with 10 c.c. of dilute
hydrochloric acid. Dilute to 100 c.c., and pass sulphuretted hydrogen.
Filter, digest the precipitate with soda, wash, and boil with dilute
sulphuric acid. Filter; the filtrate contains the cadmium and, possibly,
a small quantity of zinc, from which it is best separated by
reprecipitating with sulphuretted hydrogen.


GRAVIMETRIC DETERMINATION.

The solution containing the cadmium freed from the other metals is
precipitated with sulphuretted hydrogen in a moderately-acid solution.
The precipitate is collected on a weighed filter, and washed, first with
an acid solution of sulphuretted hydrogen, and afterwards with water. It
is dried at 100° C. and weighed. If free sulphur is suspected to be
present, extract with bisulphide of carbon, and again weigh. The residue
is cadmium sulphide, which contains 77.78 per cent. of cadmium. It is a
yellow powder insoluble in solutions of the alkalies, alkaline
sulphides, or cyanide of potassium. It dissolves readily in acid. It
cannot be ignited in a current of hydrogen without loss.


VOLUMETRIC METHOD.

The solution containing the cadmium is concentrated by evaporation, and
mixed with an excess of oxalic acid and alcohol. The precipitate is
filtered, washed with alcohol, dissolved in hot hydrochloric acid, and
titrated with permanganate of potassium.

FOOTNOTES:

[64] When chromium is present some of the iron may escape precipitation
but it can be recovered from the solution by means of ammonic sulphide.

[65]

(1) 10FeSO_{4} + 2KMnO_{4} + 8H_{2}SO_{4} = 5Fe_{2}(SO_{4})_{3} +
2MnSO_{4} + K_{2}SO_{4} + 8H_{2}O.

(2) 6FeCl_{2} + K_{2}Cr_{2}O_{7} + 14HCl = 3Fe_{2}Cl_{6} + Cr_{2}Cl_{6}
+ 2KCl + 7H_{2}O.

[66] (1) Fe_{2}Cl_{6} + SnCl_{2} = 2FeCl_{2} + SnCl_{4}.
     (2) Fe_{2}Cl_{6} + SH_{2} = 2FeCl_{2} + 2HCl + S.
     (3) Fe_{2}Cl_{6} + Na_{2}SO_{3} + H_{2}O = 2FeCl_{2}
         + Na_{2}SO_{4} + 2HCl.
     (4) Fe_{2}Cl_{6} + Zn = 2FeCl_{2} + ZnCl_{2}.

[67] 20 grams of stannous chloride and 20 c.c. of dilute hydrochloric
acid are diluted to one litre.

[68] The maximum reducing effect of zinc is obtained by exposing as
large a surface as possible of the metal in a hot concentrated solution
containing but little free acid (Thorpe).

[69] About 5 inches in diameter.

[70] 61: 60:: 59: 58.13.

The iron in the ore is, then, the same in amount as that in 58.13 c.c.
of the ferric chloride solution; and since 100 c.c. of the latter
contain 1 gram of iron, 58.13 c.c. of the same contains 0.5813 gram of
iron; and, further, if 1 gram of ore carries this amount of iron, 100
grams of ore will obviously give 58.13 grams of iron.

[71] These compounds are Ni_{2}As and Co_{2}As.

[72] With large quantities of iron the ferric precipitate should be
re-dissolved and re-precipitated. The filtrate must be added to the
original filtrate.

[73] 4KCy + NiSO_{4} = K_{2}NiCy_{4} + K_{2}SO_{4} 2KCy + AgNO_{3}
     = KAgCy_{2} + KNO_{3}
     .'. 2AgNO_{3} = Ni

[74] Zn + H_{2}SO_{4} = H_{2} + ZnSO_{4}.

[75] These 104 c.c. are equivalent to 100 c.c. of dry air at 0° C. and
760 mm.



CHAPTER XII.

TIN--TUNGSTEN--TITANIUM.


TIN.

Tin occurs in nature as cassiterite (containing from 90 to 95 per cent.
of oxide of tin), which mineral is the source from which the whole of
the tin of commerce is derived. Tin also occurs as sulphide combined
with sulphides of copper and iron in the mineral stannine or bell-metal
ore. It is a constituent of certain rare minerals, such as tantalite.

The methods of assaying tin in actual use are remarkable when compared
with those of other metals. The more strictly chemical methods are
rendered troublesome by the oxide being insoluble in acids, resembling
in this respect the gangue with which it is associated. Moreover, it is
not readily decomposed by fusion with alkalies. The oxide has first to
be reduced to metal before the tin can be dissolved. The reduction may
be performed by fusing with potassic cyanide, by heating to moderate
redness in a current of hydrogen or coal gas, or by heating to a higher
temperature with carbon. The reduced metal is only slowly dissolved by
hydrochloric acid, and although it is readily soluble in aqua regia, the
solution cannot be evaporated or freed from the excess of acids, by
boiling, without loss of tin, because of the volatility of stannic
chloride. There has long been a difficulty in getting a quick wet
method.

The process of assaying tin ores adopted in the mines of Cornwall is a
mechanical one known as "vanning," the object of which is to find the
percentage of "black tin," which, it is well to remember, is not pure
cassiterite, much less pure oxide of tin. Tin ore, as taken from the
lode, contains from 2 to 5 per cent. of cassiterite, and is mainly made
up of quartz, felspar, chlorite, schorl, and other stony minerals,
together with more or less mispickel, iron and copper pyrites, oxide of
iron, and wolfram. The cassiterite has a specific gravity (6.4 to 7.1)
considerably higher than that of the vein-stuff (2.5 to 3.0), and is
concentrated by a series of washings till it is free from the lighter
material. Those minerals which have a specific gravity approaching that
of the cassiterite are not completely removed. The mispickel and copper
and iron pyrites are converted into oxides by roasting, and are in great
part removed by a subsequent washing. The concentrated product is known
as "black tin," and in this condition is sold to the smelter. The chief
foreign matters in the black tin are silica, oxides of iron and copper,
and wolfram, with traces of manganese and niobic acid; and in certain
stream ores there may be as much as 6 or 7 per cent. of titaniferous
iron. The black tin from the mines contains from 5 to 12 per cent. of
water, and is sold and assayed wet. A series of typical samples of black
tin ranged as follows:--

  --------------------------+---------------------+-----------------
      Source of Material.   | Percentage of Metal |Specific Gravity.
                            |     in Dry Ore.     |
  --------------------------+---------------------+-----------------
   Good mine ore            |        72.0         |      6.39
   Inferior do.             |        71.5         |      6.64
   Titaniferous stream ore  |        67.0         |      6.39
   Mine ore with wolfram    |        64.5         |      6.67
   Ore from stream works    |        58.5         |      5.99
  --------------------------+---------------------+-----------------

It will be seen from these figures that black tin is a very variable
substance; and that the specific gravity is largely influenced by the
impurities; hence, it is only an indication of the percentage of metal
when the same kind of ore is dealt with.

As already pointed out, the object of vanning is to determine the
proportion of black tin in the lode stuff. The relation between the
actual content in oxide of tin and the produce got by vanning has been
tested on several occasions with results which show a fair degree of
approximation.

The following are some published results of assays of the same batch of
ore. The vanning results were obtained by a Cornish vanner of recognised
ability, and the wet assays by two London firms of the highest
standing:--

  Vanning results:
    (Average)        91 lbs. of "black tin."
  Wet Assay results:
    A                83.7 lbs. of stannic oxide.
    B                79.7 lbs.       "

The vanner reported his black tin as containing 70 per cent. of tin.
This will bring his result, if calculated as stannic oxide, to 80.9 lbs.
to the ton; which agrees with the others.

According to our experience the "van" assay agrees fairly well with the
"wet" one, if the black tin is assumed to contain 92.5 per cent. of
stannic oxide (SnO_{2}).

Vanners are, as a rule, skilful men, and show remarkable dexterity in
separating the black tin, with the help of their apparatus, which
consists simply of a shovel and a kieve of water. An account of the
process is given below. But different vanners, all good men, will get
different results working on material new to them. The black tin weighed
by the vanner is supposed to correspond in quality with the black tin
returned from the floors of the mine for which he is assaying, but this
differs materially in different mines with the nature of the gangue. The
process leaves too much to the judgment of the vanner. It is more than
probable that in practice the returns from the dressing-floors check the
assayer, instead of, as should properly be the case, the assayer
checking the returns. It is only when this last is done that any control
is had over the system of dressing. A correct assay of this ore is a
matter of some importance, because of the high price of the metal.

The method of assaying the black tin is a dry one, and consists of
mixing it with "culm," and submitting it in a black-lead crucible to the
highest temperature of a wind furnace. The sample is taken wet as it
arrives at the smelting house, and is assayed direct. The product of the
assay is examined, and a deduction of a considerable percentage is very
properly made for impurities, since the assay really determines the
percentage, not merely of tin, but of the bodies present which are
reducible at a white heat. The judgment as to how much is to be deducted
is assisted partly by an examination of the metal got from the assay,
and partly by the experience acquired in smelting similar ores. The
produce, which is that of the impure tin, is stated in parts in twenty;
thus a produce of 14 is equivalent to 70 per cent., or to 14 cwt. per
ton.

[Illustration: FIG. 57.]

MECHANICAL SEPARATION.--VANNING.

This process, which has already been referred to, is carried out as
follows:--After sampling the ore in the ordinary way, a quantity
(varying with its richness) is weighed out. Special weights are
generally used. The standard weight, marked 200, weighs about an ounce;
with poor ores this quantity is taken for an assay, but with richer ores
100 or even 50 is sufficient. The unit of weight has no special name,
but the parts in 200 are spoken of as the produce; thus, if 200 of ore
were taken and 9.5 of black tin were separated, the produce would be
9-1/2: obviously half the "produce" will give the percentage. The
weighed portion of the ore is placed on the vanning shovel. The vanner
stands in front of a tub of water (kieve) and allows 30 or 40 c.c. of
water to flow on to the ore. He then raises the shovel a little above
the surface of the water, and, holding it nearly horizontal, briskly
rotates the water by imparting to the shovel a slight circular motion,
passing into an elliptical one (front to back). This causes the finer
mud to be suspended in the liquid, which is then run off, leaving the
body of the ore in the centre of the shovel. This is repeated until the
water after standing a moment is fairly clear. About half as much water
as before is brought on; then, with a motion which is similar to the
previous one, but with a jerk added in one direction, the heavier
minerals are thrown up, and the stony matter brought back. The jerk is
produced just as the wave of water is returning. The descending wave of
water draws with it the bulkier and lighter particles of the ore, whilst
the heavier matter lying on the bottom is scarcely affected by it. The
jerky motion, however, carries it to the front of the shovel. The
lighter stuff is washed off, and the residue dried by holding the shovel
over the furnace. It now corresponds, more or less, to the stuff which
on the mine is sent to the calciner. It is swept from the shovel into a
scoop, and transferred to a hot crucible; in which it is calcined until
free from sulphur. Some vanners calcine their samples before commencing
to van. The calcined ore is shaken out of the crucible on to the shovel;
rubbed up with a hammer; and washed (as at first) to get rid of the
finer and lighter "waste." The separating motions are again gone
through; and the "head" of the best of the black tin is thrown well up
on one side of the shovel in the form of a crescent, so as to leave room
on the shovel to work with the "tailings." The quantity of water used is
kept low, to prevent this "crop" tin from being washed back again. The
tailings are then crushed to free the tin from adherent oxide of iron;
and again washed to throw up the remaining tin ore. As this tin is
finely divided, it is more difficult to bring it up, so that a vigorous
and rapid motion is required. The tailings are now washed off, and the
whole of the black tin is brought into the centre of the shovel. It
requires two or three washings more to free it from the waste it
contains. Very small quantities of water are used. The purity of the
black tin can be seen by its appearance on the shovel. The cleaned ore
is dried as before, freed from particles of iron with the aid of a
magnet, and weighed. The weighings are carried to 1/8th of the unit
used. The following example illustrates the method of calculation
adopted on the mine. A parcel of 1 ton 2 cwt. 3 qrs. of tin ore with a
produce of 45 (equal to 22-1/2 per cent.) contains 5 cwt. 0 qrs. 12 lbs.
of black tin. This result is obtained as follows:--

  ton   cwt.   qrs.
    1     2     3
                9   }
  ----------------- }
   10     4     3   } equivalent to multiplying by 45.
                5   }
  ----------------  }
    5.1   3     3     strike off the first figure to the right.
                4      multiply by 4 to reduce to quarters.
  ---------
    4    12
          3
  ---------
    4    15
   28                  multiply by 28 to reduce to pounds.
  -----
  112
   15
  -----
   12.7                strike off the first figure to the right.

Similarly, a parcel of 20 tons 10 cwt. with a produce of 9-1/2 contains
19 cwt. 1 qr. 25 lbs. of black tin. For the following information, as
well as for much of that already given about vanning, we are indebted to
Captain Reynolds, of Cook's Kitchen Mine. "To have a complete set of
tools for all vanning purposes, it will be necessary to get the
following:--A vanning shovel 14 inches long and 13 inches wide, weighing
not over 2-3/4 pounds. It is made of hammered sheet iron of the shape
shown in fig. 57. It must have a light wooden handle (preferably of
deal) 3 feet long. A bruising hammer, weighing 2-1/2 pounds, with a
handle 1 foot long. A pair of tongs (furnace) 2-1/2 feet long, made of
1/2-inch round iron. And a set of ordinary clay crucibles for calcining.
There ought to be two sets of scales and weights: the first should be
confined to weighing the powdered tin stuff, and the second ought to be
a much higher class one, for weighing the black tin obtained. The
furnace for roasting the sample should be 10 inches square and 12 inches
deep, with the fire-bars at the bottom three-quarters of an inch apart.
The water-box for vanning in should be at least 4 feet long, 2 feet 6
inches wide, and 8 inches deep."


DRY METHODS.

For the following description of the process adopted in Cornwall we are
indebted to Mr. A.K. Barnett, F.G.S., of Chyandour.

~Cornish Method.~--_Tin Ore Assay._--The ore to be smelted or assayed
should be concentrated to say not less than 50 per cent. of metallic
tin; though to obtain satisfactory results it should be brought nearer
70 per cent., as with ore containing less than 40 to 50 per cent. of
metal there will be a considerable loss both in the assaying and in the
smelting. If the ore to be operated on does not contain this quantity of
metal, then the sample (if coarse) must be reduced to a fine state, the
gangue being removed by vanning, and the ore saved for the fire assay.

The method adopted for the determination of tin in the ore is as
follows:--About 2-1/2 ounces troy (1200 grains, or about 80 grams) of
the ore to be assayed is weighed out and mixed on a flat copper pan
(shaped with a long lip) with one-fifth of its weight (240 grains, or
15.5 grams) of powdered culm (anthracite). The mixture of ore and culm
is either transferred to a black-lead crucible before the latter is put
into the furnace, or, as some prefer, it is carefully swept into a
crucible which has been imbedded in the fire. Some assayers cover their
pots with a flat cover placed loosely on, while others leave the mixture
in the open pot. The furnace, which has been previously fired to a
strong heat, is then covered, and the sample is subjected to a sharp
fire for a period of from twelve to twenty minutes. No definite time can
be stated, as, besides the strength of the fire, the quality and
condition of the ore, and the impurities associated with it, greatly
affects the time required for the complete reduction of the ore. As soon
as the mixture in the crucible has settled down to a uniform white heat,
and any very slight ebullition which may have taken place has subsided,
the crucible is gently shaken, removed from the fire (the culm-ash or
slag which covers the metal being carefully drawn aside with an iron
scraper), and the metal is poured quickly into an iron ingot-mould,
which is usually placed on a copper pan to save the culm-slag and the
adherent metal which comes out with it. The crucible is then carefully
scraped, and the scrapings, together with the contents of the mould and
pan, are transferred to a mortar. There the ingot of tin is freed from
slag and then taken to the scales. The rest, after being finely
powdered, is passed through a sieve. The flattened particles of tin
which remain on the sieve are weighed with the ingot (the _lump_, as it
is called); whilst the siftings are vanned on a shovel, and (the slag
being washed off) the fine tin is collected, dried, and weighed with the
rest: the whole gives the produce or percentage of metal in the ore.
The results of the assays are expressed in cwts. of metal in the ton of
ore. The percentage is rarely given and never used in Cornwall.
Thus--"13-1/2 Produce" would mean that the assay yielded results at the
rate of 13-1/2 cwts. of metal for one ton of the ore. Some assayers use
a little powdered fluor-spar to assist the fusion of refractory slags. A
small quantity of borax will also occasionally be of service for ores
containing silica in excess of any iron that may be present. The borax
renders the slag more fusible, and assists the formation of a larger
lump (with less fine tin in the slag) than would be obtained by the use
of culm alone.

The quality and the percentage of _pure tin_ in the metal will vary
considerably, according to the impurities that are associated with the
ore to be assayed.

The crude lump is then remelted in a small iron ladle at as low a
temperature as possible, and the fused metal is poured into a shallow
trench about 4 inches long by 3/4 of an inch wide cut in a block of
white marble. The metal will be silvery-white if the temperature
employed be correct; if too hot, the surface will show a yellow, red, or
blue colour (according to the heat employed); in such case the metal
should be remelted at a lower temperature. If the metal on cooling
remains perfectly clear and bright, then it may be assumed that the tin
is of good quality and commercially pure. A crystallised or frosted
appearance of the metal indicates the presence of some alloy, say of
iron, copper, zinc, lead, antimony, &c. The assayer who has had much
practice can readily distinguish the metal or metals that are associated
with the ore by noting the appearance of the tin on cooling; and can
fairly judge the quantity of impurity present by the amount of the
crystallisation or stain.

Whilst the foregoing method of assaying cannot lay claim to scientific
accuracy, it is by no means so imperfect as some writers would have us
believe, who state that a loss of 5 to 10 per cent. arises in the
operation. It is certainly the most ready and expeditious mode of
determining the commercial value of a parcel of tin ore, which, after
all, is the main object of all assaying operations.

The difficulty which beginners find in obtaining satisfactory results,
and any loss of metal which those not accustomed to the process may
incur, will invariably occur in the vanning of the powdered slag for the
fine tin, the rest of the operations being easy of execution, and
requiring only the ordinary care necessary for all metallurgical work.

There is no doubt that if low percentage ores containing silica are
assayed in this manner, low results are obtained, as it is impossible
to reduce the whole of the tin in the presence of free silica; with this
class of ores, care should be taken to remove some of the silica by
preliminary vanning, or some flux should be added which will combine
with the silica, and so prevent its entering into combination with the
tin. Low quality tin ores containing iron, copper, lead, zinc, antimony,
etc., combined with arsenic, sulphur, or oxygen, will give very much
higher results than the actual percentage of tin in the sample. The
other metals (being readily reduced in the presence of tin) alloy with
it, and give a hard lump difficult to fuse in the iron ladle; where the
quantity of foreign metals is large, the metal can only be melted to a
stiff pasty mass; so that (in determining the value of a ton of tin ore,
or even reporting on the percentage of tin it contains) not only must
the weight of the assay be the basis for calculation, but the quality
and character of the metal obtained must also be considered. Thus two
ores of tin might be assayed both yielding a similar _produce_, say
13-1/2 (67-1/2 per cent.), and yet one might contain 5 per cent. less
tin than the other.

If it be required to obtain the pure metal from tin ores containing the
ores of other metals associated with them, the latter must be removed by
digesting in strong hydrochloric acid, and washing. The assay may then
be conducted in the usual way, and a fairly pure lump will be obtained.

If wolfram be present in any appreciable quantity in the ore, it
considerably reduces the proportion of lump, and at the same time it
increases the fine tin (or _prillion_, as it is termed) in the assay.
This may be got rid of by boiling in aqua regia, and dissolving out the
tungstic acid which has been liberated by means of ammonia.

It will be seen that this method of assaying tin has its advantages and
its drawbacks. It is quickly performed; with ores of good quality it
gives results not to be excelled by any other process; and it gives the
smelter the actual alloy and quality of metal he may expect to get in
the smelting of the ore, which no other mode of assaying will do:
against which may be set the skill required to obtain accurate results
with the vanning shovel; the loss of metal in poor ores containing an
excess of silica; and the high results from ores containing a large
quantity of metallic impurities.

~Cyanide Method.~--Weigh up 20 grams of the ore and dry it on a scoop
over the Bunsen flame. When dry, weigh, and calculate the percentage of
water from the loss in weight. Transfer the dried ore to an evaporating
dish, and cover with 30 c.c. of hydrochloric acid; boil for 10 or 12
minutes, and then add 5 c.c. of nitric acid and boil again. Dilute with
water, and filter. Transfer the filter and its contents to an E
Battersea crucible, and calcine it for a few minutes. Cool, and weigh
the residue. The loss equals the oxides soluble in acid. Transfer the
residue to the crucible and mix it with its own weight of cyanide of
potassium; add a similar amount of "cyanide" as a cover. Place in the
furnace, and when the charge has attained the temperature of the furnace
(in from 3 to 6 minutes), remove it at once; tap the pot _vigorously_
several times, and then pour its contents quietly into a mould. Dissolve
the slag in water, clean, dry, and weigh the button of tin.


WET METHODS.

~Detection.~--Tin ore is detected by its insolubility in acids, high
specific gravity, and characteristic appearance in water. The powder is
separated from the lighter gangue by washing. It is fused in a Berlin
crucible with five times its weight of potassic cyanide at a moderately
high temperature in a muffle, or over the blowpipe. The slag is washed
off with water, and the metallic buttons or residue treated with
hydrochloric acid (not aqua regia), for some time. One portion of the
solution strikes a purple colour with chloride of gold, another portion
gives a white or grey precipitate or cloudiness with mercuric chloride.
These reactions are characteristic of tin as stannous chloride.

Metallic tin treated with nitric acid becomes converted into a white
insoluble powder (metastannic acid). Aqua regia dissolves tin readily,
forming stannic chloride, and in this solution the metal is detected by
precipitation with sulphuretted hydrogen, which gives a yellow
precipitate. Tin in solution as stannic or stannous chloride is
precipitated as metal by means of zinc.

The fact that tin forms two well-defined series of compounds is taken
advantage of in assaying (just as in the case of iron), by determining
how much of an oxidising agent is required to convert it from the
stannous into the stannic state. For example, on the addition of a
solution of permanganate of potash to a solution of stannous chloride
the oxidation goes on rapidly, and the finishing point is sharp and
distinct; but acid solutions of stannous chloride quickly take up oxygen
from that dissolved in the water used and from the air. Unfortunately,
there is no obvious sign that such oxidation has taken place, except
that (fatal to the assay) a smaller volume of the permanganate is
required. Great care is required with such solutions, both before and
during titration. The addition of an excess of ferric chloride to the
stannous solution, as soon as the whole of the tin has been dissolved,
will lessen this liability to oxidation.

~Separation.~--If the tin is present in an alloy, the substance is
boiled in an evaporating dish with dilute nitric acid until the whole of
the material is attacked. Evaporate nearly to dryness, dilute, boil for
a few minutes, and filter off the white insoluble residue. Under certain
circumstances this residue will be nearly free from other metals, in
which case it is ignited and weighed. If not known to be pure it must be
ignited, reduced in a current of hydrogen, and treated as subsequently
described.

When the tin is present as insoluble oxide in an ore, the substance is
finely powdered, and from 1 to 5 grams of it (according to its richness)
boiled with 30 c.c. of hydrochloric acid in an evaporating dish till the
oxide of iron is seen to be dissolved. Then add 1 c.c. of nitric acid
(or more if much pyrites, &c., is present) and continue the boiling till
these are decomposed; dilute and filter off, washing first with dilute
acid and afterwards with a little dilute ammonia, dry, ignite, and place
in a combustion tube (together with the filter-ash) and heat to redness
for about thirty minutes in a current of dried hydrogen.

[Illustration: FIG. 58.]

The oxide of tin is placed in a porcelain boat (fig. 58), which is then
introduced into a piece of combustion tube. The latter, wrapped in a
piece of wire gauze, is supported on a couple of iron rings, and heated
by one or two Bunsen burners in a furnace fitted up with loose
fire-brick tiles, as shown in fig. 59.

[Illustration: FIG. 59.]

When the reduction is complete the tube is allowed to cool; the boat is
removed and the tin dissolved. Add a rod of zinc to the freely-acid hot
solution, and in a few minutes decant through a filter and wash with
water, after having removed the zinc. Wash the precipitated metal back
into the beaker, and dissolve in 10 c.c. of dilute nitric acid,
evaporate off the excess of acid; dilute, boil, and filter. Wash, dry,
ignite strongly in a porcelain crucible, and weigh.

In the absence of antimony the above separation works very well, but if
this metal is present in quantity the metals precipitated on the zinc
must be covered with hydrochloric acid and treated with a few drops of
nitric. It is then warmed with iron wire until no more of the latter
dissolves. The antimony is precipitated as metal, and the tin remains in
solution as stannous chloride. The antimony is filtered off, and may be
washed with alcohol, and weighed, whilst the tin in the filtrate is
precipitated with zinc, and treated as already described.


GRAVIMETRIC METHOD.

If the tin is not already in the metallic state it is reduced to this
condition by the method given (precipitation by zinc). Treat the
finely-divided metal (washed free from chlorides) in a four-inch
evaporating dish with 10 c.c. of dilute nitric acid, cover with a
clock-glass, and apply a gentle heat until the precipitate appears of a
white colour and the metal is completely attacked. Evaporate nearly to
dryness on a water-bath; then add 50 c.c. of water, heat to boiling, and
filter. Wash with hot water, dry, transfer to a weighed porcelain
crucible, add the filter-ash, ignite strongly, and weigh. The
precipitate after ignition is stannic oxide (SnO_{2}). It is a
yellowish-white powder (darker whilst hot), insoluble in acids, and
contains 78.67 per cent. of tin. Cold dilute nitric acid dissolves tin
to a clear solution, which becomes a white enamel-like jelly on heating;
this (filtered off, washed, and dried) forms an opal-like substance,
which is converted on ignition into stannic oxide with evolution of
nitrous fumes. Stannic oxide when ignited with chlorides is more or less
completely converted into stannic chloride, which volatilises. The
presence of chlorides during the evaporation with nitric acid causes a
similar loss.

~Determination of Tin in an Alloy.~--(_Bronze._)--Take 2 grams, and
attack with 20 c.c. of dilute nitric acid in a covered beaker with the
aid of heat. Boil till the bulk is reduced by one-half, dilute with 50
c.c. of water, allow to settle for a few minutes, and filter; wash well
first with water acidulated with a little nitric acid, and afterwards
with water; dry, ignite, and weigh as stannic oxide.

~Determination of Tin in Tin Ore.~--Treat 5 grams of the dried and
finely-powdered ore with 30 c.c. of hydrochloric acid in a four-inch
evaporating dish. After the soluble oxides have been dissolved add 1 or
2 c.c. of nitric acid, boil off nitrous fumes, dilute, and filter. Dry
the filter, transfer the cleaned ore to a piece of combustion tube ten
or twelve inches long and narrowed at one end. Pass a current of
hydrogen through the tube and heat to redness for 30 minutes; cool
whilst the gas is still passing. Dissolve in 20 c.c. of dilute
hydrochloric acid and keep the solution tinted with permanganate of
potassium. When the colour of the permanganate becomes permanent dilute
to a bulk of 50 c.c. with water, filter, and wash. Heat; add a rod of
zinc weighing about 3 grams; allow to stand for a few minutes; decant
through a filter; and wash, removing the remaining zinc and returning
the tin to the beaker. Treat with 5 c.c. of dilute nitric acid, boil for
some time, take up with water, filter, wash, dry, ignite, and weigh as
stannic oxide.


VOLUMETRIC METHOD.

~Titration with Solution of Permanganate of Potassium.~--This titration
may be made either directly on the solution of stannous chloride
(prepared by dissolving the precipitated metal in hydrochloric acid), or
indirectly, on a solution of ferrous chloride (produced by the reducing
action of the precipitated metal on ferric chloride). The standard
solution of permanganate of potassium is made by dissolving 5.356 grams
of the salt in water and diluting to one litre. 100 c.c. are equivalent
to 1.00 gram of tin.

The precipitated tin is transferred to a flask; and dissolved in 10 c.c.
hydrochloric acid, with the aid of heat and in an atmosphere of carbonic
acid. The acid and metal are placed in the flask; which is then filled
with the gas, and stopped with a cork provided with a rubber valve. When
solution is complete the flask is again filled with carbonic acid. Fifty
c.c. of water freed from air and saturated with carbonic acid are then
added. This water is made by adding a gram of bicarbonate of soda and 2
c.c. of hydrochloric acid to 100 c.c. of water: the effervescence sweeps
out the dissolved oxygen. The permanganate of potassium solution is then
run in from a stop-cock burette in the usual way until a faint pink
tinge is obtained.

The following experiments show the effect of variations in the
conditions of the assay. A solution of stannous chloride equivalent in
strength to the "permanganate" was made by dissolving 19.06 grams of the
crystallised salt (SnCl_{2}.2H_{2}O.) in 50 c.c. of water and 10 c.c. of
hydrochloric acid and diluting to 1 litre with water freed from
dissolved oxygen. 100 c.c. contain 1 gram of tin. In the first
experiments tap water was used and no precautions were taken for
excluding air. Except when otherwise stated, 20 c.c. of the stannous
chloride were used in each experiment with 10 c.c. of hydrochloric acid,
and were diluted to 100 c.c. with water before titration.

~Effect of Varying Hydrochloric Acid.~

  Acid added                  1.0 c.c.  10.0 c.c.  20.0 c.c.  50.0 c.c.
  "Permanganate" required    18.8  "    18.9  "    18.8  "    18.8  "

The only effect of the increase in quantity of acid was to give the
brown of perchloride of manganese instead of the pink of permanganic
acid to mark the finishing point.

~Effect of Varying Temperature.~

  Temperature                15° C.     50° C.     70° C.     100° C.
  "Permanganate" required    18.8 c.c.  18.7 c.c.  18.6 c.c.  18.4 c.c.

~Rate of Atmospheric Oxidation.~--Solutions ready for titration were
exposed to air at the ordinary temperature for varying lengths of time
and then titrated.

  Time exposed      0 min.     5 min.    10 min.    20 min.    60 min.
  "Permanganate"
    required       18.8 c.c.  18.8 c.c.  18.8 c.c.  18.8 c.c.  18.6 c.c.

It is best to titrate at once, although the loss by oxidation is only
small after one hour's exposure.

~Effect of Varying Tin.~

  Stannous chloride
    added           1.0 c.c.  10.0 c.c.  20.0 c.c.  50.0 c.c.  100.0 c.c.
  "Permanganate"
    required        0.7  "     8.8  "    18.0  "    47.4  "     95.4  "

~Effect of Varying Bulk.~

  Bulk                       50.0 c.c.  100.0 c.c.  200.0 c.c.  500.0 c.c.
  "Permanganate" required.    9.0  "     18.3  "     17.4  "     15.1  "

The two last series show an interference, which is due to the oxygen
dissolved in the water, as may be seen from the following similar
experiments, which were, however, performed with water freed from oxygen
and in which the titrations were effected in an atmosphere of carbonic
acid.

~Effect of Varying Tin.~--A new solution of stannous chloride was used.

  Stannous chloride
    added           1.0 c.c. 10.0 c.c. 20.0 c.c. 50.0 c.c. 100.0 c.c.
  "Permanganate"
    required        1.0  "   10.0  "   19.8  "   49.6  "    99.3  "

~Effect of Varying Bulk.~

  Bulk             30.0 c.c. 50.0 c.c. 100.0 c.c. 200.0 c.c. 500.0 c.c.
  "Permanganate"
    required       19.8  "   19.8  "    19.8  "    19.8  "    19.8  "

It will be seen that in working under these conditions the results are
proportional and the method satisfactory.

~Examination of Tin Phosphide.~--(_Phosphor Tin_.)--This substance is
used in the manufacture of "phosphor bronze" and similar alloys. It is a
crystalline, imperfectly-malleable, metallic substance. It is soluble in
hydrochloric acid with effervescence; phosphoretted hydrogen, which
inflames on the addition of a drop or two of nitric acid, being evolved.
It is attacked by nitric acid, yielding a white powder of stannic
phosphate; this is not easily decomposed by ammonium sulphide or readily
soluble in hydrochloric acid.

"Phosphor-tin" is made up only of tin and phosphorus. For the estimation
weigh up 1 gram. Place in a weighed Berlin dish; and cover with 10 c.c.
of nitric acid and 3 or 4 c.c. of water. Let the reaction proceed (under
a clock-glass) on the water-bath till complete. Remove the glass;
evaporate to dryness, and ignite, at first gently over a Bunsen burner,
and afterwards in the muffle at a red heat. Cool in the desiccator, and
weigh as quickly as possible when cold. The substance contains the tin
as stannic oxide, SnO_{2}, and the phosphorus as phosphoric oxide,
P_{2}O_{5}. The increase in weight on the gram of substance taken gives
the weight of the oxygen taken up by the phosphorus and tin, and since 1
gram of tin takes up only 0.271 gram of oxygen, and 1 gram of phosphorus
takes up 1.29 gram, the proportion of tin to phosphorus can be
calculated from the increase in weight. For example, 1 gram of a sample
gave 1.3410 gram of mixed oxides, which is 0.070 gram in excess of that
which would be got with pure tin. If the substance was all phosphorus
the excess would be 1.0190 gram; consequently the proportion of
phosphorus in the substance is 0.070 / 1.019, or 6.87 per cent. The tin
is calculated by difference, 93.13 per cent.

Another method of separating and determining the phosphorus is as
follows:--Take 1 gram of the substance and add to it 15 c.c. of hot aqua
regia. Boil till dissolved, dilute, and precipitate the tin with
sulphuretted hydrogen. To the filtrate add ammonia and "magnesia
mixture." Filter; wash the precipitate with dilute ammonia; dry, ignite,
and weigh as magnesic pyrophosphate. Calculate the phosphorus, and take
the tin by difference.

A sample of phosphor tin gave--

  Tin            93.1 per cent. (by difference)
  Phosphorus      6.9    "
                -----
                100.0

~Tin Arsenide.~--This is met with in tin-smelting; it closely resembles
the phosphide, but the crystals have a duller grey appearance. It
contains simply tin and arsenic. The determination is made by treating 1
gram of the substance with nitric acid and weighing the mixed oxides of
tin and arsenic in the same manner as in the case of the phosphide. One
gram of arsenic will give 1.533 gram of arsenic oxide, As_{2}O_{5};
consequently the excess of weight of the mixed oxides over 1.271 gram
must be divided by 0.262; the result multiplied by 100 gives the
percentage of arsenic. In consequence of the higher atomic weight of
arsenic the results by this method are not so close as with the
phosphide. Each milligram of excess weight (over 1.271) represents 0.38
per cent. of arsenic, As. Both in this and in the corresponding
phosphide determination care must be taken to avoid absorption of
moisture, by allowing the oxides to cool in a desiccator and weighing
quickly.

The percentage of arsenic is better determined as follows:--Weigh up 1
gram of the substance, dissolve in aqua regia, dilute, and pass
sulphuretted hydrogen. Render alkaline with ammonia, and add ammonium
sulphide till the precipitate is dissolved. Add "magnesia mixture."
Filter off the precipitate, wash with dilute ammonia, ignite with a few
drops of nitric acid, and weigh as magnesic pyrarsenate. Calculate the
arsenic and take the tin by difference. A sample treated in this way
gave--

  Tin       96.8 per cent. by difference
  Arsenic    3.2           "
           -----
           100.0

~Examination of Black Tin.~--Dry the ore, and reduce it to a fine
powder. Weigh up 2 grams, and boil with 20 c.c. of hydrochloric acid and
2 c.c. of nitric for ten or fifteen minutes. Filter, and reserve the
filtrate.

~Tungstic Acid.~--Digest the residue with about 50 c.c. of water and a
few c.c. of dilute ammonia for a few minutes, and filter; collect the
filtrate in a weighed porcelain dish, evaporate to dryness, ignite, and
weigh as tungstic acid, WO_{3}.

~Stannic Oxide.~--Dry, ignite, and weigh the insoluble residue. Transfer
to a porcelain boat, and reduce in a current of hydrogen at a red heat
for half an hour. Allow to cool whilst the hydrogen is still passing.
Transfer the boat to a beaker, and dissolve up the tin in 10 c.c. of
hydrochloric acid and a c.c. or so of nitric. Wash out the combustion
tube with some acid and add the washing to the contents of the beaker.
Warm gently, dilute with water, and filter. Collect, dry, ignite, and
weigh the insoluble residue. Through the filtrate pass a rapid current
of sulphuretted hydrogen, allow to settle, and filter. Wash the
precipitate with hot water, dry, calcine gently; ignite with ammonium
carbonate, and weigh as stannic oxide, SnO_{2}. The insoluble residue
will in most cases retain some tin. Fuse it with fusion mixture, take up
with hydrochloric acid, filter, pass sulphuretted hydrogen through the
filtrate, collect and wash the sulphide of tin. Ignite and weigh as
stannic oxide, and add it to that previously obtained.

~Copper.~--Pass sulphuretted hydrogen through the acid filtrate obtained
in the first cleaning of the ore, collect the precipitate, and wash
first with soda solution and then with hot water. Dry, ignite, and weigh
as cupric oxide, CuO. Mix the filtrate with that from the main portion
of the sulphide of tin.

~Ferric Oxide.~--Boil off the sulphuretted hydrogen from the mixed
filtrates and peroxidise with nitric acid. Add ammonia in slight excess,
boil, filter, dry, ignite, and weigh the precipitate as ferric oxide.
This will be practically pure, but the iron in it must be determined by
dissolving and titrating. The filtrate from the iron may contain zinc,
lime, and magnesia, but rarely in quantities sufficient to be
determined.

~Silica, &c.~---The silica may be calculated from the weight of the
residue insoluble in acid, after the reduction of the tin in hydrogen,
by deducting from it the weight of the oxide of tin subsequently found.
Or it may be determined as follows:--The insoluble portion is fused with
fusion mixture, and taken up with hydrochloric acid, as already
described. On filtering, the filter will retain a portion of the silica.
The rest is recovered, after the removal of the stannous sulphide, by
evaporating to dryness, taking up with hydrochloric acid, and filtering
through the same filter. It is washed, dried, ignited, and weighed as
silica. The filtrate from the silica is boiled with a little nitric acid
and precipitated with ammonia. The precipitate is collected, washed,
ignited, and weighed as ferric oxide and alumina (but it frequently
contains oxide of titanium). When the last is present it is determined
by fusing with bisulphate of potash and extracting with cold water. The
solution is nearly neutralised with ammonia, charged with sulphurous
acid, and boiled. The precipitate is collected, washed, dried, ignited,
and weighed as oxide of titanium, TiO_{2}. The difference between this
weight and that of the combined oxides gives the ferric oxide and
alumina. The filtrate from the mixed oxides is examined for lime and
magnesia.

