Refining Precious Metal Wastes - C M Hoke

REFINING METAL WASTE
ACKNOWLEDGMENT
The list of friends and co-workers who have helped in the prepara-
tion of this book is so long that space is lacking to give each the
individual thanks that he deserves. I must, however, express my
gratitude to T. R. McDearman, Margaret Stockford, Stanislaus
Skowronski, Adolph Bregman, Edward Wichers, and the late Philip
E. Browning, for especial courtesies. I wish to thank again the
many authors and publishers who have granted permission for the
use of their writings, and the several business organizations that have
placed illustrations at my disposal. In particular I wish to mention
the frontispiece, copyrighted by the National Geographic Society,
and reproduced from the National Geographic Magazine with
special permission; and to thank Drs. Wichers and Gilchrist for
their permission to use it here.
PALISADE, NEW JERSEY. C. M. HOKE
January, 1940.
PREFACE
This book is based on twenty-four years of experience in teaching

jewelers and others how to refine their precious metal wastes.
In many cases, especially at first, the instructions were given in
person, at the jewelry factory or dental laboratory. Mostly, how-
ever, the instructions were given by mail, in the form of typed
direction sheets, supplemented by replies to such questions as came
to the student's mind. These instructions were prepared to meet
the individual requirements of the user, and were adapted to the
type of jewelry he made, the size of his shop, his own background
of education or experience, and the kind of waste he happened to
have on hand at the moment.
This book is based on these oral and written instructions, and
differs from them mainly in that it hopes to meet, in one volume,
the needs of many workers handling many different problems. It
assumes complete inexperience on the part of the reader.
For many centuries—from the days of the Egyptians up to about
1900 A.D.—the metals used in jewelry were gold and silver. Copper
and zinc were added to cheapen and to strengthen them. The
processes of melting and refining were simple, well known, and
adequately described in the literature. A young man wishing to
learn the art or trade could do so either through the printed word,
or by apprenticing himself to some older artisan.
But at about the beginning of the 20th century, platinum entered
the fields of jewelry-making and dentistry. By the time we entered
the first World War, it had burst like a nova into first magnitude
in the jewelry firmament. Its sister metals, palladium and iridium
especially, came with it. Immediately the problems of refining,
separating, and remelting the scrap metal became problems indeed.
At first, say up to about 1915, comparatively few jewelry shops in
the whole world were equipped to melt their platinum scrap, and
the task of separating the platinum from the gold scrap was equally
beyond their powers; a handful of professional refiners handled the
entire output. Reasons for this were several; preoccupation with
viii PREFACE
war was only one. The habit of secrecy among many workers was
a potent reason; the institution of apprenticeship was dying out;
the literature of platinum was scanty and so highly technical as to
discourage the average reader. Platinum was so valuable that in
many shops the proprietor did all the melting and refining himself
such as it was and excluded all workmen from the room where he
worked, thus increasing the superstitions and misinformation that
collected around the whole subject.
(A practical reason was the fact that the melting point of platinum is
much higher than that of silver or gold; so high, in fact, that an oxygen
flame is required for melting it. That is, a gold-melting furnace, using
gas or coke with compressed air, is not hot enough to melt platinum.
The fuel must be combined with oxygen oxygen such as is now pro
vided in steel cylinders in order to attain the necessary high tempera
ture. The lack of compressed oxygen was a major factor in the non-use
of platinum; its introduction at low prices was a major factor in its
popularization.)
Nor were these the only complications that beset the precious-
metal worker of the early Nineteen Hundreds. New gold alloys
appeared white golds and green golds which made refining more
difficult. Electroplating became commoner, adding its cyanide solu-
tions to the duties of the harassed refiners. Chromium and rhodium
plating did their bit to complicate matters. Stamping laws that
insisted upon definite percentages of metal and alloy added further
to his legal, as well as his ethical responsibilities.
Indeed it has been said, with truth, that there have been more
changes in precious metal technology during the last thirty years,
than during the previous thirty centuries.
Hence this book.
TABLE OF CONTENTS
PAGE
PREFACE. .vii
CHAPTER
I. THE PURPOSE OF THE BOOK 1
What the book does. What it does not do. Is a knowledge
of chemistry essential? How to use the book.
II. MATERIALS TO BE DISCUSSED 4
Jewelers' wastes. Dental wastes. Photographers' wastes.
Base metals present in precious metal wastes. Non-metallic
admixtures. Old electroplating and stripping solutions.
Economics of refining the various kinds of wastes.
III. THE SIMPLEST CASE PLATINUM FILINGS THAT CONTAIN
No GOLD II
Equipment and chemicals used. "Acquaintance experi
ments." Beginning work; sieve; magnet; caustic; heat.
The nitric acid treatment. How to fold a filter paper.
Summary. Questions and answers.
IV. ARRANGEMENT OF EQUIPMENT. DISPOSING OF FUMES . . 25
Materials for hoods. Lighting. Burners. Fans. Water
and plumbing. Storage of chemicals. Hazards.
V. THE COMMONEST CASE GOLD SCRAP THAT CONTAINS NO
PLATINUM METALS 33
The first refining. Equipment and chemicals. Acquain
tance experiments. Beginning work. (I) Preliminary
treatments. (2) Melting into a button if desired. (3)
Treatment with nitric acid. (4) Dissolving gold in aqua
regia. (5) Removing excess nitric acid. (6) Recovering
the dissolved gold with copperas. (7) Washing the fine
gold. (8) Melting the fine gold. (9) Saving the silver.
Old filter papers. What to do if you spill acids or caustic.
VI. M O R E A B O U T G O L D A L T E R N A T I V E M E T H O D S .... 50
(I) Alternative preliminary treatments. (2) Melting the
material into a button. Quartation. (3) Removing base
metals with acids. (4) Dissolving the gold in aqua regia.
(5) Evaporating off the excess nitric acid. (6) Recover
ing the dissolved gold by alternative methods: sulphur
dioxide; oxalic acid; sodium nitrite; metallic precipitants
copper, iron, zinc. (7) Washing and (8) Melting the
ix
: REFINING PRECIOUS METAL WASTES
HAPTER PAGE
fine gold. (9) Saving the silver. Summary. Questions
and answers.
VII. SOME SPECIAL CASES 65
Green gold. White golds. Rolled, filled and electroplated
gold, and gold scrap containing soft solder. Procedure
when tin and lead are present.
VIII. SILVER 74
Forms of silver encountered by the refiner. Its charac
teristics and properties; acquaintance experiments. Metallic
wastes scrap, filings. Silver chloride and its conversion
into metal. Silver in solution: the nitrate; the sulphate;
stripping and pickling solutions. Cyanide solutions of
silver or gold, including plating baths. Mirror solutions.
Photographers' wastes.
IX. IDENTIFYING METALS IN SOLUTION 88
The need for identification. Definition of "in solution."
Equipment for making tests. Testing for gold in acid
solution. In alkaline or cyanide solution. Testing for
silver. Stannous chloride solution for all precious metals.
Standard solutions of gold, platinum, palladium. Using
stannous chloride with platinum; with gold; with palla
dium; with silver. Usefulness of stannous chloride. If the
tests are inconclusive . . . Two or more precious metals.
Dimethyl glyoxime for palladium. DMG and nickel.
Other chemical tests. The glow test. Copper. Iron.
Acids and alkalies litmus paper. Other tests. A warning.
Identifying metals in solid form.
X. GETTING ACQUAINTED WITH THE PLATINUM GROUP . . 103
Especially platinum, palladium, and iridium. The six
metals and their employment in art and industry. Books
to read. Familiar chemical characteristics. Platinum and
ammonium chloride. Palladium and ammonium chloride.
Palladium and acids. Possible experiments. A little about
iridium. Questions and answers.
XI. JEWELERS FILINGS AND CLIPPINGS CONTAINING MIXTURES
O F GOLD AND P L A T I N U M I I I
Kind of materials. Equipment and chemicals. (I) Gen
eral cleaning. (2) Removing base metals. (3) Removing
gold. (4) Removing silver chloride. (5) Melting the
platinum. (6) Recovering dissolved platinum. (7) Burn
ing the orange powder and melting the sponge. (8) Re