~Sulphur.~--Rub up 5 grams of the ore with 5 grams of nitre, transfer to
a porcelain dish, and fuse over a Bunsen burner for fifteen minutes.
When cold, extract with water, and determine the sulphur volumetrically
with standard barium chloride. The sulphur may be present as sulphide or
sulphate.

~Arsenic.~--Take 5 grams, and evaporate with nitric acid; dilute, add
ammonia, pass sulphuretted hydrogen, and filter.

To the filtrate add "magnesia mixture." Collect the precipitate, ignite
with nitric acid, and weigh as magnesic pyrarsenate.

The following may be taken as an example of the composition of an impure
black tin:--

  Tungstic acid    1.8 %
  Stannic oxide   79.0
  Silica           2.6
  Titanic oxide    0.8
  Copper oxide     0.9
  Ferric oxide    13.4
  Sulphur          0.4
  Arsenic          0.3
                  ----
                  99.2

~Examination of Hardhead.~--In the smelting of tin ores a quantity of
speise, known as "hardhead," is produced. It is essentially an arsenide
of iron, carrying a considerable quantity of tin. Much of this last is
present in the form of small buttons of metal distributed through the
mass. The buttons can be seen on careful inspection, and become evident
on powdering.

In assaying the substance, a variation in the usual method of sampling
is required, because of the quantity of metal present which cannot be
powdered. After powdering as finely as possible, the coarse particles
are sifted off and weighed. The weight of the powder is also taken. The
method of working is best illustrated by an example. A sample of
hardhead weighed 155.1 grams, and gave 21.0 grams of coarse particles,
equivalent to 13.5 per cent. of the whole. The fine portion weighed 134
grams, which is equivalent to 86.5 per cent.

Thirteen and a half grams of the coarse material were dissolved in aqua
regia, and diluted with water to 1 litre. Ten c.c. of this contain 0.135
gram of the metallic portion, which is the amount contained in 1 gram of
the original hardhead. If, in a determination, 1 gram of the substance
is wanted, weigh up 0.865 gram of the powdered portion, and add to it 10
c.c. of the solution. It will be seen that these together make up 1 gram
of the original sample. The solution of the metallic portion must be
saved until the analysis is finished.

~Tin and Copper.~--Weigh up the portion of the powdered stuff equivalent
to 1 gram of the sample. Transfer to a flask, and cover with 10 c.c. of
the solution of the metallic portion and 10 c.c. of aqua regia. Boil
gently till oxidation is complete and the nitric acid for the greater
part driven off. Dilute to 100 c.c. with water, and pass sulphuretted
hydrogen for some time. Filter, wash with hot water, and rinse through
the funnel back into the flask. Digest with yellow sodium sulphide until
only a light, flocculent, black precipitate is left. Filter this off,
wash with hot water, dry, calcine, treat with a little nitric acid,
ignite, and weigh as copper oxide, CuO. The weight multiplied by 0.7983
gives the weight of copper.

The filtrate containing the tin is rendered acid with hydrochloric acid,
and filtered. The precipitate is rinsed into a half-pint beaker, covered
with 20 c.c. of hydrochloric acid, and boiled down to about 20 c.c. The
solution is filtered off from the sulphur and sulphide of arsenic,
which, after washing with hot water, is transferred to a flask labelled
"arsenic." A strip of sheet zinc (2 in. by 1 in.) is placed in the
solution. The evolution of hydrogen should be brisk. In five or ten
minutes decant off a few c.c. of the liquid, and test with sulphuretted
hydrogen for tin. If no yellowish precipitate is formed, decant off the
rest of the liquid, and wash the precipitated metal with hot water two
or three times by decantation. The metal should be in a lump; if there
are any floating particles they must be made to sink by compression with
a glass rod. Transfer the washed metal to an evaporating dish 3 or 4 in.
across, and cover with a few c.c. of hot water. Add nitric acid drop by
drop till the tin is completely attacked. Evaporate nearly to dryness,
and add a drop or two more of nitric acid and 20 c.c. of water. Boil and
filter. Wash with hot water, dry, ignite, and weigh as stannic oxide,
SnO_{2}. Calculate to metallic tin by multiplying by 0.7867.[76]

The filtrate from the first treatment with sulphuretted hydrogen will
probably no longer smell of the gas. Warm and pass the gas for a few
minutes longer. Filter off any precipitate of sulphide of arsenic, and
transfer it to the flask for "arsenic." Boil the filtrate (ignoring any
signs of a further precipitation of arsenic) with a few c.c. of nitric
acid, and separate the iron as basic acetate. Wash; reserve the filtrate
for cobalt.

~Iron.~--Rinse back the "basic acetate," precipitate into the flask, add
ammonia, dilute with water to about 100 c.c., and pass sulphuretted
hydrogen for a few minutes. Filter, and wash with hot water. Collect the
filtrate in the flask labelled "arsenic." Boil the precipitate with
dilute sulphuric acid, filter, and titrate the filtrate with the
permanganate of potassium solution after boiling off the sulphuretted
hydrogen. Report the result as iron. The sulphuric acid will not effect
complete solution, a light black residue will remain, chiefly sulphur;
this must be rinsed into the filtrate from the acetate separation. It
contains cobalt.

~Cobalt.~--The filtrate from the acetate separation will have a pink
colour. Render it ammoniacal and pass sulphuretted hydrogen. Collect
the precipitate on a filter, dry, and ignite. Dissolve in hydrochloric
acid, and evaporate nearly to dryness with an excess of nitric acid.
Dilute with 10 or 20 c.c. of water and add potash solution in slight
excess. Add acetic acid until the solution is acid and the precipitate
is quite dissolved. Add 20 or 30 c.c. of a strong solution of potassium
nitrite, and determine the cobalt, as described on pp. 254, 256. Boil
the filtrate from the cobalt, precipitate with hydrochloric acid, render
ammoniacal, and test for zinc, nickel, and manganese.

_The remainder of the tin_ will be contained in the flask labelled
"arsenic." Acidify with hydrochloric acid and filter. Rinse into a
beaker, and evaporate to a small bulk with 10 c.c. of nitric acid.
Dilute and filter. Dry the precipitate, consisting of stannic arsenate
(2SnO_{2}.As_{2}O_{5}), ignite, and weigh. Calculate the tin it contains
by multiplying by 0.4453, and add to that already found.

~Arsenic.~--This is determined in a separate portion. Weigh up a portion
of the powder equivalent to 1 gram of the hardhead, place in a pint
flask, and boil with 10 c.c. of nitric acid. When action has ceased add
10 c.c. of the solution of the metallic portion and then hydrochloric
acid (a few drops at a time) till solution is complete. Warm gently in
dissolving, but do not boil. Dilute to about 100 c.c., render alkaline
with ammonia, and add 20 c.c. of yellow ammonium sulphide. Digest at a
gentle heat for about thirty minutes, filter, and wash. Add 50 c.c. of
magnesia mixture, shake well, allow to stand for an hour, filter, and
wash with dilute ammonia. The precipitate is dissolved and then titrated
with uranium acetate, or it is evaporated with nitric acid, ignited, and
weighed as pyrarsenate of magnesia. Calculate the result to arsenic, As.

~Sulphur.~--Weigh up a portion of the powder equivalent to 2 or 3 grams
of the hardhead. Rub up in a mortar with 5 grams of nitre and fuse in a
porcelain dish for ten minutes. Extract with water, add 20 or 30 c.c.
(as the case may be) of the solution of the "metallics." Add 10 grams of
sodic acetate, and ferric chloride until the precipitate turns brown;
dilute with water to half a litre, boil, and titrate with standard baric
chloride, as described under _Sulphur_. Report as sulphur.

A sample of hardhead examined in this way gave--

  Sulphur              3.00%
  Arsenic             27.10
  Tin                 22.2
  Copper               1.64
  Iron                43.2
  Cobalt               2.6
                     ------
                      99.74

~Examination of Tin Slags.~--In tin smelting works the term "slag" is
applied to the unfused portion of the charge. It is made up of unburnt
anthracite and small lumps of slag proper together with some buttons of
metallic tin. This is rarely, if ever, assayed. The slag proper (or, as
it is generally called, "glass") is a silicate of iron, alumina, and
lime, containing from 3 to 7 per cent. of tin. It is thus examined:--The
sample after bruising on an iron plate, is reduced to a very fine powder
by grinding in an agate mortar. In this state it is in most cases
readily decomposed by hydrochloric acid.

~Determination of Tin.~--Where the percentage of tin only is required,
take 2 grams of the powdered slag and well mix with it 20 c.c. of
hydrochloric acid, and heat to boiling. Add 1 c.c. of nitric acid, allow
to stand for fifteen minutes, dilute with water, and filter. Pass a
rapid current of sulphuretted hydrogen for some time. Allow to settle,
and filter. The precipitate, after washing with hot water, is dried, and
gently calcined until the greater part of the sulphur is burnt off. It
is then strongly ignited in the muffle (or over the blowpipe) with the
addition of a small lump of ammonic carbonate. The residue is weighed as
stannic oxide (SnO_{2}); and is calculated to metallic tin by
multiplying by 0.787. The percentage on the slag is calculated in the
usual way.

The tin is always best determined in the examination of slags by a
separate assay carried out in this way. The determination of the other
constituents is thus made:--

~Silica.~--Take 2 grams of the powdered slag and cover them, in a small
evaporating dish, with 20 c.c. of hydrochloric acid; mix well by
stirring with a glass rod; and evaporate to dryness. If (as is generally
the case) tungsten is present the solution will be blue. Take up with 20
c.c. of hydrochloric acid. Add 1 c.c. of nitric acid; and reduce by
boiling to about half the bulk. Add about 20 c.c. of water, boil, and
filter. Wash the residue with hot dilute hydrochloric acid. It consists
of silica with the tungstic acid. Wash it back into the dish; and digest
with 5 or 10 c.c. of a cold solution of ammonic carbonate. Filter; and
collect the filtrate and washings in a weighed porcelain dish. Dry the
residue, ignite strongly, and weigh as silica, SiO_{2}. In certain
exceptional cases this may contain some unaltered cassiterite, which is
easily recognised by its appearance.

~Tungsten.~--The ammonic carbonate filtrate from the silica is
evaporated to dryness, ignited strongly over the blowpipe, and weighed.
The residue is tungstic acid, WO_{3}. The tungsten may be conveniently
reported in this form, although it is probably present as a lower
oxide.

~Tin.~--The acid filtrate from the silica and tungstic acid is treated
with sulphuretted hydrogen. The sulphide of tin is filtered off. Since
the percentage of tin has been already determined, this precipitate may
be neglected; or may be treated in the same way as the previous one, so
as to check the result. Since some stannic chloride will have been lost
in the evaporation, a low result may be expected. The tin should be
reported as stannous oxide; and is calculated by multiplying the
percentage of tin by 1.136.

The filtrate from the tin is boiled rapidly down to remove sulphuretted
hydrogen; and then peroxidised with 1 or 2 c.c. of nitric acid. It is
cooled, transferred to a graduated flask, and diluted with water to 200
c.c.

~Ferrous Oxide and Alumina.~--Half the filtrate from the tin (that is,
100 c.c.) is taken, nearly neutralised with soda, and treated with
sodium acetate. The basic acetate precipitate obtained on boiling is
filtered off and washed. Reserve the filtrate. The precipitate is
dissolved off the filter with hot dilute hydrochloric acid; and the
solution thus formed is treated with a slight excess of ammonia, and
boiled. The precipitate is filtered off, washed with hot water, dried,
ignited, and weighed as mixed ferric oxide and alumina. The ignited
precipitate is then dissolved with sulphuric and hydrochloric acids; and
the iron determined in the solution by titration with the solution of
stannous chloride. The iron found is calculated to and reported as
ferrous oxide, FeO (factor = 1.286). To find the alumina, which is best
estimated by difference, multiply the iron by 1.428 to get the weight of
ferric oxide, and deduct this from the weight of alumina and ferric
oxide found. This, of course, gives the alumina. A direct determination
may be made by removing the tin from the titrated solution with
sulphuretted hydrogen, filtering, nearly neutralising with ammonia, and
boiling with a few grams of hyposulphite of soda. The precipitate,
filtered, washed, and ignited, is the alumina, which is weighed. The
direct determination gives a slightly low result.

~Oxides of Zinc and Manganese.~--These are determined in the filtrate
from the basic acetate precipitate by rendering alkaline with ammonia,
and passing a current of sulphuretted hydrogen. Generally a small, but
decided, precipitate of alumina comes down, together with sulphides of
any zinc or manganese which is present. The precipitate is allowed to
settle, dried, ignited, and weighed. The metals are separately
determined in it; and the residue is counted as alumina, and added to
that already found. The mixed precipitate amounts to from 1 to 2 per
cent. of the sample.

~Lime.~--The filtrate from the last is treated with ammonic oxalate,
boiled for a few minutes, allowed to settle, and filtered. The
precipitate is washed with hot water; dried; ignited; and weighed as
carbonate, after gentle ignition; or as lime, after strong ignition in
the muffle.

~Magnesia.~--The filtrate from the lime is treated with sodic phosphate
and ammonia. It is well mixed by stirring, and allowed to stand
overnight. The precipitate is washed with dilute ammonia, dried,
ignited, and weighed as pyrophosphate.

~Soda and Potash.~--These are determined in the remaining half of the
filtrate from the tin. The solution is rendered ammoniacal with ammonia;
and treated, first with sulphuretted hydrogen, and then with ammonium
oxalate. The precipitate is filtered off and rejected. The filtrate is
evaporated in a small porcelain dish over a Bunsen burner, or on the
sand bath; and towards the close (or earlier if the evaporation is not
proceeding well) nitric acid is added. The evaporation is carried to
dryness; and the residue heated nearly to redness. The residue, which
consists of magnesia with carbonates and chlorides of the alkalies, is
extracted with water; and filtered. The filtrate is evaporated with
hydrochloric acid in a weighed platinum dish, ignited gently, and
weighed. This gives the weight of the mixed chlorides of sodium and
potassium; which are then separated and determined as described under
_Potash_.

It must be remembered when calculating the percentage that (with the
exception of the silica, tungstic acid, and tin) the determinations have
been made on 1 gram of the sample.

The following analysis will illustrate the composition of such a slag:--

  Tungstic acid                  1.3%
  Silica                        39.4
  Stannous oxide                 8.1
  Ferrous oxide                 26.2
  Alumina                       14.8
  Oxide of manganese           traces
  Lime                           7.9
  Magnesia                       0.5
  Alkalies calculated as soda    1.7
                               -----
                                99.9


TITANIUM.

Titanium only occurs as a mineral in its oxidised state, or as titanic
oxide (TiO_{2}). It is a substance which has little commercial value,
and is generally recognised as one of the rare bodies; although, in
small quantities, it is widely disseminated. It occurs in granite,
basalt, and other igneous rocks in quantities up to as much as 1 per
cent. It is also met with in clays and iron ores, and in river sands, in
which it is often associated with stream tin. The proper minerals of
titanium are rutile (TiO_{2}), titaniferous iron (titanate of iron), and
sphene (titanate and silicate of lime).

The oxide of titanium (like cassiterite and quartz) is undecomposed by
hydrochloric or nitric acid; so that it is generally found in the
residue insoluble in acids. The titanates, however, are attacked, and a
portion of the titanium dissolves; so that it must be looked for in both
the filtrate and residue. Oxide of titanium in its native form, or after
ignition, may be made soluble by fusing the finely-divided substance
with fusion mixture in a platinum dish. The resulting titanate is
dissolved out of the "melt" by cold hydrochloric acid.

The method most commonly used is fusion with bisulphate of potash. This
renders the oxide of titanium soluble in cold water. The process is as
follows:--The substance is extracted with hydrochloric and nitric acids,
and the solution reserved for further treatment; the residue is dried,
moistened with sulphuric acid, and evaporated once or twice to dryness
with hydrofluoric acid. It is then fused with bisulphate of potash, and
the "melt" extracted with cold water until all soluble matter is
removed. The solution is filtered. The residue may consist of unremoved
silica, and oxides of tantalum, niobium, and, perhaps, chromium. On the
prolonged boiling of the filtrate, the oxide of titanium (and oxide of
zirconium, if any) is precipitated.

Any titanium dissolved by the first extraction with acids is recovered
in the following way:--Sulphuretted hydrogen is passed into the acid
solution, and any precipitate that may be formed is filtered off. The
filtrate is oxidised, and the iron, aluminium, and titanium are
separated as basic acetates (see under _Iron_). The precipitate is dried
and fused with bisulphate of potash. The "melt" is extracted with cold
water, filtered if necessary, and the solution rendered first faintly
alkaline with ammonia, then very slightly acid with sulphuric acid. 30
or 40 c.c. of a saturated solution of sulphurous acid is added, and the
oxide of titanium precipitated by prolonged boiling. It is filtered off,
added to the precipitate previously got, ignited with ammonic carbonate
towards the end, and then weighed.

~Detection.~--Titanium is detected in an insoluble residue by fusing the
residue for some time in a bead of microcosmic salt. In the reducing
flame it gives a violet colour, which becomes reddish-brown if much iron
is present. In the oxidising flame it gives a colourless or whitish
bead. It is best detected in acid solutions by the deep brown or iodine
colour developed on adding hydroxyl. A solution of this can be prepared
by pouring peroxide of barium (BaO_{2}) diffused in water into dilute
hydrochloric acid (a little at a time), and keeping the acid in excess.

~Separation.~--In the usual course of an analytical separation the
hydrate of titanium will be thrown down with ferric hydrate, &c., on the
addition of ammonic chloride and ammonia. It is best separated from this
precipitate by fusion with bisulphate of potash, as already described,
but it must be remembered that the presence of much mineral acid
prevents complete precipitation when the solution is boiled. Further, if
phosphates are present, the precipitate will contain phosphoric oxide;
it may be freed from this by fusion with sodium carbonate. A very good
method of separating titanium from iron is to add tartaric acid and
ammonia to the solution, and then precipitate the iron (as sulphide)
with sulphuretted hydrogen. The filtrate contains the titanium, which is
recovered by evaporating and igniting. It may be separated from zirconia
by the action of sodium carbonate, which precipitates both; but when
concentrated, redissolves the zirconia. The separation from large
quantities of silica is best effected by evaporating with hydrofluoric
acid, which volatilises the silicon; but sulphuric acid must be present,
otherwise some titanium also will be lost, as may be seen from the
following experiments,[77] in which oxide of titanium (pure, ignited)
was evaporated to dryness with a quantity of hydrofluoric acid known by
experiment to be sufficient to volatilise 1 gram of silica.

_Without sulphuric acid_, 0.0466 gram of titanic oxide left 0.0340 gram,
showing a loss of about 25 per cent.

_With sulphuric acid_ the following results were obtained:--

  Oxide taken.    Left after Evaporation
                      and Ignition.
  0.0340 gram          0.0340 gram
  0.0414  "            0.0413  "
  0.0520  "            0.0520  "
  0.0352  "            0.0352  "


GRAVIMETRIC DETERMINATION.

The titanic hydrate thrown down by ammonia (or on boiling the solution
from the bisulphate) is collected, washed, dried, ignited strongly with
the addition of a little ammonic carbonate, and weighed. The substance
is titanic oxide (TiO_{2}), and is generally reported as such. It
contains 60.98 per cent. of titanium. It should be white, if pure
(Holland), white, yellow, or brown (Fresenius), or black (Tidy).


VOLUMETRIC METHOD.

A method has been proposed based on the reduction of titanic oxide by
zinc in hydrochloric acid solutions to the sesquioxide. The reduction is
marked by the development of a violet or green colour, the former with
chlorides and the latter when fluorides are present. The quantity of
titanium reduced is measured by titrating with permanganate of potassium
solution. The water used must be free from dissolved oxygen.


TUNGSTEN AND TUNGSTATES.

Tungsten occurs in nature only in the oxidised state, or as tungstic
acid (WO_{3}), either free, as in wolframine, or combined with oxides of
manganese and iron, as in wolfram, or with lime, as in scheelite.
Wolfram occurs associated with tin ores, the value of which is
consequently lowered. Both wolfram and scheelite are of considerable
importance as a source of tungstic acid for the manufacture of sodium
tungstate, which is used as a mordant and for some other purposes, and
as a source of metallic tungsten, which is used in steel-making.

The tungsten minerals have a high specific gravity (6 to 7.5). On
treatment with hydrochloric acid or aqua regia they are decomposed; the
yellow tungstic acid separates and remains insoluble.

Tungsten itself is insoluble in nitric acid or aqua regia; but is
converted into tungstic acid (WO_{3}) by prolonged and strong ignition
in air. Alloys containing tungsten leave tungstic acid after treatment
with nitric acid or aqua regia. Tungstic acid may be got into solution
after fusion with alkalies or alkaline carbonates. This solution gives
with hydrochloric acid a white precipitate of tungstic acid, which
becomes yellow on boiling, but the separation is not complete. Fusion
with bisulphate of potash gives a residue, which does not dissolve in
water, but is soluble in ammonic carbonate. For the assay of minerals
containing tungsten these reactions are only occasionally taken
advantage of for testing or purifying the separated tungstic acid.

~Detection.~--The minerals are easily recognised by their physical
characters, and the yellow tungstic acid separated by boiling with acids
is the best test for its presence; this, after decanting and washing,
immediately dissolves in a few drops of dilute ammonia. A solution of
tungstate acidulated with hydrochloric acid becomes intensely blue on
the addition of stannous chloride and warming. Fused in a bead of
microcosmic salt it gives a clear blue colour (reddish-brown if iron is
also present) in the reducing flame, but is colourless in the oxidising
flame.

~Solution and Separation.~--The decomposition and solution of natural
tungstates is difficult to effect owing to the separation of tungstic
acid; the method of treatment is as follows:--Boil the finely-powdered
substance with hydrochloric acid or aqua regia till it apparently ceases
to be attacked; dilute, filter, and wash with dilute hydrochloric acid.
Cover with dilute ammonia, and filter the solution, which contains
ammonic tungstate, into an evaporating dish. Treat the residue again
with acid, and again dissolve out the separated tungstic acid with
ammonia, and repeat this operation until decomposition is complete. By
this means there will be obtained--(1) a solution containing tungstate
of ammonia; (2) an insoluble residue with silicates, and oxides of tin,
niobium, tantalum, &c.; and (3) an acid solution containing the soluble
bases. The tungstate of ammonia requires simple evaporation on the
water-bath and gentle ignition in order to cause the tungstic acid to be
left in an almost pure state; possibly, it may carry a little silica.


GRAVIMETRIC DETERMINATION.

The tungstic acid is dissolved, and separated as ammonic tungstate, and,
after evaporation, is gently ignited, the heat being increased towards
the end. The residual tungstic acid is fixed, so that when the ammonia
has been driven off it may be strongly heated without loss. It is a dark
yellow or brown powder whilst hot, which becomes a light yellow on
cooling. If any reduction has taken place it will be more or less
greenish. It is weighed when cold, and is the trioxide or "tungstic
acid" (WO_{3}), which contains 79.31 per cent. of tungsten. After its
weight has been taken its purity is checked by fusing with hydric
potassic sulphate, extracting with water, and treating the residue with
ammonic carbonate. Any silica present will be left undissolved; it
should be separated and weighed, and its weight deducted from that of
the tungstic acid found.

~Determination of Tungstic Acid in Wolfram.~--Take 2 grams of the
finely-powdered sample and boil with 50 c.c. of hydrochloric acid for
half an hour, adding 5 c.c. of nitric acid towards the end. Allow to
stand overnight and boil again for 15 or 20 minutes; dilute with an
equal volume of water, and filter. Wash with dilute hydrochloric acid,
dissolve in a few c.c. of warm dilute ammonia, and dilute to 200 c.c.
with distilled water; allow to settle, and filter. Evaporate in a
weighed dish, ignite, and weigh.

The following analysis will illustrate the composition of a sample of
Cornish wolfram as brought into the market:--

  Tungstic acid                  50.1%
  Cassiterite                    10.9
  Ferrous oxide                  24.6
  Manganous oxide                 5.4
  Niobic oxide, alumina,  &c.     3.5
  Silica                          1.2
  Copper oxide                    2.7
  Zinc oxide                      0.22
  Arsenic                         0.51
  Sulphur                         0.20
                                 -----
                                 99.33


NIOBIC AND TANTALIC OXIDES.

These oxides are commonly met with in samples of wolfram and tinstone,
especially niobic. They are probably present in the form of columbite, a
niobate of iron and manganese; and tantalite, a tantalate of the same
metals.

On boiling with hydrochloric acid they are both liberated, and remain
for the greater part (all the niobic) in the insoluble residue with the
tungstic acid. On removing the latter with dilute ammonia they remain as
a white insoluble precipitate, very prone to run through the filter on
washing. They may be dissolved in hydrofluoric acid either at once or
after fusion with bisulphate of potash, and extraction with cold water.
To the solution in hydrofluoric acid gradually add a boiling solution of
acid potassium fluoride (HF, KF.). Potassic fluotantalate (soluble in
200 parts of water) separates out first, and afterwards potassic
fluoniobate (soluble in 12 parts of water). The separated salts (after
heating with sulphuric acid and washing out the potassium sulphate
formed) are ignited with ammonic carbonate, and weighed as tantalic
oxide (Ta_{2}O_{5}) and niobic oxide (Nb_{2}O_{5}) respectively.

They are both white powders. The oxide of niobium dissolved in a bead of
microcosmic salt gives a bluish colour in the reducing flame. The oxide
of tantalum dissolves in the bead, but gives no colour.

FOOTNOTES:

[76] This will give almost the whole of the tin; a further portion will
be got in subsequent work, and must be added to this result.

[77] Published by P. Holland, in the _Chemical News_, vol. lix. p. 27.



CHAPTER XIII.

MANGANESE, CHROMIUM, &c.


MANGANESE.

Manganese occurs mainly as black oxide (MnO_{2}) in the mineral
pyrolusite; and, in a less pure form, in psilomelane and wad. The value
of the ore depends rather on the percentage of available oxygen than on
the proportion of metal present. The results of assays are generally
reported as so much per cent. of the dioxide (MnO_{2}). In smaller
quantities it is very widely distributed. Manganese itself has a value
for steel-making; or, rather, for the making of spiegeleisen and
ferro-manganese, which are used in the Bessemer and Siemens processes.
For this purpose the percentage of the metal (Mn) is required.
Consequently the minerals of manganese may be considered in two
aspects--(1) as a source of oxygen; and (2) as a source of manganese.
These will require separate consideration.

The black oxide is mainly used in the preparation of chlorine,
liberation of which it brings about when treated with hot hydrochloric
acid, or with a mixture of common salt and sulphuric acid. The quantity
of chlorine which is obtained depends upon the proportion of dioxide
present;[78] and in assaying may either be measured by its equivalent of
iodine liberated, or by the oxidising effect on an acid solution of
ferrous sulphate. When the ore also carries substances which have a
reducing effect (such as ferrous compounds), such assays will give, not
the total dioxide (MnO_{2}), but less, by the amount required to oxidise
these impurities; and this is exactly what is required in valuing such
an ore for commercial purposes. Manganese compounds are characterised by
the readiness with which they may be converted into highly-oxidised
bodies. Solution of manganese in hydrochloric acid, rendered alkaline
with ammonia, yields a clear solution,[79] which rapidly takes up oxygen
from the air, forming a brown precipitate of the oxide (Mn_{2}O_{3}).
The addition of bromine or chlorine to such a solution determines the
precipitation of a still higher oxide (approximately MnO_{2}). On
treating a compound containing manganese with nitric acid and dioxide of
lead (PbO_{2}), the oxidation is carried still further, a
purple-coloured solution of permanganic acid (HMnO_{4} or
H_{2}O.Mn_{2}O_{7}) being formed. On fusing minerals containing (even
traces of) manganese with sodium carbonate in an open crucible, a green
"melt" is obtained which owes its colour to sodium manganate
(Na_{2}MnO_{4} or Na_{2}O.MnO_{3}). This salt is soluble in water,
forming a green solution; which, when rendered acid, rapidly changes
into the permanganate with the characteristic purple colour.
Permanganate of potash is a salt much used in assaying, with some
properties of which the student will have already become familiar.

Compounds of manganese, on boiling with strong hydrochloric acid, yield
manganous chloride[80] (MnCl_{2}).

The properties given above serve for the detection of manganese; the
higher oxides are distinguished by causing the evolution of chlorine
(with its peculiarly suffocating smell) when acted on with hydrochloric
acid; while the green "melt," with sodium carbonate, can be relied on
for the recognition of manganese itself. There is no dry assay of
manganese ores.


WET METHODS.

Strong hydrochloric acid is the best solvent for ores of manganese; but
where the proportion of dioxide (MnO_{2}) is required, the solution is
effected during the assay. The ore should be in a very fine state of
division before treatment with acids.

The separation of manganese from other metals is thus effected: Ignite,
in order to destroy any organic matter which may be present; dissolve in
hydrochloric acid, and evaporate to dryness, to separate silica. Take up
with hydrochloric acid, dilute, pass sulphuretted hydrogen, and filter.
Boil off the excess of gas, peroxidise the iron with a drop or two of
nitric acid, and separate the iron as basic acetate (as described under
_Iron_).[81] If the iron precipitate is bulky, it is dissolved in a
little hydrochloric acid, reprecipitated, and the filtrate added to the
original one. Neutralise with soda, and add bromine in excess; heat
gradually to boiling, allow to settle, and filter. The precipitate is
impure dioxide of manganese (containing alkalies and, possibly, cobalt
or nickel).


GRAVIMETRIC DETERMINATION.

Dissolve the precipitate in hydrochloric acid, and boil; add a slight
excess of carbonate of soda, warm, and filter. Wash with hot water, dry,
carefully ignite in an open Berlin crucible, and weigh. The substance is
the brown oxide (Mn_{3}O_{4}), and contains 72.05 per cent. of
manganese. If the percentage of dioxide is required it may be calculated
by multiplying the percentage of manganese by 1.582. It must be borne in
mind that the manganese should never be calculated to dioxide except
when it is known to exist in the ore only in that form.


VOLUMETRIC METHODS.

The two methods are based on the oxidising effect of manganese dioxide;
and if the metal does not already exist in this form it will require a
preliminary treatment to convert it. The following method due to Mr. J.
Pattinson[82] effects this: A quantity of the ore containing not more
than .25 grams of the metal (Mn), is dissolved in hydrochloric acid in a
pint beaker, and, if necessary, 3 or 4 c.c. of nitric acid are added to
peroxidise the iron, and ferric chloride is added if required, so that
there may be at least as much iron as manganese. Calcium carbonate is
added till the solution is slightly red; and next the redness is removed
by the cautious addition of acid; 30 c.c. of zinc chloride solution
(containing 15 grams of zinc per litre) are added, the liquid is brought
to boil and diluted to about 300 c.c. with boiling water; 60 c.c. of a
solution of bleaching powder (33 grams to the litre and filtered),
rendered slightly greenish by acid, are then run in and are followed by
3 grams of calcium carbonate suspended in 15 c.c. of boiling water.
During effervescence the beaker is covered, the precipitate is stirred,
and 2 c.c. of methylated spirit are mixed in. The precipitate is
collected on a large filter, washed with cold water, and then with hot,
till free from chlorine, which is tested for with starch and potassium
iodide. The acid ferrous sulphate solution (presently described) is then
measured into the beaker, and the precipitate, still in the paper,
added; more acid is added (if necessary), and the solution is diluted
and titrated. In place of bleaching powder solution, 90 c.c. of bromine
water (containing 22 grams per litre) may be used.


FERROUS SULPHATE ASSAY.

This method, which is the one commonly used, is based on the
determination of the amount of ferrous iron oxidised by a known weight
of the ore. It is known that 87 parts of the dioxide will oxidise 112
parts of ferrous iron;[83] therefore 1 gram will oxidise 1.287 gram of
ferrous iron, or 1 gram of ferrous iron oxidised will be equivalent to
0.7768 gram of the dioxide. The finely-divided substance containing the
dioxide is digested in a solution of a known quantity of iron in
sulphuric acid. The iron, of course, must be in excess, which excess is
determined when the ore is dissolved by titrating with standard
permanganate or bichromate of potash solution. The assay resolves itself
into one for the determination of ferrous iron, for which the standard
solutions and method of working described under _Iron_ are used.

The assay is as follows:--For rich ores, 2 grams of clean soft iron wire
are treated, in a pint flask, with 100 c.c. of dilute sulphuric acid and
warmed till dissolved. Carefully sample the ore, and in one portion
determine the "moisture at 100° C.;" grind the rest in a Wedgwood mortar
with a little pure alcohol until free from grit. This reduces the
substance to a finely-divided state and assists solution. Evaporate off
the alcohol and dry at 100° C., mix well, and keep in a weighing-bottle.
Weigh up 2 grams and add them to the solution of iron in the flask;
carefully wash it all down into the acid liquid. On rotating the flask
the ore will rapidly dissolve, but gentle heat may be used towards the
end to complete the solution. When the residue is clean and
sandy-looking, and free from black particles, the flask is cooled, and
the residual ferrous iron is determined by titration with
"permanganate." The iron thus found, deducted from the 2 grams taken,
will give the amount of iron peroxidised by the dioxide contained in the
2 grams of ore. This divided by 2 and multiplied by 77.68 will give the
percentage of dioxide in the sample, or multiplied by 49.41 will give
that of metallic manganese.

When the quantity of manganese or of the dioxide to be determined is
small, it is not necessary to use 2 grams of iron; 1 gram, or even less,
may be taken. The iron may be used in the form of a standard solution of
ferrous sulphate and portions measured off, thus saving the labour of
weighing.

~Determination of Dioxide in a Manganese Ore.~--Weigh up 1 or 2 grams of
the finely-powdered ore[84] and an equal weight of pure iron wire,
dissolve the wire in 50 or 100 c.c. of dilute sulphuric acid, and, when
solution is complete, add the ore and warm till it too is dissolved.
Cool and titrate the remaining ferrous iron with the permanganate or
bichromate of potassium solution.

For example, 0.7560 gram of pyrolusite and 1.000 gram of iron were taken
and treated as above; 13.9 c.c. of "permanganate" (standard 100 c.c. =
0.4920 gram iron) were required; this indicates that 0.0684 gram of iron
was left unoxidised by the ore. The iron oxidised, then, was 0.9316 gram
(1.000 - 0.0684); multiplying this by 0.7768, we find that 0.7237 gram
is the quantity of manganese dioxide which was present. This is
equivalent to 95.77 per cent.;

        0.7560 : 0.7237 :: 100 : 95.77.


IODINE METHOD.

It has been already stated that when dioxide of manganese is boiled with
strong hydrochloric acid chlorine is given off, and that the amount of
chlorine so liberated is a measure of the dioxide present. If the
chlorine is passed into a solution of potassium iodide, an equivalent of
iodine will be set free.[85] This is apparently a very indirect way of
determining how much of the dioxide is present; but the reactions are
very sharp, and the final determination of the iodine is an easy one.

[Illustration: FIG. 60.]

The finely-powdered sample of dioxide is placed in a small flask
provided with an exit tube leading into a solution of potassic iodide
(fig. 60). On adding hydrochloric acid and boiling, the chlorine evolved
is driven into the iodide solution and there absorbed; the boiling is
continued till the steam and hydrochloric acid fumes have driven the
last portions of the chlorine out of the flask and into the solution. In
this experiment there is a strong tendency for the iodide solution to
rush back into the flask. This tendency is overcome by avoiding draughts
and regulating the heat; or by placing a lump of magnesite in the flask,
which acts by evolving carbonic acid and so producing a steady outward
pressure. When the distillation is finished the tube containing the
iodine is detached and washed out into a beaker. If the solution is
strongly acid it should be almost neutralised by the cautious addition
of dilute ammonia. If crystals of iodine have separated, potassium
iodide must be added in quantity sufficient to dissolve them. The
condenser must be kept cool whilst the chlorine is passing into it.

The solution, transferred to a beaker, is titrated with a standard
solution of sodic hyposulphite (100 c.c. = 1.27 gram iodine or 0.435
gram of dioxide of manganese). In titrating, the solution should be
cold, or not warmer than 30° C. The bulk may vary from 100 to 200 c.c.;
but it is best always to work with the same volume. The "hypo" is run in
with constant agitation until the brown colour has been reduced to a
light yellow; 5 c.c. of starch solution are then added and the titration
cautiously continued until the end is reached; the finish is indicated
by a change from blue to colourless.

The assay solution may be acidified with acetic, sulphuric, or
hydrochloric acid before titrating with "hypo;" but it must be only
faintly so. An excess of acid may be nearly neutralised with ammonia
without interference, but excess of alkali is fatal. Bicarbonate of soda
must not be used in excess; it is best to avoid it altogether. The assay
solution should be titrated at once, as it weakens on standing; and the
"hypo" solution should be standardised every two or three days, as its
strength is not constant.

_The standard solution of hyposulphite of soda_ is made by dissolving 25
grams of the salt (Na_{2}S_{2}O_{3}.5H_{2}O) in water and diluting to 1
litre. 100 c.c. are equivalent to 1.27 gram of iodine.

This solution is standardised by weighing, in a small beaker, about half
a gram of iodine, to which is added a crystal or two of potassium iodide
and a few drops of water. When dissolved, the solution is diluted to 100
c.c., and titrated in the manner described. The starch solution is made
in the manner described under the iodide copper assay. 5 c.c. are used
for each titration.

In determining the effects of variations in the condition of the assay a
solution of iodine was used, which was equivalent in strength to the
"hypo" solution. It was made by dissolving 12.7 grams of iodine with 25
grams of potassium iodide in a little water and diluting to 1 litre. 100
c.c. of this solution were found (at the time of the experiments) to be
equivalent to 102.0 c.c. of the "hypo."

~Effect of Varying Temperature.~--The bulk of the solution was 100 c.c.;
20 c.c. of iodine were taken, and 5 c.c. of starch solution were added
towards the end as indicator. These conditions are also those of the
other experiments, except where otherwise stated. Iodine being volatile,
it is to be expected that with hot solutions low results will be
obtained.