covering dissolved gold. (9) Recovering silver if worth
while. (10) The Stock Pot. Summary. Questions and
answers.
TABLE OF CONTENTS xi
CHAPTER PAGE
XII. ALLOYS AND BUTTONS CONTAINING GOLD AND PLATINUM-
GROUP METALS 129
Materials handled. Jewelers buttons. Dental alloys.
Equipment and chemicals. (A) Preliminary treatments.
(B) Melting into a button if necessary. (C) Granulating
the button or rolling it thin. (D) Dissolving the metal in
aqua regia. (E) Recovering the iridium from the silver
chloride. (F) Recovering dissolved platinum. (G) Re
covering dissolved gold. (H) Recovering silver if worth
while. (I) The Stock Pot. Summary. Diagram. Silver-
rich buttons. Questions and answers.
XIII. THE STOCK POT 143
Materials handled. Equipment and chemicals. Collect
ing the precious metals with zinc. Refining the black
powder. The iridium residue. The dilute aqua regia
solution, platinum, palladium, gold. Burning the red
palladium powder. Further treatment for palladium. Dis-
posal of the final tailings. Diagram. Questions and
answers.
X I V . ALTERNATIVE METHODS FOR CHAPTERS X I , X I I , AND X I I I 156
Part I An alternative procedure for the aqua regia solu
tion in Chapter XI or Chapter XII. Part II Formic acid
method of reducing the colored powders. Part III An
alternative method of precipitating palladium, using di
methyl glyoxime. Part IV Alternative methods of refin
ing the Stock Pot. Diagram for the second scheme for
Stock Pot concentrates. Questions and answers.
XV. REPURIFICATION OF GOLD, PLATINUM, PALLADIUM, SILVER,
IRIDIUM 170
Part I. Repurifying gold: by reprecipitation; the use of
oxalic acid; by fire; by chlorine gas; by electrolysis. Part
II. Repurifying platinum. Part III. Repurifying palla-
dium: the colored powders; buttons or bars; palladium
black; palladium sponge. Part IV. Silver. Part V.
Iridium.
xii REFINING PRECIOUS METAL WASTES
CHAPTER
X V I I . RHODIUM, RUTHENIUM, OSMIUM
Their rarity. Rhodium. Ruthenium. Osmium. Refin
ing the "mysterious" metals. Some schemes of separa
tion: (I) The classic system. (2) The method of Gilchrist
and Wichers. A list of books and papers referring to the
three rarer metals.
XVIII. Low GRADE WASTES
Materials discussed. Determining the method of disposal.
Two extreme cases. Part I. The shop of moderate size.
Procedure for floor sweeps. Equipment for dressing floor
sweeps. Procedure for wash-barrel residues. For old
crucibles. Flue dust. Paper storage. Polishings, grind-
ings, old aprons, floors, etc. Old solutions. Fluxes for use
with low grade wastes. Litharge. Lead buttons. Cupella
tion. If platinum is the only precious metal. Part II. The
large plant. Amalgamation not profitable. Disposal of the
final fine dust. Sampling. Questions and answers.
XIX. LOSSES
Metal losses in the refining room; in the shop; in the
furnace room. How much loss is permissible? Figures
on losses in melting gold; in melting platinum; in polish
ing. Losses due to dishonesty. Losses of money, rather
than metal, due to mismanagement. Difficulty of esti-
mating the value of wastes. Assaying. Questions and
answers.
XX. MORE ABOUT EQUIPMENT
Where to buy equipment. How to order. When handling
large quantities. The Buchner funnel. New ideas in
chemical apparatus. More about ventilation. Crucibles.
Cleanliness in the refinery. Wash-barrels and filter-presses.
The Cottrell precipitator.
XXI. PROCESSES USED BY THE PROFESSIONAL REFINERS
Materials handled. Virgin metals. By-product metals.
Secondary metals. The Miller chlorine process. The several
electrolytic methods. Possible variations. Advantages and
disadvantages of the electrolytic methods. Can professional
methods be adapted to small-scale operation?
X X I I . SOME LARGE PROFESSIONAL REFINERIES
The Irvington Smelting and Refining Works. The Miller
process at the Homestake Mine. The Raritan Copper
Works. The United States Assay Office. The plant at
Acton, England. Some other large plants,: Copper Cliff;
Montreal East; Old Hickory; The Royal Mint at Ottawa,
Canada.
TABLE OF CONTENTS xiii
APPENDIX
PAGE
A. MELTING THE PRECIOUS METALS 293
Part I, Platinum and its group. Fuel and equipment. Melt
ing small lots of platinum scrap. Working and annealing.
Melting platinum filings. Defective platinum. Platinum
sponge and black. Large melts. Making up platinum alloys.
The electric furnace. The oxygen torch. Spectacles. Crucibles
and refractories: lime; zirconia; sand; contamination from the
crucible; other refractories. The platinum group and occluded
gases. Melting and working palladium. Melting the other
platinum metals. Part II, Melting gold and silver. Sources of
information. Melting the fine gold powder. Melting lemel.
Melting clean scrap. Deoxidizers. Fluxes. Casting precious
metals. Making up new alloys. New light on old metals.
B. HAZARDS 320
The National Safety Council. Comparison of chemical and
heat burns. First aid. Chemical eye burns. Neutralizers.
Alkalies and caustics. Hydrochloric or sulphuric acid. Hydro
fluoric acid. Nitric acid. Nitrous fumes. When handling
sulphuric acid. Shower baths. 'Cyanide poisoning. Mercury.
Hazards of the melting room: hazardous arrangement; cru
cibles; oxygen cylinders. Stains on the skin. Gas masks.
Some articles to read.
C. LAWS AND REGULATIONS 330
Federal licenses for handling gold. Regulations concerning
Government purchase of precious metals. National Stamping
Law covering gold, silver, and their alloys. Law for stamping
platinum. The Commercial Standards. Local ordinances.
Standards in foreign countries.
D. A LIST OF DEALERS 336
Buyers of precious metals. Dealers in chemical apparatus.
Dealers in supplies for jewelers. Dealers in special equip
ment and machinery.
E. BOOKS TO READ 345
Some sources of information. A list of books. A list of articles.
INDEX 355
REFINING PRECIOUS METAL WASTES
CHAPTER I
THE PURPOSE OF THE BOOK
What the book does. What it does not do. Is a knowledge of chemistry
essential? How to use the book.
• # * •
When a jeweler makes a piece of jewelry, he starts with a sheet of
metal and he hammers and carves it and drills holes in it, and files
off the rough edges; possibly he engraves it; and finally he polishes
it. He winds up with a piece of jewelry that weighs one-half or
maybe only one-tenth as much as the original sheet. The rest is
now in the form of clippings, scrap, filings and the minute particles
that are now mixed with the polishing materials all contaminated
with more or less dirt.
When a dentist makes a denture he does much the same thing;
he starts with gold or some other metal, usually of high intrinsic
value, and when he is done, a good part of the original metal is now
in the form of scrap, grindings, and the like.
These precious metal wastes are useless until refined. In most
cases, especially when the proportion of dirt is small, the task is
easy. In some cases the mixture is of such a nature as to make it
wise for him to sell the stuff to a professional refiner. In a few cases
the proportion of dirt is so high that the owner might as well throw
the thing away.
The purpose of this book is to teach the jeweler or dental tech-
nician how to handle all these kinds of waste. The equipment
needed will be described, the processes will be explained, and he
will be told how to decide whether a given type of waste is too com-
plex for him to refine or not.
2 REFINING PRECIOUS METAL WASTES
This book does not cover the refining of virgin ore or crude metal, nor
any phase of mining engineering.
It does not cover assaying which is the analysis of a small weighed
sample, made to learn its ingredients and their proportions.
It concerns itself only with secondary refining that is, the refining
of precious metals that have already been in use and are now waiting
to be returned to further employment in art, science, or industry.
And while the book does describe fully the processes used by such
huge plants as the United States Mints and others who handle gold
by the ton and platinum by the hundredweight, its main concern is
with the small plant the jewelry factory of from one to a hundred
men, and the average dental laboratory and it covers the recovery
of gold, silver, and the platinum-group metals from the wastes that
such plants produce.
IS A KNOWLEDGE OF CHEMISTRY ESSENTIAL?
For the task that we have set ourselves, the answer is no. Many
excellent refiners have no formal chemical training. Of course by
the time they learn to refine they have acquired a knowledge of the
behavior of several metals and acids, and the worker with a knowl-
edge of chemistry can progress more rapidly and with greater assur-
ance. However, this book is written for the man without a knowl-
edge of chemistry.
It is pertinent to digress here and recall the origin of chemistry. . . .

It began with man's dream of making gold out of base metals. In the
pursuit of this dream he experimented with every substance he could lay
hands on, combining them, heating them, cooling them, crushing them,
cursing them and blessing them, and learning a great deal about them.
Chemistry of today is an outgrowth of these efforts. The student of
this book will have some of the same mental adventures that were en-
joyed by the Alchemists, but, it is hoped, with a greater monetary reward.
HOW TO USE THE BOOK
We shall begin with easy tasks tasks that are easy to perform
and easy to understand. The assumption is that the reader has
never even seen the processes of refining precious metals. It is
taken for granted that he hopes to learn how to handle even the
more complex mixtures normally found in the jewelry shop or
dental laboratory; therefore we shall go step by step to the more
complex cases. Finally it is assumed that he has eyes with normal
THE PURPOSE OF THE BOOK 3
color-perceptions, a fairly steady hand, and will perform in order

the various "acquaintance" tests, without too great impatience to
reach the more interesting chapters in the back of the book. He
must of course obtain the required equipment and chemicals.
Precautions against the hazard of fumes, acids, poisons, etc., will
be brought to his attention.
After performing the acquaintance tests, the reader will take
small quantities of filings, clippings, etc., and will read about the
first procedure in the instructions, and perform it. Then he will
read about the next step, and perform it. Soon he will have clean
metal, ready to melt up and use again, or to be sold.
The process of learning can be carried out best with small quan-
tities of material. Later, after the process and its reasons are familiar,
the reader will handle larger and larger quantities of material at a
time; this means larger vessels, and sometimes different ones, all of
which are described in their place.
* * * *
There are three agencies used in refining: acids and other chemi-
cals; heat; and the electric current. Most refining is done with acids
the so-called wet way which will occupy a large part of this book.
The dry way, which means heating the metal sufficiently to melt it,
and then removing impurities by means of fluxes, has fewer appli-
cations today than it had a generation ago. Electrolytic refining
may be the method of the future, and is now an accepted method
for handling large quantities of materials. All three of these will
be described, so that the reader may choose the agency best suited
to his needs.
* # • *
"No matter how a text is written, one reader will yearn for more
details while another will be bored by the tedious explicitness." These
words of truth can be found in the Introduction to J. Harry Howard's
Handbook for the Amateur Lapidary, and they seem so applicable that
we must here echo his request that the reader take this into account when
tempted to criticise the book. The author has purposely gone into every
small detail necessary to instruct the newcomer who knows nothing of
chemistry or laboratory technique. From the chemist, who may skip
these elementary instructions, we ask patience with the beginners.
CHAPTER H
MATERIALS TO BE DISCUSSED
Jeweler's wastes. Dental wastes. Photographer's wastes. Base metals