  Temperature        15°        20°        40°        60°        80°
  "Hypo" required  20.4 c.c.  20.4 c.c.  20.1 c.c.  19.2 c.c.  15.5 c.c.

These show that the temperature should not much exceed 20°.

~Effect of Exposure of the Iodine Solution.~--Twenty c.c. of the iodine
were diluted to 100 c.c., and exposed for varying lengths of time in
open beakers at the ordinary temperature, and then titrated.

  Time exposed       --         1 day      2 days    3 days
  "Hypo" required  20.4 c.c.  16.1 c.c.  13.6 c.c.  9.4 c.c.

~Effect of Varying Bulk.~--These experiments were carried out in the
usual way, bulk only varying.

  Bulk             100.0 c.c.  200.0 c.c.  300.0 c.c.  500.0 c.c.
  "Hypo" required   20.4  "     20.4  "     20.4  "     20.4  "

~Effect of Varying Acid.~--These experiments were under the usual
conditions, the bulk being 100 c.c. The results were--

  Acetic acid          --        1.5 c.c.  30.0 c.c.
  "Hypo" required    20.4 c.c.  20.7  "    20.7  "

  Hydrochloric acid    --        1.5 c.c.  15.0 c.c.
  "Hypo" required    20.4 c.c.  20.6  "    20.9  "

  Sulphuric acid       --        0.5 c.c.  20.0 c.c.
  "Hypo" required    20.4 c.c.  20.7  "    15.2  "[86]

  Nitric acid          --        0.5 c.c.  10.0 c.c.
  "Hypo" required    20.4 c.c.  21.5  "    could not be titrated.

In the application of this titration to the assay of manganese ores,
hydrochloric and hydriodic acids are the only ones likely to be present.

~Effect of Alkalies.~--On theoretical grounds the presence of these is
known to be inadmissible. A solution rendered faintly alkaline with
ammonia required only 11.2 c.c. of "hypo;" and another, with 0.5 gram of
caustic soda, required 4.0 c.c. instead of 20.4 c.c. as in neutral
solutions.

~Effect of nearly Neutralising Hydrochloric Acid Solutions with
Ammonia.~--Provided care is taken not to add excess of ammonia, this
has a good effect, counteracting the interference of excess of acid.
Thus 20 c.c. of iodine (as before) required 20.4 c.c. of "hypo;" with 15
c.c. of hydrochloric acid 20.7 c.c. were required, but with 15 c.c. of
acid, nearly neutralised with dilute ammonia 20.4 c.c. were used.

~Effect of the Addition of Starch.~--The addition of varying quantities
of starch has no effect, provided it is added when the titration is
nearly finished, as the following experiments show:--

  Starch added       1.0 c.c.   5.0 c.c.  10.0 c.c.  50.0 c.c.
  "Hypo" required   20.4  "    20.4  "    20.4  "    20.5  "

But if the starch is added before the titration, the results are liable
to error.

  Starch added       1.0 c.c.  50.0 c.c.
  "Hypo" required   20.4  "    24.0  "

The starch should be used fresh, and is best made on the day it is used;
after four days the finishing point is not so good.

~Effect of Varying Potassium Iodide.~--An excess of iodide is always
required to keep the iodine in solution; a larger excess has little
effect.

  Iodide added       --        1 gram    20 grams
  "Hypo" required  20.4 c.c.  20.5 c.c.  20.6 c.c.

The 20 c.c. of iodine used, itself contained 0.5 gram of potassium
iodide.

~Effect of Foreign Salts.~--

  Bicarbonate of soda
    added                --        0.5 gram    1.5 gram    5.0 grams
  "Hypo" required      20.4 c.c.  18.2 c.c.   17.1 c.c.   16.0 c.c.

The solution obviously must be free from bicarbonate of soda. This
should be remembered, since when titrating arsenic assays with iodine it
must be present; and students must avoid confounding the two titrations.

In some other experiments, in which 10 grams each of the salts were
taken, the following results were obtained:--

  Salt added          --         AmCl      AmNO_{3}    Am_{2}SO_{4}
  "Hypo" required   20.4 c.c.  20.5 c.c.   20.3 c.c.     20.2 c.c.

  Salt added         NaCl       NaNO_{3}   Na_{2}SO_{4}
  "Hypo" required   20.3 c.c.  20.4 c.c.   20.4 c.c.

~Effect of Varying Iodine.~--

  Iodine added     1.0 c.c.  10.0 c.c.  20.0 c.c.  50.0 c.c.   100.0 c.c.
  "Hypo" required  1.3  "    10.2  "    20.4  "    51.0  "     102.0  "

~Determination of Dioxide in a Manganese Ore.~--Weigh up 0.25 to 0.3
gram of the powdered ore; place in a flask, cover with 10 c.c. of
hydrochloric acid, and close the flask with a paraffined cork, and bulbs
(as shown in fig. 60), having previously charged the bulb with 5 grams
of potassium iodide in strong solution. Heat the flask, and boil
cautiously for about fifteen minutes. Wash the contents of the bulbs
into a large beaker, nearly (but not quite) neutralise with dilute
ammonia, and titrate with the standard "hypo."

As an example, 0.2675 gram of pyrolusite was taken, and required 60.3
c.c. of standard "hypo" (100 c.c. equal 1.185 gram iodine, or 0.4042
gram MnO_{2}), which equals 0.2437 gram of the dioxide or 91.1 per cent.


COLORIMETRIC METHOD.

When compounds of manganese free from chlorides are boiled with nitric
acid and dioxide of lead,[87] the manganese is converted into
permanganic acid, which is soluble and tints the solution violet. The
depth of colour depends on the amount of manganese present, and this
should not much exceed 10 milligrams. A quantity of substance containing
not more than this amount of manganese should be boiled for a few
minutes with 25 c.c. of a solution containing 5 c.c. of nitric acid, and
10 or 20 c.c. of dilute sulphuric acid, with 2 or 3 grams of lead
dioxide. Filter through asbestos, wash by decantation with dilute
sulphuric acid, make up with distilled water[88] to a definite bulk, and
take a measured portion for the colorimetric determination.

The standard solution of manganese is made by dissolving 0.1435 gram of
permanganate of potash (KMnO_{4}) in a little water acidulated with
nitric acid, and diluting to 1 litre. One c.c. will contain 0.05
milligram of manganese.


PRACTICAL EXERCISES.

1. What percentage of manganese (Mn) is contained in permanganate of
potash (KMnO_{4})?

2. Ten c.c. of a solution of permanganate of potash is found to oxidise
10 c.c. of an acid solution of ferrous sulphate. The manganese is
determined in the titrated solution by precipitation as dioxide and
titrating. How much of the ferrous solution will be oxidised in the
second titration?

3. What weight of potassium iodide would be just sufficient to absorb
the chlorine evolved by 0.5 gram of pure dioxide of manganese?

4. What weight of iron must be dissolved up so as to have an excess of
0.25 gram after oxidation by 1 gram of pure dioxide?

5. What weight of the brown oxide, Mn_{2}O_{4} will be left on igniting
1 gram of the pure dioxide?


CHROMIUM.

Chromium occurs in nature chiefly as chromite or chrome iron ore
(FeO_{2}Cr_{2}O_{3}, with more or less MgO and Al_{2}O_{3}), which is
the chief ore. It is a constituent of some silicates, and is frequently
met with in very small quantities in iron ores. It occurs as chromate in
crocoisite (PbCrO_{4}), and some other rare minerals.

The metal is used in steel-making. Steel containing about 0.5 per cent.
of it is rendered very hard; but its chief value is in its salts, the
chromates. These are highly-coloured compounds, generally red or yellow.
Some of the insoluble chromates are used as pigments; chromate of lead
or chrome-yellow is the most important. The soluble chromates, those of
soda and potash, are valuable chemicals, and are largely used in the
preparation of pigments, dyeing and tanning, and as oxidising agents.

Chromium forms two important classes of compounds--chromic salts,
corresponding to the oxide Cr_{2}O_{3}, and chromates, which contain the
trioxide CrO_{3}. Solutions of chromic salts are green, whilst those of
the chromates are yellow. Chromates are reduced to chromic salts by the
action of most reducing agents in the presence of an acid; and this
property is used in assaying for the volumetric determination of ferrous
iron, &c. The chromates in solution are more stable than other similar
oxidising agents, and consequently are generally used in the laboratory
as one of the standard oxidising agents for volumetric analysis. They
have the disadvantage of requiring an outside indicator. Bichromate of
potash (K_{2}Cr_{2}O_{7}) is the salt generally used for this purpose.

Chromic salts are oxidised to chromate by fusion with "fusion mixture"
and nitre, or by treating with chlorine in an alkaline solution.

Chromic salts closely resemble those of ferric iron, and in the ordinary
course of analysis chromic hydrate (green) is precipitated together with
ferric hydrate, alumina, &c., on the addition of ammonic chloride and
ammonia. The ignited oxide, Cr_{2}O_{3}, however, is not reduced on
heating to redness in a current of hydrogen.

~Detection.~--Chromium is detected by fusing the powdered substance with
"fusion mixture" and nitre. The melt is extracted with water and
filtered. The filtrate is acidified with acetic acid, and treated with a
few drops of a solution of lead acetate. A yellow precipitate indicates
chromium. Substances containing chromium impart a green colour to the
borax bead in both flames. Small quantities of chromate in neutral
solution can be found by the dark or violet-red colouration imparted
thereto on boiling with a dilute decoction of logwood.

~Solution and Separation.~--Chromates and chromic salts are generally
soluble in water or dilute acids. Chrome iron ore, however, and ignited
chromic oxide are insoluble; and the former presents considerable
difficulty on attempting to open up by the usual methods. A large number
of mixtures have been tried in order to get all the chromium in a
soluble form. Among these are the following. One part of the very
finely-powdered ore is fused with any of these mixtures.

  (1) 10 parts of bisulphate of potash.
  (2)  5 parts of bisulphate of potash and 5 parts of potassium fluoride.
  (3)  5 parts of hydric potassic fluoride.
  (4) 12 parts of bisulphate of potash; and, afterwards, with 6 parts of
       carbonate of soda and 6 parts of nitre.
  (5)  8 parts of borax; afterwards, with carbonate of soda till it ceases
       to effervesce; then, with 3 parts of carbonate of soda and 3 of
       nitre.
  (6)  4 parts of borax and 6 parts of fusion mixture.
  (7) 12 parts of caustic potash.
  (8) 10 parts of caustic soda and 30 of magnesia.
  (9)  5 parts of caustic soda and 3 of magnesia.
  (10) 2 parts of carbonate of soda and 1 of lime.
  (11) 6 parts of soda-lime and 2 of chlorate of potash.
  (12)   Sodium peroxide.

Of these, numbers 1, 2, and 3 yield the chromium in a form soluble in
dilute acids, as chromic salt. The rest in a form soluble in water, as
potassium or sodium chromate.

On boiling an insoluble chromium compound with chlorate of potash and
nitric acid, the chromium passes into solution as chromate. This method,
however, does not answer for chrome iron ore. In the fusion methods the
ore must be very finely powdered, well mixed with the fluxes, and
subjected to a prolonged fusion in a platinum vessel at a high
temperature. Undecomposed particles require re-fusion.

The aqueous extract containing the chromate is ready for volumetric
work, except in those cases where nitre has been used. For gravimetric
work the solution is acidified with hydrochloric acid, then mixed with
ammonia in slight excess, boiled, and filtered. The filtrate is
acidified with hydrochloric acid, and treated with sulphuretted
hydrogen, warmed, rendered slightly alkaline with ammonia, and the gas
again passed. The chromium is precipitated as chromic hydrate mixed with
sulphur from the reduction with sulphuretted hydrogen. It is filtered
off, washed with hot water, and ignited. It is weighed as chromic oxide.


GRAVIMETRIC DETERMINATION.

The solution containing the chromium, freed from other metals and earths
and in the form of (green) chromic salt, is heated to boiling. If any
chromate is present reduce it with sodium sulphite or sulphuretted
hydrogen. Add ammonia in slight excess, boil till the liquid is free
from a red tint, and allow to settle for a few minutes. Filter, wash
with hot water, dry, and ignite strongly in a loosely-covered crucible.
Cool, and weigh. The substance is chromic oxide, Cr_{2}O_{3}, and
contains 68.62 per cent. of chromium. It is a dark-green powder
insoluble in acids.

When, as is generally the case, the chromium exists altogether as
chromate (phosphates and arsenates being absent) it is best to proceed
as follows:--Render the solution acid with acetic acid, then add sodium
acetate to the solution and heat nearly to boiling; next treat with a
slight excess of acetate of lead, and boil. Allow to settle, and filter.
Wash the precipitate with hot water, dry in the water-oven or at a low
temperature. Transfer the precipitate to a weighed Berlin crucible, burn
the filter separately, ignite below redness, cool in the desiccator, and
weigh. The substance is lead chromate, PbCrO_{4}, and contains 16.1 per
cent. of chromium, or 23.53 per cent. of chromic oxide (Cr_{2}O_{3}).


VOLUMETRIC METHOD.

This is based on the oxidation of ferrous iron by the solution
containing the chromium as chromate. A known weight of iron (0.5, 1, or
1.5 gram, according to the quantity of chromate) is dissolved in 50 c.c.
of dilute sulphuric acid. The solution containing the chromate is added,
and the remaining ferrous iron titrated with the permanganate or
bichromate of potassium solution, as described under _Iron_. The iron
thus found is deducted from that taken, and the difference gives the
iron oxidised by the chromate. This multiplied by 0.3101 gives the
chromium, Cr, and when multiplied by 0.4529 gives the chromic oxide,
Cr_{2}O_{3}.


COLORIMETRIC METHOD.

Small quantities of chromium may be determined, after conversion into
chromate, colorimetrically. The solution, which should not contain more
than a few milligrams in 100 c.c., is acidified with acetic acid and
compared against an equal volume of water rendered acid with acetic
acid and tinted with a standard bichromate of potassium solution. This
standard bichromate is made by dissolving 2.827 grams of the salt in
water and diluting to 1 litre. One c.c. will contain 1 milligram of
chromium, Cr. The manner of working this assay is the same as that
adopted in the other colorimetric processes.

~Determination of Chromium in Steel.~[89]--Weigh up 2.4 grams, dissolve
in hydrochloric acid, and evaporate to dryness. Fuse with sodium
carbonate and nitre, extract with water, and make up to 301 c.c. Take
250 c.c. of the clear liquor, boil with hydrochloric acid, add sodium
phosphate, and then ammonia in slight excess. Heat till clear. Filter
off the precipitate, dissolve it in hydrochloric acid, and evaporate to
dryness. Take up with a little acid, filter, and precipitate with a
slight excess of ammonia. Wash, ignite, and weigh as chromium phosphate
(3Cr_{2}O_{3},2P_{2}O_{5}), which contains 42.2 per cent. of chromium.


VANADIUM.

Vanadium occurs in certain rare minerals, such as vanadinite
(3Pb_{3}(VO_{4})_{2}.PbCl_{2}), a vanadate of lead; mottramite, a
vanadate of copper and lead; and dechenite, a vanadate of lead and zinc.
It is occasionally found in iron and copper ores and in the slags from
them. In Spanish copper-precipitates it is found along with chromium,
and is probably derived from the iron used for precipitating. The
vanadates, like the chromates, are coloured compounds, generally yellow
or red. On reduction, blue solutions are got. In their general reactions
vanadates resemble phosphates.

Vanadium is detected by the red colouration produced by passing
sulphuretted hydrogen into ammoniacal solutions for some time. On adding
an acid to the filtered solution a brown precipitate of the sulphide is
produced. This gives with borax a colourless bead in the oxidising, and
a green one in the reducing, flame.

It is separated by fusing the ore with potassic nitrate, extracting with
water and precipitating with baric chloride. The precipitate is boiled
with dilute sulphuric acid, filtered, neutralised with ammonia, and
saturated with ammonic chloride. Ammonium vanadate separates out. It is
filtered off, ignited, and weighed as vanadic oxide, V_{2}O_{5},
containing 56.18 per cent. of vanadium.


MOLYBDENUM.

Molybdenum occurs in nature chiefly as molybdenite (MoS_{2}); it also
occurs in wulfenite, a molybdate of lead (PbMoO_{4}), and in molybdic
ochre (MoO_{3}).

Molybdate of ammonia is an important reagent in the determination of
phosphates, the manufacture of which compound is the chief purpose to
which molybdenum is applied.

Iron and copper ores frequently contain molybdenum, sometimes in
quantity; consequently it is met with in slags and pig-iron.

Molybdenum forms several series of salts. In those corresponding to the
lower oxides it is basic; but the trioxide (MoO_{3}) is the acid oxide
which forms a series of salts called the molybdates. All molybdenum
compounds are converted into the trioxide by boiling with nitric acid.
The trioxide is a white powder readily dissolved by ammonia. It fuses at
a red heat, and volatilises freely in contact with air. It is slightly
soluble in water.

Molybdates are easily reduced, with the production of coloured
solutions, by most reducing agents. Sulphuretted hydrogen first produces
a blue tint, and then precipitates a brown sulphide. The precipitation
as sulphide is only complete on prolonged treatment; a green colour
indicates that some molybdenum still remains in solution. The
precipitated sulphide is soluble in ammonium sulphide.

~Detection.~--Molybdenum is detected by its behaviour with sulphuretted
hydrogen. Molybdenite can only be mistaken for graphite, from which it
is easily distinguished by yielding sulphur dioxide on roasting, and by
giving, on charcoal, a yellowish white incrustation, which becomes blue
on touching it for a moment with the reducing flame. The borax-bead is
colourless in the oxidising, and dark-brown in the reducing, flame.


GRAVIMETRIC DETERMINATION.

The solution containing the molybdate is neutralised and treated with an
excess of mercurous nitrate. The precipitate is allowed to settle for
some time, filtered, and washed with a dilute solution of mercurous
nitrate. Then it is dried, transferred to a weighed Berlin crucible
containing ignited oxide of lead, mixed, ignited, and weighed. The
increase in weight gives the amount of trioxide, MoO_{3}. This contains
66.7 per cent. of molybdenum.


URANIUM.

Uranium occurs chiefly as pitchblende, which is an impure oxide
(U_{3}O_{8}). It is also found as sulphate in uranochre, johannite, &c.;
and as phosphate in the uranites, torbernite (hydrated phosphate of
uranium and copper), and autunite (hydrated phosphate of uranium and
lime). It also occurs in some rarer minerals.

The oxide is used for colouring glass; and the nitrate and acetate are
used as reagents. "Uranium yellow," used for enamel painting, is sodium
uranate. The uranates, in which the oxide of uranium acts as an acid,
are mostly insoluble and of secondary importance.

Uranium forms two families of salts, uranous and uranic; corresponding
to the oxides UO_{2} and UO_{3} respectively. The former are generally
green and the latter yellow. Uranous salts are converted into uranic by
boiling with nitric acid or other oxidising agents. Uranic salts, on the
other hand, are easily reduced by sulphuretted hydrogen, stannous
chloride or zinc. This property is made use of in determining the
quantity of uranium in pure solutions by titrating with permanganate of
potassium solution as in the case with iron.

~Detection.~--The most characteristic reaction of the uranium compounds
is their behaviour in the presence of alkaline carbonates in which they
are freely soluble; even ammonium sulphide will not precipitate uranium
from these solutions. On neutralising the carbonate with an acid a
uranate of the alkali is precipitated. Ammonia or sodic hydrate (free
from carbonates) give yellow precipitates, which are insoluble in excess
of the reagent, but are soluble in acids. Ferrocyanide of potassium
gives a reddish-brown precipitate. Uranium colours the borax-bead
yellowish-green in the oxidising, and green in the reducing, flame.

~Solution and Separation.~--The compounds of uranium are soluble in
acids. Powder the substance and evaporate with an excess of nitric acid.
Take up with hydrochloric acid, dilute, pass sulphuretted hydrogen, and
filter. Peroxidise the filtrate with a little nitric acid, add an excess
of ammonic carbonate and some ammonium sulphide, and filter. Render the
solution acid, boil; and precipitate the uranium by means of ammonia.
Filter off, and wash it with dilute ammonic chloride. Ignite, and weigh
as protosesqui-oxide, U_{3}O_{8}.


GRAVIMETRIC DETERMINATION.

The solution containing the uranium free from other metals is, if
required, first peroxidised by boiling with nitric acid. Ammonia in
slight excess is added to the nearly-boiling solution. A yellow
precipitate is formed, which is filtered off hot and washed with a
dilute solution of ammonium chloride. The precipitate is dried and
ignited; and weighed as U_{3}O_{8}, which contains 84.8 per cent. of
uranium.


VOLUMETRIC METHOD.

This is based on the precipitation of uranium as phosphate from acetic
acid solutions and the recognition of complete precipitation by testing
with potassic ferrocyanide; it is the converse of the process for the
volumetric determination of phosphate.

_The standard solution of phosphate_ is prepared by dissolving 29.835
grams of hydric sodic phosphate (Na_{2}HPO_{4}.12H_{2}O) in water and
diluting to 1 litre. 100 c.c. will be equivalent to 2 grams of uranium.

Take 1 gram of the sample (or, if poor in uranium, 2 grams) and separate
the uranium as described. Dissolve the precipitate in nitric acid and
evaporate to a small bulk, add 2 grams of sodium acetate, dilute with
water to 100 c.c., and boil. Titrate the boiling solution with the
sodium phosphate till it ceases to give a brown colouration with
potassium ferrocyanide. Calculate the percentage in the usual way.

FOOTNOTES:

[78] MnO_{2} + 4HCl = MnCl_{2} + Cl_{2} + 2H_{2}O.

[79] Provided a sufficiency of ammonic chloride is present.

[80] With some silicates, &c., a preliminary fusion with sodium
carbonate will be necessary.

[81] Instead of sodium acetate, ammonium succinate can be used.

[82] _Journ. Soc. Chem. Industry_, vol. x. p. 333.

[83] MnO_{2} + 2FeSO_{4} + 2H_{2}SO_{4} = Fe_{2}(SO_{4})_{3} + MnSO_{4}
+ 2H_{2}O.

[84] If the ore is very rich, a smaller quantity (0.75 or 1.5 gram) must
be taken; otherwise the iron will be insufficient.

[85] MnO_{2} + 4HCl = MnCl_{2} + 2H_{2}O + Cl_{2}.
      Cl_{2} + 2KI  = 2KCl + I_{2}.

[86] Iodine probably lost by volatilisation.

[87] Obtained as a brown powder by digesting red lead with nitric acid
and filtering.

[88] The water for dilution and the dilute sulphuric acid used for
washing should be previously tested, to see they have no reducing
action, with dilute permanganate of potassium solution.

[89] Arnold and Hardy, _Chemical News_, vol. lvii. p. 153.



CHAPTER XIV.

EARTHS, ALKALINE EARTHS, ALKALIES.


ALUMINA.

Alumina, the oxide of aluminium (Al_{2}O_{3}), is found in nature fairly
pure in the mineral corundum; transparent and coloured varieties of
which form the gems sapphire and ruby. A coarser compact variety
contaminated with oxide of iron constitutes emery. Compounded with
silica, alumina forms the base of clays and many rock-forming minerals.
China clay (or kaolin) is used as a source of alumina. Bauxite, hydrated
alumina, is also used for the same purpose--that is, for the preparation
of sulphate of alumina. The mineral cryolite is a fluoride of aluminium
and sodium.

Corundum is characterised by a high specific gravity (4.0) and extreme
hardness. By these it is distinguished from felspar and similar
minerals, which it somewhat resembles in general appearance.

Aluminium is used for a variety of small purposes: it is white, light,
and very tenacious; but owing to the difficulty of its reduction it is
expensive.

Aluminium forms one series of salts which closely resemble those of
ferric iron. It forms an interesting series of double sulphates, known
as the alums. Common potash alum is
Al_{2}(SO_{4})_{3},K_{2}SO_{4},24H_{2}O.

~Detection.~--Alumina is not precipitated from its acid solution by
sulphuretted hydrogen, but it is thrown down by ammonia (with the other
earths) as a white hydrate, soluble in soda and insoluble in ammonic
carbonate. Filtered off and ignited, it assumes, after treatment with
nitrate of cobalt before the blowpipe, a blue colour which is
characteristic. With natural compounds containing metallic oxides this
colour is masked. It is more satisfactory to make a separation in the
wet way and to test the ignited oxide.

~Separation and Solution.~--If the substance is insoluble in
hydrochloric acid it is finely powdered and fused with "fusion mixture"
with the help, in the case of corundum (which is very refractory) of a
little caustic soda or potash. The method of working is the same as
that described for opening up silicates. See under _Silica_. Corundum
cannot be powdered in Wedgwood, or even agate, mortars; since it rapidly
wears these away and becomes contaminated with their powder. It is best
to use a hard steel mortar and to extract the metallic particles from
the bruised sample with a magnet or dilute acid.

When the substance has been completely attacked and dissolved, it is
evaporated to dryness with an excess of hydrochloric acid on the
water-bath to render any silica present insoluble. The residue is
extracted with hydrochloric acid and freed from the second group of
metals by means of sulphuretted hydrogen. The filtrate from this (after
removing the sulphuretted hydrogen by boiling) is nearly neutralised,
and treated with 8 or 10 grams of hyposulphite of soda[90] in solution.
It is then boiled till the sulphurous oxide is driven off. The
precipitate is filtered off, ignited, and weighed as alumina.

It is sometimes more convenient to proceed as follows:--After boiling
off the sulphuretted hydrogen peroxidise the iron with a little nitric
acid, add a solution of ammonic chloride, and then ammonia in very
slight excess; boil, filter, wash, ignite, and weigh the oxides. These
generally consist of ferric oxide and alumina. It is a common practice
to determine the iron, calculate it to ferric oxide, and so to estimate
the alumina indirectly. This may be done either by igniting in a current
of hydrogen and estimating the iron by the weight of oxygen lost; or, by
dissolving with sulphuric and hydrochloric acids, and determining the
iron volumetrically. It should be borne in mind that these oxides will
also contain any phosphoric oxide that happened to be in the mineral.

In general analyses of samples containing alumina, it may be contained
in both the soluble and insoluble portions. In these cases it is better
to fuse the sample with "fusion mixture" before treatment with acids.
The alumina in the fused mass will exist in a state soluble in acids.


GRAVIMETRIC DETERMINATION.

Solutions containing alumina free from the other metals are diluted to a
convenient bulk and heated nearly to boiling. Add chloride of ammonium,
and then ammonia in slight excess; boil, allow to settle, filter, and
wash with hot water. Dry the precipitate, and ignite in a platinum or
porcelain crucible at the strongest heat. Cool, and weigh. The substance
is alumina, Al_{2}O_{3}, which contains 52.94 per cent. of aluminium.
It is only in special cases, such as the analysis of metals and alloys,
that it is reported as aluminium. The percentage of alumina is generally
given.

Ignited alumina is difficultly soluble in acids; it is not reduced by
hydrogen at a red heat. Ignited with ammonium chloride portions are
volatilised.

~Direct Determination of Alumina in the Presence of Iron.~--The iron and
alumina are precipitated as hydrates by ammonia. The precipitate is
dissolved in hydrochloric acid and the iron reduced to the ferrous
state. It is then added to a hot solution of potash or soda. The
solution is boiled till the precipitate settles readily, filtered, and
washed with hot water. The alumina is contained in the filtrate, which
is acidified with hydrochloric acid and the alumina precipitated
therefrom as hydrate with ammonia, as just described.

~Determination of Alumina in the Presence of Phosphates and Iron.~--For
details, see a paper by R.T. Thomson in the "Journal of the Society of
Chemical Industry," v. p. 152. The principles of the method are as
follows:--If the substance does not already contain sufficient
phosphoric oxide to saturate the alumina, some phosphate is added. The
iron is reduced to the ferrous state and phosphate of alumina
precipitated in an acetic acid solution. It is purified by
reprecipitation, ignited, and weighed as phosphate
(Al_{2}O_{3},P_{2}O_{5}), which contains 41.8 per cent. of alumina,
Al_{2}O_{3}.


EXAMINATION OF CLAYS.

~Moisture.~--Take 5 grams of the carefully-prepared sample and dry in
the water-oven till the weight is constant.

~Loss on Ignition.~--Weigh up 2 grams of the sample used for the
moisture determination, and ignite in a platinum-crucible to redness,
cool, and weigh.

~Silica and Insoluble Silicates.~--Weigh up another 2 grams of the dried
sample, and place them in a platinum dish; moisten with water, and cover
with 20 c.c. of sulphuric acid. Evaporate and heat gently to drive off
the greater portion of the free acid. Allow to cool; and repeat the
operation. Extract by boiling with dilute hydrochloric acid, filter,
wash, dry, ignite, and weigh. The quantity of insoluble silicates is
determined by dissolving out the separated silica with a strong boiling
solution of sodium carbonate. The residue (washed, dried, and ignited)
is weighed, and reported as "sand."

~Alumina and Ferrous Oxide.~--To the filtrate from the silica add "soda"
solution till nearly neutral, and then sodium acetate. Boil and filter
off the precipitate. Reserve the filtrate. Dissolve the precipitate in
hydrochloric acid, and dilute to exactly 200 c.c. Divide into two parts
of 100 c.c. each. In one determine the iron by reducing and titrating in
the way described under volumetric iron. Calculate the percentage as
ferrous oxide, unless there are reasons to the contrary, also calculate
its weight as ferric oxide. To the other portion add ammonia in slight
excess, and boil. Filter, wash with hot water, dry, ignite, and weigh as
mixed alumina and ferric oxide. The weight of the ferric oxide has
already been determined in the first portion: deduct it, and the
difference is the weight of alumina.

~Lime.~--To the reserved filtrate, concentrated by evaporation, add
ammonium oxalate and ammonia; boil, filter, ignite strongly, and weigh
as lime.

~Magnesia~ is separated from the filtrate by adding sodium phosphate. It
is weighed as magnesium pyrophosphate.

~Potash and Soda.~--These are determined in a fresh portion of the
sample by Lawrence Smith's method, as described on page 333.


THORIA.

This is an oxide of thorium, ThO_{2}. It is only found in a few rare
minerals. It is a heavy oxide, having, when strongly ignited, a specific
gravity of 9.2. In the ordinary course of analysis it will be separated
and weighed as alumina. It is separated from this and other earths by
the following method. The solution in hydrochloric acid is nearly
neutralised and then boiled with sodium hyposulphite. The thoria will be
in the precipitate. It is dissolved, and the solution heated with
ammonium oxalate in excess. The precipitate is thorium oxalate, which is
washed with hot water, dried, and ignited. It is then weighed as thoria,
ThO_{2}. Thoria which has been ignited is not readily soluble in acids.


ZIRCONIA.

The oxide of zirconium, ZrO_{2}, is found in the mineral zircon, a
silicate of zirconia, ZrSiO_{4}. When heated intensely it becomes very
luminous, and is used on this account for incandescent lights.

In the ordinary course it is thrown down by ammonia with the other
earths, from which it is thus separated:--The hydrates precipitated in
the cold, and washed with cold water, are dissolved in hydrochloric
acid, nearly neutralised with soda, and precipitated by boiling with
hyposulphite of soda. Dissolve; and from the hydrochloric acid solution
precipitate the thoria (if any) with ammonium oxalate. To the filtrate
add carbonate of ammonia, which will precipitate any titanium present.
The zirconia will be in solution, and is recovered by precipitating with
potassium sulphate, or by evaporating the solution and igniting. It is
separated from alumina by taking advantage of its insolubility in
potassic hydrate.

It is estimated in zircons in the following way:--The powdered substance
is fused with bisulphate of potash, and extracted with dilute sulphuric
acid. The residue is fused with caustic soda and extracted with water.
The portion not dissolved, consisting of zirconate of soda, is dissolved
in hydrochloric acid. The solution is diluted, filtered if necessary,
and treated with ammonia in excess. The precipitate is filtered off,
washed with hot water, dried, ignited, and weighed as zirconia, ZrO_{2}.
This is a white powder, which is insoluble in acids; even in
hydrofluoric acid it is only slightly attacked.


CERIUM.

Cerium occurs as silicate (together with the oxides of lanthanum,
didymium, iron and calcium) in the mineral cerite, which is its chief
source. It also occurs as phosphate in monazite, and as fluoride in
fluocerite. The oxalate is used in medicine. Cerium forms two classes of
salts corresponding to the oxides, cerous oxide (Ce_{2}O_{3}) and ceric
oxide (CeO_{2}). Compounds of cerium with volatile acids yield dioxide
on ignition; and this, on solution in hydrochloric acid, yields cerous
chloride and chlorine.

In the ordinary course cerium is thrown down along with alumina and the
other earths by ammonia. It is separated by dissolving the hydrates in
hydrochloric acid, and oxidizing with chlorine water. On treating with
oxalic acid, cerium, lanthanum, and didymium are precipitated as
oxalates, which on ignition are converted into oxides. These are soluble
in acids. Their solution in hydrochloric acid is nearly neutralised;
acetate of soda is then added, and an excess of sodium hypochlorite. On
boiling, the cerium is precipitated as dioxide, which is filtered off,
ignited, and weighed.

Cerium is detected by giving with borax a bead which is yellow in the
oxidising, and colourless in the reducing flame. Traces of cerium
compounds boiled with dioxide of lead and nitric acid will give a yellow
solution.


LANTHANUM AND DIDYMIUM

occur together with cerium in cerite, and are separated with that metal
as oxalates, as described under _Cerium_.

Didymium salts have a rose or violet colour, and impart (when in
sufficient quantity) the same colour to the borax bead. Solutions have a
characteristic absorption-spectrum.

The separation of lanthanum and didymium in the solution from which the
cerium has been precipitated is effected by precipitating them together
as oxalates, igniting, and dissolving in dilute nitric acid. This
solution is then evaporated to dryness and ignited, for a few minutes,
just below redness. A subnitrate of didymium is formed, and remains as
an insoluble residue on extracting with hot water. The separated salts
are treated with ammonia and ignited, and weighed as oxides (La_{2}O_{3}
and Di_{2}O_{3}).


YTTRIA.

Yttria is found in gadolinite and some other rare minerals. It is
precipitated along with the other earths by ammonia. It is distinguished
by the insolubility of its hydrate in potash, by the insolubility of its
oxalate in oxalic acid, and by not being precipitated by hyposulphite of
soda or potassium sulphate. Further, it is precipitated by potash in the
presence of tartaric acid as an insoluble tartrate. This reaction
distinguishes the members of the yttria group from most of the other
earths. The other members of the group closely resemble it, and amongst
them are erbia, terbia, ytterbia, scandia, &c.


BERYLLIA.

The oxide of beryllium, BeO (also known as glucina), occurs in nature
mainly as silicate. Beryl, the green transparent variety of which is the
emerald, is the best known of these. It is a silicate of alumina and
beryllia.[91] Some other minerals in which it occurs are phenakite,
euclase, and chrysoberyl.

In the ordinary course of analysis, beryllia will be precipitated with
alumina, &c., by ammonic hydrate. It is distinguished by the solubility
of its hydrate in ammonic carbonate, by not being precipitated by
boiling with sodium hyposulphite, and by not being precipitated by
ammonic sulphide from an ammonic carbonate solution.

The analysis of silicates containing beryllia is thus effected. The
finely powdered substance is fused with twice its weight of potassium
carbonate; and the "melt" is extracted with water, and evaporated with a
slight excess of sulphuric acid to render the silica insoluble. Treat
with water, filter, and evaporate the filtrate until a crust is formed.
Potash alum crystallises out. The liquor is poured off into a warm
strong solution of ammonium carbonate. Ferric hydrate and alumina will
be precipitated. They are filtered off, re-dissolved, and again
precipitated in ammonic carbonate solution; the combined filtrates are
boiled for some time, and acidified slightly with hydrochloric acid. The
carbon dioxide is boiled off, and the beryllia is then precipitated as
hydrate with ammonia. The hydrate is washed with hot water, dried,
ignited, and weighed as beryllia, BeO.

Beryllia has a specific gravity of 3.08. It is white, infusible, and
insoluble in water. After ignition, it is insoluble in acids, except
sulphuric, but is rendered soluble by fusion with alkalies.

Beryllia, in a solution of carbonate of ammonia, is precipitated as
carbonate on boiling in proportion as the carbonate of ammonia is
volatilised. The hydrate is dissolved by a boiling solution of ammonic
chloride, ammonia being evolved.


THE ALKALINE EARTHS.


LIME.

Lime is an oxide of calcium, CaO. It occurs abundantly in nature, but
only in a state of combination. The carbonate (CaCO_{3}), found as
limestone, chalk, and other rocks, and as the minerals calcite and
arragonite, is the most commonly occurring compound. The hydrated
sulphate, gypsum (CaSO_{4}.2H_{2}O), is common, and is used in making
"plaster of Paris." Anhydrite (CaSO_{4}) also occurs in rock masses, and
is often associated with rock salt. Phosphate of lime, in the forms of
apatite, phosphorite, coprolite, &c., is largely mined. Lime is a
component of most natural silicates. Calcium also occurs, combined with
fluorine, in the mineral fluor (CaF_{2}). In most of these the acid is
the important part of the mineral; it is only the carbonate which is
used as a source of lime.

Lime, in addition to its use in mortars and cements, is valuable as a
flux in metallurgical operations, and as a base in chemical work on a
large scale. A mixture of lime and magnesia is used in the manufacture
of basic fire-bricks.

Carbonate of lime on ignition, especially when in contact with reducing
substances, loses carbonic acid, and becomes lime. This is known as
"quicklime"; on treatment with water it becomes hot, expands, and falls
to a powder of "slaked lime" or calcium hydrate (CaH_{2}O_{2}). The
hydrate is slightly soluble in water (0.1368 gram in 100 c.c.), forming
an alkaline solution known as lime-water. Calcium hydrate is more
generally used suspended in water as "milk of lime."

As a flux it is used either as limestone or as quicklime. Silica forms
with lime a compound, calcium silicate, which is not very fusible; but
when alumina and other oxides are present, as in clays and in most rocky
substances, the addition of lime gives a very fusible slag.

~Detection.~--Calcium is detected by the reddish colour which its salts
impart to the flame. It is best to moisten with hydrochloric acid (or,
in the case of some silicates, to treat with ammonium fluoride) before
bringing the substance into the flame. When seen through a spectroscope,
it shows a large number of lines, of which a green and an orange are
most intense and characteristic. Calcium is detected in solution (after
removal of the metals by treatment with sulphuretted hydrogen and
ammonium sulphide) by boiling with ammonium oxalate and ammonia. The
lime is completely thrown down as a white precipitate. Lime is
distinguished from the other alkaline earths by forming a sulphate
insoluble in dilute alcohol, but completely soluble in a boiling
solution of ammonium sulphate.

Lime compounds are for the most part soluble in water or in dilute
hydrochloric acid. Calcium fluoride must be first converted into
sulphate by evaporation in a platinum dish with sulphuric acid.
Insoluble silicates are opened up by fusion with "fusion mixture," as
described under _Silica_.

~Separation.~--The separation of lime is effected by evaporating with
hydrochloric acid, to separate silica; and by treating with sulphuretted
hydrogen, to remove the second group of metals. If the substance
contains much iron, the solution is next oxidised by boiling with a
little nitric acid; and the iron, alumina, &c., are removed as basic
acetates. The filtrate is treated with ammonia and sulphuretted
hydrogen, and allowed to settle. The filtrate from this is heated to
boiling, treated with a solution of ammonium oxalate in excess, boiled
for five or ten minutes, allowed to settle for half an hour, and
filtered. The precipitate contains all the lime as calcium oxalate.


GRAVIMETRIC DETERMINATION.

The precipitate of calcium oxalate is washed with hot water, dried,
transferred to a weighed platinum crucible, and ignited at a
temperature not above incipient redness. This ignition converts the
oxalate into carbonate, with evolution of carbonic oxide, which burns at
the mouth of the crucible with a blue flame.[92] Generally a small
quantity of the carbonate is at the same time converted into lime. To
reconvert it into carbonate, moisten with a few drops of ammonic
carbonate solution, and dry in a water-oven. Heat gently over a Bunsen
burner, cool, and weigh. The substance is calcium carbonate (CaCO_{3}),
and contains 56 per cent. of lime (CaO). It is a white powder, and
should show no alkaline reaction with moistened litmus-paper.

Where the precipitate is small, it is better to ignite strongly over the
blowpipe, and weigh directly as lime. With larger quantities, and when
many determinations have to be made, it is easier to make the
determination volumetrically.


VOLUMETRIC METHODS.

These are carried out either by dissolving the oxalate at once in dilute
sulphuric acid, and titrating with permanganate of potassium solution;
or by calcining it to a mixture of lime and carbonate, and determining
its neutralising power with the standard solutions of acid and alkali.

~Titration with Permanganate of Potassium Solution.~--This solution is
made by dissolving 5.643 grams of the salt in water, and by diluting to
1 litre; 100 c.c. are equivalent to 0.5 gram of lime. The solution is
standardised by titrating a quantity of oxalic acid about equivalent to
the lime present in the assay; 0.5 gram of lime is equivalent to 1.125
gram of crystallised oxalic acid. The standardising may be done with
iron. The standard found for iron multiplied by 0.5 gives that for lime.

The process is as follows:--The calcium oxalate (having been
precipitated and washed, as in the gravimetric process) is washed
through the funnel into a flask with hot dilute sulphuric acid, boiled
till dissolved, diluted to 200 c.c. with water, and heated to about 80° C.
The standard solution of "permanganate" is then run in, (not too
quickly, and with constant shaking) until a permanent pink tinge is
produced. The c.c. used multiplied by the standard, and divided by the
weight of the substance taken, will give the percentage of lime.

~Estimation of Lime by Alkalimetry.~--The methods of determining the
amount of an alkali or base by means of a standard acid solution, or,
conversely, of determining an acid by means of a standard alkaline
solution, are so closely related that they are best considered under
one head. The same standard solution is applicable for many purposes,
and, consequently, it is convenient to make it of such strength that one
litre of it shall equal an equivalent in grams of any of the substances
to be determined. Such solutions are termed _normal_. For example, a
solution of hydrochloric acid (HCl = 36.5) containing 36.5 grams of real
acid per litre, would be normal and of equivalent strength to a solution
containing either 17 grams of ammonia (NH_{3} = 17) or 40 grams of sodic
hydrate (NaHO = 40) per litre. It will be seen in these cases that the
normal solution contains the molecular weight in grams per litre; and,
if solutions of these strengths be made, it will be found that they
possess equal neutralising value.

If, now, a solution containing 98 grams of sulphuric acid (H_{2}SO_{4} =
98) per litre be made, it will be found to have twice the strength of
the above solution, that is, 100 c.c. of the soda would only require 50
c.c. of the acid to neutralise it. The reason for this will be seen on
inspecting the equations:--

  NaHO + HCl = NaCl + H_{2}O.
  2NaHO + H_{2}SO_{4} = Na_{2}SO_{4} + 2H_{2}O.

Acids like sulphuric acid are termed bibasic, and their equivalent is
only half the molecular weight. Thus, a normal solution of sulphuric
acid would contain 49 grams (98/2) of real acid per litre. Similarly,
lime and most of the bases are bibasic, as may be seen from the
following equations; hence their equivalent will be half the molecular
weight.

  2HCl + CaO = CaCl_{2} + H_{2}O.
  2HCl + MgO = MgCl_{2} + H_{2}O.

_The standard normal solution of hydrochloric acid_ is made by diluting
100 c.c. of the strong acid to one litre with water. This will be
approximately normal. In order to determine its exact strength, weigh up
3 grams of recently ignited pure sodium carbonate or of the ignited
bicarbonate. Transfer to a flask and dissolve in 200 c.c. of water; when
dissolved, cool, tint faintly yellow with a few drops of a solution of
methyl orange, and run in the standard "acid " from a burette till the
yellow changes to a pink. Read off the number of c.c. used, and
calculate to how much sodium carbonate 100 c.c. of the "acid" are
equivalent. If the "acid" is strictly normal, this will be 5.3 grams. It
will probably be equivalent to more than this. Now calculate how much
strictly normal "acid" would be equivalent to the standard found. For
example: suppose the standard found is 5.5 gram of sodium carbonate,
then--

        5.3 : 5.5 :: 100 : _x_
         (where _x_ is the quantity of normal "acid" required).
                           _x_ = 103.8 c.c.

To get the "acid" of normal strength, we should then add 3.8 c.c. of
water to each 100 c.c. of the standard solution remaining. Suppose there
were left 930 c.c. of the approximate "acid," 35.3 c.c. of water must be
added and mixed. It should then be checked by another titration with
pure sodium carbonate.

_The standard solution of semi-normal "alkali."_ The best alkali for
general purposes is ammonia, but, since it is volatile (especially in
strong solutions), it is best to make it of half the usual strength, or
_semi-normal_. One litre of this will contain 8.5 grams of ammonia
(NH_{3}), and 100 c.c. of it will just neutralise 50 c.c. of the normal
"acid." Take 100 c.c. of dilute ammonia and dilute with water to one
litre. Run into a flask 50 c.c. of the standard "acid," tint with methyl
orange, and run in from a burette the solution of ammonia till
neutralised. Less than 100 c.c. will probably be used. Suppose 95 c.c.
were required, there should have been 100, hence there is a deficiency
of five. Then, for each 95 c.c. of standard "ammonia" left, add 5 c.c.
of water, and mix well. 100 c.c. will now be equivalent to 50 c.c. of
the "acid."

As an example of the application of this method, we may take the
determination of lime in limestone, marble, and similar substances.

~Determination of Lime in Limestone.~--Weigh up 1 gram of the dried
sample, and dissolve in 25 c.c. of normal acid, cool, dilute to 100
c.c., and titrate with the semi-normal solution of alkali (using
methyl-orange as an indicator). Divide the c.c. of alkali used by 2,
subtract from 25, and multiply by 0.028 to find the weight of lime. This
method is not applicable in the presence of other carbonates or oxides,
unless the weight of these substances be afterwards determined and due
correction be made.


STRONTIA.

Strontia, the oxide of strontium (SrO), occurs in nature as sulphate, in
the mineral celestine (SrSO_{4}), and as carbonate in strontianite
(SrCO_{3}). It is found in small quantities in limestones, chalk, &c.

Strontia is used in sugar-refining, and for the preparation of coloured
lights.

~Detection.~--It is detected by the crimson colour which its compounds
(when moistened with hydrochloric acid) impart to the flame. The
spectrum shows a large number of lines, of which a red, an orange, and a
blue are most characteristic.

It resembles lime in many of its compounds, but is distinguished by the
insolubility of its sulphate in a boiling solution of ammonium sulphate,
and by the insolubility of its nitrate in alcohol. From baryta, which it
also resembles, it is distinguished by not yielding an insoluble
chromate in an acetic acid solution, by the solubility of its chloride
in alcohol, and by the fact that its sulphate is converted into
carbonate on boiling with a solution formed of 3 parts of potassium
carbonate and 1 of potassium sulphate.

It is got into solution in the same manner as lime. The sulphate should
be fused with "fusion mixture," extracted with water, and thoroughly
washed. The residue will contain the strontia as carbonate, which is
readily soluble in dilute hydrochloric or nitric acid.

~Separation.~--It is separated (after removal of the silica and metals,
as described under _Lime_) by adding ammonia and ammonia carbonate, and
allowing to stand for some hours in a warm place. In the absence of
baryta or lime it is filtered off, and weighed as strontium carbonate,
which contains 70.17 per cent. of strontia. It is separated from baryta
by dissolving in a little hydrochloric acid, adding ammonia in excess,
and then acidifying with acetic acid, and precipitating the baryta with
potassium bichromate, as described under _Baryta_. The strontia is
precipitated from the filtrate by boiling for some time with a strong
solution of ammonic sulphate and a little ammonia. Fifty parts of
ammonic sulphate are required for each part of strontia or lime present.
The precipitate is filtered off, and washed first with a solution of
ammonic sulphate, and then with alcohol. It is dried, ignited and
weighed as strontium sulphate.


GRAVIMETRIC DETERMINATION.

The determination of strontia in pure solutions is best made by adding
sulphuric acid in excess and alcohol in volume equal to that of the
solution. Allow to stand overnight, filter, wash with dilute alcohol,
dry, ignite at a red heat, and weigh as sulphate (SrSO_{4}). This
contains 56.4 per cent. of strontia (SrO); or 47.7 per cent. of
strontium.


BARYTA.

Baryta, oxide of barium (BaO), commonly occurs in combination with
sulphuric oxide in the mineral barytes or heavy spar (BaSO_{4}), and in
combination with carbon dioxide in witherite (BaCO_{3}). These minerals
are not unfrequently found in large quantity (associated with galena and
other metallic sulphides) in lodes. Small isolated crystals of these are
frequently found in mining districts. Barium is a constituent of certain
mineral waters. The minerals are recognised by their high specific
gravity and their crystalline form.

Compounds of barium are often used by the assayer, more especially the
chloride and hydrate. The salts are, with the exception of the sulphate,
generally soluble in water or hydrochloric acid. In such solutions
sulphuric acid produces a white precipitate of baric sulphate, which is
practically insoluble in all acids.

The dioxide (BaO_{2}) is used for the preparation of oxygen. On strong
ignition it gives up oxygen, and is converted into baryta (BaO), which,
at a lower temperature, takes up oxygen from the air, re-forming the
dioxide.

~Detection.~--Barium is detected by the green colour its salts,
especially the chloride, give to the flame. This, viewed through the
spectroscope, shows a complicated spectrum, of which two lines in the
green are most easily recognised and characteristic. The salts of barium
give no precipitate with sulphuretted hydrogen in either acid or
alkaline solution, but with sulphuric acid they at once give a
precipitate, which is insoluble in acetate of soda. In solutions
rendered faintly acid with acetic acid, they give a yellow precipitate
with bichromate of potash. These reactions are characteristic of barium.

Baryta is got into solution in the manner described under _Lime_; but in
the case of the sulphate the substance is fused with three or four times
its weight of "fusion mixture." The "melt" is extracted with water,
washed, and the residue dissolved in dilute hydrochloric acid.

~Separation.~--The separation is thus effected:--The solution in
hydrochloric acid is evaporated to dryness, re-dissolved in hot dilute
hydrochloric acid, and sulphuric acid is added to the solution till no
further precipitate is formed. The precipitate is filtered off, and
digested with a solution of ammonium acetate or of sodium hyposulphite
at 50° or 60° C. to dissolve out any lead sulphate. The residue is
filtered off, washed, dried, and ignited. The ignited substance is mixed
with four or five times its weight of "fusion mixture," and fused in a
platinum-dish over the blowpipe for a few minutes. When cold, it is
extracted with cold water, filtered, and washed. The residue is
dissolved in dilute hydrochloric acid, and (if necessary) filtered. The
solution contains the barium as baric chloride mixed, perhaps, with
salts of strontium or lime. To separate these, ammonia is added till the
solution is alkaline, and then acetic acid in slight excess. Chromate of
baryta is then thrown down, by the addition of bichromate of potash, as
a yellow precipitate. It is allowed to settle, filtered and washed with
a solution of acetate or of nitrate of ammonia. It is dried, ignited
gently, and weighed. It is BaCrO_{4}, and contains 60.47 per cent. of
baryta.


GRAVIMETRIC DETERMINATION.

The gravimetric determination of baryta, when lime and strontia are
absent, is as follows:--The solution, if it contains much free acid, is
nearly neutralised with ammonia, and then diluted to 100 or 200 c.c. It
is heated to boiling, and dilute sulphuric acid is added till no further
precipitation takes place. The precipitate is allowed to settle for a
few minutes, decanted through a filter, and washed with hot water; and,
afterwards, dried, transferred to a porcelain crucible, and strongly
ignited in the muffle or over the blowpipe for a few minutes. It is then
cooled, and weighed as sulphate of baryta (BaSO_{4}). It contains 65.67
per cent. of baryta (BaO).

In determining the baryta in minerals which are soluble in acid, it is
precipitated direct from the hydrochloric acid solution (nearly
neutralised with ammonia) by means of sulphuric acid. The precipitated
baric sulphate is digested with a solution of ammonic acetate; and
filtered, washed, ignited, and weighed.


VOLUMETRIC DETERMINATION.

The principle and mode of working of this is the same as that given
under the Sulphur Assay; but using a standard solution of sulphuric acid
instead of one of barium chloride. The standard solution of sulphuric
acid is made to contain 32.02 grams of sulphuric acid (H_{2}SO_{4}), or
an equivalent of a soluble alkaline sulphate, per litre. 100 c.c. will
be equal to 5 grams of baryta.

Five grams of the substance are taken, and the baryta they contain
converted into carbonate (if necessary). The carbonate is dissolved in
dilute hydrochloric acid. Ten grams of sodium acetate are added, and the
solution, diluted to 500 c.c., is boiled, and titrated in the manner
described.

Lead salts must be absent in the titration, and so must strontia and
lime. Ferrous salts should be peroxidised by means of permanganate or
chlorate of potash. Other salts do not interfere.


MAGNESIA.

Magnesia, the oxide of magnesium (MgO) occurs in nature in the rare
mineral periclase (MgO); and hydrated, as brucite (MgH_{2}O_{2}). As
carbonate it occurs in large quantity as magnesite (MgCO_{3}), which is
the chief source of magnesia. Mixed with carbonate of lime, it forms
magnesian limestone and dolomite. It is present in larger or smaller
quantity in most silicates; and the minerals, serpentine, talc, steatite
and meerschaum are essentially hydrated silicates of magnesia. Soluble
magnesian salts occur in many natural waters; more especially the
sulphate and the chloride. Kieserite (MgSO_{4}.H_{2}O) occurs in
quantity at Stassfurt, and is used in the manufacture of Epsom salts.

~Detection.~--Magnesia is best detected in the wet way. Its compounds
give no colour to the flame, and the only characteristic dry reaction is
its yielding a pink mass when ignited before the blowpipe (after
treatment with a solution of cobalt nitrate). In solution, it is
recognised by giving no precipitate with ammonia or ammonic carbonate in
the presence of ammonic chloride, and by giving a white crystalline
precipitate on adding sodium phosphate or arsenate to the ammoniacal
solution.

Magnesia differs from the other alkaline earths by the solubility of its
sulphate in water.

Magnesia is dissolved by boiling with moderately strong acids; the
insoluble compounds are fused with "fusion mixture," and treated as
described under _Silicates_.

~Separation.~--It is separated by evaporating the acid solution to
dryness to render silica insoluble, and by taking up with dilute
hydrochloric acid. The solution is freed from the second group of metals
by means of sulphuretted hydrogen, and the iron, alumina, &c., are
removed with ammonic chloride, ammonia, and ammonic sulphide. The
somewhat diluted filtrate is treated, first, with ammonia, and then with
carbonate of ammonia in slight excess. It is allowed to stand for an
hour in a warm place, and then filtered. The magnesia is precipitated
from the filtrate by the addition of an excess of sodium phosphate and
ammonia. It is allowed to stand overnight, filtered, and washed with
dilute ammonia. The precipitate contains the magnesia as
ammonic-magnesic phosphate.

In cases where it is not desirable to introduce sodium salts or
phosphoric acid into the assay solution, the following method is
adopted. The solution (freed from the other alkaline earths by ammonium
carbonate) is evaporated in a small porcelain dish with nitric acid. The
residue (after removing the ammonic salts by ignition) is taken up with
a little water and a few crystals of oxalic acid, transferred to a
platinum dish, evaporated to dryness, and ignited. The residue is
extracted with small quantities of boiling water and filtered off; while
the insoluble magnesia is washed. The filtrate contains the alkalies.
The residue is ignited, and weighed as magnesia. It is MgO.


GRAVIMETRIC DETERMINATION.

The solution containing the magnesia is mixed with chloride of ammonium
and ammonia in excess. If a precipitate should form, more ammonic
chloride is required. Add sodium phosphate solution in excess, stir and
allow to stand overnight. Filter and wash the precipitate with dilute
ammonia. Dry, transfer to a platinum or porcelain crucible, and ignite
(finally at intense redness); cool, and weigh. The substance is magnesic
pyrophosphate (Mg_{2}P_{2}O_{7}), and contains 36.04 per cent. of
magnesia.


VOLUMETRIC METHOD.

The magnesia having been precipitated as ammonic-magnesic phosphate,
which is the usual separation, its weight can be determined
volumetrically by the method of titration described under _Phosphates_.

The same standard solution of uranium acetate is used. Its standard for
magnesia is got by multiplying the standard for phosphoric oxide by
0.5493. For example, if one hundred c.c. are equivalent to 0.5 gram of
phosphoric oxide, they will be equivalent to (0.5 × .5493) 0.2746 gram
of magnesia. The method of working and the conditions of the titration
are the same as for the phosphate titration. The quantity of substance
taken for assay must not contain more than 0.1 or 0.2 gram of magnesia.
After precipitating as ammonic-magnesic phosphate with sodium phosphate,
and well washing with ammonia, it is dissolved in dilute hydrochloric
acid, neutralised with ammonia, and sodic acetate and acetic acid are
added in the usual quantity. The solution is boiled and titrated.


EXAMINATION OF A LIMESTONE.

~Silica and Insoluble Silicates.~--Take one gram of the dried sample and
dissolve it in 10 c.c. of dilute hydrochloric acid; filter; wash, dry,
and ignite the residue.

~Organic Matter.~--If the residue insoluble in hydrochloric acid shows
the presence of organic matter, it must be collected on a weighed filter
and dried at 100°. On weighing, it gives the combined weights of organic
and insoluble matter. The latter is determined by igniting and weighing
again. The organic matter is calculated by difference.

~Lime.~--Where but little magnesia is present, this is determined by
titration with standard acid. Take one gram, and dissolve it in 25 c.c.
of normal hydrochloric acid. Tint with methyl-orange and titrate with
semi-normal ammonia. Divide the quantity of ammonia used by 2, deduct
this from 25, and multiply the remainder by 2.8. This gives the
percentage of lime. Where magnesia is present, the same method is
adopted, and the magnesia (which is separately determined) is afterwards
deducted. The percentage of magnesia found is multiplied by 1.4, and the
result is deducted from the apparent percentage of lime got by
titrating.

~Magnesia.~--Dissolve 2 grams of the limestone in hydrochloric acid, and
separate the lime with ammonia and ammonium oxalate. The filtrate is
treated with sodium phosphate, and the magnesia is weighed as
pyrophosphate, or titrated with uranium acetate.

~Iron.~--Dissolve 2 grams in hydrochloric acid, reduce, and titrate with
standard permanganate of potassium solution. This gives the total iron.
The ferrous iron is determined by dissolving another 2 grams in
hydrochloric acid and at once titrating with the permanganate of
potassium solution.

~Manganese.~--Dissolve 20 grams in hydrochloric acid, nearly neutralise
with soda, add sodium acetate, boil, and filter. To the filtrate add
bromine; boil, and determine the manganese in the precipitate. See page
300.

~Phosphoric Oxide.~--This is determined by dissolving the ferric acetate
precipitate from the manganese separation in hydrochloric acid, adding
ammonia in excess, and passing sulphuretted hydrogen. Filter and add to
the filtrate "magnesia mixture." The precipitate is collected, washed
with ammonia, ignited, and weighed as pyrophosphate.


THE ALKALIES.

The oxides of sodium, potassium, lithium, cæsium, and rubidium and
ammonia are grouped under this head. Of these cæsia and rubidia are
rare, and lithia comparatively so. They are easily distinguished by
their spectra. They are characterised by the solubility of almost all
their salts in water, and, consequently, are found in the solutions from
which the earths and oxides of the metals have been separated by the
usual group re-agents.

The solution from which the other substances have been separated is
evaporated to dryness, and the product ignited to remove the ammonic
salts added for the purpose of separation. The residue contains the
alkali metals generally, as chlorides or sulphates. Before determining
the quantities of the particular alkali metals present, it is best to
convert them altogether, either into chloride or sulphate, and to take
the weight of the mixed salts. It is generally more convenient to weigh
them as chlorides. They are converted into this form, if none of the
stronger acids are present, by simply evaporating with an excess of
hydrochloric acid. Nitrates are converted into chlorides by this
treatment. When sulphates or phosphates are present, the substance is
dissolved in a little water, and the sulphuric or phosphoric acid
precipitated with a slight excess of acetate of lead in the presence of
alcohol. The solution is filtered, and the excess of lead precipitated
with sulphuretted hydrogen. The filtrate from this is evaporated to
dryness with an excess of hydrochloric acid, and the residue, consisting
of the mixed chlorides, is gently ignited and weighed. In many cases
(such as the analysis of slags and of some natural silicates where the
percentage of alkalies is small) the percentage of soda and potash
(which most commonly occur) need not be separately determined. It is
sufficient to report the proportion of mixed alkalies; which is thus
ascertained:--Dissolve the ignited and weighed chlorides in 100 c.c. of
distilled water, and titrate with the standard solution of silver
nitrate (using potassic chromate as indicator) in the manner described
under _Chlorine_. The c.c. of silver nitrate used gives the weight in
milligrams of the chlorine present. Multiply this by 0.775, and deduct
the product from the weight of the mixed chlorides. This will give the
combined weight of the alkalies (Na_{2}O and K_{2}O) present. For
example, 0.0266 gram of mixed chlorides required on titrating 14.2 c.c.
of silver nitrate, which is equivalent to 0.0142 gram of chlorine. This
multiplied by 0.775 gives 0.0110 to be deducted from the weight of the
mixed chlorides.

  Mixed chlorides  0.0266 gram
  Deduction        0.0110   "
                   ------
  Mixed alkalies   0.0156   "

Assuming this to have been got from 1 gram of a rock, it would amount to
1.56 per cent. of "potash and soda."

The relative proportions of the potash and soda can be ascertained from
the same determination. Sodium and potassium chlorides have the
following composition:--

  Sodium    39.38        Potassium  52.46
  Chlorine  60.62        Chlorine   47.54
            -----                   -----
           100.00                  100.00

The percentage of chlorine in the mixed chlorides is calculated. It
necessarily falls somewhere between 47.5 and 60.6 per cent., and
approaches the one or the other of these numbers as the proportion of
the sodium or potassium preponderates. Each per cent. of chlorine in
excess of 47.5 represents 7.63 per cent. of sodium chloride in the mixed
chlorides. The percentage of potash and soda in the substance can be
calculated in the usual way. Sodium chloride multiplied by 0.5302 gives
its equivalent of soda (Na_{2}O), and potassium chloride multiplied by
0.6317 gives its equivalent of potash (K_{2}O).

The weight of sodium chloride in the mixed chlorides is also calculated
thus:--Take the same example for illustration. Multiply the chlorine
found by 2.103. This gives--

  (0.0142×2.103) = 0.02987.

From the product deduct the weight of the mixed chlorides found--

  Product               0.02987
  Mixed chlorides       0.02660
                        -------
  Difference            0.00327

The difference multiplied by 3.6288 gives the weight of sodium chloride
in the mixture. In this case it equals 0.0118 gram. The potassium
chloride is indicated by the difference between this and the weight of
the mixed chlorides. It equals 0.0148 gram. We have now got--

  Sodium chloride          0.0118 gram
  Potassium chloride       0.0148   "

from 1 gram of the rock taken. Multiplying these by their factors we
have (Soda = 0.0118×0.5302; Potash 0.0148×0.6317)--

  Soda    = 0.625 per cent.
  Potash  = 0.935     "

~Concentration of the Alkalies.~--With the exception of magnesia, all
the other bases are separated from the alkalies in the ordinary course
of work without the addition of any re-agent which cannot be removed by
simple evaporation and ignition. Consequently, with substances soluble
in acids, successive treatment of the solution with sulphuretted
hydrogen, ammonia, ammonic sulphide, and ammonic carbonate, filtering,
where necessary, will yield a filtrate containing the whole of the
alkalies with ammonic salts and, perhaps, magnesia.

The filtrate is evaporated in a small porcelain dish, with the addition
of nitric acid towards the finish. It is carried to dryness and ignited.
The residue is taken up with a little water, treated with a few crystals
of oxalic acid, and again evaporated and ignited. The alkaline salts are
extracted with water, and filtered from the magnesia into a weighed
platinum dish. The solution is then evaporated with an excess of
hydrochloric acid, ignited at a low red heat, and weighed. The residue
consists of the mixed alkaline chlorides.

For substances (such as most silicates and similar bodies) not
completely decomposed by acids, Lawrence Smith's method is generally
used. This is as follows:--Take from 0.5 to 1 gram of the finely
powdered mineral, and mix, by rubbing in the mortar, with an equal
weight of ammonium chloride. Then mix with eight times as much pure
calcium carbonate, using a part of it to rinse out the mortar. Transfer
to a platinum crucible, and heat gently over a Bunsen burner until the
ammonic chloride is decomposed (five or ten minutes). Raise the heat to
redness, and continue at this temperature for about three quarters of an
hour. The crucible must be kept covered. Cool, and turn out the mass
into a 4-inch evaporating dish; wash the crucible and cover with
distilled water, and add the washings to the dish; dilute to 60 or 80
c.c., and heat to boiling. Filter and wash. Add to the filtrate about
1.5 gram of ammonium carbonate; evaporate to about 40 c.c., and add a
little more ammonic carbonate and some ammonia. Filter into a weighed
platinum dish, and evaporate to dryness. Heat gently, to drive off the
ammonic chloride, and ignite to a little below redness. Cool and weigh.
The residue consists of the mixed alkaline chlorides.

~Separation of the Alkali-Metals from each other.~--Sodium and lithium
are separated from the other alkali-metals by taking advantage of the
solubility of their chlorides in the presence of platinic chloride; and
from one another by the formation of an almost insoluble lithic
phosphate on boiling with a solution of sodium phosphate in a slightly
alkaline solution. Cæsium, rubidium, and potassium yield precipitates
with platinic chloride, which are somewhat soluble, and must be
precipitated from concentrated solutions. Cæsium and rubidium are
separated from potassium by fractional precipitation with platinum
chloride. Their platino-chlorides, being less soluble than that of
potassium, are precipitated first. One hundred parts of boiling water
dissolve 5.18 of the potassium platino-chloride, 0.634 of the rubidium
salt, and 0.377 of the corresponding cæsium compound. The separation of
lithium, cæsium, and rubidium is seldom called for, owing to their
rarity. The details of the separation of potassium from sodium are
described under _Potassium_. Ammonia compounds are sharply marked off
from the rest by their volatility, and it is always assumed that they
have been removed by ignition; if left in the solution, they would count
as potassium compounds. They will be considered under _Ammonia_.


SODIUM.

Sodium is the commonest of the alkali metals. It is found in nature
chiefly combined with chlorine as "common salt" (NaCl). This mineral is
the source from which the various compounds of sodium in use are
prepared. Sodium occurs abundantly as nitrate (NaNO_{3}) in Chili
saltpetre, and as silicate in various minerals, such as albite (or
soda-felspar).

It occurs as fluoride in cryolite (Na_{3}AlF_{6}), and as carbonate in
natron, &c. Sulphates are also found. Sodium is very widely diffused,
few substances being free from it.

The detection of sodium is easy and certain, owing to the strong yellow
colour its salts impart to the flame; this, when viewed by the
spectroscope, shows a single yellow line.[93] The extreme delicacy of
this test limits its value, because of the wide diffusion of sodium
salts. It is more satisfactory to separate the chloride, which may be
recognised by its taste, flame coloration, fusibility, and negative
action with reagents. The chloride dissolved in a few drops of water
gives with potassium metantimoniate, a white precipitate of the
corresponding sodium salt.

Sodium salts are dissolved out from most compounds on treatment with
water or dilute acids. Insoluble silicates are decomposed and the alkali
rendered soluble by Lawrence Smith's method, which has just been
described. The separation of the sodium from the mixed chlorides is
effected in the following way:--The chlorides are dissolved in a little
water and the potassium separated as platino-chloride. The soluble
sodium platino-chloride, with the excess of platinum, is boiled, mixed
with sulphuric acid, evaporated to dryness, and ignited. On extracting
with water, filtering, evaporating, and igniting, sodium sulphate is
left, and is weighed as such.

It is more usual, and quite as satisfactory, to calculate the weight of
the sodium chloride by difference from that of the mixed chlorides, by
subtracting that of the potassium chloride, which is separately
determined. For example, 1 gram of a rock gave--Mixed chlorides, 0.0266
gram, and 0.0486 gram of potassic platino-chloride. This last is
equivalent to 0.0149 gram of potassium chloride.

  Mixed chlorides found          0.0266
  Deduct potassium chloride      0.0149
                                 ------
  Leaves sodium chloride         0.0117

The weight of sodium chloride found, multiplied by 0.5302, gives the
weight of the soda (Na_{2}O).


GRAVIMETRIC DETERMINATION.

The solution, which must contain no other metal than sodium, is
evaporated in a weighed platinum crucible or dish. Towards the finish an
excess, not too great, of sulphuric acid is added, and the evaporation
is continued under a loosely fitting cover. The residue is ignited over
the blowpipe, a fragment of ammonic carbonate being added towards the
end, when fumes of sulphuric acid cease to be evolved. This ensures the
removal of the excess of acid. The crucible is cooled in the desiccator,
and weighed. The substance is sulphate of soda (Na_{2}SO_{4}), and
contains 43.66 per cent. of soda (Na_{2}O), or 32.38 per cent. of sodium
(Na).


VOLUMETRIC METHODS.

There are various methods used for the different compounds of sodium.
There is no one method of general application. Thus with "common salt"
the chlorine is determined volumetrically; and the sodium, after
deducting for the other impurities, is estimated by difference.

With sodic carbonate and caustic soda, a given weight of the sample is
titrated with standard acid, and the equivalent of soda estimated from
the alkalinity of the solution.

With sodium sulphate, a modification of the same method is used. To a
solution of 3.55 grams of the salt contained in a half-litre flask, 250
c.c. of a solution of baryta water is added. The volume is made up to
500 c.c. with water. The solution is mixed and filtered. Half of the
filtrate is measured off, treated with a current of carbonic acid, and
then boiled. It is transferred to a half-litre flask, diluted to the
mark, shaken up, and filtered. 250 c.c. of the filtrate, representing a
quarter of the sample taken, is then titrated with standard acid. The
standard acid is made by diluting 250 c.c. of the normal acid to 1
litre. The c.c. of acid used multiplied by 2 gives the percentage. A
correction must be made to counteract the effect of impurities in the
baryta as well as errors inherent in the process. This is small, and its
amount is determined by an experiment with 3.55 grams of pure sodium
sulphate.


EXAMINATION OF COMMON SALT.

~Moisture.~--Powder and weigh up 10 grams of the sample into a platinum
dish. Dry in a water oven for an hour, and afterwards heat to bare
redness over a Bunsen burner. Cool, and weigh. The loss gives the water.

~Chlorine.~--Weigh up two separate lots of 1 gram each; dissolve in 100
c.c. of water, and determine the chlorine by titrating with the standard
silver nitrate solution, using chromate of potash as indicator. See
_Chlorine_.

~Insoluble Matter.~--Dissolve 10 grams of the salt in water with the
help of a little hydrochloric acid. Filter off the sediment, wash,
ignite, and weigh. This residue is chiefly sand. Dilute the nitrate to
500 c.c.

~Lime.~--Take 250 c.c. of the filtrate, render ammoniacal and add
ammonium oxalate; wash, dry, and ignite the precipitate. Weigh as lime
(CaO).

~Magnesia.~--To the filtrate from the lime add phosphate of soda. Allow
to stand overnight, filter, wash with dilute ammonia, dry, ignite, and
weigh as pyrophosphate.

~Sulphuric Oxide.~--To the remaining 250 c.c. of the filtrate from the
"insoluble," add an excess of barium chloride. Collect, wash, dry,
ignite, and weigh the barium sulphate.

~Sodium.~--It is estimated by difference.

The following may be taken as an example:--

  Moisture             0.35
  Insoluble matter     0.40
  Lime                 0.40
  Magnesia             0.05
  Sulphuric oxide      0.60
  Chlorine            59.60
  Sodium              38.60
                     ------
                     100.00


POTASSIUM.

Potassium occurs in nature as chloride, in the mineral sylvine (KCl),
and more abundantly combined with magnesium chloride, in earnallite
(KCl.MgCl_{2}.6H_{2}O). It occurs as nitrate in nitre (KNO_{3}), and as
silicate in many minerals, such as orthoclase (or potash-felspar) and
muscovite (or potash-mica).

Potassium compounds are detected by the characteristic violet colour
they impart to the flame. The presence of sodium salts masks this tint,
but the interference can be neutralised by viewing the flame through a
piece of blue glass. Viewed through the spectroscope, it shows a
characteristic line in the red and another in the violet. These,
however, are not so easy to recognise or obtain as the sodium one.
Concentrated solutions of potassium salts give a yellow crystalline
precipitate with platinum chloride, and a white crystalline one with the
acid tartrate of soda. For these tests the solution is best neutral.
These tests are only applicable in the absence of compounds other than
those of potassium and sodium.


GRAVIMETRIC DETERMINATION.

This process serves for its separation from sodium. Take 1 gram of the
sample and dissolve it in an evaporating dish with 50 c.c. of water.
Acidify with hydrochloric acid in quantity sufficient (if the metals are
present as chlorides) to make it acid, or, if other acids are present,
in at least such quantity as will provide the equivalent of chlorine.
Add 3 grams of platinum, in solution as platinum chloride, and evaporate
on a water-bath to a stiff paste, but not to dryness. Moisten with a few
drops of platinic chloride solution without breaking up the paste by
stirring. Cover with 20 c.c. of strong alcohol, and wash the crystals as
much as possible by rotating the dish. Allow to settle for a few
moments, and decant through a filter. Wash in the same way two or three
times until the colour of the filtrate shows that the excess of the
platinum chloride used is removed. Wash the precipitate on to the filter
with a jet of alcohol from the wash-bottle; clean the filter-paper,
using as little alcohol as possible. Dry in the water-oven for an hour.
Brush the precipitate into a weighed dish, and weigh it. It is potassium
platino-chloride (K_{2}PtCl_{6}), and contains 16.03 per cent. of
potassium, or 30.56 per cent. of potassium chloride (KCl), which is
equivalent to 19.3 per cent. of potash (K_{2}O).

If the filter-paper is not free from precipitate, burn it and weigh
separately. The excess of weight over that of the ash will be due to
platinum and potassic chloride (Pt and 2KCl). This multiplied by 1.413
will give the weight of the potassic platino-chloride from which it was
formed. It must be added to the weight of the main precipitate.

The mixed alkaline chlorides obtained in the usual course of analysis
are treated in this manner; the quantity of platinum added must be about
three times as much as the mixed chlorides weigh.


VOLUMETRIC METHODS.

These are the same as with soda.

~Examination of Commercial Carbonate of Potash.~--The impurities to be
determined are moisture, silica, and insoluble matter, chlorine,
sulphuric oxide, and oxide of iron. These determinations are made in the
ways described under the examination of common salt.

The ~potassium~ is determined by converting it into chloride and
precipitating with platinum chloride, &c., as just described.

~Available Alkali.~--Weigh up 23.5 grams of the sample, dissolve in
water, and make up to 500 c.c. Take 50 c.c., tint with methyl orange,
and titrate with the normal solution of acid. The c.c. of acid used
multiplied by 2 gives the percentage of available alkali calculated as
potash (K_{2}O).

~Soda.~--This is calculated indirectly in the following way:--Deduct
from the potassium found the quantity required for combination with the
chlorine and sulphuric oxide present, and calculate the remainder to
potash (K_{2}O). The apparent surplus excess of available alkali is the
measure of the soda present.

~Carbon Dioxide.~--The c.c. of acid used in the available alkali
determination, multiplied by 2.2 and divided by 2.35, gives the
percentage of carbon dioxide.


LITHIUM.

Lithia, the oxide of lithium (Li_{2}O), occurs in quantities of 3 or 4
per cent. in various silicates, such as lepidolite (or lithia-mica),
spodumene, and petalite. It also occurs as phosphate in triphyline. It
is a constituent of the water of certain mineral springs. A spring at
Wheal Clifford contained as much as 0.372 gram of lithium chloride per
litre. In small quantities, lithia is very widely diffused.

The ~Detection~ of lithia is rendered easy by the spectroscope; its
spectrum shows a red line lying about midway between the yellow sodium
line and the red one of potassium. It also shows a faint yellow line.
The colour of the flame (a crimson) is characteristic.

The reactions of the lithium compounds lie between those of the alkalies
and of the alkaline earths. Solutions are not precipitated by tartaric
acid nor by platinic chloride. The oxide is slowly soluble in water.
The carbonate is not freely soluble. Lithia is completely precipitated
by sodic phosphate, especially in hot alkaline solutions.