present in precious metal wastes. Non-metallic admixtures. Old elec
troplating and stripping solutions. Economics of refining the various
kinds of wastes.
• • * #
The simplest and cleanest kinds of precious metal waste are old
jewelry, the metal parts of old dentures, and the clean scrap metal
that is produced in the manufacture of jewelry or dentures. The
refining of this material is easy because there is little or no dirt to
be removed, and the job is profitable for the same reason.
Old jewelry may be made of Sterling silver, or of base metal plated
with silver; or of karat gold in which pure gold is alloyed with
more or less of other metals; or of base metal plated with gold; or
of platinum, which also is alloyed with other metals. Dental restora-
tions such as bridge work, plates, fillings, etc., are often made of
high quality gold; sometimes of platinum; sometimes of alloys in
which gold, platinum, silver, copper, and other metals are com-
bined; sometimes of metals as inexpensive as stainless steel.
Non-metallic substances are often found in conjunction with
these precious materials: the crystal of the watch-case; the porce-
lain tooth in the denture; the sapphire in the ring.
In the jewelry factory and the dental laboratory we find metal-
bearing wastes generated in the processes of manufacture, and it is
with these that we shall concern ourselves most, partly because they
are the stuff on which the refiner spends most of his time, and partly
because the processes involved are more complex.
These wastes are sometimes of quite high intrinsic value, as in
the case of the sprues chipped from castings, or the little shavings
produced by the engraver's tools, or the clippings that fall on the
jeweler's bench. The neater the workman, the higher the value per
ounce; that is, the less trash, the fewer match ends, bits of paper,
cigarette ashes, etc., that have to be removed.
4
MATERIALS TO BE DISCUSSED 5
In other cases these wastes contain much dirt. For example, when
jewelry is polished, the dust and the used polishing materials are
collected (usually by an exhaust system) and saved for refining.
The particles of precious metal are too small, as well as too dirty,
to be distinguished by the unaided eye, but they are large enough
Jeweler at work. Small particles of platinum, mixed with bits of steel from the
saw, are being produced.
Courtesy of Byard A. Brogan.
to be worth recovery in most cases. Then we have old plating

baths, for example the solutions used in gilding or silver plating.
Photographic wastes are not legitimate parts of the jeweler's prob
lem, but now and then are brought to the small refinery, and are
therefore discussed here briefly.
Some shops handle only gold and silver. Some handle only plati
num. Most jewelry shops and dental laboratories handle all three,
as well as palladium and iridium. In a few instances we find os
mium and rhodium and ruthenium rare and expensive metals of
the platinum group in addition.
Finally we have very low grade wastes, including the stuff swept
up from the floor, the deposit scraped out of the flue over the melting
furnace, the discarded crucibles in which metal has been melted, and
the settlings from the water in which the workmen wash their hands.
All of these are treated for the recovery of their values in the well-
ordered jewelry factory and dental laboratory. All of these and
more will be discussed in this book.
(h) Cadmium a metal closely related to zinc in many of its

properties is sometimes found as a component of gold solders.
• # * #
Crude gold ore sometimes contains arsenic, selenium, tellurium,
and other of the less common elements. These do not come within
the scope of this book, and materials containing them should not
be refined according to the methods here given.
NON-METALLIC ADMIXTURES
(a) Those that will burn are numerous and varied: dust, paper,
match ends, scraps of tobacco, charcoal, the wax used by dental me-
chanics, shellac, the strings and cloth used in polishing, wood rubbed
from the bench or peg, and, if we are truthful, we must include nail
parings, hair, and bits of food, from the filings swept up from the
worker's bench.
(b) Those that will not burn include an occasional precious stone,
bits of glass, sand from the sand-blasting machine, emery from the
files, carborundum, plaster of paris, dentists' investment compounds,
vulcanite, fragments of porcelain teeth, rouge and other polishing
materials, pieces of crucible, and grit from city dust.
Floor sweeps are not unlike bench sweeps in composition, except
that the proportion of dirt is higher. The settlings from wash-
barrel wastes contain much soap and soap scum.
Polishings contain the lint from the buffs, the rouge, pumice and
other materials used in polishing, atmospheric dust, and rather
small quantities of precious metals.
OLD ELECTROPLATING AND STRIPPING SOLUTIONS
Solutions for pickling, stripping or plating gold and silver are

common in most jewelry shops, including many retail stores and
repair shops. Sometimes platinum, palladium, and rhodium baths
are used. When these solutions become foul or exhausted, they
must be treated for the recovery of their valuable content.
ECONOMICS OF REFINING THE VARIOUS KINDS OF WASTES
Precious metal wastes may be divided into high grade and low
grade. They may also be divided into (a) wastes whose only
precious metals are those of the platinum group; (b) those whose
precious elements are gold and silver only; and (c) those that con
tain both gold and platinum-group metals.
Up to the time when the United States went off the gold standard,
most shops sold their high grade wastes of group (b) those con
taining gold and silver to the United States Assay Offices or Mints,
that being the best way to dispose of lots sufficiently high grade to
meet the Treasury Department's requirements.
Since the United States went off the gold standard, the regulations
have been drastically modified. The reader is advised to apply fre
quently for current rulings, which are subject to change from time
to time; these are obtainable from any Federal Reserve Bank, or
from the Mint or Assay Office of his district.
At present it pays most shops to refine all their high grade wastes
of whatever composition.
Wastes of groups (a) and (c), in which platinum is present,
cannot be sold profitably to the Government, which is not authorized
by Congress to buy platinum or its group. It can accept them, as
constituents of gold or silver shipments, but cannot pay for them.
In fact the refining charges are increased when these unwanted
metals are present. In spite of this fact much platinum is turned in,
mostly in shipments where it is a minor constituent, in sufficient
quantity to net the Government many thousands of dollars a year.
The owner of platinum-bearing wastes, therefore, must decide
whether to refine them himself or sell them to a professional refiner.
Group (a), containing platinum but no gold, is so simple to refine,
as will shortly be seen, that it is rarely shipped out. Group (c),
however, calls for special knowledge and some little equipment
knowledge that this book will give, and equipment that this book
will describe. It also represents the most profitable branch of the
refiner's art.
Let us compare the expense of selling a lot of high grade material to a

refiner, and doing it yourself. Suppose you have $1000 worth of filings,
old dentures, jewelry, or the like. If you sell it to a refiner you must
consider several costs:
(1) The interest on the cost of the metal while it is idle. Two months
at 5% per annum would be $8.34.
(2) An assay should be made, which will cost the same for a small lot
as for a large one. If platinum group metals are present, the cost will be
considerably more than for gold and silver alone.
(3) Cost of packing, shipping, and insurance during transit.
(4) The refiner's legitimate fee, and his profit.
Some refiners buy dirty filings at so much per ounce, then sell
clean metal at a higher rate per ounce. The difference in the case
of platinum-group metals is apt to be considerable, this sum being
in effect the refining charge per ounce. Sometimes the refiner
simply returns a certain sum, without information as to what it
represents.
Now, if you refine this $1000 worth of metal yourself, in your
own shop, you (I) cut down its period of idleness; (2) cut out the
assay entirely; (3) cut out the shipping expense and chance of loss
during shipment. And your expense for acids and labor is generally
less than the refiner's charge plus his legitimate profit.
In a well-managed shop, the total expense is much less. Even
when the metal is in the form of dirty mixed filings, the cost of
refining is generally less than the cost of a good assay. And the
factor of idleness is important a shop can do business with a smaller
capital investment when it can promptly return its scrap to usable
condition.
There are reasons why the small shop, in particular, should do
its own refining. A shop that has only small amounts of filings, etc.,
is apt to wait a long time before sending them away; this means a
relatively high loss through idleness. Then it has to pay just as
much for an assay as a big shop; or maybe it makes the mistake of
selling them without an assay, which is a false economy if there
ever was one. The expense of shipping is nearly the same as for a
large lot, the insurance only a little less. And if the refiner charges
a flat fee, this means that the expense per ounce on a small lot is
much more than for a large one.
So far we have talked of high grade wastes, clean scrap, bench
filings, old dentures, etc. Home refining for these has been found
profitable in such a large proportion of cases that it is advised for
all except a few very oddly situated establishments. Thus a jeweler
in an office building with no gas, no suitable sink, and no way to
dispose of fumes, may not afford to rent extra space elsewhere simply
for refining. On the other hand the work might well be done at
his residence; there are many excellent basement and roof refineries.
But when we come to low grade wastes such as polishings, floor
sweeps, and old crucibles, then we begin to question seriously the
economics of refining. In general it has been found that even the
smallest shop should take certain measures to dress and reduce the
bulk of such materials before shipping them to a professional re
finer. And even the largest plant, that always burns, sieves, grinds
and samples these materials, will need to question the economics of

the actual ultimate refining of the final low-grade dust. (This will
be discussed again in Chapter XVIII.)
In general, the larger the shop the more economically it can operate.
However, there are limits to economy even when there are no limits on
the supply of material. To explain: It has long been known that there
is gold in sea water. Some years ago the Dow Chemical Company built
an enormous plant to extract bromine (used in the manufacture of ethyl
gasoline) from sea water. Sea water contains about 61 grams (almost
2 Troy ounces) of bromine per ton of water, but they manage to extract
over 15,000 pounds of it every day, at a good profit. Naturally these
chemists considered the recovery of the gold in the sea water, since they
can extract it with the same operation as that used for bromine simply
by adding one or two more steps. There was no question as to the
presence of the gold, which is there to the extent of about 0.002 gram
per ton of water. So they worked for some weeks, handled 12 tons of
sea water, spent the value of many ounces of gold, and recovered a tiny
scrap of metal weighing about 1/10,000 of a gram. They labored over a
mountain and brought forth much less than a mouse.
• # • •
The first time you refine, handle a small quantity of material, and keep
your attention centered upon the process. The third or fourth time,
however, that you carry through a given procedure, make written note
of the following things: the exact weight of the material you start with;
the amount of time and chemicals you consume; and the weight of pure
metals recovered. The apparent loss will be made up of the following;
base metal and trash; metal tied up in old filter papers or in solutions laid
aside for later treatment; and actual true loss.
As time goes on, you will find that the time and chemicals consumed
will become less much less. The proportion of immediate recovery will
become higher and higher, and the actual loss and the proportion of
metal tied up will become lower; the process therefore will be more and
more profitable as your skill increases.
CHAPTER III
THE SIMPLEST CASE PLATINUM FILINGS THAT