In its determination the mixed alkaline chlorides obtained in the
separation of the alkalies are dissolved in water, a solution of soda is
added in slight excess, and the lithia precipitated with _sodic_
phosphate. Before filtering, it is evaporated to dryness and extracted
with hot water rendered slightly ammoniacal. The residue is transferred
to a filter, dried, ignited, and weighed. The precipitate is lithium
phosphate (3Li_{2}O, P_{2}O_{5}), and contains 38.8 per cent. of lithia.
The separation of lithia from magnesia is not given by the usual
authorities. Wohler recommends evaporating the solution to dryness with
carbonate of soda. On extracting the residue with water, the lithia
dissolves out and is determined in the filtrate. One hundred parts of
water dissolve, at the ordinary temperature, 0.769 parts of lithium
carbonate (Li_{2}CO_{3}); the basic magnesia compound is almost
insoluble in the absence of carbon dioxide and ammonium salts.


CAESIUM.

The oxide of caesium, caesia (Cs_{2}O), is found associated with lithia
in lepidolite, &c., and, together with rubidium, in many mineral waters.
The mineral pollux is essentially a silicate of alumina and caesia; it
contains 34.0 per cent. of the latter oxide.

Caesium is best detected by the spectroscope, its spectrum being
characterised by two lines in the blue and one in the red; the latter is
about midway between the lithium and sodium lines.

If not detected by the spectroscope, or specially looked for, caesia
would, in the ordinary course of work, be separated with the potash and
weighed as potassium platino-chloride.

Caesia is separated from all the other alkalies by adding to the acid
solution of the mixed chlorides a strongly acid cold solution of
antimonious chloride. The acid used must be hydrochloric. The caesium is
precipitated as a white crystalline precipitate (CsCl.SbCl_{3}), which
is filtered off, and washed, when cold, with strong hydrochloric acid;
since it is decomposed by water or on warming. The precipitate is washed
into a beaker, and treated with sulphuretted hydrogen; after filtering
off the sulphide of antimony, the solution leaves, on evaporation, the
caesium as chloride.


RUBIDIUM.

Rubidium occurs widely diffused in nature, but in very small quantities.
It is generally associated with caesium.

It is detected by the spectroscope, which shows two violet lines and two
dark red ones. Like caesium, it is precipitated with platinic chloride,
and in the ordinary course of work would be weighed as potassium. It is
separated from potassium by fractional precipitation with platinic
chloride. Rubidium platino-chloride is much less soluble than the
potassium salt.


AMMONIUM.

It is usual to look upon the salts of ammonia as containing a compound
radical (NH_{4} = Am), which resembles in many respects the metals of
the alkalies. Ammonium occurs in nature as chloride in sal ammoniac
(AmCl), as sulphate in mascagnine (Am_{2}SO_{4}), as phosphate in
struvite (AmMgPO_{4}.12H_{2}O). Minerals containing ammonium are rare,
and are chiefly found either in volcanic districts or associated with
guano. Ammonia and ammonium sulphide occur in the waters of certain
Tuscan lagoons, which are largely worked for the boracic acid they
contain. The crude boracic acid from this source contains from 5 to 10
per cent. of ammonium salts. It is from these that the purer forms of
ammonium compounds of commerce known as "from volcanic ammonia" are
derived. But the bulk of the ammonia of commerce is prepared from the
ammoniacal liquors obtained as bye-products in the working of certain
forms of blast furnaces and coke ovens, and more especially in
gas-making.

Ammonia hardly comes within the objects of assaying; but it is largely
used in the laboratory, and the assayer is not unfrequently called on to
determine it. Ammonium salts are mostly soluble in water. In strong
solutions they give a yellow precipitate of ammonium platino-chloride on
the addition of chloride of platinum; and with the acid tartrate of soda
yield a white precipitate of hydric ammonic tartrate. These reactions
are similar to those produced with potassium compounds.

Heated with a base, such as lime or sodic hydrate, ammonium salts are
decomposed, yielding ammonia gas (NH_{3}), which is readily soluble in
water. The solution of this substance is known as ammonic hydrate or
"ammonia."

They are volatilised on ignition; either with, or without, decomposition
according to the acid present. This fact is of importance in analytical
work; since it allows of the use of alkaline solutions and reagents
which leave nothing behind on heating. It must be remembered, however,
that, although ammonic chloride is volatile, it cannot be volatilised in
the presence of substances which form volatile chlorides without loss of
the latter. For example: ferric oxide and alumina are thus lost,
volatilising as chlorides; and there are some other compounds (notably
ammonic magnesic arsenate) which on heating to redness suffer reduction.
The presence of ammonic chloride in such cases must be avoided.

~Detection.~--Compounds of ammonium are detected by their evolving
ammonia when mixed or heated with any of the stronger bases. The ammonia
is recognised by its odour, by its alkaline reaction with litmus paper,
and by yielding white fumes, when brought in contact with fuming acid.
In consequence of the use of ammonium salts and ammonia as reagents, it
is necessary to make a special test for and determination of
ammonium.[94] In the ordinary course of work it will be "lost on
ignition." The determination presents little difficulty, and is based on
the method used for its detection.

[Illustration: FIG. 61.]

~Solution and Separation.~--Although ammonium salts are soluble in
water, there is no necessity for dissolving them. The compound
containing the ammonia is boiled with an alkaline solution; and the
liberated ammonia condensed and collected. The substance is weighed out
into a flask of about 200 c.c. capacity. The flask is closed with a
rubber cork perforated to carry a 20 c.c. pipette and a bulb exit tube.
The latter is connected with a receiver, which is a small flask
containing dilute hydrochloric acid (fig. 61). The flask containing the
substance is corked, and the greater part of the soda solution is run in
from the pipette. The solution is then boiled. The ammonia volatilises,
and is carried over into the hydrochloric acid, with which it combines
to form ammonic chloride. The distillation is carried on gently until
the bulk of the liquid is driven over. The ammonia in the receiver will
be mixed only with the excess of hydrochloric acid. This separation is
used in all determinations.


GRAVIMETRIC DETERMINATION.

The contents of the flask are transferred to a weighed platinum dish,
and evaporated on the water-bath. It is dried until the weight is
constant. The chloride of ammonium remains as a white mass which, after
cooling in a desiccator, is weighed. It contains 33.72 per cent. of
ammonium (NH_{4}), or 31.85 per cent. of ammonia (NH_{3}). On heating
over the Bunsen burner it is completely volatilised, leaving no residue.


VOLUMETRIC DETERMINATION.

Weigh up 1.7 gram of the substance and place it in the flask. Measure
off 50 c.c. of the normal solution of acid, place them in the receiver,
and dilute with an equal volume of water. Run in through the pipette (by
opening the clip) 20 c.c. of a strong solution of soda, boil until the
ammonia has passed over, and then aspirate a current of air through the
apparatus. Disconnect the receiver, and tint its contents with methyl
orange. Titrate the residual acid with a semi-normal solution of alkali.
Divide the c.c. of the "alkali" solution used by 2, and deduct from the
50 c.c. The difference will give the number of c.c. of the normal acid
solution neutralised by the ammonia distilled over. Each c.c. of "acid"
so neutralised, represents 1 per cent. of ammonia in the sample. If the
results are to be reported as ammonium, 1.8 gram of the sample is taken
instead of 1.7 gram.


COLORIMETRIC DETERMINATION.

This is effected by means of "Nessler's" reagent, which strikes a brown
colour with traces of ammonia, even with a few hundredths of a milligram
in 100 c.c. of liquid. With larger quantities of ammonia the reagent
gives a precipitate. This reagent is a strongly alkaline solution of
potassic mercuric iodide; and is thus made:--


_Nessler's solution_: Dissolve 17 grams of mercuric chloride in 300 c.c.
of water; and add the solution to one of 35 grams of potassium iodide in
100 c.c. of water until a permanent precipitate is produced. Both
solutions must be cold. Then make up to a litre by adding a 20 per cent.
solution of potash. Add more of the mercuric chloride (a little at a
time) until a permanent precipitate is again formed. Allow to settle,
decant, and use the clear liquor. Four or five c.c. are used for each
100 c.c. of liquid to be tested.

_A Standard Solution of Ammonia_ is made by dissolving 0.315 gram of
ammonic chloride in water, and diluting to 100 c.c. Ten c.c. of this are
taken and diluted to 1 litre. One c.c. contains 0.01 milligram of
ammonia (NH_{3}).

In working, the solution containing the ammonia is diluted to a definite
volume, and to such an extent that 50 c.c. of it shall not contain more
than 0.02 or 0.03 milligram of ammonia. Fifty c.c. of it are transferred
to a Nessler glass and mixed with 2 c.c. of Nessler's reagent. The
colour is noted, and an estimate made as to the amount of ammonia it
indicates. A measured quantity of the standard ammonia, judged to
contain about as much ammonia as that in the assay, is then put into
another Nessler glass. It is diluted to 50 c.c. with water, and mixed
with 2 c.c. of "Nessler." After standing a minute or two, the colours in
the two glasses are compared. If the tints are equal, the assay is
finished; but if the standard is weaker or stronger than the assay,
another standard, containing more or less ammonia, as the case may be,
must be prepared and compared with the assay. Two such experiments will
generally be sufficient; but, if not, a third must be made. The addition
of more standard ammonia to the solution to which the "Nessler" has
already been added does not give a satisfactory result.

When the ammonia in 50 c.c. has been determined, that in the whole
solution is ascertained by a suitable multiplication. By 10, for
example, if the bulk was 500 c.c., or by 20 if it was a litre.

Distilled water is used throughout. It must be free from ammonia; and is
best prepared by distilling an ammonia-free spring water.

FOOTNOTES:

[90] Al_{2}Cl_{6} + 3Na_{2}S_{2}O_{3} + 3H_{2}O = Al_{2}(HO)_{6} + 6NaCl
+ 3S + 3SO_{2}

[91] 3BeO,Al_{2}O_{3},6SiO_{2}

[92] CaC_{2}O_{4} = CaCO_{3}+CO.

[93] Resolved into two with a powerful spectroscope.

[94] Ammonium compounds are frequently produced when dissolving metals
in nitric acid; or when nitrates are heated in the presence of the
metals.



PART III--NON-METALS.



CHAPTER XV.

OXYGEN AND OXIDES.--THE HALOGENS.


OXYGEN.

Oxygen occurs in nature in the free state, forming 23 per cent. by
weight, or 21 per cent. by volume of the atmosphere; but, since it is a
gas, its presence is easily overlooked and its importance
underestimated. Except in the examination of furnace-gases, &c., the
assayer is not often called upon to determine its quantity, but it forms
one of his most useful reagents, and there are many cases where he
cannot afford to disregard its presence. It occurs not only in the air,
but also dissolved in water; ordinary waters containing on an average
0.00085 per cent. by weight, or 0.85 parts per 100,000.

Chemically, it is characterised by its power of combining, especially at
high temperatures, with the other elements, forming an important class
of compounds called oxides. This combination, when rapid, is accompanied
by the evolution of light and heat; hence oxygen is generally called the
supporter of combustion. This property is taken advantage of in the
operation of calcining, scorifying, cupelling, &c. The importance of a
free access of air in all such work is seen when it is remembered that 1
litre of air contains 0.2975 gram of oxygen, and this quantity will only
oxidise 0.1115 gram of carbon, 0.2975 gram of sulphur, or 3.849 grams of
lead.

Oxidation takes place at the ordinary temperature with many substances.
Examples of such action are seen in the weathering of pyrites, rusting
of iron, and (in the assay office) the weakening of solutions of many
reducing agents.

For methods of determining the percentage of oxygen in gases, for
technical purposes, the student is referred to Winkler & Lunge's
"Technical Gas Analysis."


OXIDES.

Oxides are abundant in nature, almost all the commonly occurring bodies
being oxidised. Water (H_{2}O) contains 88.8 per cent. of oxygen;
silica, lime, alumina, magnesia, and the other earths are oxides, and
the oxides of the heavier metals are in many cases important ores; as,
for example, cassiterite (SnO_{2}), hæmatite (Fe_{2}O_{3}), magnetite
(Fe_{3}O_{4}), and pyrolusite (MnO_{2}). In fact, the last-named mineral
owes its value to the excess of oxygen it contains, and may be regarded
as an ore of oxygen rather than of manganese.

Most of the metals, when heated to redness in contact with air, lose
their metallic lustre and become coated with, or (if the heating be
prolonged) altogether converted into, oxide. This oxide was formerly
termed a "calx," and has long been known to weigh more than the metal
from which it was obtained. For example, one part by weight of tin
becomes, on calcining, 1.271 parts of oxide (putty powder). The student
will do well to try the following experiments:--Take 20 grams of tin and
heat them in a muffle on a scorifier, scraping back the dross as it
forms, and continuing the operation until the whole of the metal is
burnt to a white powder and ceases to increase in weight.[95] Take care
to avoid loss, and, when cold, weigh the oxide formed. The oxide should
weigh 25.42 grams, which increase in weight is due to the oxygen
absorbed from the air and combined with the metal. It can be calculated
from this experiment (if there has been no loss) that oxide of tin
contains 21.33 per cent. of oxygen and 78.67 per cent. of tin. Oxidation
is performed with greater convenience by wet methods, using reagents,
such as nitric acid, which contain a large proportion of oxygen loosely
held. Such reagents are termed oxidising agents. Besides nitric acid,
permanganate of potash, bichromate of potash, and peroxide of hydrogen
are largely used for this purpose. One c.c. of nitric acid contains as
much oxygen as 2.56 litres of air, and the greater part of this is
available for oxidising purposes. Try the following experiment:--Take 2
grams of tin and cover in a weighed Berlin dish with 20 c.c. of dilute
nitric acid, heat till decomposed, evaporate to dryness, ignite, and
weigh. The 2 grams of tin should yield 2.542 grams of oxide. The
increase in weight will be proportionally the same as in the previous
experiment by calcination, and is due to oxygen, which in this case has
been derived from the nitric acid.

The percentage of oxygen in this oxide of tin (or in any of the oxides
of the heavier metals) may be directly determined by heating such oxides
in a current of hydrogen, and collecting and weighing the water formed.

It is found by experiment that 88.86 parts by weight of oxygen,
combining with 11.14 parts of hydrogen, form 100 parts of water; so that
from the weight of water formed it is easy to calculate the amount of
oxygen the oxide contained.

[Illustration: FIG. 62.]

Take 1 gram of the dried and powdered oxide and place it in a warm dry
combustion tube. Place the tube in a furnace, and connect at one end
with a hydrogen apparatus provided with a sulphuric acid bulb for drying
the gas, and at the other with a weighed sulphuric acid tube for
collecting the water formed. The apparatus required is shown in fig. 62.
Pass hydrogen through the apparatus, and, when the air has been cleared
out, light the furnace. Continue the heat and current of hydrogen for
half an hour (or longer, if necessary). Allow to cool. Draw a current of
dry air through the weighed tube. Weigh. The increase in weight gives
the amount of water formed, and this, multiplied by 0.8886, gives the
weight of the oxygen. The percentage of oxygen thus determined should be
compared with that got by the oxidation of the metal. It will be
practically the same. The following results can be taken as examples:--

Twenty grams of tin, calcined as described, gave 25.37 grams of oxide.

Two grams of tin, oxidised with nitric acid and ignited, gave 2.551
grams of oxide.

One gram of the oxide of tin, on reduction in a current of hydrogen,
gave 0.2360 gram of water (equivalent to 0.2098 gram of oxygen), and
left 0.7900 gram of metal.

Ten grams of ferrous sulphate gave, on strong ignition, 2.898 grams of
ferric oxide (Fe_{2}O_{3})[96] instead of 2.877.

The student should similarly determine the percentage of oxygen in
oxides of copper and iron. The former oxide may be prepared by
dissolving 5 grams of copper in 50 c.c. of dilute nitric acid,
evaporating to dryness, and strongly igniting the residue. The oxide of
iron may be made by weighing up 10 grams of powdered ferrous sulphate
(= to 2.014 grams of iron) and heating, at first gently, to drive off the
water, and then at a red heat, until completely decomposed. The weight
of oxide, in each case, should be determined; and the percentage of
oxygen calculated. Compare the figures arrived at with those calculated
from the formula of the oxides, CuO and Fe_{2}O_{3}.

It would be found in a more extended series of experiments that the same
metal will, under certain conditions, form two or more oxides differing
among themselves in the amount of oxygen they contain. These oxides are
distinguished from one another by such names as "higher" and "lower
oxides," "peroxides," "protoxides," "dioxides," &c.

The oxides may be conveniently classified under three heads:--

(1) _Those that are reduced to metal by heat alone_, such as the oxides
of mercury, silver, platinum, gold, &c.;

(2) _Those which are reduced by hydrogen at a red heat_, which includes
the oxides of the heavy metals;

(3) _Those which are not reduced by these means_, good examples of which
are silica, alumina, the alkalies, and the alkaline earths.

Another important classification is into acid, basic and neutral oxides.
The oxides of the non-metallic elements, such as sulphur, carbon,
phosphorus, &c., are, as a rule, acid; and the more oxygen they contain,
the more distinctly acid they are. The oxides of the metals are nearly
all basic; and, as a rule, the less oxygen they contain, the more
distinctly basic they are.

The basic oxides, which are soluble in acids, give rise to the formation
of salts when dissolved therein. During the solution, water is formed,
but no gas is evolved. The oxide dissolved in each case neutralizes an
equivalent of the acid used for solution.[97] The basic properties of
many of these can be taken advantage of for their determination. This is
done in the case of soda, potash, lime, &c., by finding the quantity of
acid required to neutralize a given weight of the substance.

There are some oxides which, under certain conditions, are acid to one
substance (a stronger base) and basic to another (a stronger acid). For
example, the oxides of lead and of tin, as also alumina, dissolve in
caustic soda, acting as acids; whilst, on the other hand, they combine
with sulphuric or hydrochloric acid, playing the part of bases.

The oxides known as "earths," when ignited, are many of them insoluble
in acids, although easily dissolved before ignition.

It is common in complete analyses of minerals to meet with cases in
which the sum total of the elements found falls short of the amount of
ore taken; and here oxygen must be looked for. For example, this occurs
in the case of a mixture of pyrites with oxide of iron, or in a mixture
of sulphides and sulphates. The state in which the elements are present,
and the percentage (say of sulphides and sulphates) can in many cases be
determined; but this is not always required. When the difference between
the sum total and the elements found is small, it is reported as "oxygen
and loss." When, however, it is considerable, the oxygen may be reported
as such; and its amount be either determined directly in the way already
described, or calculated from the best determination that can be made of
the relative amounts of oxides, sulphides, sulphates, &c., present. Such
cases require a careful qualitative analysis to find out that the
substance is present; and then the separation of each constituent is
made as strictly as possible. These remarks apply especially to ores of
the heavy metals. The separation of the constituents is effected with
suitable solvents applied in proper order. The soluble sulphates, for
example, are extracted with water; the oxides by the dilute acids or
alkalies in which they are known to be soluble. The oxygen in the
sulphates and oxides thus obtained is estimated by determining the
sulphur and metals in the solutions, and calculating the amount of
oxygen with which they combine. The metals of the earths and alkalies
are almost invariably present as oxides, and are reported as such;
except it is known that they are present in some other form, such as
fluoride or chloride. Thus, silica, alumina, lime, water, &c., appear in
an analysis; even in those cases where "oxygen and loss" is also
mentioned. As an example of such a report, take the following analysis
of Spanish pyrites:--

  Sulphur           49.00
  Iron              43.55
  Copper             3.20
  Arsenic            0.47
  Lead               0.93
  Zinc               0.35
  Lime               0.10
  Silica, &c.        0.63
  Water              0.70
  Oxygen and loss    1.07
                    -----
                   100.00

The following example will illustrate the mode of calculating and
reporting. A mineral, occurring as blue crystals soluble in water, and
found on testing to be a mixed sulphate of iron and copper, gave on
analysis the following results:--

  Water                44.51 per cent.
  Sulphuric oxide      28.82    "
  Copper                8.44    "
  Ferrous iron         11.81    "
  Ferric iron           0.38    "
  Zinc                  0.28    "
                       -----
                       94.24

There is here a deficiency of 5.76 per cent. due to oxygen. Nothing else
could be found, and it is known that in the sulphates the metals exist
as oxides. By multiplying the weight of the copper by 1.252, the weight
of copper oxide (CuO) will be ascertained; in this case it equals 10.57
per cent. The ferrous iron multiplied by 1.286 will give the ferrous
oxide (FeO); in this case 15.19 per cent. The ferric iron multiplied by
1.428 will give the ferric oxide (Fe_{2}O_{3}); in this case 0.54 per
cent. The zinc multiplied by 1.246 will give the zinc oxide (ZnO); in
this case it equals 0.35 per cent. The analysis will be reported as--

  Water                 44.51
  Sulphuric oxide       28.82
  Copper oxide          10.57 equal to copper 8.44%
  Ferrous oxide         15.19
  Ferric oxide           0.54
  Zinc oxide             0.35
                        -----
                        99.98

The following (A) is an analysis of a sample of South American copper
ore, which will serve as a further illustration. The analysis showed the
presence of 6.89 per cent. of ferrous oxide, and some oxide of copper.

The analysis (B) is that of an ore from the same mine after an imperfect
roasting. It will be seen that the carbonates have been converted into
sulphates. If the total sulphur simply had been determined, and the
sulphate overlooked, the "oxygen and loss" would have been 5.65 per
cent., an amount which would obviously require an explanation.

                         A.             B.
  Water                  0.25           0.59
  Organic matter         0.54            --
  Sulphur               29.50          21.33
  Copper                10.92           9.80 {Copper        9.57
                                             {Copper oxide  0.28
  Iron                  32.09          39.73   {Iron         34.32
                                               {Ferric oxide  7.73
  Lead                   0.35           0.12
  Zinc                   0.86           0.69
  Cobalt                 0.06           0.11
  Lime                   5.25           7.69
  Magnesia               2.33           2.55
  Sulphuric oxide        1.00           5.30
  Carbon dioxide         8.87            --
  "Insoluble silicates"  5.12           8.38
  Oxygen and loss        2.86           2.47
                        -----  Potash   0.15
                       100.00  Soda     1.09
                                       -----
                                      100.00


WATER.

Water occurs in minerals in two forms, free and combined. The term
"moisture" ought, strictly, to be limited to the first, although, as has
already been explained, it is more convenient in assaying to apply the
term to all water which is driven off on drying at 100° C. The combined
water is really a part of the mineral itself, although it may be driven
off at a high temperature, which varies with the base. In some cases a
prolonged red heat is required; whilst with crystallised salts it is
sometimes given off at the ordinary temperatures. This latter
phenomenon, known as efflorescence, is mostly confined to artificial
salts.

The determination of the combined water may often be made by simply
igniting the substance from which the moisture has been removed. The
quantity of water may be determined, either indirectly by the loss, or
directly by collecting it in a calcium chloride tube, and weighing. In
some cases, in which the loss on ignition does not give simply the
proportion of combined water, it can be seen from the analysis to what
else the loss is due; and, after a proper deduction, the amount of water
can be estimated. For example, 1 gram of crystallised iron sulphate was
found to contain on analysis 0.2877 gram of sulphuric oxide; and on
igniting another gram, 0.2877 gram of ferric oxide was left. As the salt
is known to be made up of ferrous oxide, sulphuric oxide, and combined
water, the combined water can be thus calculated: 0.2877 gram of ferric
oxide is equal to 0.2589 gram of ferrous oxide,[98] and consequently,
the loss on ignition has been diminished by 0.0288 gram, which is the
weight of oxygen absorbed by the ferrous oxide during calcining. The
loss on ignition was 0.7123 gram, to which must be added 0.0288 gram;
hence 0.7411 gram is the weight of the combined sulphuric oxide and
water present. Deducting the weight of sulphuric oxide found, 0.2877
gram, there is left for combined water 0.4534 gram. The composition of 1
gram of the dry salt is then:--

  Water             0.4534
  Sulphuric oxide   0.2877
  Ferrous oxide     0.2589
                    ------
                    1.0000

The following is another example:--A sample of malachite lost on
ignition 28.47 per cent., leaving a residue which was found on analysis
to be made up of oxide of copper (equal to 70.16 per cent. on the
mineral), and silica and oxide of iron (equal to 1.37 per cent.). Carbon
dioxide and water (but nothing else) was found to be present, and the
carbon dioxide amounted to 19.64 per cent.; deducting this from the loss
on ignition, we have 8.82 as the percentage of water present. The
analysis was then reported as follows:--

  Cupric oxide              70.16 equal to 56.0% copper.
  Silica and ferric oxide    1.37
  Carbon dioxide            19.64
  Water                      8.82
                            -----
                            99.99

[Illustration: FIG. 63.]

~Direct Determination of Combined Water.~--Transfer about 3 grams of the
substance to a piece of combustion tube (8 or 10 inches long), attached
(as in fig. 63) at one end to a ~U~-tube containing sulphuric acid, and
at the other end to a calcium chloride tube. The last is weighed
previous to the determination. The tube should be warmed to ensure
complete dryness, and must be free from a misty appearance. Aspirate a
current of air through the apparatus, heat the mineral by means of a
Bunsen burner, cautiously at first, and afterwards to redness (if
necessary). The water is driven off and condenses in the calcium
chloride tube, which is afterwards cooled and weighed. The increase in
weight is due to the water. If the substance gives off acid products on
heating, it is previously mixed with some dry oxide of lead or pure
calcined magnesia.


EXAMINATION OF WATERS.

The assayer is occasionally called on to test water for the purpose of
ascertaining the nature and quantity of the salts contained in it, and
whether it is or is not fit for technical and drinking purposes.

In mineral districts the water is generally of exceptional character,
being more or less charged, not only with earthy salts, but also
frequently with those of the metals. Distilled water is only used by
assayers in certain exceptional cases, so that by many it would be
classed among the rarer oxides. Water of ordinary purity will do for
most purposes, but the nature and quantity of the impurities must be
known.

The following determinations are of chief importance:--

~Total Solids at 100° C.~--Where simply the amount is required, take 100
c.c. and evaporate on the water-bath in a weighed dish; then dry in the
water-oven, and weigh.

~Total Solids Ignited.~--The above residue is very gently ignited
(keeping the heat well below redness), and again weighed. A larger loss
than 4 or 5 parts per 100,000 on the water requires an explanation.

~Chlorine.~--Take 100 c.c. of the water in a porcelain dish, add 2 c.c.
of a 5 per cent. solution of neutral potassic chromate, and titrate with
a neutral standard solution of nitrate of silver, made by dissolving
4.789 grams of crystallised silver nitrate in distilled water, and
diluting to 1 litre. The addition of the nitrate of silver is continued
until the yellow of the solution assumes a reddish tint. The reaction is
very sharp. Each c.c. of nitrate of silver used is equal to 1 part by
weight of chlorine in 100,000 of water. At inland places this rarely
amounts to more than 1 in 100,000; but near the sea it may amount to 3
or 5. More than this requires explanation, and generally indicates
sewage pollution.

~Nitric Pentoxide (N_{2}O_{5}).~--It is more generally reported under
the heading, "nitrogen as nitrates." Take 250 c.c. of the water and
evaporate to 2 or 3 c.c.; acidulate with a few drops of dilute sulphuric
acid, and transfer to a nitrometer (using strong sulphuric acid to wash
in the last traces). The sulphuric acid must be added to at least twice
the bulk of the liquid. Shake up with mercury. The mercury rapidly
flours, and nitric oxide is given off (if any nitrate is present). The
volume of the nitric oxide (corrected to normal temperature and
pressure), multiplied by 0.25, gives the parts of nitrogen per 100,000;
or, multiplied by 0.965, will give the nitric pentoxide in parts per
100,000. In well and spring waters the nitrogen may amount to 0.3 or 0.4
parts per 100,000; or in richly cultivated districts 0.7 or 0.8 parts
per 100,000. An excess of nitrates is a suspicious feature, and is
generally due to previous contamination.

~Ammonia.~--Take 500 c.c. of the water and place them in a retort
connected with a Liebig's condenser. Add a drop or two of a solution of
carbonate of soda and distil over 100 c.c.; collect another 50 c.c.
separately. Determine the ammonia in the distillate colorimetrically
(with Nessler's solution, as described under _Ammonia_) and compare with
a standard solution of ammonic chloride containing 0.0315 gram of
ammonic chloride in 1 litre of water. One c.c. contains 0.01 milligram
of ammonia. The second distillate will show little, if any, ammonia in
ordinary cases. The amounts found in both distillates are added
together, and expressed in parts per 100,000.

Waters (other than rain and tank waters) which contain more than 0.003
per 100,000 are suspicious.

~Organic Matter.~--The organic matter cannot be determined directly; but
for ordinary purposes it may be measured by the amount of permanganate
of potassium which it reduces, or by the amount of ammonia which it
evolves on boiling with an alkaline permanganate of potassium solution.

A. _Albuminoid Ammonia._--To the residue left after distilling the
ammonia add 50 c.c. of a solution made by dissolving 200 grams of potash
and 8 grams of potassium permanganate in 1100 c.c. of water, and rapidly
boiling till the volume is reduced to 1 litre (this should be kept in a
well stoppered bottle, and be occasionally tested to see that it is free
from ammonia). Continue the distillation, collecting 50 c.c. at a time,
until the distillate is free from ammonia. Three or four fractions are
generally sufficient. Determine the ammonia colorimetrically as before.
If the total albuminoid ammonia does not exceed 0.005 in 100,000, the
water may be regarded as clean as regards organic matter; if it amounts
to more than 0.015, it is dirty.

B. _Oxygen Consumed._--A standard solution of permanganate of potash is
made by dissolving 0.395 gram of the salt in water and diluting to 1
litre. Each c.c. equals 0.1 milligram of available oxygen. The following
are also required:--1. A solution of sodium hyposulphite containing 1
gram of the salt (Na_{2}S_{2}O_{3}.5H_{2}O) in 1 litre of water. 2.
Dilute sulphuric acid, made by adding one part of the acid to three of
water, and titrating with the permanganate solution till a faint pink
persists after warming for several hours. 3. Starch paste. 4. Potassium
iodide solution.

Take 250 c.c. of the water in a stoppered bottle, add 10 c.c. of
sulphuric acid and 10 c.c. of the permanganate, and allow to stand in a
warm place for four hours. Then add a few drops of the solution of
potassium iodide, and titrate the liberated iodine with "hypo," using
starch paste towards the end as an indicator. To standardise the
hyposulphite, take 250 c.c. of water and 10 c.c. of sulphuric acid, and
a few drops of potassium iodide; then run in 10 c.c. of the
"permanganate" solution, and again titrate; about 30 c.c. of the "hypo"
will be used. The difference in the two titrations, divided by the last
and multiplied by 10, will give the c.c. of permanganate solution used
in oxidising the organic matter in the 250 c.c. of water. Each c.c.
represents 0.04 parts of oxygen in 100,000.

~Metals.~--These may for the most part be estimated colorimetrically.

~Lead.~--Take 100 c.c. of the water in a Nessler tube, and add 10 c.c.
of sulphuretted hydrogen water, and compare the tint, if any, against a
standard lead solution, as described under _Colorimetric Lead_. Report
in parts per 100,000.

~Copper.~--Proceed as with the last-mentioned metal; but, if lead is
also present, boil down 500 c.c. to about 50 c.c., then add ammonia,
filter, and estimate the copper in the blue solution, as described under
_Colorimetric Copper_.

~Iron.~--Take 50 c.c., or a smaller quantity (if necessary), dilute up
to the mark with distilled water, and determine with potassium
sulphocyanate, as described under _Colorimetric Iron_.

~Zinc.~--Zinc is the only other metal likely to be present; and, since
it cannot be determined colorimetrically, it must be separately
estimated during the examination of the "total solids."

~Examination of "Total Solids."~--Evaporate 500 c.c. to dryness with a
drop or two of hydrochloric acid. Take up with hydrochloric acid,
filter, ignite, and weigh the residue as "silica." To the filtrate add a
little ammonic chloride and ammonia, boil and filter, ignite, and weigh
the precipitate as "oxide of iron and alumina." Collect the filtrate in
a small flask, add a few drops of ammonium sulphide or pass sulphuretted
hydrogen, cork the flask, and allow to stand overnight; filter, wash,
and determine the zinc gravimetrically as oxide of zinc. If copper or
lead were present, they should have been previously removed with
sulphuretted hydrogen in the acid solution. To the filtrate add ammonic
oxalate and ammonia, boil for some time, allow to stand, filter, wash,
ignite, and weigh as "lime." Evaporate the filtrate with nitric acid,
and ignite. Take up with a few drops of dilute hydrochloric acid, add
baric hydrate in excess, evaporate, and extract with water. The residue
contains the magnesia; boil with dilute sulphuric acid, filter,
precipitate it with phosphate of soda and ammonia, and weigh as
pyrophosphate. The aqueous extract contains the alkalies with the excess
of barium. Add sulphuric acid in slight excess, filter, evaporate, and
ignite strongly. The residue consists of the sulphates of the alkalies
(which are separately determined, as described under _Potash_).

~Sulphuric Oxide (SO_{3}).~--Take 200 c.c. and boil to a small bulk with
a little hydrochloric acid, filter (if necessary), add baric chloride
solution in slight excess to the hot solution, filter, ignite, and weigh
as baric sulphate.

~Carbon Dioxide (free).~--Carbon dioxide exists in waters in two forms,
free and combined. The latter generally occurs as bicarbonate, although
on analysis it is more convenient to consider it as carbonate, and to
count the excess of carbon dioxide with the free. The method is as
follows:--To determine the free carbon dioxide, take 100 c.c. of the
water, place them in a flask with 3 c.c. of a strong solution of calcium
chloride and 2 c.c. of a solution of ammonic chloride, next add 50 c.c.
of lime-water. The strength of the lime-water must be known. Make up to
200 c.c. with distilled water, stop the flask, and allow the precipitate
to settle. Take out 100 c.c. of the clear solution with a pipette, and
titrate with the standard solution of acid.[99] The number of c.c.
required, multiplied by two, and deducted from that required for the 50
c.c. of lime-water, and then multiplied by 0.0045, will give the carbon
dioxide present other than as normal carbonates.

~Carbon Dioxide combined~ as normal carbonate.--100 c.c. of the water
are tinted with phenacetolin or lacmoid; then heated to near boiling,
and titrated with standard acid. The number of c.c. used, multiplied by
0.0045, will give the weight in grams of the combined carbon dioxide.

~Free Acid.~--In some waters (especially those from mining districts)
there will be no carbonates. On the contrary, there may be free mineral
acid or acid salts. In these cases it is necessary to determine the
amount of acid (other than carbon dioxide) present in excess of that
required to form normal salts. This is done in the following way:--Make
an ammoniacal copper solution by taking 13 grams of copper sulphate
(CuSO_{4}.5H_{2}O), dissolving in water, adding solution of ammonia
until the precipitate first formed has nearly dissolved, and diluting to
1 litre. Allow to settle, and decant off the clear liquid. The strength
of this solution is determined by titrating against 10 or 20 c.c. of the
standard solution of sulphuric acid (100 c.c. = 1 gram H_{2}SO_{4}). The
finishing point is reached as soon as the solution becomes turbid from
precipitated cupric hydrate. At first, as each drop falls into the acid
solution, the ammonia and cupric hydrate combine with the free acid to
form ammonic and cupric sulphates; but as soon as the free acid is used
up, the ammonia in the next drop not only precipitates an equivalent of
cupric hydrate from the solution, but also throws down that carried by
itself. This method is applicable in the presence of metallic sulphates
_other than ferric_. The standardising and titration should be made
under the same conditions. Since sulphuric acid and sulphates are
predominant in waters of this kind, it is most convenient to report the
acidity of the water as equivalent to so much sulphuric acid.

~Dissolved Oxygen.~--For the gasometric method of analysing for
dissolved oxygen, and for the Schützenberger's volumetric method, the
student is referred to Sutton's "Volumetric Analysis." The following is
an easy method of estimating the free oxygen in a water:--Take 20 c.c.
of a stannous chloride solution (about 20 grams of the salt with 10 c.c.
of hydrochloric acid to the litre); add 10 c.c. of hydrochloric acid,
and titrate in an atmosphere of carbon dioxide with standard
permanganate of potassium solution (made by dissolving 1.975 gram of the
salt in 1 litre of water: 1 c.c. equals 0.5 milligram of oxygen). A
similar titration is made with the addition of 100 c.c. of the water to
be tested. Less permanganate will be required in the second titration,
according to the amount of oxygen in the water; and the difference,
multiplied by 0.5, will give the weight of the oxygen in milligrams.
Small quantities of nitrates do not interfere.

In REPORTING the results of the analysis, it is customary to combine the
acids and bases found on some such principle as the following:--The
sulphuric oxide is calculated as combined with the potash, and reported
as potassic sulphate (K_{2}SO_{4}); the balance of the sulphuric oxide
is then apportioned to the soda, and reported as sulphate of soda
(Na_{2}SO_{4}); if any is still left, it is reported as calcium sulphate
(CaSO_{4}), and after that as magnesic sulphate (MgSO_{4}). When the
sulphuric oxide has been satisfied, the chlorine is distributed, taking
the bases in the same order, then the nitric pentoxide, and lastly the
carbon dioxide. But any method for thus combining the bases and acids
must be arbitrary and inaccurate. It is extremely improbable that any
simple statement can represent the manner in which the bases and acids
are distributed whilst in solution; and, since different chemists are
not agreed as to any one system, it is better to give up the attempt,
and simply state the results of the analysis. This has only one
inconvenience. The bases are represented as oxides; and, since some of
them are present as chlorides, the sum total of the analysis will be in
excess of the actual amount present by the weight of the oxygen
equivalent to the chlorine present as chloride. The following is an
example of such a statement:--

                                                Parts per 100,000.
  Total solids, dried at 100° C.                      28.73
  Chlorine                                             1.70
  Nitrogen as nitrate                                  0.03
  Ammonia                                              0.001
  Albuminoid ammonia                                   0.004
  "Oxygen consumed" in 4 hours                         0.01

The solids were made up as under:--

                                            Per 100,000 of the Water.
  Potash                                               0.38
  Soda                                                 2.01
  Magnesia                                             1.44
  Lime                                                10.55
  Ferric oxide                                         0.01
  Silica                                               0.30
  Sulphuric oxide                                      3.69
  Nitrogen pentoxide                                   0.11
  Carbon dioxide                                       8.38
  Chlorine                                             1.70
  Volatile and organic matter                          0.66
                                                      -----
                                                      29.23
  Less oxygen equivalent to chlorine found             0.39
                                                      -----
                                                      28.84

For the preparation of distilled water, the apparatus shown in fig. 64
is convenient for laboratory use. It consists of a copper retort heated
by a ring gas-burner, and connected with a worm-condenser.