CONTAIN NO GOLD
Equipment and chemicals. "Acquaintance experiments." Beginning work;

sieve; magnet; caustic; heat. The nitric acid treatment. How to fold a
filter paper. Summary. Questions and answers.
* * * *
This type of waste consists primarily of bits of platinum or iridio
platinum produced by the file or saw, and larger pieces that were
clipped or sawed out. It is found in shops that make only high
grade platinum jewelry. It is rarely found in dental laboratories.
Now and then we find a jewelry shop making both gold and plati-
num articles, in which a given workman may handle only one kind
of goods, and the man who works on platinum only will assume
that his filings are truly free from gold.
Now and then he is right, especially if his bench is in a separate
room. But usually his filings will contain a little gold carried in by
the tools, or by his hands in their contact with door-knobs and
faucets that are handled by the gold workers. Moreover, most
platinum solders contain gold, and filings normally contain bits of
solder. Accordingly this type of filings is rather rare. However, it
is so obviously the simplest case and the most logical to begin with,
that we shall describe it fully.
Mixed with the bits of platinum there will be steel particles also,
from the files and saws; brass or steel binding wire; the usual or
ganic dirt such as paper, tobacco, hair, emery, etc.; and possibly
other kinds of dust. One unexpected contamination that sometimes
occurs is nickel in the form of filings put in by a dishonest work
man as a substitute for platinum, with the purpose of cheating his
employer. Shellac from the diamond-setter's bench is often present.
If your material contains pure palladium, or alloys containing as
much as 40% palladium, do not use the methods of this chapter;
save it for refining by the methods described in Chapter XII.
The student is advised to read this chapter to the end, without attempt-
ing to remember the steps, and without any great effort to understand
them. He should then return to this point, and provide himself with
the necessary articles; then read one paragraph at a time, do what it
says; then read the next paragraph, do what it says; and so on. Points
that seem obscure at the first reading will clear up when put into
operation.
EQUIPMENT AND CHEMICALS USED
These are obtainable at almost any chemical supply house; noth-

ing of an unusual nature is needed. For purpose of identification
we shall give the catalogue numbers used by Eimer and Amend, of
New York City. Prices are subject to change, and are F.O.B. New
York City; higher prices prevail throughout the West and South.
Other dealers all over the world have access to this catalogue, and
can provide you with similar articles if you mention these numbers
as being taken from Eimer and Amend's 85th Anniversary Cata-
logue of Laboratory Apparatus and Supplies.
Apparatus:
A porcelain casserole, such as the one illustrated, will serve many
purposes. Eimer and Amend Catalogue No. 19656, the 61/2inch
size, costs $2.20. A second and smaller casserole, for example the
31/2 inch size at 6o£, will be used in making tests and handling
Casserole; stirring rod;

thin asbestos disc; one type
of heater.
small samples. You may need additional casseroles later, especially

if you refine more than one batch of material at a time.
A magnet; No. 25746, 6 inches, costs $0.35.
Wash bottle No. 18952, size 1000 cc, $0.70.
A glass funnel; No. 23632, size 5 inches, costs $0.25 Filter paper
to fit it is necessary. E. & A. Nos. 23150 and 23180 are good kinds,
and the 25 cm. size fits the 5 inch funnel; 100 sheets cost $0.55 (Later
THE SIMPLEST CASE
Magnet. Wash-bottle. Funnel.
One kind of filter paper. One kind of funnel support.
Test Test tube Spoon- Test tube

tube. clamp. spatula. brush.
you will want additional and larger funnels, with filter paper to fit
each size funnel.)
A funnel support; No. 31937 holds two funnels and costs $0.90
Home-made supports are good if made to slide up and down, and
if so built that even when the funnel is full of liquid they will not
tip over.
A dozen glass test tubes, size 6 inches by 3/4 inch or smaller,
No. 32124; about$.060the dozen.
Test tube support No. 31989 at $.60.
Test tube clamp No. 20260 or equivalent, about $0.15.
Support for test tubes. Pitcher. Wire gauze with asbestos center.
Sieve. One type of burner. Frying pan.
Test tube brush No. 19058 or equivalent, about $0.10.

Glass stirring rod No. 24562; a 6 inch or 8 inch length costs only
a few cents. You will need several, of assorted lengths.
A useful tool for stirring, taking samples, making tests, etc., is a
porcelain spoon-spatula; No. 31280,16 1/2 cm. long, $0.50.
Paper towels to wipe your hands and mop up spilled solutions.
A sieve. Many shops already have a series of sieves with different
meshes, such as No. 30920 to No. 30960. If not, buy a rice strainer
at a department store. Also a frying pan, and two or three plain
glass pitchers of the kind used for ice-water. Those straight across
the top are best, because a funnel can be rested on them. The two-
quart size is useful.
THE SIMPLEST CASE 15
You will need a stove or heater such as the one in the first sketch,
with one or two burners. A piece of thin asbestos to go between
the dish and the flame is helpful, or better, a square of wire gauze
with asbestos center, such as No. 33090, 6 inches, at $0.15.
A Bunsen or Meker burner is useful for many purposes; No. 19240
at $0.30 to $0.50 according to the type of gas you use when ordering
be sure to specify.
CHEMICALS: Nitric acid, chemically pure. A 7 lb. bottle costs about
$2.00, and an allowance will be made for the empty container. Large
shops save money by buying acids in carboys.
You will use considerable ordinary tap water.
If your material contains much shellac or grease, you will need
some caustic soda, in sticks, lumps, or flakes. (Caustic lye, also
called potash, serves the same purpose.) It need not be chemically
pure; the ordinary household kind at $0.10 a can is excellent.
. "ACQUAINTANCE EXPERIMENTS"
Let us become acquainted with our materials through exposing

them, in turn, to the reagents that are used in refining. We cannot
too strongly urge the reader to make every one of these tests, him-
self, working in a good light and noting down in writing the effects
that he observes.
To describe these effects in words is easy, but for the man who seri-
ously wants to learn the nature of these metals, wordy descriptions are a
waste of time. He must see for himself what happens. Half an hour
spent making these tests with his own hands will tell him more than a
thousand words in print.
(1) Platinum. Place some clean platinum scrap in a casserole or
test tube and cover it with C.P. nitric acid, full strength. . . Noth-
ing happens. . . Warm the vessel gently. . . Nothing happens to
the platinum; the acid simply boils away. You are safe to use your
best platinum ring for this test, if its stones will not be injured by
acid and if it is truly platinum. (Iridio-platinum, the alloy gener
ally used in good jewelry, behaves so much like pure platinum, that
in most of this book we shall make no distinction between the two.)
You will find that diluting the acid produces no unexpected change
in its action.
(2) Copper. Repeat with copper scrap. Note the speed with
which the acid goes to work, even when cold, giving off a brown
gas of extremely unpleasant nature. Thin wire will disappear within

a few seconds. Notice the effect of heating the acid. Diluting the
acid slows matters up, but produces no other conspicuous change.
The resulting liquid is green of a color that we associate with
copper compounds and contains copper nitrate.
(3) Nickel. Repeat with a piece of pure nickel. (The $0.05 coin
known as a nickel is not suitable; it is mostly copper, alloyed with
enough nickel to make it white.) Nickel, like copper, dissolves
readily in nitric acid, either concentrated or dilute, giving off the
same brown fumes. The liquid remaining is again green, and con-
tains nickel nitrate.
(4) Similar tests with brass, gold, silver and zinc, will suggest
themselves and will be instructive. Now is the time, also, to begin
to read the trade publications, Government reports, books, and other
literature, some of it obtainable without charge, that are mentioned
in Chapter X and Chapter E of the Appendix.
BEGINNING WORK: SIEVE; MAGNET; CAUSTIC; HEAT
Let us start with a small quantity of filings or scraps. A few

pennyweight will be enough; do not Use more than five ounces. To
many workers this will seem a ridiculously small quantity, because
they expect to refine many ounces of such goods every day. It
would be a small quantity to refine, perhaps, but it is an excellent
quantity to learn on.
Begin by sieving the filings. This will remove bits of paper,
match ends, etc. Some large pieces of clean metal, and perhaps a
stone or two, may appear and can be returned to their proper places.
Next, spread the filings out on hard-surfaced paper, or in a flat
dish, and pass a magnet through them, back and forth, to collect
iron and steel. Brush the steel particles into a separate vessel, and
you will note that some platinum has been carried along with them.
You can remedy this by a second treatment of the steel particles,
spreading them out thinly and again letting the magnet pick them
up. Return the platinum particles to the main dish.
It is not possible to get rid of all steel by this treatment, partly be-
cause some kinds of steel are not attracted to the magnet, but we shall
remove any remaining steel in a subsequent treatment.
By this time you have observed the nature of the dirt that is
present, and its character will determine your next step.
If the filings contain much shellac or grease, place them in a

casserole, cover well with water, and add about a half ounce of
lye or caustic. Use care; heat is generated when caustic and water
combine, and there may be some spattering. Bring the solution
to a boil, and stir with a glass rod.
(This caustic is bad for the skin, so do not get it on your hands. If
you should spill it on your hands, wash them at once with plenty of
water and they will not be hurt.)
The porcelain used in chemical ware will, if properly treated,
resist heat amazingly well. Place a piece of thin asbestos or wire
gauze between it and the flame; do not use thick asbestos because
it wastes time and heat. Have your dish perfectly clean and dry
on the outside, and it will not crack. Get into the habit of wiping
the outside of the dish every time you put it on the fire.
Stir the material frequently with a glass rod or porcelain spoon.
After the solution has boiled a few minutes, remove it from the
fire, fill the dish with water, let the filings settle, and pour off the
liquid into a pitcher, leaving the filings in the casserole. Let this
liquid settle well, and save any metal that you may have poured
over; then throw the liquid away.
Fill the casserole again with hot water, to wash the filings. Stir
well, then pour off this wash-water. Repeat until the filings no
longer feel slippery. Let the wash waters settle before you throw
them away, to be sure that you do not throw away any metal.
If your filings contain only a little shellac or grease, they will not
need to be boiled in caustic. In that case, simply burn out the wood,
paper, etc., by heating them to redness in a frying pan.
Do not use both the caustic and the frying pan. Use one or the
other, according to the kind of dirt present.
THE NITRIC ACID TREATMENT
Next, dissolve base metals, such as brass, nickel, etc., by boiling the
filings in nitric acid, chemically pure. The filings should be in the
casserole. Pour on enough nitric acid to cover them and add a little
water. Heavy brown fumes will appear, which are poisonous and
very disagreeable, and will damage your machinery, so do this work
outdoors, or near a good flue. Or have an exhaust fan in your
window, and let the fumes go out the window. Ways of disposing
of fumes are discussed fully in the next chapter.
HOW TO FOLD A FILTER PAPER
Use a paper of the right size. Fold it in half.
Fold again in right angles; this gives Open it up, with three thicknesses on
you a pie-shaped piece of four thicknesses. one side and one on the other side.
Fit it into the DRY funnel, and then Press the paper in with your two hands,
wet it with water. (This shows one way pressing it down smoothly so that the
of using the wash-bottle. Hold your liquid will run through rapidly. It pays
thumb against the stopper.) to use paper of proper thickness.
Add the acid slowly, or it might boil over. If there is much

base metal, there will be heavy fumes. If there is very little base
metal, there will be only light fumes. Stir well with the glass rod.
This may stand overnight, or longer. But if you are in a hurry,
heat the acid to boiling, and the base metals will all dissolve in a
few minutes. Stir often, so that the acid may reach all parts of
the filings.
While the acid is acting, let us learn how to fold a filter paper,
following the illustrations on page 18.
It pays to work with a filter till it fits, by folding wider or nar-
rower pie-shaped pieces, even if you waste several papers. If it
does not fit the liquid will run through very slowly.
Large filters are apt to break at the point. So fold a second small
filter, and fit it as a cap over the point of the big one, with the three-
thickness side of one covering the single thickness of the other.
Use filter paper of the right size. It should not reach the top of
the funnel. If so big that it sticks up beyond the funnel, it is im-
possible to wash it down properly, so trim it off with scissors.
Let the acid work all it will, then dilute it, pour it off, and either
filter it through the filter paper, or let it settle. It is best to use the
filter paper and funnel; you may pour over some filings, and the
filter paper will catch them. Let the paper dry and put it away in
a jar or box, which we shall call your "paper storage." Eventually
you will refine this and recover the metal that it contains.
When pouring, hold the glass rod or porcelain spoon against the
lip of the dish, and pour down it. In that way you avoid splashing
acid and staining the outside of your dish.
Pour against the side of the filter. Do not let the liquid fall
heavily against the thin point of your filter paper.
Throw the acid away. Leave the filings in the dish. Add fresh
acid, heat, and see if it changes color and gives off fumes; if ,so,
that means that there are still base metals for the acid to work on.
Stir well.
Continue this treatment until there is no more action between the
acid and the metal; that is, continue until all base metals are
dissolved.
Wash off the acid from the filings carefully, with plenty of hot
water. Wash several times. Then let the filings dry. In most cases
they are now ready to be remelted, for use in making up new
jewelry. Methods of melting are discussed in the Appendix.
Sometimes it is plain that sand or emery still remains, mixed with
the filings. If the quantity is small, it will do no harm, but will
flow to one side when the metal is molten. If the quantity is large,
it may become troublesome; this will be discussed later, in Chap-
ter XVI.
* * • *
SUMMARY
To repeat the procedure, in one sentence, is the removal of

unwanted materials from the platinum, which itself remains un-
changed. Thus we sieve out paper or trash, remove iron with a
magnet, remove shellac and grease with caustic (or burn it out in
a frying pan), and remove base metals with nitric acid. Then we
wash, dry, and melt the clean platinum.
* * * *
QUESTIONS AND ANSWERS ON CHAPTER III
The following are questions that have been asked on this chapter, with
replies.
Q. You tell us to filter the nitric acid solution, but you did not tell us
to filter the caustic solution.
A. Acid solutions, unless very concentrated, are easy to filter. Caustic
solutions are difficult to filter; if concentrated they destroy the filter
paper; if somewhat diluted they toughen it and make it almost imper-
vious. Only when very dilute are they filtered with ease. Usually you
save time by setting any kind of a solution aside to settle, then pouring
off the top part, and filtering only the last portion. This is especially
true with caustic solutions.
# * # *
Q. Why not use aqua regia instead of simple nitric acid ?
A. Because aqua regia (a mixture of nitric and hydrochloric acids)
dissolves platinum. The action is slow, especially if cold, but none the
less it does dissolve platinum. Some old-fashioned workers do not real-
ize this, but you can easily prove it by trying it on some clean filings;
notice how the color of the acid darkens.
As you will learn later, if your filings contain gold, you would use
aqua regia to remove it. But that is a different story.
# # # #
Q. I have read somewhere that steel, especially stainless steel, will
not dissolve in nitric acid. Yet here we are told to get rid of it with
nitric acid.
A. An excellent question, and one that can be answered best by a
simple acquaintance experiment. Take a piece of clean steel, either stain-
less steel or ordinary steel. Place it in a clean dry dish, and pour a little
full strength nitric acid over it. In most cases nothing will happen, at
least for a while. Stainless steel, especially, will remain unaffected in
what is called the "passive" state for some time. But if you sprinkle
a few platinum filings over the steel, or if you touch it with the sharp
edge of a piece of platinum scrap, the acid will go to work at once,
evolving brown fumes and dissolving the steel speedily. Sometimes a
mere touch with a glass rod will start it. Dilute acids also will go to
work at once. Chromium is another metal that will lie passive in strong
acid, but will dissolve in dilute acid, or will be galvanized into action
by the touch of a piece of platinum.
# # # #
Q. What are those brown fumes?
A. Mostly oxides of nitrogen products of the interaction of nitric
acid with the base metals. These same fumes will be seen again and
again when nitric acid is interacting with other substances.
# # # #
Q. Is there any other acid that would do this work as well as nitric?
A. You could use hydrochloric acid in some cases. It dissolves iron
and steel readily, also tin. But it does not dissolve copper or nickel, both
of which are rather apt to be present. Sulphuric acid is another pos-
sibility, but nitric acid is best.
# # # #
Q. Would it not be easier simply to melt the platinum filings and let
the flame burn out the impurities? When gold filings are melted into
a button, the dirt is burned out or goes into the flux. Certainly the oxygen
flame is hot enough to burn out most impurities.
A. Actual experiments many times repeated show that it is much
easier to remove the dirt with magnet and acid than to try to burn it out
with the oxygen flame. Even small amounts of impurities make plati-
num brittle, apparently by forming compounds such as platinum carbide.
This point will be discussed again in Chapter A of the Appendix.
# # # #
Q. You say for me to boil the filings in nitric acid, to which a little
water may be added. How much water is advised?
A. Diluted acid is more efficient per unit of acid in the sense that it
wastes less of the brown gas which is an active part of the reaction. But
it works less rapidly. However, the matter is not as simple as it appears,
because as soon as the acid starts to work it begins to dilute itself. Since
the concentration changes every second, it seems idle to be too dogmatic
about the exact concentration at the start.
The concentrated acid sometimes works so rapidly that it boils over;
this tendency may be reduced by adding water. Frankly it is hard to say
how much water to add, because this tendency to boil over is dependent
upon the size of the pieces of metal as well as upon the number of them,
and these factors change with every batch.
If you use much water you increase your acid efficiency but decrease
the speed of operation, and you have more liquid to filter and store.
Some workers use ten parts of water to one of acid; others use ten parts
acid to one part water. The results are the same in the end. We our-
selves use about equal quantities of water and acid.
# # # #
Q. You say that it is impossible to remove all the steel and iron with
the magnet that the remainder is removed by the acid. Then, why
bother to use the magnet why not let the acid take care of all the steel ?
A. When you make your acquaintance experiments with copper and
nickel, you observe that the acid will dissolve all of the copper, and all of
the nickel. If you make the same experiment with steel, you will find
that nitric acid will, under certain circumstances, dissolve steel, but that
it always leaves a small black residue, undissolved, behind it. This
residue is carbon, a constituent of all kinds of steel, appearing as a fine
powder when steel is dissolved. When you remove your steel by the
magnet, you remove both iron and carbon; whatever steel is acted on by
acid, leaves a residue of carbon mixed with the platinum filings. This
residue, if small, does no harm; but if copious it might cause trouble.
* * • *
Q. Why are we told to use chemically pure acid? Why wouldn't
ordinary commercial acid be good enough?
A. This will be answered later, in the chapter on gold scrap. In gen-
eral it may be said that the chemically pure ("C.P.") acids are more
concentrated than the others, therefore go further and cost less in the
end. But there are other reasons as well which will be discussed in
Chapter V.
Q. Should I use distilled water in this work?