[Illustration: FIG. 64]


PRACTICAL EXERCISE.

A mineral, on analysis, gave the following results:--Water, 44.94 per
cent.; sulphuric oxide, 28.72 per cent.; ferrous iron, 13.92 per cent.;
ferric iron, 0.35 per cent.; copper, 6.1 per cent. The mineral was
soluble in water, and showed nothing else on testing. How would you
report the analysis? Calculate the formula for the salt.


THE HALOGENS.

There is a group of closely allied elements to which the name halogen
(salt-producer) has been given. It comprises chlorine, bromine, iodine,
and fluorine. These elements combine directly with metals, forming as
many series of salts (chlorides, bromides, iodides, and fluorides),
corresponding to the respective oxides, but differing in their formulæ
by having two atoms of the halogen in the place of one atom of oxygen.
For example, ferrous oxide is FeO and ferrous chloride is FeCl_{2}, and,
again, ferric oxide is Fe_{2}O_{3}, whilst ferric chloride is
Fe_{2}Cl_{6}. These salts differ from the carbonates, nitrates, &c., in
containing no oxygen. Consequently, it is incorrect to speak of such
compounds as chloride of potash, fluoride of lime, &c., since potash and
lime are oxides. It is important to bear this in mind in reporting
analyses in which determinations have been made, say, of chlorine,
magnesia, and potash, or of fluorine, silica, and alumina. It is
necessary in all such cases to deduct from the total an amount of oxygen
equivalent to the halogen found, except, of course, where the base has
been determined and recorded as metal. Compounds containing oxides and
fluorides, &c., do not lend themselves to the method of determining the
halogen by difference. For example, topaz, which, according to Dana, has
the formula Al_{2}SiO_{4}F_{2}, would yield in the ordinary course of
analysis--

  Alumina          55.4%
  Silica           32.6
  Fluorine         20.6
                  -----
                  108.6

The oxygen equivalent to 20.6 per cent. fluorine may be found by
multiplying the percentage of fluorine by 0.421; it is 8.7 per cent.,
and must be deducted. The analysis would then be reported thus:--

  Alumina                                 55.4%
  Silica                                  32.6
  Fluorine                                20.6
                                         -----
                                         108.6
  Less oxygen equivalent to fluorine       8.7
                                         -----
                                          99.9

Take as an illustration the following actual analysis by F.W. Clarke and
J.S. Diller:--

  Alumina                     57.38%
  Silica                      31.92
  Fluorine                    16.99
  Potash                       0.15
  Soda                         1.33
  Water                        0.20
                             ------
                             107.97
  Deduct oxygen equivalent     7.16
                             ------
                             100.81

In calculating the factor for the "oxygen equivalent," divide the weight
of one atom of oxygen (16) by the weight of two atoms of the halogen;
for example, with chlorine it would be 16/71 or 0.2253; with bromine,
16/160 or 0.1000; with iodine, 16/254 or 0.063; and with fluorine, 16/38
or 0.421.


CHLORINE AND CHLORIDES.

Chlorine occurs in nature chiefly combined with sodium, as halite or
rock salt (NaCl). With potassium it forms sylvine (KCl), and, together
with magnesium, carnallite (KCl.MgCl_{2}.6H_{2}O). Of the metalliferous
minerals containing chlorine, kerargyrite, or horn silver (AgCl), and
atacamite, an oxychloride of copper (CuCl_{2}.3Cu(HO)_{2}.) are the most
important. Apatite (phosphate of lime) and pyromorphite (phosphate of
lead) contain a considerable amount of it. Chlorine is a gas of a
greenish colour, possessing a characteristic odour, and moderately
soluble in water. It does not occur native, and is generally prepared by
the action of an oxidising agent on hydrochloric acid. It combines
directly with metals at the ordinary temperature (even with platinum and
gold), forming chlorides, which (except in the case of silver) are
soluble.

It is important in metallurgy, because of the extensive use of it in
extracting gold by "chloridising" processes. It is also used in refining
gold.

~Detection.~--Compounds containing the oxides of chlorine are not found
in nature, because of the readiness with which they lose oxygen. By
reduction they yield a chloride; the form in which chlorine is met with
in minerals. In testing, the compound supposed to contain a chloride is
boiled with water, or, in some cases, dilute nitric acid. To the clear
solution containing nitric acid a few drops of nitrate of silver
solution are added. If, on shaking, a white curdy precipitate, soluble
in ammonia, separates out, it is sufficiently satisfactory evidence of
the presence of chlorides.

~Solution and Separation.~--The chlorides are generally soluble in
water, and are got into solution by extracting with warm dilute nitric
acid. Or, if insoluble, the substance is fused with carbonate of soda,
extracted with water, and the filtrate acidified with nitric acid. For
the determination, it is not necessary to obtain the solution of the
chloride free from other acids or metals. If tin, antimony, mercury, or
platinum is present, it is best to separate by means of sulphuretted
hydrogen. The chloride is determined in the solution after removal of
the excess of the gas. Where traces of chlorides are being looked for, a
blank experiment is made to determine the quantity introduced with the
reagents. One hundred c.c. of ordinary water contains from 1 to 3
milligrams of chlorine. On the addition of nitrate of silver to the
nitric acid solution, chloride of silver separates out. This is free
from other substances, except, perhaps, bromide and iodide.


GRAVIMETRIC DETERMINATION.

Freely mix the solution containing the chloride with dilute nitric acid,
filter (if necessary), and treat with nitrate of silver. Heat nearly to
boiling, and, when the precipitate has settled, filter, and wash with
hot distilled water. Dry, and transfer to a weighed Berlin crucible.
Burn the filter-paper separately, and convert any reduced silver into
chloride by alternate treatment with drops of nitric and of hydrochloric
acid. Add the main portion to this, and heat cautiously till the edges
of the mass show signs of fusing (about 260°). Cool in the desiccator
and weigh. The substance is chloride of silver (AgCl), and contains
24.73 per cent. of chlorine.

The precipitated chloride is filtered and washed as soon as possible
after settling, since on exposure to light it becomes purple, and loses
a small amount of chlorine.


VOLUMETRIC METHOD.

There are several volumetric methods; but that based on the
precipitation of silver chloride in neutral solution, by means of a
standard solution of silver nitrate (using potassium chromate as
indicator), is preferred. Silver chromate is a red-coloured salt; and,
when silver nitrate is added to a solution containing both chloride and
chromate, the development of the red colour marks off sharply the point
at which the chloride is used up. Silver chromate is decomposed and
consequently decolorised by solution of any chloride. The solution for
this method must be neutral, since free acid prevents the formation of
the red silver chromate. If not already neutral, it is neutralised by
titrating cautiously with a solution of soda. In a neutral solution,
other substances (such as phosphates and arsenates) also yield a
precipitate with a solution of nitrate of silver; and will count as
chloride if they are not removed.

_The Standard Solution of Nitrate of Silver_ is made by dissolving 23.94
grams of the salt (AgNO_{3}) in distilled water, and diluting to 1
litre; 100 c.c. are equal to 0.5 gram of chlorine.

The _indicator_ is made by adding silver nitrate to a strong neutral
solution of yellow chromate of potash (K_{2}CrO_{4}), till a permanent
red precipitate is formed. The solution is allowed to settle, and the
clear liquid decanted into a stoppered bottle labelled "chromate
indicator for chlorine."

Standardise the silver nitrate by weighing up 0.5 gram of pure sodium
chloride (or potassium chloride). Transfer to a flask and dissolve in
distilled water; dilute to 100 c.c. Fill an ordinary burette with the
standard silver solution, and (after adjusting) run into the flask a
quantity sufficient to throw down the greater part of the chlorine. Add
a few drops of the chromate indicator and continue the addition of the
silver nitrate until the yellow colour of the solution becomes
permanently tinted red, after shaking. This shows that the chlorine is
all precipitated, and that the chromate is beginning to come down. The
further addition of a couple of drops of the silver solution will cause
a marked difference in the tint. Read off the quantity run in, and
calculate the standard. One gram of sodium chloride contains 0.6062 gram
of chlorine; and 1 gram of potassium chloride contains 0.4754 gram.

For the determination of small quantities of chloride (a few
milligrams), the same method is used; but the standard solution is
diluted so that each c.c. is equal to 1 milligram of chlorine; and the
chromate indicator is added before titrating. The standard solution is
made by measuring off 200 c.c. of the solution described above, and
diluting with distilled water to 1 litre.


BROMINE AND BROMIDES.

Bromine closely resembles chlorine in the nature of its compounds. It
does not occur free in nature, but is occasionally found in combination
with silver as bromargyrite (AgBr) and, together with chloride, in
embolite. It mainly occurs as alkaline bromides in certain natural
waters. Nearly all the bromine of commerce is derived from the mother
liquors of salt-works--_i.e._, the liquors from which the common salt
has been crystallised out. Bromine combines directly with the metals,
forming a series of salts--the bromides. In ordinary work they are
separated with, and (except when specially tested for) counted as,
chlorides. They are detected by adding chlorine water to the suspected
solution and shaking up with carbon bisulphide. Bromine colours the
latter brown.


IODINE AND IODIDES.

Iodine does not occur in nature in the free state; and iodides are rare,
iodargyrite or iodide of silver (AgI) being the only one which ranks as
a mineral species. Iodates are found associated with Chili saltpetre,
which is an important source of the element.

Iodine and Iodides are largely used in the laboratory, and have already
been frequently referred to. It is used as an oxidising agent in a
similar manner as permanganate and bichromate of potash, especially in
the determinations of copper, arsenic, antimony, and manganese.

Iodine is not readily soluble in water; but dissolves easily in a
concentrated solution of potassium iodide. Its solutions are strongly
coloured; a drop of a dilute solution colours a large volume of water
decidedly yellow; on the addition of starch paste, this becomes blue.
The delicacy of this reaction is taken advantage of in titrations to
determine when free iodine is present. The blue colour may be
alternately developed and removed by the addition of iodine (or an
oxidising agent) and hyposulphite of soda (or some other reducing
agent). In decolorising, the solution changes from blue or black to
colourless or pale yellow according to circumstances. Sometimes the
solution, instead of remaining colourless, gradually develops a blue
which recurs in spite of the further addition of the reducing agent. In
these cases the conditions of the assay have been departed from, or (and
this is more often the case) there is some substance present capable of
liberating iodine.

Iodine forms a series of salts--the iodides--resembling in many respects
the chlorides. These can be obtained by direct combination of the metals
with iodine.

~Detection.~--Free iodine is best recognised by the violet vapours
evolved from the solution on heating, and by the blue or black colour
which it strikes on the addition of starch paste. Iodides are detected
by boiling with strong solutions of ferric sulphate or chloride. Iodine
is liberated, distilled over, and collected. Chlorine also liberates
iodine from iodides; and this reaction is frequently made use of in
assaying. A process based on this is described under _Manganese_. All
substances which liberate chlorine on boiling with hydrochloric acid
(dioxides, bichromates, permanganates, &c.) are determined in a similar
way.

~Solution and Separation.~--Most iodides are soluble in water or dilute
acids. The separation is effected by distilling the substance with
solution of ferric sulphate, and collecting the vapour in a dilute
solution of sulphurous acid or arsenite of soda. On the completion of
the distillation, the iodine will be in the distillate as iodide; and
the gravimetric determination is made on this.


GRAVIMETRIC DETERMINATION.

To the solution containing the iodine, as iodide, and which is free from
chlorides (and bromides), add a little dilute nitric acid and nitrate of
silver till no further precipitate is produced. Filter off, wash with
hot water, and dry. Clean the filter-paper as much as possible, and burn
it. Collect the ash in a weighed porcelain crucible, add the main
portion, and heat to incipient fusion; cool, and weigh. The substance is
silver iodide, and contains 54.03 per cent. of iodine.


VOLUMETRIC METHOD.

This is for the titration of free iodine, and is practically that which
is described under _Manganese_. The substance to be determined is
distilled with ferric sulphate, and the iodine is collected in a
solution of potassium iodide, in which it readily dissolves. If flaky
crystals separate out in the receiver, more potassium iodide crystals
are added. When the distillation is finished, the receiver is
disconnected, and its contents washed out into a beaker and titrated
with "hypo." The standard solution of "hypo" is made by dissolving 19.58
grams of hyposulphite of soda (Na_{2}S_{2}O_{3}.5H_{2}O) in water and
diluting to 1 litre; 100 c.c. are equal to 1 gram of iodine. To
standardise the solution, weigh up 0.25 gram of pure iodine in a small
beaker. Add 2 or 3 crystals of potassium iodide; cover with water; and,
when dissolved, dilute to 50 or 100 c.c. Titrate, and calculate the
standard.


FLUORINE AND FLUORIDES.

Fluorine is frequently met with as calcium fluoride or fluor-spar
(CaF_{2}). It occurs less abundantly as cryolite (Na_{3}AlF_{6}), a
fluoride of aluminium and sodium, which is used in glass-making. Certain
other rarer fluorides are occasionally met with. Fluorine is also found
in apatite, and in some silicates, such as topaz, tourmaline, micas, &c.

Hydrofluoric acid is used for etching glass and opening up silicates. It
attacks silica, forming fluoride of silicon (SiF_{4}), which is
volatile. Silica is by this means eliminated from other oxides, which,
in the presence of sulphuric acid, are fixed. The commercial acid is
seldom pure, and generally weak; and the acid itself is dangerously
obnoxious. The use of ammonium fluoride (or sodium fluoride) and a
mineral acid is more convenient. Determinations of this kind are made in
platinum dishes enclosed in lead or copper vessels in a well-ventilated
place. Fluor-spar is useful as a flux in dry assaying; it renders slags,
which would otherwise be pasty, quite fluid. Fluorides generally are
fusible, and impart fusibility to substances with which they form weak
compounds. Their fluxing action does not depend on the removal of
silicon as fluoride.

~Detection.~--Fluorides in small quantity are easily overlooked unless
specially sought for. In larger amounts they are recognised by the
property hydrofluoric acid has of etching glass. A watch-glass is
warmed, and a layer of wax is melted over the convex side. When cold,
some lines are engraved on the waxed surface with any sharp-pointed
instrument. The substance to be tested is powdered; and moistened, in a
platinum dish, with sulphuric acid. The watch-glass is filled with cold
water and supported over the dish. The dish is then carefully warmed,
but not sufficiently to melt the wax. After a minute or two, the glass
is taken off, and the wax removed. If the substance contained fluorine,
the characters will be found permanently etched on the glass. An equally
good, but more rapid, test is to mix the powdered substance with some
silica, and to heat the mixture in a test tube with sulphuric acid.
Silicon fluoride is evolved, and, if a moistened glass rod is held in
the tube, it becomes coated with a white deposit of silica, formed by
the decomposition of the silicon fluoride by the water. This is also
used as a test for silica; but in this case the substance is mixed with
a fluoride, and the experiment must obviously be carried out in a
platinum vessel.

~Separation and Determination.~--The determination of fluorine is
difficult. In the case of fluorides free from silicates (such as
fluor-spar), it is determined indirectly by decomposing a weighed
portion with sulphuric acid, evaporating, igniting, and weighing the
residual sulphate. The increase in weight multiplied by 0.655 gives the
weight of fluorine.

In the presence of silica this method does not answer, because of the
volatilisation of silicon fluoride. In these cases Wöhler adopted the
following plan, which resembles that for the indirect determination of
carbon dioxide. Mix the weighed substance in a small flask with powdered
silica and sulphuric acid. The mouth of the flask is closed with a cork
carrying a tube which is filled, the first half with calcium chloride
and the second half with pumice coated with dried copper sulphate. The
apparatus is weighed quickly, and then warmed till decomposition is
complete. A current of dry air is aspirated for a minute or two; and the
apparatus again weighed. The loss in weight gives that of the silicon
fluoride (SiF_{4}), which, multiplied by 0.7307, gives the weight of
fluorine.

Fresenius uses the same reaction; but collects and weighs the silicon
fluoride. The finely powdered and dried substance is mixed with ten or
fifteen times its weight of ignited and powdered silica. The mixture is
introduced into a small dry flask connected on one side with a series of
drying-tubes, and on the other with an empty tube (to condense any
sulphuric acid). To this last is joined a drying-tube containing
chloride of calcium and anhydrous copper sulphate. This is directly
connected with a series of three weighed tubes in which the fluoride of
silicon is collected. The last of these is joined to another
drying-tube. The first weighed tube contains pumice and cotton wool,
moistened with water; the second tube contains soda-lime as well as (in
the upper half of the second limb) fused calcium chloride between plugs
of wool; the third tube is filled half with soda-lime and half with
fused calcium chloride. The distilling-flask containing the substance
mixed with silica is charged with 40 or 50 c.c. of sulphuric acid, and
placed on the hot plate. Alongside it is placed a similar dry flask
containing a thermometer, and the temperature in this is kept at 150° or
160° C. A current of air is sent through the tubes during the operation,
which takes from one to three hours for from 0.1 to 1 gram of the
substance. A correction is made by deducting 0.001 gram for every hour
the dried air has been passed through. The increase in weight of the
three tubes gives the weight of the silicon fluoride.

Penfield uses a similar arrangement, but passes his silicon fluoride
into an alcoholic solution of potassium chloride. Silica and potassium
silico-fluoride are precipitated, and hydrochloric acid is set
free.[100] The acid thus liberated is titrated, with a standard solution
of alkali, in the alcoholic solution, and from the amount of free acid
found the fluorine is calculated. The weight of hydrochloric acid (HCl)
found, multiplied by 1.562, gives the weight of the fluorine. With this
method of working, fewer ~U~-tubes are required. The exit tube from the
flask is bent so as to form a small ~V~, which is kept cool in water;
this is directly connected with the ~U~-tube containing the alcoholic
solution of potassium chloride. The flask with the assay is heated for
about two hours, and a current of dry air is aspirated throughout the
determination. Fluoride of silicon is a gas not easily condensed to a
liquid: but is immediately decomposed by water or moist air.

FOOTNOTES:

[95] This will require two or three hours to thoroughly complete. It is
best to powder the oxide first produced, and recalcine.

[96] No magnetic oxide was formed.

[97] For example:--

CaO + 2HCl = CaCl_{2} + H_{2}O.

PbO + H_{2}SO_{4} = PbSO_{4} + H_{2}O.

MgO + 2HNO_{3} = Mg(NO_{3})_{2} + H_{2}O.

Al_{2}O_{3} + 6HCl = Al_{2}Cl_{6} + 3H_{2}O.

Fe_{2}O_{3} + 3H_{2}SO_{4} = Fe_{2}(SO_{4})_{3} + 3H_{2}O.

[98] Fe_{2}O_{3}: 2FeO:: 0.2877: 0.2589.

[99] 100 c.c. contain 1 gram of sulphuric acid.

[100] 3SiF_{4} + 4KCl + 2H_{2}O = 2K_{2}SiF_{6} + SiO_{2} + 4HCl.



CHAPTER XVI.

SULPHUR AND SULPHATES.


Sulphur occurs native in volcanic districts, and is mined in Sicily,
Italy, and California in considerable quantities. Combined with metals
(sulphides), it is common in all mineral districts. Iron pyrites
(FeS_{2}) is the most abundant source of this element. Sulphates, such
as gypsum, are fairly common, but have no value so far as the sulphur in
them is concerned. In coal it exists as an impurity, occurring partly as
a constituent of organic compounds.

Sulphur, whether free or combined with metals, forms, on burning,
sulphurous oxide (SO_{2}), which by the action of oxidising agents and
water is converted into sulphuric acid. It forms two oxides, sulphurous
(SO_{2}) and sulphuric (SO_{3}), which combine with bases to form
sulphites and sulphates. Sulphites are of little importance to the
assayer, and are converted into sulphates by the action of nitric acid
and other oxidising agents.

The native sulphides, when acted on with hydrochloric acid, give off
sulphuretted hydrogen; with nitric acid or aqua regia, sulphates are
formed, and more or less sulphur separated.

Sulphur is detected in sulphides by the irritating odour of sulphurous
oxide given off on roasting, by the evolution of sulphuretted hydrogen
when treated with hydrochloric acid, or by a white precipitate of barium
sulphate formed when barium chloride is added to the aqua regia
solution.

~Dry Assay.~--There is no method of general application. Free or native
sulphur may be volatilised, condensed, and weighed, but pyrites only
gives up a portion of its sulphur when heated in a closed vessel, while
most sulphides, and all sulphates, give up none at all.

In the determination of sulphur in brimstone, 10 grams of the substance
are taken, placed in a small porcelain dish, heated over a Bunsen burner
in a well-ventilated place, and ignited. When the sulphur has been
completely burnt off, the residue (which consists chiefly of sand) is
collected and weighed. In a separate portion the moisture and arsenic
are determined; the amounts of these are deducted from the loss in the
first experiment. The difference, multiplied by 10, gives the percentage
of sulphur.


WET METHODS.

~Solution.~--All sulphates, excepting those of lead, barium, strontium,
and lime, are soluble in water or dilute acid. All sulphides, except
cinnabar, are converted into sulphates by the action of nitric acid at a
gentle heat; or, better, by the action of a mixture of three volumes of
nitric acid and one volume of hydrochloric acid. This last attacks
cinnabar as well. With most substances it is difficult to convert the
whole of the sulphur into sulphuric acid. The sulphur separates out at
first as a dark spongy mass, which (on continued treatment) changes to
light-coloured flakes. When the solution becomes concentrated and the
temperature rises sufficiently, the sulphur fuses into one or more
honey-coloured globules which, owing to the small surface they oppose to
the acid, are very slowly oxidised. It is not desirable to assist the
formation of these globules; therefore, the temperature is kept as low
as possible, and strong nitric acid is used. When such globules form, it
is best to allow the solution to cool, when the globules will solidify.
They can then be filtered off and picked out from the insoluble residue,
dried, weighed, ignited, and again weighed, the loss being counted as
sulphur. With iron pyrites this difficulty seldom occurs.

Metallic sulphides when fused with an excess of nitre are completely
oxidised. If the ore is rich in sulphur, some inert body (such as sodium
chloride, or, better, sodium carbonate) is added to dilute the action.
With pure sulphur, the action is so energetic as to cause an explosion,
so that care should be taken. With burnt ores (incompletely calcined
pyrites), there is sufficient oxide of iron present to prevent too rapid
action.

These fusions with nitre are best conducted in a platinum dish covered
with a piece of platinum foil. The ore is ground with the nitre to
ensure complete mixing. The heat need not be excessive, so that a single
Bunsen burner placed beneath the dish will suffice; if the bottom of the
dish is seen to be red-hot, it is sufficient. On cooling and extracting
with water, the sulphur will pass into solution as potassium sulphate,
which is then filtered off from the insoluble oxides of iron, copper,
&c. The filtrate, after having been treated with a large excess of
hydrochloric acid, evaporated to dryness, and re-dissolved in water, is
ready for the determination.

Lead sulphate may be dissolved by boiling with ammonium acetate. The
insoluble sulphates of barium, strontium, and lime, are decomposed by
fusing with 4 or 5 times their weight of "fusion mixture." The alkaline
sulphates are then dissolved out with water, and filtered off from the
insoluble residue. The filtrate is rendered acid with hydrochloric acid.

~Separation.~--The determination of the sulphuric acid in these
solutions by precipitation with barium chloride also serves as a
separation; but in hot acid solutions containing copper, and more
especially iron salts, the baric sulphate has a strong tendency to carry
down amounts of those bodies, varying, no doubt, with the conditions of
the precipitation. Boiling hydrochloric acid fails to completely extract
them. Moreover, the use of hot concentrated hydrochloric acid causes a
loss by dissolving barium sulphate. Nitric acid and nitrates must be
decomposed by prolonged boiling and evaporation with hydrochloric acid.
The iron may be removed by adding a slight excess of ammonia to the
faintly acid solution, filtering off, and washing the precipitated
ferric hydrate with hot water. By slightly acidulating the filtrate with
hydrochloric acid, it will be rendered ready for the determination.


GRAVIMETRIC METHOD.

This assay is one of those which strikingly shows the necessity of
getting the assay solution under proper conditions, in order to obtain
satisfactory results. The method has been repeatedly investigated, and
the conclusion arrived at, "that it can be correct only by accident."
Yet there are many chemists who get good results, and place considerable
faith in its accuracy. This can only be due to differences in the manner
of working. It is generally understood that nitric acid or nitrates must
be absent; and our experience fully confirms this. Precipitations in
nitrate solutions are worthless, as the following experiments show. In
each experiment the bulk of the solution was 150 c.c. The solutions
contained 10 grams of nitre, were freely acid with hydrochloric acid,
and were precipitated (while boiling) with slight excess of baric
chloride.

  Sulphuric acid taken  0.020 gram  0.050 gram  0.100 gram
     "           found  0.019   "   0.047   "   0.098   "
     "           taken  0.500   "   1.004   "   1.000   "
     "           found  0.526   "   1.126   "   1.126   "

All the precipitates were boiled with hydrochloric acid, and thoroughly
washed before weighing. The results of some other experiments on this
subject are given under "sulphur" in the "examination of commercial
copper," page 207.

The solution having been obtained free from nitrates and chlorates (and
containing but little free hydrochloric acid), is largely diluted,
heated to boiling, and precipitated with a moderate excess of a solution
of chloride of barium (8 parts of the crystallized barium chloride are
sufficient for 1 of sulphur). It is allowed to settle for half-an-hour,
and then decanted through a filter. The precipitate is shaken up with
boiling water, rendered slightly acid, filtered, washed, dried, ignited,
and weighed. The ignited precipitate, when pure, is white, and is not
decomposed at a red heat; it is barium sulphate (BaSO_{4}), and contains
13.73 per cent. of sulphur, or 34.33 per cent. of sulphuric oxide
(SO_{3}).

~Determination of Sulphur in Pyrites.~--Weigh up half a gram of the
dried and powdered sample, and treat with 10 c.c. of a mixture of 3
volumes of nitric acid and 1 volume of hydrochloric acid, occasionally
heating. Evaporate to dryness, treat with 5 c.c. of hydrochloric acid,
and again evaporate; take up with 1 c.c. of hydrochloric acid and 100
c.c. of hot water, filter through a small filter, and wash. The residue
may contain sulphates of lead, barium, or lime; it must be separately
examined, if the total sulphur is wanted. The filtrate is heated, and
rendered slightly alkaline with ammonia. Filter off the precipitated
ferric hydrate through a quick filter, and wash with hot water. If
necessary, evaporate the bulk to about 200 c.c., render faintly acid
with hydrochloric acid, and add 20 c.c. of solution of barium chloride;
allow to stand for half-an-hour, and decant through a filter. Wash with
hot water, dry, ignite, and weigh. Pure pyrites contains 53.33 per cent.
of sulphur.


VOLUMETRIC METHOD.

This is based upon the easy conversion of all sulphur compounds into
sulphates by fusion with nitre or by oxidation with nitric acid; and on
the determination of the sulphate formed by titration in an acetic acid
solution with baric chloride.[101] The finishing point is determined by
filtering off portions of the assay solution, and testing with sulphuric
acid. A slight excess of baric chloride will cause a precipitate.

The process may be divided into--(1) the preparation of the solution,
and (2) the titration.

~Preparation of the Solution.~--Weigh up from 1 to 5 grams of the dried
and powdered substance, and mix intimately with 4 grams of powdered
nitre; clean out the mortar with another gram of nitre, and add this as
a cover. Heat in a platinum crucible for fifteen minutes at a low
temperature; cool, and extract with water in an evaporating-dish about
9 inches across, and holding 700 or 800 c.c. Add 10 grams of sodium
acetate and 10 c.c. of acetic acid, and dilute to half a litre. Boil.
The solution is ready for titrating. Substances which lose sulphur on
heating (such as pyrites) are thus treated:--Weigh up 1 gram, and
evaporate nearly to dryness with 10 c.c. each of nitric and hydrochloric
acids. Take up with 10 c.c. of hydrochloric acid, and again boil down to
a small bulk; dilute and transfer to a 9-inch evaporating-dish; add 10
grams of sodium acetate and 5 c.c. of acetic acid, dilute to half a
litre, and boil. The solution is ready for titrating. Sulphates may be
dissolved up in the dish itself with the help of a c.c. or so of
hydrochloric acid; sodium acetate and acetic acid are then added; and,
after dilution and boiling, the solutions are at once titrated.

The solution before titration must contain no free mineral acid, but 5
or 10 c.c. of acetic acid should be present. It must contain 10 grams of
sodium acetate, or sufficient to convert any free mineral acid into its
corresponding sodic salt; or, if chlorides, nitrates or sulphates of the
metals are present, sufficient to decompose them. If a precipitation
occurs, as is the case with ferric salts, &c., the solution is titrated
with the precipitate in it.

~The Titration.~--_The standard solution of barium chloride_ is made by
dissolving 76.25 grams of the crystallized salt (BaCl_{2}.2H_{2}O) in
distilled water, and diluting to 1 litre. 100 c.c. will equal 1 gram of
sulphur. As indicator, use dilute sulphuric acid. The strength of the
solution may be checked by the titration of 5 grams of ferrous sulphate
(oxidized with permanganate of potassium or a few drops of nitric acid),
which should require 57.5 c.c. of the barium chloride solution; or any
pure sulphate of known composition can be used; anhydrous salts should
be preferred.

[Illustration: FIG. 65.]

Fill an ordinary 100 c.c. burette with the solution of barium chloride.
The evaporating dish containing the assay solution is placed on a round
burner (as shown in fig. 65), and the solution is kept steadily boiling.
An ordinary Bunsen-burner flame will cause bumping, and should not be
used. Run in the standard solution in quantity known to be
insufficient; then withdraw a portion of about 2 c.c., with a pipette,
and filter through a fine filter-paper into a test tube. Run in another
0.5 c.c. of the standard solution, and withdraw and filter into a test
tube another portion of 2 c.c.; and continue this operation until
half-a-dozen or more portions have been drawn off. The test tubes should
be arranged in order in a stand resting on a piece of paper, so that
each test tube representing 0.5 c.c. of the standard baric chloride may
have its value recorded beneath it (fig. 66). Add to each test tube 3
drops of dilute sulphuric acid; that which shows the first appearance of
a precipitate marks the point at which the titration is complete.
Suppose, for example, that the test tube marked 48.5 c.c. shows no
precipitate, while that at 49.0 c.c. shows one, it is evident that the
finishing point lies between these readings. With a little practice, one
can judge from the appearance of the precipitate in the 49 c.c. tube,
whether 1/4 c.c. should be deducted or not.

[Illustration: FIG. 66.]

It is better to add dilute sulphuric acid, and to watch for the
appearance of a precipitate in the test tube, than to add baric chloride
and to look for its non-appearance; besides, baric chloride is much less
likely to be present in a test tube as impurity than sulphates are. In
this way the chance of error from what are termed "accidental causes" is
diminished.

The following experiments show the effect of variation in the conditions
of titration:--

Make _a standard solution of sulphuric acid_ by diluting 43.65 grams of
sulphuric acid (sp. g. 1.6165) to 1 litre: 100 c.c. will contain 1 gram
of sulphur. An equivalent solution may be made by dissolving 100.62
grams of sodium sulphate crystals (Na_{2}SO_{4}.10H_{2}O), or 86.88
grams of ferrous sulphate (FeSO_{4}.7H_{2}O), in water (oxidising the
latter), and diluting to 1 litre.

The order in which these experiments are given is that in which they
were made in an investigation into the conditions under which the
titration could most accurately be effected.

~Effect of Hydrochloric and Nitric Acids.~--The titrations were
performed in the manner already described, but sodic acetate and acetic
acid were absent. Twenty c.c. of the standard solution of sulphuric acid
were used.

  Hydrochloric acid present     0.0 c.c.  1.0 c.c.  2.0 c.c.  5.0 c.c.
  "Baric chloride" required    20.0  "   20.0  "   19.7  "   12.5  "

  Nitric acid present           0.0 c.c.  1.0 c.c.  2.0 c.c.  5.0 c.c.
  "Baric chloride" required    20.0  "   19.5  "   18.0  "   10.0  "

These show clearly the interference of free mineral acids, although very
dilute hydrochloric acid (1 c.c. in 500 of water) has no effect.

~Effect of Acetic and Citric Acids.~--A similar series of experiments
with these acids gave the following results:--

  Acetic acid present          0.0 c.c.  5.0 c.c. 50.0 c.c. 100.0 c.c.
  "Baric chloride" required   20.0  "   20.0  "   20.0  "    20.0  "

  Citric acid present          0   gram  1 gram    5 grams
  "Baric chloride" required   20.0 c.c. 20.0 c.c. 20.0 c.c.

These acids do not interfere.

~Effect of Sodic Acetate and Acetic Acid.~--In each of these experiments
5 c.c. of acetic acid was present.

  Sodium acetate added         0 gram    1 gram    10 grams   50 grams
  "Baric chloride" required   20.0 c.c. 20.0 c.c.  20.0 c.c.  20.0 c.c.

As sodic acetate and acetic acid did not interfere, it became desirable
to make some experiments on the _finishing point_. The first object
sought for was the smallest amount of the standard baric chloride in 500
c.c. of water, required to give an indication when tested in the manner
already described.

                                                    Baric Chloride
         Conditions of Assay Solution.                required.

  Water only                                           0.05 c.c.
  With 10 grams of sodium acetate and 5 c.c. of
    acetic acid                                        0.05  "
  The same with 5 grams of nitre                       0.10  "
  Like the last, but with 5 grams of salt instead of
    nitre                                              0.10 "

These show that as small an amount of baric chloride solution as is
equal to only 0.000002 gram of sulphur in the 2 c.c. of solution tested
yields a decided precipitate on the addition of 3 drops of sulphuric
acid.

To determine whether the same finishing point is obtained on testing the
filtered portions in the test tubes with baric chloride as is obtained
on testing with sulphuric acid, a titration was made with 20 c.c. of
standard solution of sulphuric acid, together with the usual quantities
of sodic acetate and acetic acid; and two lots of 2 c.c. each were
filtered into two sets of test tubes after each addition of the standard
baric chloride. To one series 3 drops of baric chloride solution were
added, and to the other 3 drops of sulphuric acid. The results were--

                                 With Dilute          With Baric
  "Baric Chloride" added.      Sulphuric Acid.    Chloride Solution.

        19.5   c.c.                Clear                Cloudy
        19.75   "                  Clear                Cloudy
        20.0    "                  Finished             Finished
        20.25   "                  Cloudy               Clear
        20.5    "                  Cloudy               Clear

The two methods of testing give the same result. But this balance is
disturbed in the presence of much nitre, the indications with baric
chloride being disturbed by an opalescence for some c.c. beyond the
finishing point. In solutions containing free hydrochloric or nitric
acid, a precipitate is obtained with either baric chloride or sulphuric
acid.

~Effect of Varying Sulphur.~--In these and the subsequent experiments
the titrations were performed in the presence of 10 grams of sodic
acetate and 10 c.c. of acetic acid in the manner already described.

  Standard sulphuric
     acid used        5.0 c.c.  10.0 c.c.  20.0 c.c.  50.0 c.c. 100.0 c.c.

  "Baric chloride"
    required          5.0  "    10.0  "    20.0  "    50.0  "   100.0  "

~Effect of Varying Temperature.~--With 5 c.c. of standard sulphuric acid
titrated at 15° C., 5 c.c. of baric chloride were required; but with
larger quantities the results were altogether unsatisfactory when
titrated cold.

~Effect of Varying Bulk.~--

  Bulk               100.0 c.c.   200.0 c.c.   500.0 c.c.   1000.0 c.c.
  "Baric chloride"
    required          20.0  "      20.0  "      20.0  "       20.5  "

Considerable variation in bulk has no effect, but 500 c.c. is the most
convenient volume to work with. It is well to occasionally replace the
water boiled off during titration.

~Effect of Foreign Salts.~--In all these experiments 20 c.c. of
"sulphuric acid" were used, and the titration was performed in the
ordinary way.

  _Sodic chloride_ added      0 gram      5 grams    10 grams
  "Baric chloride" required   20.0 c.c.   20.0 c.c.   19.7 c.c.

  _Ammonic chloride_ added    0 gram      5 grams    10 grams
  "Baric chloride" required   20.0 c.c.   20.0 c.c.   19.5 c.c.

  _Calcic chloride_ added     0 gram      1 gram     2 grams    5 grams
  "Baric chloride" required   20.0 c.c.   20.0 c.c.  19.2 c.c.  19.0 c.c.

  _Zinc chloride_ added       0  gram     1 gram     3 grams    5 grams
  "Baric chloride" required   20.0 c.c.   20.0 c.c.  20.0 c.c.  20.0 c.c.

  _Ferrous chloride_ added    0 gram      1 gram     3 grams    5 grams
  "Baric chloride" required   20.0 c.c.   19.7 c.c.  19.5 c.c.  19.0 c.c.

  _Ferric chloride_ added     0 gram      1 gram     3 grams    5 grams
  "Baric chloride" required   20.0 c.c.   20.0 c.c.  20.0 c.c.  20.0 c.c.

  _Copper chloride_ added     0 gram      1 gram     3 grams    5 grams
  "Baric chloride" required   20.0 c.c.   20.0 c.c.  20.0 c.c.  20.0 c.c.

  _Potassic Nitrate_ added    0 gram      1 gram     5 grams   10 grams
  "Baric chloride" required   20.0 c.c.   20.0 c.c.  20.0 c.c.  19.0 c.c.

  _Potassic Nitrite_ added    0 gram      1 gram     5 grams
  "Baric chloride" required   20.0 c.c.   20.0 c.c.  20.0 c.c.

  _Sodic phosphate_ added     0 gram      1 gram
  "Baric chloride" required   20.0 c.c.   22.5 c.c.

  _Sodic arsenate_ added      0 gram     1 gram
  "Baric chloride" required   20.0 c.c.  20.5 c.c.

In the absence of ferric salts, phosphates and arsenates count as
sulphur.

In two series of experiments for determining the effect of varying
amounts of sulphur in the form of ferrous sulphate, we obtained the
following results:--In the first series the assay solution was prepared
in the manner we have described for _Pyrites_; and in the second series,
by fusion with nitre.