A. No. Ordinary tap water is satisfactory in all the processes de-
scribed in this book. It contains impurities, but these do no harm.
* * • *
Q. Why is it they use oxygen when melting platinum? Isn't it the
same as compressed air, only in a tank? What is the advantage?
A. No, the oxygen in the tank is not just the same as compressed air.
Air, ordinary air that we breathe, is a mixture of one part oxygen to about
four parts of nitrogen, plus small amounts of other gases. That one part
of oxygen about 20 percent of the whole is the only part that produces
heat or sustains life the other 80 percent simply dilutes the oxygen.
As a result, when you compress air you compress 20 percent useful
oxygen and 80 percent useless nitrogen. It is as though you bought coal
in which one-fifth of the lumps were coal, and the other four-fifths slate.
The slate (similar to nitrogen) is useless in itself, and also tends to cool
the fire through taking up heat itself.
The stuff in the oxygen tank is almost 100 percent pure oxygen.
You know how compressed air from your blower will increase the
temperature of a flame. Adding the same amount of oxygen (at the
same pressure in pounds) will increase the heat much more. With
oxygen you will get a white heat in much less time, and the temperature
will go higher.
Ordinary gas with compressed air simply will not melt platinum, no
matter how much you use nor how long you try. The same gas, plus a
little oxygen from a tank, will melt it easily.
One reason why a compressed air and gas flame is so noisy is because
of the rush of the useless nitrogen. An oxygen-gas flame that will do
the same work will be smaller and much less noisy.
* * * *
Q. I have been advised not to melt platinum because the fumes that
come off are very dangerous. Is that true?
A. Pure clean platinum melts easily under the oxygen flame, without
unpleasant fumes, and it is perfectly safe to do this work without a flue.
See Chapter A of the Appendix. Impure platinum, however, often
produces fumes, as many substances are volatile at the temperature at
which platinum melts. Crude platinum ore, as obtained from the mines,
often contains osmium and ruthenium, and these metals when heated
strongly produce fumes that are both unpleasant and dangerous. This
point will come up again in Chapter XVII and elsewhere. However,
crude platinum is always "refined at least once, possibly several times,
before being used in jewelry or dentistry, and all the osmium and
ruthenium are normally removed. Accordingly you are not apt to be
exposed to any hazard when melting your platinum scrap, filings, or
recovered metal.
* * # *
Q. Why is it I have so much trouble with these test tubes ? They
are always breaking, or the liquid spurts out when I heat it, and
they never look clean.
A. Practice with a little plain water until you get the knack. Do
not put too much liquid in the tube an inch to an inch and a half
is enough, and experienced workers use less. Have the tube per-
fectly clean and dry on the outside before heating it. Use a small
flame such as that of a Bunsen or Meker burner, and hold the tube
in a slanting position, being careful not to let the flame touch the
glass above the level of the liquid. While it is warming shake the
tube gently so that steam will not accumulate in one part of the
tube. Remove it from the flame as soon as it boils. Hold it with
a wooden or wire clamp; do not use your bare fingers.
Immediately after finishing the test, wash the tube well with
plain hot water. Never let material dry inside it. The test tube
brush with a sponge on the end is useful, but must be kept scrupu-
lously clean itself. Invert the clean tube so that it will be dry when
you need it again.
Most pictures of chemists at work show them holding a test tube.
You will soon come to understand the popularity of this modest
piece of apparatus. The larger size about 6 inches by 3/4 inch is
preferred by beginners because it is easy to clean and less apt to
boil over, but the smaller size say 4 inches by 1/2 inch is more
economical of material.
CHAPTER IV
ARRANGEMENT OF EQUIPMENT. DISPOSING OF

FUMES
Disposing of fumes. Material for hoods. Lighting. Burners. Fans and

blowers. Water and plumbing. Storage of chemicals. Hazards.
• * # *
In Chapter III we produced a few fumes and handled a minimum
of equipment. Let us now consider both of these subjects more
deeply. When a shop has very little refining to do, as in a repair
shop, or one that handles platinum only, the space given over to
refining will be small. Once a month, perhaps, the end of a bench
will be cleared off, dishes and bottles will be brought from beneath
a cupboard, and the task will be completed in a few hours. For
such shops the main problem is fume disposal, and several inex-
pensive plans are being used. One man will place a wooden box
outside a window, and heat the dish with a small movable gas
burner, electric hot plate, or even by frequently-changed hot bricks.
Another will work only when the weather is such that he can operate
outdoors. Another will shut himself up in a small room where
there is no machinery to be damaged, open the window, and hope
that the fumes will drift out.
This book, however, is written mainly for the shop that makes
refining a routine process, not an occasional experiment. While it
offers many suggestions for the smaller shops, its main concern is
for the average user.
Many jewelry factories and dental laboratories already have a
good part of the equipment required for refining, especially the
tables, heaters, running water, possibly furnaces, and perhaps a
means of disposing of fumes.
You will need a table about 3 by 6 feet, more or less. Set it waist
high; that is easy to work at and leaves room below for cabinets,
a drawer or two, and space for storing jars or crocks.
25
Many different materials have been proposed for table tops; prob-
ably wood is as good as any. Stone tops are handsome, but unless
covered with a rubber pad they are certain to cause the fracture of
many a dish and flask. Metal tops are useless in this work.
DISPOSING OF FUMES
Every jewelry shop needs a method of carrying off acid fumes.

Acid fumes are produced when gold or silver or platinum is refined,
also when gold or silver is cut down for the preparation of plating
solutions; also when articles are boiled up in sulphuric acid for the
removal of fire coat; also when enamel is dissolved in hydrofluoric
acid; and in several other processes well known to the jewelry maker.
One inexpensive device for removing

fumes, suitable to the small establish-
ment. The fan is detachable and
portable. It should be placed low,
close to the work table.
Most shops already have a hood or flue for this purpose, but many
of them are both inefficient and unhandy. The reason is that some
of them were made in imitation of the canopies that jewelers have
over their melting furnaces.
When we recall that the job of boiling acid is quite different from
the job of melting gold, we see that the hood requirements for the
two processes must be different also.
Thus: (I) melting gold furnishes a high temperature, enough to
produce a draft up the chimney, while boiling acid does not. (2)
The fumes from acids will destroy metals, while furnace fumes do
not. (3) The furnace works admirably in a dark corner of the
shop, while the acid hood should be lighted by daylight if possible,
or by powerful and well-placed bulbs. (4) It is sometimes necessary
to let the work with acids stand overnight; the melting furnace is
ARRANGEMENT OF EQUIPMENT 27
almost never in operation overnight. This means that the acid

hood alone should be so fixed as to permit closing up at night, with
all openings fume-tight. . . And there are other differences.
What are the requirements of an acid hood?
First, it must carry off fumes.
Second, its walls must be of some material that will not scale off
and drop down, but that will last reasonably well.
Third, if it can possibly be arranged, its platform should be large
enough for carrying on all of the work that evolves fumes. Some
solutions must be left to stand, and while standing they give off
fumes; so build your hood large enough to accommodate these
solutions as well as all others. A working space of 3 by 6 feet is
excellent for a small shop, and when not actively in use with fuming
solutions, can be used as a storage space for material that might
possibly give off fumes, and so is always in profitable operation.
Fourth, it must be extremely well lighted.
Fifth, it must be so arranged that if a vessel of acid breaks or boils
over, there will be neither loss nor damage.
Sixth, the platform must be at a comfortable height for a stand-
ing worker; waist high is about right.
Seventh, the gas cocks, if any, should be placed outside the hood,
for easy adjustment without opening the hood door.
In addition, the door or window (or windows) should open and
close easily, and the whole arrangement be easily cleaned.
Some plants are lucky enough to have a tall smoke stack that
furnishes enough draft for an acid hood. But in most cases a fan
or blower must be utilized.
If only a small fume hood is available, place your work table near
it, with one gas burner thereon, and have one or two burners in
the hood.
The chemical supply houses sell hoods made of slate, glass, etc.,
that are handsome, efficient, and well designed. However, they are
apt to be expensive, and possibly more ponderous than the jewelry
factory really needs. It is suggested that a study of the catalogue of
such a supply house would be useful, as it would provide ideas for
arrangement and design.
Another suggestion is that the jeweler visit some neighboring school or

college, where chemistry is taught. Most teachers of chemistry are happy
to display their equipment to any inquirer that shows a normal interest,
and will take pleasure in explaining to a visitor how the fume-hoods

work, their advantages or faults, and so on.
Also, it is usually wise when building a hood to employ a contractor
who has had experience in such work; your neighboring laboratory,
whether scholastic or commercial, is usually happy to give you the name
of the firm that installed its hoods.
The frontispiece of this book illustrates a large hood, designed for a
busy laboratory.
MATERIALS FOR HOODS
The chemists normally use glass-walled hoods, with flues of terra

cotta or stoneware. Wire fire-proof glass is excellent. The back
walls are usually stone.
The jeweler may not wish to use such expensive materials. Glass
set in wood frames will serve excellently for years, especially if the
A standard chemical fume

hood, with glass sides.
Courtesy of the Laboratory
Furniture Co., Inc.
wood be coated with acid-proof paint. Back walls of Transite or

other fire-proof composition are moderate in cost and will give years
of service.
The platform of the fume hood, because of its closeness to the
heaters, is usually made of cement or stone. If wood be used, place
asbestos or stone slabs under the burners. Keep some distance be-
tween the heaters and the glass sides or windows of the hood; other
wise the glass may crack.
A hood with a floor-space of say 3 by 6 feet would require two
windows. These should be similar to ordinary windows with
counter-weights, running up and down in grooves, and fitting prop
erly on all sides.
The objections to sheet metal are two the fumes cause it to flake
off and drop flakes into the solutions; and sheet metal cuts off the
light.
The platform of the hood should be smooth and easily cleaned
not made of rough brick. Sometimes a dish of acid is spilled; let
the platform be made with a slight slope, and with a slight raised
edge across the front, and at the lowest corner a hole should be
bored, and underneath the hole an earthenware crock should be
stationed.
LIGHTING
The lighting of the hood is easy if its walls are of clear glass.
Since most of these operations call for close watchfulness on the
worker's part, see to it that the lighting is extremely good. So-called
"daylight" bulbs are advised, as they do not obscure the colors of
the solutions.
BURNERS
The arrangement of the gas burners is important. Run the gas

supply-pipe along the front of the hood, underneath its platform.
Bore holes through the platform, some three or four inches back
from the front of the hood, just large enough for flexible gas hose
to pass through. The gas cocks are then situated underneath the
hood, quite outside it, so that they can be manipulated without open
ing the window or exposing the hand to heat or vapors.
Most workers prefer movable heaters instead of stoves that are
piped in permanently, because movable ones are more easily cleaned.
The little stoves used in cooking are cheap and satisfactory. Unless
your insurance rules forbid rubber tubings for connections, we ad
vise them; flexible metal tubing is not so safe when acids are present.
Electric heaters are not satisfactory for work involving strong
acids and corrosive fumes; their essential element is a coil of wire
that is affected by the acids in a short time.
FANS AND BLOWERS
There are fans and blowers on the market made especially for use
with corrosive fumes; their motors are either specially enclosed and
ventilated, or skilfully placed so as to be outside the reach of the
fumes. These models are not unduly expensive.
Practically all the fumes occurring in these processes are heavy
fumes that tend to settle downward. Remind your contractor of
that when he designs your hood he will see to it that there is a
good draft at the bottom of the hood. Sometimes this is accom
The width of the slots at top

and bottom should be estab-
lished by trial.
plished by locating the fan low; sometimes a baffle plate is so ar-

ranged as to draw air through a slot near the bottom of the hood,
Usually there is another slot higher up, to take care of such fumes
as do ascend.
Remember that when the fan is working, in order to provide
good draft the windows must not be entirely closed. Leave a small
opening for fresh air to enter as the foul air is pulled out. But
there will be many times when you will wish to shut the hood up
tight and leave it so. . . Because of these considerations we some-
times find a small port or slit that can be easily closed at will.
Even with the best ventilating equipment there may be some
escape of acid fumes. Accordingly the space used for refining
should, if possible, contain no electric switchboards and as little
machinery as possible.
Let us repeat most chemical laboratories, whether in college or indus-
trial plant, are open to visitors, and the chemist in charge will be genu-
inely pleased to escort a guest through his establishment, explaining the
operation of hoods, drying ovens, steam baths, filters, and many other
devices.
WATER AND PLUMBING
Running water should be available if possible; otherwise provide

a supply jar and a good vessel for slops. Do not let the spent solu-
tions from the refining run into your settling barrels they will
injure the plumbing and may cause losses in other ways. Remem
ber that acids corrode the metals used in plumbing; therefore wash
the sink out well at frequent intervals. Though much strong acid
is used in the refining, comparatively little of it is poured down the
sink until after it has been neutralized or otherwise weakened.
Even so, frequent flushing is advised.
STORAGE OF CHEMICALS
In large shops nitric and hydrochloric acids are purchased in car-

boys holding 5 gallons; these should be supported in a cradle device
that permits tipping. The small shops will buy acid by the gallon.
It is convenient to have a smaller bottle, say 16 ounces or about
500 cc, in addition, for use when your task calls for only a small
quantity of acid, as the large bottles are not easy to pour from.
Dealers normally grant an allowance on the return of empty bottles
or carboys.
Dry chemicals should be stored in crocks, jars such as Mason jars,
or in clean wooden boxes, or left in the original container if it has
a good cover.
Filter papers should be stored in a drawer or box within easy reach.
HAZARDS
With one very minor exception (sodium chlorate) the chemicals

used in refining do not present any fire hazard. (Indeed, some of
them would make excellent fire extinguishers.) But since many of
them are corrosive in their nature, they should be stored where wan-
dering feet will not kick them, nor aimless wheelbarrows fracture
their bottles.
The refining processes do not produce explosive materials. When
your contractor is designing your fume hood, he will probably ask you
if you are apt to have explosives, that he may govern his design accord-
ingly. There is just one possibility of explosion: if gas should be per-
mitted to escape from the gas burner, it will provide the hazard that is
present in every gas cook-stove, but no more.
In general, when chemicals are spilled upon hands or clothing,

the first thing to do is to wash them off with plenty of plain water.
After that, neutralize an acid stain with a little weak ammonia, or
bicarbonate of soda. Do not attempt to neutralize until after you
have removed all acid possible with plain water! If you spill caustic
or lye on skin or clothing, again wash off thoroughly with plenty of
Ihe best immediate

treatment For~
CHEMICAL
water; after which it may be well to neutralize with a little vinegar,

lemon juice, or extremely dilute hydrochloric acid solution. Keep
a jar of clean water beside you, in which you can quickly place
your hands in case of need. Use paper towels, and if they show
stains of solutions that may contain precious metal, put them in the
"paper storage" for ultimate refining.
In Chapter XX will be found a further discussion of equipment and

its care. In Chapter B of the Appendix will be found additional informa-
tion on the personal hazards to which the refiner is exposed, with sug-
gestions on how to minimize them.
* *
CHAPTER V
THE COMMONEST CASE—REFINING GOLD SCRAP

THAT CONTAINS NO PLATINUM METALS
The first refining. Equipment and chemicals. "Acquaintance experi-

ments." Beginning work, (I) Preliminary treatments. (2) Melting into
a button, if desired. (3) Treatment with nitric acid. (4) Dissolving
gold with aqua regia. (5) Removing excess nitric acid. (6) Recovering
the dissolved gold with copperas. (7) Washing the fine gold. (8) Melt-
ing the fine gold. (9) Saving the silver. Old filter papers. What to do
if you spill acids or caustic.
* # # #
This type of scrap is found in almost every shop or laboratory
that handles precious metals, so we shall discuss fully the different
methods of refining it. First we shall describe a method of general
applicability, and in later chapters mention the possible modifica-
tions. The reader should in time acquaint himself with several pro-
cedures, then adopt the one that best fits his needs.
For our first refining, let us handle only high grade gold, such
as clean yellow dental gold, clean sprues clipped from castings, or
bench clippings and filings, or old solid gold jewelry. For this
first refining let us carefully exclude all platinum-group metals, or
alloys that may contain them; also anything that might contain soft
solder (this means most cheap jewelry, rolled and filled goods, and
jewelry that has been repaired); amalgam and mercury; all white-
metal dental scrap; and all green gold of 16 karat or higher. In
later chapters we shall take care of these materials now being
excluded.
The procedure will differ with the purposes of the worker. Thus,
sometimes he wants merely to get his gold into a clean button of
good working qualities, for use in making up new jewelry. Ulti-
mately, however, he will separate the metals to obtain fine gold and
fine silver, each free from alloy of any kind. The first case is not
true refining; it should be described as a cleaning or re-melting.
33
Filings will be contaminated with the same sorts of dirt that

were in the platinum filings described in Chapter III emery, bind
ing wire and the like, possibly plaster of paris and vulcanite. The
metal will consist of gold alloyed with copper, silver, and probably
zinc, cadmium and other base metals. The only precious metals
present are gold and silver.
For your first refining handle 50 dwt. or less; 10 dwt. is an excel
lent quantity to begin on. After the methods become familiar, you
will find it easier to handle much larger quantities.
EQUIPMENT AND CHEMICALS
We shall use all the equipment listed in Chapter III: frying pan,
magnet, sieve, casserole, funnel, filter paper, funnel support, test
tubes, stirring rod, wash bottle, glass pitcher, burner or stove, wire
gauze or thin asbestos. In addition, the following:
An evaporating dish. A small shop will want Eimer & Amend's
No. 21966 or its equivalent, 8 1/2 inches across, at $1.20. A larger
shop will also need the 10 1/2 inch size at $1.80, and probably others
both larger and smaller. A large evaporating dish is shown in the
foreground of the frontispiece; also in the hands of one of the
workers.
Graduate glass, No. 24697, 16 oz., $1.10. A large shop might want
the 32 oz. size instead; $1.80.
Pyrex glass beakers, with lip, are especially useful when handling
small lots of material. No. 17614. The 2000 cc. size (something over
two quarts) will be used in the first refining, and you will find use
eventually for other sizes as well; a nest consisting of the following
sizes 150 cc, 250, 400, 600, 800, 1000, 1500, and 2000 cc. will cost
about $3.75. The smallest of these is about the size of a teacup.
A sand bath is an inexpensive piece of apparatus; it is merely a
flat shallow pan of sheet metal, filled with an inch of clean sand.
No. 17288, 6 inches across, is about$0.15.Kitchen ware such as a pie-
pan, is also adequate, or a pan can be shaped up out of a piece of
sheet metal. A more expensive device that is used for the same
purpose is a steam bath a metal container filled with hot water,
its top carrying circular openings upon which dishes of different
sizes can rest. Sometimes the water is heated by live steam, some-
times by a gas burner underneath the bath. E. & A. No. 17352,
water bath or steam bath, is one of several models; $30.50. The
THE COMMONEST CASE 35
Graduate glass. Beaker. Glass cylinder.
One type of steam bath. Earthenware crock.
Stoneware crock. Sand bath.