  Sulphur added                    0.050 gram     0.100 gram     0.200 gram
  "Baric chloride" required (1)    5.0 c.c.      10.0 c.c.      20.0 c.c.
         "          "       (2)    4.7  "        10.0  "        20.0  "

  Sulphur added                    0.500 gram     1.000 gram
  "Baric chloride" required (1)   50.0 c.c.     100.0 c.c.
         "          "       (2)   50.0  "       100.0  "

More than 5 grams of nitre must not be used in an assay; and, since the
requisite amount of nitre considerably exceeds that sufficient to
oxidise the sulphur, not more than 0.5 gram of unoxidised sulphur should
be present in the portion of the sample weighed up for determination.
When the amount of sulphur present is not known within reasonable
limits, the test portions may be tried with a drop of baric chloride
solution instead of sulphuric acid, so that the diminishing quantity of
precipitate may give warning of an approach to the finishing point.

~Determination of Sulphur in Blende.~--Weigh up 1 gram of dried and
powdered blende, and mix and fuse with 5 grams of nitre in the manner
described. Place the dish and its contents in the titrating-dish,
extract with water, add 10 grams of sodium acetate and 10 c.c. of acetic
acid, remove and wash the platinum-dish, and dilute to 500 c.c.; boil
and titrate. In the example, duplicate determinations required (a)
32.0 c.c., (b) 32.25 c.c., giving an average of 32.1 per cent. of
sulphur.

~Determination of Sulphur in Chalcopyrite~ (Yellow Copper Ore).--Take 1
gram of the finely-powdered sample, and 5 grams of nitre. Sprinkle a
little of the nitre in a small Wedgwood mortar, place the ore on it, and
cover with 2 or 3 grams more of the nitre. Rub up together, and transfer
to a small porcelain dish; clean out the mortar with the rest of the
nitre, and add to the contents of the dish. Cover with a piece of
platinum foil, and heat gently with a Bunsen burner till the nitre melts
and the stuff shows signs of deflagrating; remove the heat, and allow
the action to go on by itself for a minute or so, then heat over the
Bunsen burner for 10 minutes. Cool; transfer the whole to the
titrating-dish; boil with 500 c.c. of water; remove the small dish and
foil; add sodic acetate and acetic acid, and titrate.

For example, 1 gram required 34.5 c.c. of "barium chloride" (standard =
1.005 gram S), which is equivalent to 34.7 per cent. sulphur. The
theoretical percentage is 34.8.

~Determination of Sulphur in Chalcocite~ (Grey Copper Ore).--Proceed as
in the last experiment but, since the action with nitre is more
moderate, no special precautions need be taken on heating. A platinum
dish may be used.

An example which was heated for 30 minutes required 20.5 c.c. of the
barium chloride solution. This is equivalent to 20.6 per cent. of
sulphur. The theoretical yield is 20.2 per cent.

~Determination of Sulphur in Pyrites.~--Take 1 gram of the
finely-powdered sample, cover with 10 c.c. of nitric acid, and, when
action has ceased, evaporate to a small bulk. Add 3 or 4 c.c. of
hydrochloric acid, and again evaporate to a paste. Take up with 1 or 2
c.c. of dilute hydrochloric acid, dilute with water, transfer to a
titrating-dish, add 10 grams of sodic acetate and 5 c.c. of acetic acid,
and dilute with water to 500 c.c. Boil and titrate.

An example with 1 gram of a pure crystallized pyrites required 52.7 c.c.
of the barium chloride solution, which is equivalent to 53.0 per cent.
of sulphur. Theory requires 53.3 per cent. of sulphur.

~Determination of Sulphur in Mispickel.~--Take 1 gram of the powdered
ore and evaporate with 10 c.c. of nitric acid, and take up with 3 or 4
c.c. of hydrochloric acid. If any globules of sulphur remain, again
evaporate with nitric acid. Dilute, and transfer to the titrating-dish.
Add 10 grams of sodic acetate, dilute with water, boil, and titrate. The
mispickel carries (according to theory) exactly sufficient iron to
precipitate the arsenic as ferric arsenate in an acetic acid solution,
so no more iron need be added. The ferric arsenate will separate out as
a yellowish-white flocculent precipitate.

An example required, in duplicate experiment, 18.5 c.c. and 18.7 c.c. of
barium chloride, equivalent to 18.7 per cent. of sulphur. The formula,
FeS_{2}.FeAs_{2}, requires 19.6 percent., but the sulphur generally
varies considerably from this amount.

~Determination of Sulphur in Burnt Ores.~--Take 5 grams of the dried and
powdered ore, and rub up with 4 grams of nitre; transfer to the
platinum-dish; clean out the mortar with another gram of nitre, and add
this as a cover. Heat, and extract with water as before; add the sodium
acetate and acetic acid; and titrate. Burnt ores carry from 2.5 to 5 per
cent. of sulphur. A series of four determinations gave:--

  "Baric Chloride" Required.  Percentage of Sulphur.  Gravimetric Results.
           12.6 c.c.                2.52 %                 2.45 %
           29.9  "                  5.98 "                 5.84 "
           18.1  "                  3.62 "                 3.53 "
           22.0  "                  4.40 "                 4.43 "

For ores carrying less than 1 per cent. of sulphur, take 10 grams for
the assay.

~Determination of Sulphuric Oxide (SO_{3}) in Sulphates.~--When the
sulphur exists in the sample received by the assayer in an oxidised
state as sulphate, it is usual to report it in terms of sulphuric oxide
(SO_{3}). In this case, the metal must also be reported as oxide. For
example, an analysis of copper sulphate would be thus reported:--

  Oxide of copper (CuO)         31.8 %
  Sulphuric oxide (SO_{3})      32.1 "
  Water                         36.1 "
                               -----
                               100.0

The percentage of sulphur multiplied by 2.5 gives the percentage of
sulphuric oxide. Thus a sample of copper sulphate containing 12.85 per
cent. of sulphur will contain 12.85 × 2.5 or 32.12 per cent. of
sulphuric oxide.

In minerals and metallurgical products, it is common to find the sulphur
in both conditions--_i.e._, as sulphate and sulphide. Generally in these
the percentage of sulphur only is wanted; but this will depend entirely
on commercial requirements, and not on the fancy of the assayer.
Soluble sulphates are determined separately by extracting with small
quantities of cold water, so as to avoid the separation of basic
sulphates, or, if the sulphides present are not at the same time
attacked, by dilute hydrochloric acid. Lead sulphate may be extracted by
boiling with ammonic acetate; whilst barium, strontium, and, perhaps,
calcium sulphate, will be mainly found in the residue insoluble in
acids.

Weigh up from 2 to 5 grams of the material according to the amount of
sulphur judged to be present, and dissolve them in the titrating-dish
with 1 c.c. of hydrochloric acid and 50 c.c. of water. Add 10 grams of
sodic acetate, and 10 c.c. of acetic acid; dilute, boil, and titrate. In
the case of ferric salts, half the quantity of acetic acid will be
better, as then the ferric iron will be precipitated, and a colourless
solution will be left, in which the end reaction is more readily
distinguished.

Determined in this way, 5 gram samples of the following salts gave the
results indicated below:--

                      "Barium Chloride"
       Salt.              Required.       Sulphuric Oxide.
  Copper sulphate         64.25 c.c.         32.12 %
  Magnesium sulphate      65.25  "           32.62 "
  Zinc sulphate           56.25  "           28.12 "
  Ferrous sulphate        58.25  "           29.12 "
  Sodium sulphate         51.25  "           25.60 "

~Determination of Sulphuric Oxide in Barytes~ (Heavy spar).--Fuse 2
grams of the powdered mineral with 5 grams of "fusion mixture" for five
minutes; and, when cold, extract with water. Filter, acidulate the
filtrate with an excess of 10 c.c. of acetic acid, dilute, boil, and
titrate. For example, a transparent crystallised sample required 27.0
c.c. of barium chloride, which is equivalent to 13.6 per cent. of
sulphur, or 34.0 per cent. of sulphuric oxide. Theory requires 34.3 per
cent. of the latter. Since both carbonate of soda and potash are liable
to contain sulphates, a blank determination should be made on 5 grams of
the "fusion mixture," and the amount found be deducted from that got in
the assay.


PRACTICAL EXERCISES.

1. The price of sulphur in an ore being 4-1/2d. per unit in the
northern markets, what would be the price of a ton of ore containing 49
per cent. of sulphur? What would be the effect on the price of an error
of 0.25 per cent. in the assay?

2. Pyrites carries 50 per cent. of sulphur, and on calcining yields 70
per cent. of its weight of burnt ore. Supposing the burnt ore carries
3.5 per cent. of sulphur, what proportion of the sulphur will have been
removed in the calcining?

3. How would blende compare with pyrites as a source of sulphur for
sulphuric acid making?

4. How would you determine the percentage of sulphuric oxide in a sample
of gypsum? What is sulphuric oxide, and what relation does it bear to
sulphur?

5. A mineral contains 20.7 per cent. of water, 32.4 per cent. of lime,
and 18.6 per cent. of sulphur. What is its probable composition? What
experiment would you try to determine the accuracy of your conclusion?


SELENIUM

occurs in nature combined with copper, mercury, and lead, in certain
rare minerals. In small quantities it is found in many ores. It is
detected in solution by the red precipitate produced on boiling the acid
solution with sodium sulphite. This reaction is used for its
determination.

~Solution.~--The solution is effected by boiling with nitric acid or
aqua regia, or by fusing with nitre. To separate the selenium, the
solution is evaporated with an excess of hydrochloric acid and a little
sodium or potassium chloride. This destroys any nitric acid that may be
present, and reduces selenic acid (H_{2}SeO_{4}) to selenious
(H_{2}SeO_{3}). The solution is diluted with water, and treated with a
solution of sulphite of soda. It is warmed, and at last boiled. The
selenium separates as a red precipitate, which (on boiling) becomes
denser and black. It is collected on a weighed filter, washed with hot
water, dried at 100° C., and weighed as pure selenium.

Selenium can be precipitated with sulphuretted hydrogen as a sulphide,
which is readily soluble in ammonium sulphide. This sulphide may be
oxidised with hydrochloric acid and chlorate of potash; and the selenium
separated in the manner described.


TELLURIUM.

Tellurium occurs in nature, native, and in combination with gold,
silver, bismuth and lead. It is sometimes met with in assaying gold
ores. It may be detected by the purple colour it imparts to strong
sulphuric acid when dissolved in the cold, and by the black precipitate
of metallic tellurium which its solutions yield on treatment with a
reducing agent. Telluric acid is reduced to tellurous (with evolution of
chlorine) on boiling with hydrochloric acid.

~Solution~ is effected by boiling with aqua regia, or by fusing with
nitre and sodium carbonate.

~Separation.~--Tellurium closely resembles selenium in its reactions. It
is separated and determined in the same way. Like it, it forms a
sulphide soluble in ammonium sulphide. It is distinguished from selenium
by the insolubility, in a solution of cyanide of potassium, of the metal
precipitated by sodium sulphite; whereas selenium dissolves, forming a
soluble potassic seleno-cyanide.[102]

For the determination, solution is effected by fusing with nitre and
sodium carbonate, dissolving out the tellurate of potash with water, and
boiling with hydrochloric acid. Tellurous compounds are formed, with
evolution of chlorine; and the solution, on treating with a reducing
agent (such as sulphurous acid or stannous chloride), yields metallic
tellurium; which is washed, dried at 100° C., and weighed.

FOOTNOTES:

[101] BaCl_{2} + Na_{2}SO_{4} = BaSO_{4} + 2NaCl.

[102] Se + KCy = KCySe.



CHAPTER XVII.

ARSENIC, PHOSPHORUS, NITROGEN.


ARSENIC.

The chief source of the arsenic of commerce is arsenical pyrites, or
mispickel, which contains about 45 per cent. of arsenic (As). Arsenic
also occurs as a constituent of several comparatively rare minerals;
and, as an impurity, it is very widely distributed. White arsenic is an
oxide of arsenic, and is obtained by roasting arsenical ores, and
refining the material (crude arsenic), which condenses in the flues.
Arsenic itself is volatile, and many of its compounds have the same
property. It forms two well-defined series of salts, corresponding to
the oxides: arsenious oxide (As_{2}O_{3}), and arsenic oxide
(As_{2}O_{5}). These combine with bases to form arsenites and arsenates
respectively. Boiling with nitric acid converts the lower into the
higher oxide; and powerful reducing-agents, such as cuprous chloride,
have the opposite effect.

Arsenic may be detected by dissolving the substance in hydrochloric
acid, or in aqua regia (avoiding an excess of nitric acid), and adding a
little of this solution to the contents of a small flask in which
hydrogen is being made by the action of zinc and hydrochloric acid. The
ignited jet of hydrogen assumes a blue colour if arsenic is present, and
a cold porcelain dish held in the flame (fig. 67) becomes coated with a
dark deposit of metallic arsenic. Antimony produces a similar effect,
but is distinguished by the insolubility of its deposit in a cold
solution of bleaching-powder.

[Illustration FIG. 67.]

Arsenites are distinguished by the volatility of the chloride; by
decolorising a solution of permanganate of potassium, and by immediately
giving a yellow precipitate with sulphuretted hydrogen. Arsenates are
distinguished (after converting into soda salts by boiling with
carbonate of soda and neutralising) by giving with nitrate of silver a
red precipitate, and with "magnesia mixture" a white crystalline one.

~Dry Assay.~--There is no dry assay which is trustworthy. The following
method is sometimes used to find the proportion of arsenious oxide in
"crude arsenic":--Weigh up 5 grams of the dried sample, and place them
in a clean dry test-tube about 6 inches long. Tie a small filter-paper
over the mouth of the tube, so as to prevent air-currents. Heat the tube
cautiously so as to sublime off the white arsenic into the upper part of
the tube. Cut off the bottom of the test-tube by wetting whilst hot.
Scrape out the arsenic and weigh it. The weight gives an approximate
idea of the quantity, and the colour of the quality, of the white
arsenic obtainable from the sample. Some workers (sellers) weigh the
residue, and determine the white arsenic by difference. In determining
the percentage of moisture in these samples, the substance is dried on a
water-bath or in a water-oven.


WET METHODS.

~Solution.~--Where, as in crude arsenic, the substance is arsenious
oxide (As_{2}O_{3}) mixed with impurities, the arsenic is best got into
solution by warming with caustic soda, and neutralising the excess with
hydrochloric acid; it will be present as sodium arsenite. Metals and
alloys are acted on by means of nitric acid; or the arsenic may be at
the same time dissolved and separated by distilling with a strongly-acid
solution of ferric chloride, in the way described under _Volumetric
Methods_.

With minerals, mattes, &c., solution is thus effected:--The
finely-powdered substance is mixed (in a large platinum or porcelain
crucible) with from six to ten times its weight of a mixture of equal
parts of carbonate of soda and nitre. The mass is then heated gradually
to fusion, and kept for a few minutes in that state. When cold, it is
extracted with warm water, and filtered from the insoluble residue. The
solution, acidified with nitric acid and boiled, contains the arsenic as
sodium arsenate. With mispickel, and those substances which easily give
off arsenic on heating, the substance is first treated with nitric acid,
evaporated to dryness, and then the residue is treated in the way just
described.

When the arsenic is present as arsenite or arsenide, distillation with
an acid solution of ferric chloride will give the whole of the arsenic
in the distillate free from any metal except, perhaps, tin as stannic
chloride. With arsenates, dissolve the substance in acid and then add an
excess of soda. Pass sulphuretted hydrogen into the solution; warm, and
filter. Acidulate the filtrate, and pass sulphuretted hydrogen. Decant
off the liquid through a filter, and digest the precipitate with ammonic
carbonate; filter, and re-precipitate with hydrochloric acid and
sulphuretted hydrogen. Allow to stand in a warm place, and filter off
the yellow sulphide of arsenic. Wash it into a beaker, clean the
filter-paper (if necessary) with a drop or two of dilute ammonia;
evaporate with 10 c.c. of dilute nitric acid to a small bulk; dilute;
and filter off the globules of sulphur. The filtrate contains the
arsenic as arsenic acid.


GRAVIMETRIC METHOD.

Having got the arsenic into solution as arsenic acid, and in a volume
not much exceeding 50 c.c., add about 20 c.c. of dilute ammonia and 20
c.c. of "magnesia mixture." Stir with a glass rod, and allow to settle
overnight. Filter, and wash with dilute ammonia, avoiding the use of
large quantities of wash water. Dry, transfer the precipitate to a
Berlin crucible, and clean the filter-paper thoroughly. Burn this paper
carefully and completely; and add the ash to the contents of the
crucible, together with 4 or 5 drops of nitric acid. Evaporate with a
Bunsen burner, and slowly ignite, finishing off with the blow-pipe or
muffle. Cool, and weigh. The ignited precipitate is pyrarsenate of
magnesia (Mg_{2}As_{2}O_{7}), and contains 48.4 per cent. of arsenic
(As).

Instead of igniting the precipitate with nitric acid, it may be
collected on a weighed filter-paper, dried at 100° C., and weighed as
ammonic-magnesic arsenate (2AmMgAsO_{4}.H_{2}O), which contains 39.5 per
cent. of arsenic. The results in this case are likely to be a little
higher. The drying is very tedious, and is likely to leave behind more
water than is allowed for in the formula. In a series of determinations
in which the arsenic was weighed in both forms, the results were:--

  Ammonic-magnesic     Arsenic     Magnesium Pyrarsenate   Arsenic
  Arsenate in grams.   in grams.         in grams.         in grams.
        0.0080          0.0032            0.0065            0.0031
        0.0400          0.0158            0.0330            0.0160
        0.0799          0.0316            0.0633            0.0306
        0.1600          0.0632            0.1287            0.0623
        0.4000          0.1580            0.3205            0.1551
        0.7990          0.3156            0.6435            0.3114


VOLUMETRIC METHODS.

There are two methods: one for determining the arsenic in the lower, and
the other in the higher state of oxidation. In the first-mentioned
method this is done by titrating with a standard solution of iodine; and
in the latter with a solution of uranium acetate. Where the arsenic
already exists as arsenious oxide, or where it is most conveniently
separated by distillation as arsenious chloride, the iodine method
should be used; but when the arsenic is separated as ammonic-magnesic
arsenate or as sulphide, the uranium acetate titration should be
adopted.


IODINE PROCESS.

This is based on the fact that sodium arsenite in a solution containing
an excess of bicarbonate of soda is indirectly oxidised by iodine to
sodium arsenate,[103] and that an excess of iodine may be recognised by
the blue colour it strikes with starch. The process is divided into two
parts--(1) the preparation of the solution, and (2) the titration.

~Preparation of the Solution.~--For substances like crude arsenic, in
which the arsenic is present as arsenious oxide, the method is as
follows:--Take a portion which shall contain from 0.25 to 0.5 gram of
the oxide, place in a beaker, and cover with 10 c.c. of sodic hydrate
solution; warm till dissolved, put a small piece of litmus paper in the
solution, and render acid with dilute hydrochloric acid. Add 2 grams of
bicarbonate of soda in solution, filter (if necessary), and dilute to
100 c.c. The solution is now ready for titrating.

[Illustration: FIG. 68.]

Where the arsenic has to be separated as arsenious chloride, the process
is as follows:[104]--Weigh up 1 gram of the finely-powdered ore (metals
should be hammered out into a thin foil or be used as filings), and
place in a 16-ounce flask provided with a well-fitting cork, and
connected with a ~U~-tube, as shown in the drawing (fig. 68). The
~U~-tube should contain 2 or 3 c.c. of water, and is cooled by being
placed in a jar or large beaker of cold water. The water used for
cooling should be renewed for each assay.

Pour on the assay in the flask 50 c.c. of a "ferric chloride mixture,"
made by dissolving 600 grams of calcium chloride and 300 grams of ferric
chloride in 600 c.c. of hydrochloric acid, and making up to 1 litre with
water.

Firmly cork up the apparatus, and boil over a small Bunsen-burner flame
for fifteen or twenty minutes, but avoid evaporating to dryness.
Disconnect the flask, and pour away its contents at once to prevent
breakage of the flask by their solidification. The arsenic will be
condensed in the ~U~-tube, together with the greater part of the
hydrochloric acid; transfer the distillate to a beaker washing out the
tube two or three times with water; add a small piece of litmus paper;
neutralise with ammonia; render faintly _acid_ with dilute hydrochloric
acid; add 2 grams of bicarbonate of soda in solution; and dilute to 250
c.c. The solution is now ready for titrating.

The arsenic comes over in the early part of the distillation, as will be
seen from the following experiment, made on 1 gram of copper
precipitate; in which experiment the distillate was collected in
separate portions at equal intervals, and the arsenic in each portion
determined:--

      Time              Iodine        Equivalent to Arsenic
   Distilling.         Required.       in the Distillate.

  5 minutes           12.0 c.c.         0.0450 gram
  5    "               0.17   "         0.0005  "
  5    "               0.0    "
  5    "               0.0    "
  To dryness           0.0    "

The volume of each distillate was about 5 c.c.

In this operation the metals are converted into chlorides by the action
of ferric chloride, which gives up a part of its chlorine, and becomes
reduced to the ferrous salt. The calcium chloride does not enter into
the chemical reaction, but raises the temperature at which the solution
boils, and is essential for the completion of the distillation.[105] Two
experiments with material containing 3.48 per cent. of arsenic gave--(1)
with ferric chloride alone, 2.74 per cent.; and (2) with the addition of
calcium chloride, 3.48 per cent.

It is always necessary to make a blank determination with 1 gram of
electrotype copper, to find out the amount of arsenic in the ferric
chloride mixture.[106] Unfortunately, a correction is always required.
This amounts to about 0.15 per cent. of arsenic on each assay, even when
the mixture has been purified; and this constitutes the weakness of the
method, since, in some cases, the correction is as much as, or even
greater than, the percentage to be determined.

The acid distillate containing the arsenious chloride may be left for an
hour or so without much fear of oxidation; but it is safer to neutralise
and then to add the bicarbonate of soda, as the following experiments
show. Several portions of a solution, each having a bulk of 100 c.c.,
were exposed for varying lengths of time, and the arsenic in each
determined.

   _____________________________________________________________________
  |               |                           |                         |
  |               |     Acid Solutions.       | Neutralised Solutions.  |
  | Time Exposed. | "Iodine"   Arsenic Found. |"Iodine"  Arsenic Found. |
  |               | Required.                 | Required                |
  |_______________|___________________________|_________________________|
  |               |                           |                         |
  |     --        | 18.2 c.c. = 0.0136 gram   | 18.1 c.c. = 0.0136 gram |
  | 1 hour        | 18.2  "   = 0.0136   "    | 18.2  "   = 0.0136   "  |
  | 2 hours       | 17.7  "   = 0.0133   "    | 18.0  "   = 0.0135   "  |
  | 4   "         | 17.5  "   = 0.0131   "    | 18.4  "   = 0.0138   "  |
  | 5   "         | 17.0  "   = 0.0127   "    | 18.3  "   = 0.0137   "  |
  |_______________|___________________________|_________________________|

~The Titration.~--Make a _standard solution of iodine_ by weighing up in
a beaker 16.933 grams of iodine and 30 grams of potassium iodide in
crystals; add a few c.c. of water, and, when dissolved, dilute to 1
litre: 100 c.c. will equal 0.500 gram of arsenic.

A solution of starch similar to that used in the iodide-copper assay
will be required. Use 2 c.c. for each assay. Variations in the quantity
of starch used do not interfere; but the solution must be freshly
prepared, as after seven or eight days it becomes useless.

To standardise the iodine solution, weigh up 0.3 gram of white arsenic;
dissolve in caustic soda; neutralise; after acidulating, add 2 grams of
bicarbonate of soda and 2 c.c. of the starch solution, and dilute to 200
c.c. with cold water. Fill a burette having a glass stop-cock with the
iodine solution, and run it into the solution of arsenic, rapidly at
first, and then more cautiously, till a final drop produces a blue
colour throughout the solution. Calculate the standard in the usual way.
White arsenic contains 75.76 per cent. of arsenic.

The following experiments show the effect of variation in the conditions
of the titration:--

Make a solution of arsenic by dissolving 6.60 grams of white arsenic in
100 c.c. of sodic hydrate solution; render slightly acid with
hydrochloric acid; add 10 grains of bicarbonate of soda, and dilute to 1
litre: 100 c.c. will contain 0.50 gram of arsenic.

~Effect of Varying Temperature.~--The reaction goes on very quickly in
the cold, and, since there is no occasion for heating, all titrations
should therefore be carried out cold.

~Effect of Varying Bulk.~--In these experiments, 20 c.c. of arsenic
solution were taken, 2 grams of bicarbonate of soda and 2 c.c. of starch
solution added, and water supplied to the required bulk. The results
were:--

  Bulk               50.0 c.c.  100.0 c.c.  250.0 c.c.  500.0 c.c.
  "Iodine" required  20.0  "     20.0  "     20.0  "     20.0  "

Considerable variation in bulk does not interfere.

~Effect of Varying Bicarbonate of Soda.~--This salt must be present in
each titration in considerable excess, to prevent the interference of
free acid. The bicarbonate must be dissolved without heating, as neutral
carbonates should be avoided.

  Bicarbonate added   1 gram     2 grams    5 grams   10 grams
  "Iodine" required  20.1 c.c.  20.0 c.c.  20.1 c.c.  20.0 c.c.

These results show that large variation in the quantity of bicarbonate
has no effect.

~Effect of Free Acid.~--In these experiments, the arsenic taken, the
starch, and the bulk were as before, but no bicarbonate was added. In
one case the solution was rendered acid with 5 c.c. of acetic acid, and
in the other with 5 c.c. of hydrochloric acid; in both cases the
interference was strongly marked, and no satisfactory finishing point
could be obtained. This was much more marked with the hydrochloric acid.

~Effect of Foreign Salts.~--The process for getting the arsenic into
solution will exclude all metals except tin, but the solution will be
charged with sodium or ammonium salts in the process of neutralising, so
that it is only necessary to see if these cause any interference. The
alkaline hydrates, including ammonia, are plainly inadmissible, since no
free iodine can exist in their presence. Monocarbonates similarly
interfere, but to a much less extent; hence the necessity for rendering
the assay distinctly acid before adding the bicarbonate of soda.

With 20 c.c. of arsenic solution; and with bulk, soda, and starch as
before, the results obtained were:--

                                               "Iodine" required.
  With 20 grams of ammonic chloride                 20.0 c.c.
   "   20 grams of sodium chloride                  20.0  "
   "   20 grams of sodium acetate                   20.0  "
   "    0.050 gram of tin, as stannic chloride      19.6  "
  Without any addition                              20.0  "

The interference of the stannic salt is probably mechanical, the
precipitate carrying down some arsenious acid.

~Effect of Varying Arsenic.~--With bulk, starch, and soda as before, but
with varying arsenic, the results were:--

  Arsenic added      1.0 c.c. 10.0 c.c. 20.0 c.c. 50.0 c.c. 100.0 c.c.
  "Iodine" required  1.1  "    9.9  "   20.0  "   50.0  "   100.0  "

~Determination of Arsenic in Metallic Copper.~--Put 1 gram of the copper
filings, freed from particles of the file with a magnet, into a
16-oz.-flask; and distil with the ferric chloride mixture, as above
described. Neutralise the distillate; acidify; add bicarbonate of soda
and starch; dilute; and titrate with the standard solution of
iodine.[107] Make a blank determination with 1 gram of electrotype
copper, proceeding exactly as with the assay; and deduct the amount of
arsenic found in this experiment from that previously obtained.

Working in this way on a copper containing 0.38 per cent. of arsenic and
0.80 per cent. of antimony, 0.38 per cent. of arsenic was found.

~Determination of White Arsenic in Crude Arsenic.~--Weigh out 1 gram of
the dried and powdered substance (or 0.5 gram if rich), and digest with
10 c.c. of a 10 per cent. solution of soda; dilute to about 50 c.c., and
filter. Render faintly acid with hydrochloric acid, and filter (if
necessary); add 2 or 3 grams of bicarbonate of soda in solution, then 5
c.c. of starch, and titrate the cold solution with the standard solution
of iodine.

The following is an example:--

  1 gram of crude arsenic required 53.7 c.c. "Iodine;"
    100 c.c. "Iodine" = 0.6000 gram white arsenic;
      100 : 53.7 :: 0.6 : 0.3222, or 32.2 per cent.

With the test-tube method of dry assaying, this same sample gave results
varying from 33 to 35 per cent. of white arsenic, which (judging from
its appearance) was impure.


URANIC ACETATE PROCESS.

This may be looked upon as an alternative to the gravimetric method. It
is applicable in all cases where the arsenic exists in solution as
arsenic acid or as arsenate of soda. The process may be considered in
two parts: (1) the preparation of the solution, and (2) the titration.

~Preparation of the Solution.~--If the arsenic has been separated as
sulphide, it is sufficient to attack it with 10 or 15 c.c. of nitric
acid, and to heat gently till dissolved, avoiding too high a temperature
at first. Afterwards continue the heat till the separated sulphur runs
into globules, and the bulk of the acid has been reduced to 3 or 4 c.c.
Dilute with 20 or 30 c.c. of water; put in a piece of litmus paper; and
add dilute ammonia until just alkaline. Then add 5 c.c. of the sodium
acetate and acetic acid solution (which should make the solution
distinctly acid); dilute to 150 c.c., and heat to boiling. The solution
is ready for titrating.

When the arsenic exists in a nitric acid solution mixed with much
copper, it is separated in the way described under _Examination of
Commercial Copper_ (Arsenic and Phosphorus), pages 208, 209.

If the arsenic has been separated as ammonium-magnesium arsenate, and
phosphates are known to be absent; dissolve the precipitate (after
filtering, but without washing) in dilute hydrochloric acid. Add dilute
ammonia till a slight precipitate is formed, and then 5 c.c. of the
sodium acetate and acetic acid solution; dilute to 150 c.c., and heat to
boiling. Titrate.

If phosphates are present (which will always be the case if they were
present in the original substance, and no separation with sulphuretted
hydrogen has been made), the phosphorus will count in the subsequent
titration as arsenic (one part of phosphorus counting as 2.4 parts of
arsenic). It will be necessary to dissolve the mixed arsenate and
phosphate of magnesia in hydrochloric acid. Add about four or five times
as much iron (as ferric chloride) as the combined phosphorus and arsenic
present will unite with, and separate by the "basic acetate" process as
described under PHOSPHORUS in the _Examination of Commercial Copper_,
page 209. Obviously, when phosphates are present, it is easier to
separate the arsenic as sulphide than to precipitate it with the
"magnesia mixture."

~The Titration.~--The _standard solution of uranium acetate_ is made by
dissolving 34.1 grams of the salt (with the help of 25 c.c. of acetic
acid) in water; and diluting to 1 litre. The water and acid are added a
little at a time, and warmed till solution is effected; then cooled, and
diluted to the required volume: 100 c.c. will equal 0.50 gram of
arsenic.

The _sodic acetate and acetic acid solution_ is made by dissolving 100
grams of sodic acetate in 500 c.c. of acetic acid, and diluting with
water to 1 litre. Five c.c. are used for each assay.

The solution of potassic ferrocyanide used as _indicator_ is made by
dissolving 10 grams of the salt in 100 c.c. of water.

To standardise the solution of uranium acetate, weigh up a quantity of
white arsenic (As_{2}O_{3}) which shall be about equivalent to the
arsenic contained in the assay (0.1 or 0.2 gram); transfer to a flask,
and dissolve in 10 c.c. of nitric acid with the aid of heat. Evaporate
to a small bulk (taking care to avoid the presence of hydrochloric
acid); dilute with water; add a small piece of litmus paper; render
faintly alkaline with ammonia; then add 5 c.c. of the sodic acetate
mixture; dilute to 150 c.c.; and heat to boiling.

Fill an ordinary burette with the uranium acetate solution, and run into
the assay a quantity known to be insufficient. Again heat for a minute
or two. Arrange a series of drops of the solution of ferrocyanide of
potassium on a porcelain slab, and, with the help of a glass rod, bring
a drop of the assay solution in contact with one of these. If no colour
is produced, run in the uranium acetate, 1 c.c. at a time, testing after
each addition, till a brown colour is developed. It is best to overdo
the assay, and to count back. It is not necessary to filter off a
portion of the assay before testing with the "ferrocyanide," since the
precipitate (uranic arsenate) has no effect.

The following experiments show the effect of variation in the conditions
of titration. Make a solution of arsenic acid by dissolving 4.95 grams
of arsenious acid (As_{2}O_{3}) in a covered beaker with 35 c.c. of
nitric acid; evaporate down to 7 or 8 c.c.; and dilute with water to 1
litre: 100 c.c. will contain 0.375 gram of arsenic. Use 20 c.c. for each
experiment.

~Effect of Varying Temperature.~--It is generally recommended to titrate
the boiling solution, since it is possible that the precipitation is
only complete on boiling. Low results are obtained in a cold solution,
the apparent excess of uranium acetate striking a colour at once; on
boiling, however, it ceases to do so; consequently, the solution should
always be boiled directly before testing.

In four experiments made in the way described, but with 20 c.c. of a
solution of arsenic acid stronger than that given (100 c.c. = 0.5 gram
As), the results at varying temperatures were:--

  Temperature            15° C.       30° C.       70° C.       100° C.
  "Uranium" required    18.0 c.c.    18.5 c.c.    18.5 c.c.    18.7 c.c.

~Effect of Varying Bulk.~--These experiments were like those last
mentioned, but were titrated boiling, and the volume was varied:--

  Bulk                  50.0 c.c.  100.0 c.c.   200.0 c.c.   300.0 c.c.
  "Uranium" required    14.0  "     14.0  "      14.5  "      15.0  "

Considerable variations in bulk are to be avoided.

~Effect of Varying Sodium Acetate.~--These experiments were carried out
like those last noticed, but the bulk was 150 c.c., and varying amounts
of sodic acetate were added in excess of the quantity used in the
experiments previously described:--

  Sodic acetate added    0 gram       1 gram     10 grams     20 grams
  "Uranium" required    14.5 c.c.    14.5 c.c.   16.0 c.c.    18.0 c.c.

It is evidently important that the quantity of this salt present in each
titration be measured out, so as to avoid variation.

~Effect of Varying the Sodium Acetate and Acetic Acid Solution.~--Acetic
Acid also affects the results, but in the opposite direction, by
preventing the precipitation of uranium arsenate. With varying volumes
of the solution now under notice, the results were:--

  Solution added         0.0 c.c.     5.0 c.c.   10.0 c.c.    15.0 c.c.
  "Uranium" required    14.5   "     14.5   "    14.5   "     14.0  "
  Solution added        20.0  "      30.0   "    40.0   "     50.0  "
  "Uranium" required    13.2  "      10.0   "     6.0   "      2.0  "

These show that the quantity ordered (5 c.c.) must be adhered to.

~Effect of Foreign Salts.~--In these experiments, 10 grams of the salt
(the effect of which it was desired to determine) were added to a
solution in other respects resembling those previously used:--

  Salt added           {Ammonic    Ammonic     Ammonic     Magnesium
                       {sulphate   nitrate     chloride    sulphate
  "Uranium" required    15.5 c.c.  15.5 c.c.   15.3 c.c.   15.3 c.c.

Without any addition, 15.0 c.c. were required; and in another
experiment, in which 30 grams of ammonic salts were present, 15.6 c.c.
of uranium solution were required. Such variations in the amount of
ammonic salts as occur in ordinary working are unimportant.

Phosphates, of course, interfere. In fact, the uranium acetate solution
can be standardised by titrating with a known weight of phosphate, and
calculating its equivalent of arsenic. Thus, in an experiment with 0.6
gram of hydric sodic phosphate (Na_{2}HPO_{4}.12H_{2}O), equivalent to
0.05195 gram of phosphorus, or 0.1256 gram of arsenic, 23.25 c.c. of a
solution of uranium acetate were required. The same solution
standardised with white arsenic gave a standard of which 100 c.c. =
0.5333 gram arsenic. On this standard the 0.6 gram of sodic phosphate
should have required 23.5 c.c.

Experiments in which 0.1 gram of bismuth and 0.1 gram of antimony were
present with 0.1 gram of arsenic, showed no interference on the
titration. Ferric or aluminic salts would remove their equivalent of
arsenic, and, consequently, must be removed before titrating.

~Effect of Varying Arsenic.~--Varying amounts of metallic arsenic were
weighed up and dissolved in nitric acid, &c., and titrated:--

  Arsenic taken   0.010 gram   0.050 gram   0.100 gram   0.200 gram
  Arsenic found   0.010  "     0.050  "     0.100  "     0.197  "

These experiments show that the method yields good results within these
limits.

~Determination of Arsenic in Mispickel.~--Weigh up 1 gram of the dried
and powdered ore, and evaporate to near dryness with 20 c.c. of dilute
nitric acid. Make up to 100 c.c. with water, and pass sulphuretted
hydrogen to reduce the ferric iron to the ferrous state, then add 20
c.c. of dilute ammonia, and again pass sulphuretted hydrogen. Warm,
filter, and evaporate the filtrate to drive off the excess of ammonia;
then add 10 c.c. of nitric acid, and boil down till the sulphide of
arsenic at first precipitated is dissolved; neutralise; add 5 c.c. of
sodium acetate and acetic acid solution; transfer to a pint flask, boil,
and titrate.

For example, an impure sample of ore required, in duplicate assay of
half a gram each, when treated in the above-mentioned way, 39.6 and 39.5
c.c. of the uranium acetate solution (100 c.c. = 0.537 gram of arsenic),
equivalent to 0.2114 gram of arsenic, or 42.3 per cent.

An alternative method is as follows. Powder the ore very finely and
weigh up .5 gram. Place in a 2-3/4 inch berlin dish and add strong
nitric acid, one drop at a time until the action ceases; with care there
need be no very violent reaction. Dry over a water bath. Cover with 2
grams of nitre and over this spread 5 grams of a mixture of equal parts
of nitre and carbonate of soda. Fuse in a muffle or over a large gentle
blow-pipe flame for 4 or 5 minutes. This will spoil the dish. Allow to
cool and boil out in a larger dish with 100 c.c. of water. Filter and
wash into an 8 oz. flask. Acidify the liquor with nitric and boil down
to about 100 c.c. The acid should not be in too large excess, but an
excess is needed to destroy nitrites. Neutralise with soda or ammonia.
Add 5 c.c. of the mixture of sodium acetate and acetic acid. Titrate
with uranium acetate.

~Determination of Arsenic (As) in Crude Arsenic.~--The method given
under the iodine titration simply determines that portion of the arsenic
which is present in the substance as arsenious oxide or white arsenic.
The following method will give the total arsenic in the sample. It would
be incorrect to report this as so much per cent. of arsenious oxide,
although it may be reported as so much per cent. of arsenic equivalent
to so much per cent. of white arsenic, thus:--

  Arsenic                          30.0 per cent.
  Equivalent to white arsenic      39.6    "

The equivalent of white arsenic is calculated by multiplying the
percentage of arsenic by 1.32. The method of determining the percentage
of arsenic is as follows:---Boil 1 gram of the sample with 10 c.c. of
nitric acid. When the bulk of the solution has been reduced to one-half,
and red fumes are no longer evolved, dilute with a little water, and
filter into a flask. Neutralise the filtrate, add 5 c.c. of sodic
acetate solution, boil and filter. The precipitate (ferric arsenate) is
transferred to a small beaker, treated with 5 c.c. of dilute ammonia,
and sulphuretted hydrogen passed through it. The iron sulphide is
filtered off, and the filtrate evaporated with an excess of nitric acid.
When the solution is clear, it is neutralised, and 1 or 2 c.c. of sodic
acetate solution having been added, is then mixed with the first
filtrate. The solution is boiled and titrated.

A sample treated in this way required 49.2 c.c. of the uranium acetate
solution (100 c.c. = 0.537 gram of arsenic), equivalent to 26.4 per
cent.

~Determination of Arsenic in Brimstone.~--Take 10 grams of the
substance, and powder in a mortar; rub up with 10 c.c. of dilute ammonia
and a little water; rinse into a pint flask; pass a current of
sulphuretted hydrogen; and warm on a hot plate for a few minutes.
Filter, acidulate the filtrate with sulphuric acid; filter off the
precipitate; attack it with 10 c.c. of nitric acid; and proceed as in
the other determinations.


PRACTICAL EXERCISES.

1. Mispickel contains 45.0 per cent. of arsenic, to how much white
arsenic will this be equivalent?

2. How would you make a standard solution of iodine so that 100 c.c.
shall be equivalent to 1 gram of white arsenic?

3. What weight of arsenic is contained in 1 gram of pyrarsenate of
magnesia, and what weight of ammonic-magnesic arsenate would it be
equivalent to?

4. The residue, after heating 10 grams of crude arsenic, weighed 0.62
gram. What information does this give as to the composition of the
substance? If another 10 grams of the substance, heated on a water-bath,
lost 0.43 gram, what conclusions would you draw, and how would you
report your results?

5. If a sample of copper contained 0.5 per cent. of arsenic, and 1 gram
of it were taken for an assay, how much standard uranium acetate
solution would be required in the titration?


PHOSPHORUS AND PHOSPHATES.

Phosphorus rarely occurs among minerals except in its highest oxidized
state, phosphoric oxide (P_{2}O_{5}), in which it occurs abundantly as
"rock phosphate," a variety of apatite which is mainly phosphate of
lime. Phosphates of most of the metallic oxides are found. Phosphoric
oxide in small quantities is widely diffused, and is a constituent of
most rocks. Its presence in varying amounts in iron ores is a matter of
importance, since it affects the quality of the iron obtainable from
them.

Phosphorus occurs in alloys in the unoxidized state. It is directly
combined with the metal, forming a phosphide. In this manner it occurs
in meteoric iron. The alloy phosphor-bronze is made up of copper, tin,
zinc, and phosphorus.

Phosphates are mined in large quantities for the use of manure
manufacturers, and for making phosphorus.

Phosphorus and arsenic closely resemble each other in their chemical
properties, more especially those which the assayer makes use of for
their determination. Phosphorus forms several series of salts; but the
phosphates are the only ones which need be considered. Pyrophosphate of
magnesia, which is the form in which phosphoric oxide is generally
weighed, differs from the ordinary phosphate in the proportion of base
to acid. Metaphosphates differ in the same way. If these are present, it
must be remembered they act differently with some reagents from the
ordinary phosphates, which are called orthophosphates. They are,
however, all convertible into orthophosphates by some means which will
remove their base, such as fusion with alkaline carbonates, boiling with
strong acids, &c.[108]

Phosphides are converted into phosphates by the action of nitric acid or
other oxidizing agents. Dilute acids, when they act on the substance,
evolve phosphuretted hydrogen (PH_{3}). The student should be on his
guard against losing phosphorus in this manner.

There is no dry assay for phosphorus. All assays for it are made either
gravimetrically or volumetrically.

The separation of phosphoric oxide is made as follows:--The ore or metal
is dissolved in acid and evaporated, to render the silica insoluble. It
is taken up with hydrochloric acid, diluted with water, and treated with
sulphuretted hydrogen. The filtrate is boiled, to get rid of the excess
of gas, and treated with nitric acid, to peroxidize the iron present. If
the iron is not present in more than sufficient quantity to form ferric
phosphate with all the phosphorus present, some ferric chloride is
added. The iron is then separated as basic acetate. The precipitate will
contain the phosphorus, together with any arsenic acid not reduced by
the sulphuretted hydrogen. The precipitate should have a decided brown
colour. The precipitate is washed, transferred to a flask, and treated
first with ammonia, and then with a current of sulphuretted hydrogen.
The filtrate from this (acidulated with hydrochloric acid, and, if
necessary, filtered) contains the phosphorus as phosphoric acid. This
method is not applicable in the presence of alumina, chromium, titanium,
or tin, if the solution is effected with nitric acid. The precipitate
obtained by the action of nitric acid on tin retains any phosphoric or
arsenic oxide that may be present.

A method of separation more generally applicable and more convenient to
work is based on the precipitation of a yellow phospho-molybdate of
ammonia,[109] by the action of an excess of ammonic molybdate upon a
solution of a phosphate in nitric acid. Dissolve the substance by
treatment with acid, and evaporate to dryness. Take up with 10 c.c. of
nitric acid, and add 20 grams of ammonic nitrate, together with a little
water. Next put in the solution of ammonium molybdate solution in the
proportion of about 50 c.c. for each 0.1 gram of phosphoric oxide judged
to be present. Warm to about 80° C., and allow to stand for an hour.
Filter, and wash with a 10 per cent. solution of ammonic nitrate. It is
not necessary that the whole of the precipitate be placed on the filter;
but the beaker must be completely cleaned. Dissolve the precipitate off
the filter with dilute ammonia, and run the solution into the original
beaker. Run in from a burette, slowly and with stirring, "magnesia
mixture," using about 15 c.c. for each 0.1 gram of phosphoric oxide.
Allow to stand for one hour. The white crystalline precipitate contains
the phosphorus as ammonium-magnesium phosphate.

Phosphate of lead is decomposed by sulphuric acid; the lead is converted
into the insoluble lead sulphate, and the phosphoric acid is dissolved.
Phosphate of copper and phosphate of iron may be treated with
sulphuretted hydrogen; the former in an acid, and the latter in an
alkaline, solution. Phosphate of alumina is generally weighed without
separation of the alumina, since this requires a fusion. In all cases
the aim is to get the phosphoric oxide either free, or combined with
some metal whose phosphate is soluble in ammonia.

Joulie's method of separation is as follows:--One to ten grams of the
sample are treated with hydrochloric acid, and evaporated to dryness
with the addition (if any pyrites is present) of a little nitric acid.
The residue is taken up with hydrochloric acid, cooled, transferred to a
graduated flask, and diluted to the mark. It is then shaken up, filtered
through a dry filter, and a measured portion (containing about 0.05 gram
of phosphoric acid) transferred to a small beaker. Ten c.c. of a
citric-acid solution of magnesia[110] is added, and then an excess of
ammonia. If an immediate precipitate is formed, a fresh portion must be
measured out and treated with 20 c.c. of the citrate of magnesia
solution and with ammonia as before. The beaker is put aside for from
two to twelve hours. The precipitate is then filtered off and washed
with weak ammonia; it contains the phosphorus as ammonium-magnesium
phosphate.


GRAVIMETRIC DETERMINATION.

If the phosphate is not already in the form of ammonic-magnesic
phosphate, it is converted into this by the addition to its solution of
an excess of ammonia and "magnesia mixture." In order to get the
precipitate pure, the "magnesia mixture" is run in gradually (by drops)
from a burette, with constant stirring. A white crystalline precipitate
at once falls, if much phosphorus is present; but, if there is only a
small quantity, it may be an hour or two before it shows itself. The
solution is best allowed to rest for twelve or fifteen hours (overnight)
before filtering. The presence of tartaric acid should be avoided; and
the appearance of the precipitate should be crystalline. The solution is
decanted through a filter, and the precipitate washed with dilute
ammonia, using as little as may be necessary. The precipitate is dried,
transferred to a weighed Berlin or platinum crucible; the filter-paper
is carefully burnt, and its ash added to the precipitate, which is then
ignited, at first gently over a Bunsen burner, and then more strongly
over the blowpipe or in the muffle. The residue is a white mass of
magnesium pyrophosphate containing 27.92 per cent. of phosphorus, or
63.96 per cent. of phosphoric oxide.


VOLUMETRIC METHOD.

Instead of separating and weighing this compound, the phosphoric oxide
in it can be determined by titration. In many cases the ore may be
dissolved and immediately titrated without previous separation. It is
better, however, to carry the separation so far as to get phosphoric
acid, an alkaline phosphate, or the magnesia precipitate. It may then be
prepared for titration in the following way:--The precipitate in the
last case (without much washing) is dissolved in a little hydrochloric
acid, and the solution in any case rendered fairly acid. Dilute ammonia
is added till it is just alkaline, and then 5 c.c. of the sodic acetate
and acetic acid mixture (as described under the Arsenic Assay). This
should yield a clear distinctly-acid solution. It is diluted to 100 or
150 c.c., heated to boiling, and titrated with the uranium acetate
solution, using that of potassic ferrocyanide as indicator.

The _standard solution_ required is made by dissolving 35 grams of
uranium acetate in water with the aid of 25 c.c. of acetic acid, and
diluting to 1 litre.

An _equivalent solution of phosphoric oxide_ is made by dissolving 25.21
grams of crystallised hydric disodic phosphate (HNa_{2}PO_{4}.12H_{2}O)
in water, and making up to 1 litre. 100 c.c. will contain 0.5 gram of
phosphoric oxide (P_{2}O_{5}), or 0.2183 gram of phosphorus. In making
this solution, transparent crystals only must be used. The uranium
acetate solution is only approximately equivalent to this, so that its
exact standard must be determined.

_Sodic Acetate and Acetic Acid Solution._--It is the same as that
described under _Arsenic_.[111] Use 5 c.c. for each assay.

The following experiments show the effect of variation in the conditions
of the titration:--

~Effect of Varying Temperature.~--The solution should be titrated while
boiling. This is especially necessary for the last few c.c. in order to
get a decided and fixed finishing point.

  Temperature            15° C.       30° C.       70° C.       100° C.
  "Uranium" required    18.0 c.c.   19.2 c.c.   19.0 c.c.   18.9 c.c.

~Effect of Varying Bulk.~--

  Bulk                 50.0 c.c.  100.0 c.c.  200.0 c.c.  300.0 c.c.
  "Uranium" required   18.8  "     18.9  "     19.0  "     19.3  "

Variation in bulk affects the results; therefore, a constant bulk should
be adhered to.

~Effect of Varying Sodium Acetate and Acetic Acid Solution.~--

  Sodium acetate
    and acetic
    acid solution      0.0 c.c.  1.0 c.c.  5.0 c.c.  10.0 c.c. 20.0 c.c.
  "Uranium" required  18.9  "   18.9  "   19.0  "    18.8  "   17.5  "

As in the titration with arsenates, an excess is dangerous to the assay;
a definite quantity (5 c.c.) should, therefore, be used.

~Effect of Foreign Salts.~--Besides the sodium acetate, &c., added, the
only salts likely to be present are those of ammonia and magnesia. In
three experiments, in one of which no foreign salts were introduced,
while in the other two 5 grams of ammonic chloride and of magnesium
sulphate respectively were added, there were required:--

  With ammonic chloride       18.8 c.c. "Uranium" solution
  With magnesium sulphate     19.0  "          "
  Without foreign salts       18.9  "          "

~Effect of Varying Phosphate.~--

  "Phosphate" solution added  10.0 c.c.  20.0 c.c.  50.0 c.c.  100.0 c.c.
  "Uranium" required           9.8  "    18.9  "    47.6  "     94.5  "

The quantity of phosphoric oxide in the assay solution for the
conditions of titration should not be much less than 0.05 gram. For
smaller quantities the uranium solution should be diluted to half its
strength, and the assay solution concentrated by reducing its bulk to 50
c.c. and using 2.5 c.c. of the sodium acetate and acetic acid solution.

~Determination of Phosphoric Oxide in Apatite.~--Weigh up 0.5 gram of
the dried and powdered sample, and dissolve it in 5 c.c. of hydrochloric
acid. Evaporate to a paste, add 5 c.c. of the sodic acetate and acetic
acid solution, dilute to 100 c.c. with water, boil, and titrate with
uranium acetate solution.

In an example, 0.5 gram of apatite required 37.4 c.c. of uranium acetate
solution (standard equal to 0.5291 gram of phosphoric oxide). The sample
therefore contained 0.1979 gram of P_{2}O_{5}, equal to 39.58 per cent.

~Determination of Phosphoric Oxide in an Iron Ore.~--Take 10 grams, boil
with 50 c.c. of hydrochloric acid, and evaporate to a paste; take up
with 10 c.c. of dilute hydrochloric acid, and dilute with water to 400
c.c. Pass sulphuretted hydrogen for nearly a quarter of an hour; warm,
and filter. Boil off the excess of gas; cool, add ammonia till nearly
neutral, and then a few drops of ferric chloride solution, and 4 or 5
grams of sodium acetate, with a drop or two of acetic acid. Boil and
filter. Dissolve the precipitate in hot dilute hydrochloric acid, and
add citro-magnesia mixture and ammonia; allow to stand overnight;
filter, ignite, and weigh.

In an example, 10 grams of ore gave 28.5 milligrams of magnesic
pyrophosphate, which is equivalent to 0.18 per cent. of phosphoric
oxide.

~Determination of Phosphorus in Iron.~--Take from 2 to 10 grams
(according to the amount of phosphorus present), and dissolve in aqua
regia, keeping the nitric acid in excess; evaporate to dryness and take
up with hydrochloric acid, boil, dilute, and filter. Add 10 c.c. of
nitric acid, nearly neutralise with ammonia, render acid with 3 or 4
c.c. of nitric acid, and add 10 or 20 c.c. of ammonic molybdate
solution. Heat for some time, allow to settle, filter, and wash the
precipitate with a solution of ammonic nitrate. Dissolve the precipitate
in dilute ammonia, nearly neutralise with dilute hydrochloric acid, and
add first "magnesia mixture," and then ammonia; allow to stand
overnight; filter, wash with dilute ammonia, dry, ignite, and weigh as
magnesic pyrophosphate. Calculate to phosphorus.


PRACTICAL EXERCISES.

1. Ten grams of an iron yielded 12 milligrams of pyrophosphate of
magnesia. What percentage of phosphorus did the metal contain?

2. Ten grams of an iron ore gave 12 milligrams of pyrophosphate. What
percentage of phosphoric oxide did it contain?

3. What weight of apatite 3Ca_{3}(PO_{4})_{2}.CaClF would require 50
c.c. of standard uranium solution (100 c.c. equal to 0.5 gram of
P_{2}O_{5})?

4. You have reason to believe that a precipitate which has been weighed
as magnetic pyrophosphate contains some arsenate. How would you
determine the amount of phosphate really present?

5. Twenty c.c. of a solution of sodic phosphate containing 0.100 gram of
P_{2}O_{5} was found to require a solution containing 0.700 gram of
hydrated uranium acetate in a titration. The precipitate contains 80.09
per cent. uranium oxide and 19.91 per cent. of phosphoric oxide. What
percentage of uranium oxide was contained in the uranic acetate?


NITROGEN AND NITRATES.

Nitrogen occurs in nature in the free state, and forms about four-fifths
of the atmosphere. In combination, as nitrate, it is found in nitre
(KNO_{3}), and Chili saltpetre (NaNO_{3}), minerals which have a
commercial importance. The latter occurs in beds, and is extensively
worked for use as a manure and in the preparation of nitric acid.

Nitrogen is mainly characterised by negative properties, although many
of its compounds are very energetic bodies. It is a gas, present
everywhere, but so inactive that the assayer can always afford to ignore
its presence, and, except in testing furnace gases, &c., he is never
called on to determine its quantity.

The nitrates are an important class of salts, and may be looked on as
compounds of the bases with nitric pentoxide (N_{2}O_{5}). They are,
with the exception of a few basic compounds, soluble in water, and are
remarkable for the ease with which they give up their oxygen. The
alkaline nitrates fuse readily, and lose oxygen with effervescence
forming nitrites; while at a higher temperature they yield more oxygen
and lose their nitrogen, either as a lower oxide or as nitrogen. The
nitrates of the metals, on heating, leave the oxide of the metal. It is
as yielders of oxygen that nitrates are so largely used in the
manufacture of explosives. Gunpowder contains from 65 to 75 per cent. of
potassium nitrate (nitre).

Nitrates are best detected and determined by their yielding nitric oxide
when treated with sulphuric acid and a suitable reducing agent, such as
ferrous sulphate, mercury, or copper. Nitric oxide is a colourless gas
very slightly soluble in water. It combines at once with oxygen, on
mixing with the air, to form brown "nitrous fumes," and dissolves in a
solution of ferrous sulphate, producing a characteristic blackish-brown
colour. It is this colour which affords the best and most easily-applied
test for nitrates. The substance suspected to contain nitrates is
dissolved in about 1 c.c. of water, and treated with an equal volume of
strong sulphuric acid. After cooling, a solution of ferrous sulphate is
poured on its surface, so as to form a layer resting on it. On standing,
a brown or black ring is developed where the liquids join, if any
nitrate or nitrite is present. Nitrites are distinguished from nitrates
by effervescing and yielding brown fumes when treated with a little
dilute sulphuric acid.

The separation of nitrates is in many cases difficult. Generally, on
treating the substance with water, the nitrate will be in the solution,
and is filtered off from any insoluble matter. In the exceptional cases
it is got into solution by treating with a boiling solution of sodium
carbonate; the nitrate will contain it as an alkaline nitrate.

Since, however, in their determination, nitrates are never separated and
weighed as such, the difficulty of separating them has little
importance. Usually, the determination can be made on the original
aqueous solution, and it is never necessary to do more than remove any
special substance which has a bad effect; and this is easily done by the
usual reagents.


GRAVIMETRIC DETERMINATION.

It follows from what has been said that there is no direct gravimetric
determination. The percentage of nitrogen pentoxide (N_{2}O_{5}) in a
comparatively pure nitrate is sometimes determined indirectly in the
following way:--Place in a platinum-crucible 4 or 5 grams of powdered
and cleaned quartz. Ignite, cool in a desiccator, and weigh with the
cover. Mix 1 gram of the dried and powdered salt with the quartz in the
crucible by stirring with a stout platinum-wire. Cover the crucible, and
heat in a Bunsen-burner flame at scarcely visible redness for
half-an-hour. Cool and weigh. The loss in weight gives the amount of
nitrogen pentoxide. Sulphates and chlorides in moderate quantity do not
interfere. The following is an example of the process:--

  Crucible and sand        26.6485 grams
  Nitre taken               1.0000   "
                           -------
                           27.6485   "
  Weight after ignition    27.1160   "
                           -------
  Loss on ignition          0.5325   "

This is equal to 53.25 per cent. of nitrogen pentoxide.


VOLUMETRIC DETERMINATION.

This is based on the oxidising action of nitric acid, or of nitrates in
acid solutions on ferrous salts. The pentoxide (N_{2}O_{5}) of the
nitrate is reduced to nitric oxide (NO), so that 336 parts of iron
peroxidised represent 108 parts of nitric pentoxide as oxidising
agent.[112] The quantity of iron peroxidised is determined by taking a
known quantity of ferrous salt, oxidizing with a weighed sample of
nitrate, and then determining the residual ferrous iron by titration
with bichromate or permanganate of potassium solution. The difference
between the ferrous iron taken and that found, gives the amount oxidized
by the nitrate. The speed with which nitric oxide takes up oxygen from
the air, and thus becomes capable of oxidising more iron, renders some
precautions necessary; ferrous chloride should, therefore, be used,
since it is easier to expel nitric oxide (by boiling) from solutions of
a chloride than it is from those of a sulphate. The process is as
follows:--Dissolve 2 grams of thin soft iron wire in 50 c.c. of
hydrochloric acid in a flask provided with an arrangement for
maintaining an atmosphere of carbon dioxide. When the iron has
dissolved, allow the solution to cool, and add 0.5 gram of the nitrate.
Heat gently for a few minutes, and then boil until the nitric oxide is
expelled. An atmosphere of carbon dioxide must be kept up. Dilute with
water, and titrate the residual iron with standard solution of
bichromate of potassium. The standard "bichromate" is made by dissolving
17.5 grams of the salt (K_{2}Cr_{2}O_{7}) in water, and diluting to 1
litre: 100 c.c. equal 2 grams of iron. Deduct the weight of iron found
from the 2 grams originally taken, and multiply by 0.3214. This gives
the weight of the pentoxide in the sample. In an example, 0.5 gram of
nitre was taken, and 59.4 c.c. of the "bichromate" solution were
required. The 59.4 c.c. thus used are equivalent to 1.198 gram of iron.
This leaves 0.822 gram as the quantity oxidised by the nitre, which,
multiplied by 0.3214, gives 0.2642 gram for the nitrogen pentoxide, or
52.8 per cent.


GASOMETRIC METHOD.

This is based upon the measurement of the nitric oxide evolved on
shaking up a weighed quantity of the nitrate with sulphuric acid over
mercury in a nitrometer. Each c.c. of nitric oxide obtained, when
reduced to normal temperature and pressure, is equivalent to:--

  0.627 milligram  of nitrogen.
  1.343     "      of nitric oxide.
  2.418     "      of nitric pentoxide.
  2.820     "      of nitric acid.
  3.805     "      of sodium nitrate.
  4.523     "      of potassium nitrate.

In working on substances not rich in nitrates, an ordinary nitrometer
(fig. 69) is used; but in the assay of sodium nitrate, nitroglycerine,
&c., an instrument provided with a bulb having a capacity of 100 c.c. is
employed.

[Illustration: FIG. 69.]

The plan of working is as follows:--The "measuring tube" is filled with
mercury until it reaches up into the tap, and the levelling-tube is
placed so that it contains an inch or two of mercury. If the nitrate is
in solution, 2 or 3 c.c. of the liquid (dilute liquids are brought to
this bulk by evaporation) are measured into the cup. The levelling-tube
is lowered a little, and the tap cautiously opened until all but the
last drop of the liquid has run in. The cup is then rinsed with 2 or 3
c.c. of sulphuric acid, which is run in in the same way, and the
operation is repeated with another lot of acid. The measuring-tube is
now taken from the clamp, and shaken for two or three minutes, until no
more gas is given off. It is replaced, and the mercury-level in the two
tubes adjusted. Then it is allowed to stand until the froth has
subsided, and the gas has cooled to the temperature of the room. The
volume of the gas is then read off. In adjusting the level, account must
be taken of the sulphuric acid in the measuring-tube; this is allowed
for by having the mercury higher in the other tube by, say, 1 mm. for
each 6.5 mm. of sulphuric acid, or it is counterpoised by an equal
height of sulphuric acid in the levelling-tube, in which case the two
mercury-levels are made to correspond. On opening the tap after reading
off the volume, there should be no change in the level of the mercury.
If it should rise or fall a little, a slight increase or decrease (say
0.1 c.c.) is made to the volume previously read off.

In working with nitrate of soda, &c., in the bulb nitrometer, it is
necessary to take a quantity of the substance which will yield more than
100 and less than 150 c.c. of the gas.

FOOTNOTES:

[103] Na_{3}AsO_{3} + H_{2}O + 2I = Na_{3}AsO_{4} + 2HI. The acid is at
once neutralised.

[104] Mr. Thomas Gibb is the originator of this ingenious process.

[105] By taking hold of the water present, it may prevent the
dissociation of arsenious chloride.

[106] It is difficult to get ferric chloride free from arsenic; but the
following treatment will remove 80 or 90 per cent. of the arsenic
contained in the commercial material:--Dissolve 2 or 3 lbs. of ferric
chloride with the smallest amount of water that will effect solution
with the addition of 100 c.c. of hydrochloric acid; add a solution of
sulphurous acid in quantity sufficient to reduce 2 or 3 per cent. of the
iron to the ferrous state; allow to stand a week; and then boil, to
remove the hydrochloric acid added. Nitric acid, which is prejudicial,
is also removed by this treatment.

[107] When the amount of arsenic to be estimated is small (as in refined
coppers), it is better to use a weaker solution of iodine. This is made
by diluting 200 c.c. of the standard solution with water to 1 litre.
Each c.c. will equal 0.1 per cent., if 1 gram of the metal has been
taken for the assay.

[108] The constitution of these phosphates may be thus illustrated--

      Magnesic meta-phosphate MgO.P_{2}O_{5}.
      Magnesic pyro-phosphate 2MgO.P_{2}O_{5}.
      Magnesic ortho-phosphate 3MgO.P_{2}O_{5}.

[109] The composition of which is--

      MoO_{2}        90.74,
      P_{2}O_{5}      3.14,
      (NH_{4})_{2}O   3.57,
      H_{2}O          2.55 = 100.00.

[110] This is made by adding 27 grams of magnesium carbonate (a little
at a time) to a solution of 270 grams of citric acid in 350 c.c. of warm
water; and, when dissolved, adding 400 c.c. of dilute ammonia, and
making up the bulk to 1 litre; 20 c.c. of the solution is sufficient for
0.1 gram of P_{2}O_{5}, although more will be required if much iron or
alumina is present.

[111] For the details of the titration, the student is referred to the
same place.

[112] N_{2}O_{5} + 6FeO = 3Fe_{2}O_{3} + 2NO.



CHAPTER XVIII.

SILICON, CARBON, BORON.


SILICON AND SILICATES.

In assaying, more especially products direct from the mine, there is
always found, when the rock is siliceous, a quantity of white
sandy-looking substance, insoluble in acids, which is sometimes
accompanied by a light gelatinous material very difficult to filter.
This is variously described as "insoluble," "sand," "insoluble
silicates," "gangue," or "rocky matter." It may be pure quartz; but
oftener it is mixed with silicates from the rock containing the mineral.
Some silicates, but not many, are completely decomposed by boiling with
hydrochloric acid or aqua regia; and others are partly so, they yield a
gelatinous precipitate of silica which greatly interferes with the
filtering. It is a common practice with assayers to carry the first
attack of the sample with acids to dryness, and to take up with a fresh
portion of acid. By this means the separated silica becomes granular and
insoluble, and capable of being filtered off and washed with comparative
ease.

This residue may be ignited and weighed; and be reported as so much per
cent. of "silica and silicates insoluble in acids." Unless specially
wanted, a determination of its constituents need not be made. When
required, the analysis is best made on the ignited residue, and
separately reported as "analysis of the insoluble portion."

Silicon only occurs in nature in the oxidised state; but the oxide
generally known as silica (SiO_{2}) is common, being represented by the
abundant minerals--quartz, flint, &c. Silica, combined with alumina,
lime, oxide of iron, magnesia and the alkalies, forms a large number of
rock-forming minerals. Most rock masses, other than limestones, contain
over 50 per cent. of silica. The following are analyses of some of the
commoner silicates; but it must be noted that these minerals often show
great variation in composition. This is more especially true of
chlorite, schorl, hornblende and augite.

[Table has been split into two because of its width--Transcriber]

------------------+--------+------------+------------+-------+----------------
                  |        |            |   Ferric   |Ferrous|
                  | Silica |  Alumina   |   Oxide,   | Oxide,|    Fluorine,
                  |SiO_{2}.|Al_{2}O_{3}.|Fe_{2}O_{3}.|  FeO. |    Water &c.
------------------+--------+------------+------------+-------+----------------
Potash-felspar    |  65.2  |    18.2    |     0.2    |  --   |
Soda-felspar      |  67.0  |    19.2    |     --     |  0.3  |
Lime-felspar      |  43.3  |    35.4    |     --     |  1.3  |
Potash-mica       |  45.7  |    33.7    |     3.1    |  --   |F (0.8)
                  |        |            |            |       |  H_{2}O (4.9)
Magnesia-mica     |  39.1  |    15.4    |     7.1    |  --   |F (0.7)
Hornblende        |  40.6  |    14.3    |     5.8    |  7.2  |
Augite            |  50.0  |     3.7    |     2.4    |  6.6  |MnO (0.1)
Almandine (Garnet)|  39.7  |    19.7    |     --     | 39.7  |MnO (1.8)
Chlorite (Peach)  |  32.1  |    18.5    |     --     |  --   |H_{2}O (12.1)
Schorl            |  37.0  |    33.1    |     9.3    |  6.2  |B_{2}O_{3} (7.7)
                  |        |            |            |       |  F (1.5)
China-clay        |  46.7  |    39.6    |     --     |  --   |H_{2}O (13.4)
Talc              |  61.7  |     --     |     --     |  1.7  |H_{2}O (3.8)
Serpentine        |  42.9  |     --     |     --     |  3.8  |H_{2}O (12.6)
Olivine           |  39.3  |     --     |     --     | 14.8  |
------------------+--------+------------+------------+-------+----------------

------------------+-----+--------+-------+--------+------------------------
                  |     |        |       |        |
                  |Lime,|Magnesia|Potash | Soda,  |    Fluorine,
                  |CaO. |  MgO.  |K_{2}O.|Na_{2}O.|    Water, &c.
------------------+-----+--------+-------+--------+------------------------
Potash-felspar    | --  |  --    | 14.7  |  1.5   |
Soda-felspar      | 1.2 |  1.8   |  2.2  |  7.2   |
Lime-felspar      |17.4 |  0.35  |  0.5  |  0.9   |
Potash-mica       | --  |  1.1   |  7.5  |  2.8   |F (0.8) H_{2}O (4.9)
                  |     |        |       |        |
Magnesia-mica     | --  | 23.6   |  7.5  |  2.6   |F (0.7)
Hornblende        |12.5 | 14.0   |  1.5  |  1.6   |
Augite            |22.8 | 13.5   |  --   |  --    |MnO (0.1)
Almandine (Garnet)| --  |  --    |  --   |  --    |MnO (1.8)
Chlorite (Peach)  | --  | 36.7   |  --   |  --    |H_{2}O (12.1)
Schorl            | 0.5 |  2.6   |  0.7  |  1.4   |B_{2}O_{3} (7.7) F (1.5)
                  |     |        |       |        |
China-clay        | --  |  --    |  --   |  --    |H_{2}O (13.4)
Talc              | --  | 31.7   |  --   |  --    |H_{2}O (3.8)
Serpentine        | --  | 40.5   |  --   |  --    |H_{2}O (12.6)
Olivine           | --  | 45.8   |  --   |  --    |
------------------+-----+--------+-------+--------+------------------------

Silicon, from a chemical point of view, is an interesting body. It
combines with iron to form a silicide; and is present in this condition
in cast iron. Only in the case of the analysis of this and similar
substances is the assayer called on to report the percentage of
_silicon_. Silicon is readily converted into silica by the action of
oxidizing agents. Silica forms only one series of salts--the
silicates--which have in many cases a complex constitution; thus there
are a large number of double silicates, which vary among themselves, not
only in the relation of base to acid (which is the essential
difference), but also in the ratio of the bases between themselves
(which varies with almost every specimen).

Silica is detected by heating the substance with a fluoride and
sulphuric acid in a platinum-crucible. On holding a rod, moistened with
a drop of water, over the escaping fumes, the white crust of silica
formed on the drop of water shows its presence. The insolubility of a
fragment of the mineral in a bead of microcosmic salt, is also a very
good test; the fragment, on prolonged heating, does not lose its angular
form.

There is no dry assay for this substance, nor volumetric method; when
the determination is required, it is carried out gravimetrically and,
generally, by the following plan.

If the sample contains oxides, sulphides, &c., in any quantity, these
are first dissolved out by treatment with acid, evaporated to dryness,
taken up with hydrochloric acid, and filtered. The dried residue is
treated in the same way as the silicates. Some silicates are completely
decomposed by such treatment; but it saves time (unless one is sure that
no undecomposable silicate is present) to treat these in the same way as
the others. On the other hand, there are some silicates which are only
attacked with difficulty even by fusion with alkaline carbonates;
consequently, it is always well to have the substance reduced to the
finest state of division by careful powdering, as this greatly assists
the subsequent action. With very hard silicates, the grinding away of
the mortar in this operation will be perceptible; the foreign matter
thus introduced must not be ignored. Previously igniting the substance
sometimes assists the powdering; but it is best to use a steel mortar.
The particles of steel can be removed by a magnet, or, where the nature
of the substance will allow it, by boiling with a little dilute
hydrochloric acid.

The dried and powdered material is intimately mixed with four times its
weight of "fusion mixture" in a platinum-crucible or dish. It is then
moderately heated over a Bunsen burner, and afterwards more strongly
fused over a blast, or enclosed in a clay crucible in the wind-furnace.
The action is continued until the fused mass is perfectly tranquil. With
very refractory substances, the action must be long continued at a high
temperature. When sufficiently cold, the crucible is examined to see
that no particles of foreign matter are adhering to its outer surface.
It is then transferred to a five- or six-inch evaporating-dish, where
its contents are acted upon with warm water for some time. The "melt"
will slowly dissolve, but the solution should be hastened by keeping the
liquid moderately acid with hydrochloric acid. When the "melt" has
dissolved, clean and remove the platinum-dish, and evaporate the
solution to a paste. Continue the evaporation to dryness on a water-bath
(not on the hot plate), and whilst drying stir with a glass rod, feeling
at the bottom of the dish for any unfused particles, which, if present,
can be detected by their grittiness. If there is much grit, it will be
necessary to repeat the assay; but with a small quantity it will only be
necessary to refuse the grit and silica after ignition.

During solution of the "melt" and evaporation (which may be carried on
together), a clear solution will not be obtained, a flocculent silica
will separate out, and towards the end of the evaporation the mass will
get gelatinous. The drying of the jelly must be finished on the
water-bath; first, because at this temperature the silica is rendered
insoluble in hydrochloric acid, whilst the solubility of the alumina,
iron, &c., is unaffected, which would not be the case at a much higher
temperature; and second, because the gelatinous residue requires very
cautious drying to prevent loss from spirting.

When dry, the substance is moistened, and heated with strong
hydrochloric acid, and the sides of the dish are washed down with water.
The silica is washed by decantation two or three times with hydrochloric
acid and hot water, before being thrown on to the filter. The filtrate
is again evaporated to dryness, taken up with a little hydrochloric acid
and water and again filtered. The residue on the filter is silica. The
two lots of silica are washed free from chlorides with hot water, dried
on an air-bath, transferred to a platinum-crucible, ignited gently at
first, at last strongly over the blast or in a muffle, cooled in a
desiccator, and weighed.

The white powdery precipitate is silica (SiO_{2}), and its weight,
multiplied by 100, and divided by the weight of ore taken, gives the
percentage of silica in the sample. Where the percentage of silicon is
wanted, which is very rarely the case, it is got by multiplying this
result by 0.4667. It is always necessary to examine the purity of the
body weighed as silica. This is done by re-fusing the material weighed,
and re-determining the silica in it; or, better, by mixing a weighed
portion in a platinum-dish with a little strong sulphuric acid, covering
with hydrofluoric acid, and evaporating. In the latter case, the silica
will be converted into fluoride, which will be driven off, and the
impurities will be left behind as sulphates of barium, phosphate and
oxide of tin, titanium, &c. This must be weighed and deducted from the
weight of the silica. In a complete examination of a silicate it should
be treated with the precipitate containing alumina, ferric oxide, &c.


EXAMINATION OF SILICATES.

The student interested in the analysis of rocks and rock-forming
minerals is advised to consult a valuable paper by Dr. W.F. Hillebrand
in the _Bulletin of the United States Geological Survey_, _No._ 148, to
which I am very largely indebted in the revision of the following pages.

~Moisture.~--Five grams of the powdered sample is dried between
watch-glasses in the water-oven for two hours, or till its weight is
constant; and the loss is reported as water lost at 100° C. The rest of
the determinations are made on this dried mineral.

~Combined Water, &c.~--Weigh up 1 gram of the substance, and ignite
over the blowpipe for some time in a platinum-crucible, cool in a
desiccator, and weigh. Record the loss as "loss on ignition," not as
"combined water."

~Silica.~--The ignition should have been performed in an oxidising
atmosphere in a muffle or over a slanting blowpipe flame; this will
ensure the oxidation of any pyrites or other sulphide present, which if
unoxidised would injure the crucible in the next operation. The ignited
residue is mixed with 6 or 7 grams of anhydrous sodium carbonate. This
reagent should be the purest obtainable, but its purity should be
checked, or rather its impurities should be determined by running a
"check" or "blank" assay with 10 grams of it through the stages of the
analysis; the impurities will be chiefly silica, alumina and lime, and
altogether they ought not to exceed 1 milligram. The crucible with the
mixture is heated at first gently over a Bunsen and afterwards more
strongly in an oxidising atmosphere in a muffle or over the blowpipe.
The fused mass is allowed to cool in the crucible, and is then dissolved
out in a basin with water and a small excess of hydrochloric acid. After
the removal and cleaning of the crucible, the liquor is evaporated
almost to dryness. Dr. Hillebrand advises stopping short of complete
dryness. The residue is taken up with a little hydrochloric acid and
water and filtered and washed. The liquor, including the washings, is
again evaporated and taken up with water and a little acid. Usually
about 1 per cent. of silica will be thus recovered. It is to be filtered
off and washed and added to the main silica. The filtrate is reserved.
The silica, thoroughly washed, is dried and ignited at a high
temperature for twenty or thirty minutes. It is then weighed in a
platinum crucible. After weighing it is treated with hydrofluoric acid
and a little sulphuric, carefully evaporated and ignited strongly. The
residue, which in extreme cases may amount to 2 or 3 per cent. of the
rock, is weighed and deducted from the weight of the impure silica. It
is retained in the crucible.

~Alumina, &c.~--The filtrate from silica is treated by the basic acetate
method. That is, it is first treated by a cautious addition of a
solution of soda, almost to the point of producing a precipitate, in
order to neutralise the excess of acid; 2 or 3 grams of sodium acetate
are added, and the whole boiled for a minute or so. The precipitate is
filtered off and washed only slightly. Save the filtrate. The
precipitate is dissolved in hydrochloric, or, perhaps better, in nitric
acid; and is reprecipitated by adding an excess of ammon