price may seem high, but it will be a good investment, as you will
have many occasions to use it. Its great advantage is that with it
work can go on without supervision, and without danger of
spattering.
Large jars of earthenware, stoneware, or glass of various sizes, will
be needed. For storing dilute acid solutions, and for water or slops,
the earthenware crocks sold in department stores will serve; the
1 gallon size at about $0. 65 and the 5 gallon size at $1.50 are handy.
Strong acids destroy cheap earthenware; accordingly get stone-
ware or glass jars for strong solutions and those containing much
value, and for cases where trouble would come from admixtures of
substances that have soaked into the ware. A good acid-proof crock
is E. & A.'s No. 25637; the 5 gallon size is $6.50. Heavy Pyrex glass
jars, cylinder shaped, E. & A. No. 25592, are especially desirable
because of their transparency. Sizes and prices are as follows: two
gallons, 6 by 18 inches, $3.25; four gallons, 8% by 18 inches, $4.50;
five gallons, 12 by 12 inches, $6.00.
For melting your metal a furnace or blow-pipe is needed, also
crucibles and a flux such as borax. Rolls, or an anvil and hammer,
are useful but not always essential.
The chemicals used in Chapter III will be used here also. These
include caustic or lye, and nitric acid. In addition: hydrochloric
acid, preferably the chemically pure; a 6 lb. bottle costs about $1.20;
ferrous sulphate crystals, U.S.P., called "copperas"; these are pale
green when fresh, and should be kept in a tight jar or box, as they
deteriorate on exposure to air; they cost about $0.20 a pound in 5 lb.
lots. Sulphuric acid, concentrated; a pound bottle is about $0.25.
"ACQUAINTANCE EXPERIMENTS"
Gold. Obtain a scrap of fine gold (it will not be wasted). Also
some 14-k or 12-k yellow or red gold, and a bit of low-karat yellow
gold, preferably 6-k or lower. This can be made on the spot by
melting a little high grade gold into a button with the proper
quantity of brass.
In Chapter III we learned what nitric acid does to certain base
metals and to platinum. Repeat these experiments, using gold. The
fine gold and the 18-k gold will show no change. The low-karat
gold will first darken, then seem to dissolve, giving off the usual
brown fumes. But on closer inspection you will see that a heavy
dark powder remains undissolved. What happens is this: the base
metals dissolve; they constitute about three-fourths of the bulk;
the remaining one-fourth that is gold remains as this brown powder.
Save this material, which is valuable; it will be handled again soon.
The solution obtained will probably be of a green color, because
most low-karat gold is alloyed with copper.
Silver. Obtain a bit of fine silver, and repeat the tests with it.
Many shops have no fine silver, and will probably use Sterling or
coin silver. Nitric acid acts promptly on either, even without being
warmed, producing the usual brown fumes. Sterling and coin silver
yield a green liquid, the green color being again due to the copper
they contain. Fine silver yields a colorless solution, which, if ex
posed to light, will gradually blacken. Place it to one side in a little
bottle or corked test tube, labeled "Silver Nitrate," as it will be
needed again shortly.
Aqua Regia. This name means "royal water," and was applied by
the Alchemists to a mixture of nitric acid and hydrochloric acid,
because it was the first solvent known that would attack the "noble"
metal, gold. Prepare a little by mixing the two acids, preferably in
a graduate glass; use one part nitric acid to four parts hydrochloric.
Note that bubbles appear; in hot weather they are conspicuous. Do
not stopper the mixture or these bubbles will blow the stopper out or
possibly rupture the bottle.
Aqua Regia and Gold. Cover a small piece of fine gold with a
small quantity of aqua regia. Note the fumes; they resemble the
brown fumes of nitric acid, but are not quite the same. Fine gold
dissolves promptly, leaving no residue. Note the effect of heating.
Notice also that the color of the solution is a clear yellow not green.
Repeat, using karat gold, such as 14-k or 6-k. Note that after
the aqua regia has acted all it will, and the bit of metal has gone to
pieces, there is a residue, usually of a gray color. This is silver chlo
ride the result of interaction between the hydrochloric acid and
the silver that was present as an alloying element. Note also that
the solution is not clear yellow, but greenish yellow. The green is
due to dissolved copper.
Save these solutions, labeling them "Aqua regia gold solution."
They contain a mixture of substances, including gold chloride, cop
per chloride and copper nitrate, unused hydrochloric acid, and un
used nitric acid.
Aqua Regia and Silver. Cover a small piece of silver (fine or al-
loyed) with a little aqua regia. If desired, warm the vessel. Most
beginners are surprised at the result. They think of aqua regia as a
powerful acid which it is and so expect it to dissolve silver. It
does affect it; it changes its color and soon coats it with a firm ad
herent layer of gray or whitish substance. This is silver chloride.
This can be rubbed off, exposing a fresh surface to the onslaught of
the acid, which may in time convert all the metal into the grayish
silver chloride, but it would be a slow process.
Let us learn more about silver chloride. In a test tube take a
few drops of the silver nitrate that you saved from a previous test,
and add a single drop of hydrochloric acid. . . Notice the white
cloud that forms. Shake the test tube and see how the cloud seems
to fill the whole bulk of the liquid. Set it aside for a few hours
and notice how the white cloud settles. Place it in sunlight for a
time, and notice how the white stuff darkens, going through lav
ender and gray into black. (Incidentally, this change is the basis
of the science of photography.)
Again take a few drops of the silver nitrate solution in a clean
test tube, and this time add a single crystal of table salt. (Table
salt is sodium chloride made up of sodium and chlorine in com
bination.) As the crystal dissolves, a white cloud will again form;
this again is silver chloride, formed by a rearrangement of the mole
cules in the solutions. It is identical with the stuff formed when
hydrochloric acid reacted with silver or with silver nitrate. It can
be formed in other ways as well.
These acquaintance tests cover only a few of the many that could be
made. You will learn other characteristics of the metals as you go along,
and your own ingenuity will suggest other acquaintance tests that you
can make from time to time. Beginners usually take too large a sample;
use only a few drops of solution, and a small bit of metal, and you will
save both time and materials.
BEGINNING WORK
In outline, the method consists of the following steps:

(1) Preliminary treatments as required.
(2) If necessary, melting the material into a button and rolling it
thin, or granulating it.
(3) Removing as much base metal as possible with nitric acid.
(4) Dissolving the gold in aqua regia.

(5) Removing excess nitric acid.
(6) Recovering the dissolved gold with copperas.
(7) Washing the fine gold.
(8) Melting it.
(9) Saving the silver, if worth while.
(I) PRELIMINARY TREATMENTS
You first want to get your material in such form that it will be
easy for the nitric acid to reach the base metals and dissolve them.
The treatment will differ for different kinds of material.
Old gold jewelry, clippings, and old dentures should be ham
mered flat and rolled thin, the thinner the better. Every minute at
the rolls saves many minutes at the acids. Feed the pieces into the
rolls with a little scoop whose sides have been cut out so that they
will fit against the rolls closely; or put them on a piece of paper
and feed in paper and all.
Cut the rolled pieces up into short lengths, twisting them so they
will not lie flat in the acid; then anneal.
Filings, or "lemel" as they are often called, always contain dirt.
Lemel is derived from the French word "limaille," which means
filings. In Chapter III you were told how to sieve filings, how to
remove iron with a magnet, burn out trash, and dissolve grease and
shellac by boiling the material in caustic soda or lye solution. These
same operations are usually carried out on gold filings.
(2) MELTING INTO A BUTTON, IF DESIRED
This may or may not be necessary. It is often done with very

dirty filings, or with dental wastes that contain much carborundum
or emery; also in some other cases that will be discussed later. One
way of melting dirty filings is to mix them with their weight of
some flux, such as crushed borax glass and soda ash, with or with
out nitre, and melt them in a gas furnace, using a sand crucible.
Another and quicker method is to use an oxygen-gas torch, the kind
used when melting platinum, with a shallow open sand crucible.
If you have melted your filings into a button, the next step is to
get the button into such form that the acid can act upon it readily.
Sometimes it is so soft that you can roll it easily. Cut the rolled
strips up short and twist them so that they will not lie flat in the
acids. Roll them very thin; a few minutes at the rolls will save hours
in later steps. Anneal.
Usually it is easier to granulate the button than to roll it. Remelt
the button in a clean crucible with a little flux, and pour it while
molten into a big bucket or tub of water. Pour it in a thin stream,
Granulating gold. The molten metal is being poured into water, Later the small
pellets will be captured in the strainer.
and the molten metal will break up into fine shots or pellets when
it hits the water. Collect these pellets carefully.
(3) TREATMENT WITH NITRIC ACID
The metal, whether filings, scrap, old jewelry rolled thin, or pel-
lets, should now be placed in a casserole. Cover it with nitric acid.
Add the acid slowly, as it will tend to boil over. Cover well. Add
a little water, say one part of water to four or five of acid. Some
people use much more water.
Heavy, poisonous brown fumes appear, which are disagreeable
and bad for machinery. Therefore do this work near an exhaust
fan, or outdoors, or in a hood, or outside a window.
This treatment dissolves all the copper, silver, and base metals
that the acid can reach. The thinner your clippings and the lower
the karat, the more base metal the acid can reach.
As you learned in your acquaintance tests, nitric acid works rap-
idly on low-karat gold such as 6-k or under, which soon crumbles
to pieces. The copper and silver with which it is alloyed go into
solution, and the particles of gold are left as a brown powder.
Gold that assays 12-k or higher does not crumble. The acid dis-
solves most of the solder, and a piece of jewelry will drop apart for
that reason, but the 12-k gold will not be acted upon much. Later
we treat this gold with aqua regia, in which it dissolves promptly
and easily. As for gold between 6-k and 12-k: the effect of this
treatment will depend upon the strength of the acid, the character
of the alloy, and the thickness of the pieces.
The amount of acid needed will therefore depend upon the kind
of metal you are refining. So if your material is mostly high grade
gold, you will not use up much nitric acid in this process.
With low-grade material, the tendency to boil over is much
greater. Stir the acid well with a glass rod or porcelain spoon as
soon as you see there is no danger of boiling over. The treatment
with nitric acid must continue until there are no more brown fumes.
After the acid has acted a while, you may want to heat it to
boiling. See that the whole bottom of the dish is clean and dry,
and then place it on a gas stove, with a thin piece of asbestos be-
tween it and the flame. Or, heat the dish in a pan of sand or on
the steam bath. A thin sheet of metal, or a square of strong metal
screen (wire gauze) can be used instead of the asbestos, to protect
the dish from the direct flame.
The metal may be left in the acid overnight, or longer. But do
not let it evaporate or boil dry. Also keep out dust and dirt.
Here is a way to keep fumes from getting into the room. Set the
casserole or other vessel into a big flat pan or tray containing a half-inch
of water. Now invert over the dish a bucket, or a crock, letting it rest
in the water. A glass cake-cover is excellent for this job. This holds the
fumes in, and they are absorbed by the water.
When there are no more brown fumes on stirring, add a good

quantity of water, preferably hot, to dilute your acid, and pour it
off through a clean filter paper folded as already described. Leave
A trap for fumes.
the metal in the dish. Let the liquid run through into a big jar or
crock which we shall call the Silver Jar. This liquid contains dis-
solved silver, copper, zinc, nickel, etc.
Sometimes there is so little silver in this solution that it does not pay
to bother with it; if so pour it down the sink, being careful to flush the
sink out well with plenty of water, to protect the plumbing.
But if you wish to recover silver, store this solution in a big jar or
crock (not bottle), and keep it covered. Usually the solution is not
strongly acid by this time, partly because the acid has been killed, partly
because it has been diluted with wash water; therefore ordinary inexpen-
sive earthenware jars are suitable for storing it. Directions for saving
the silver will be given later See Chapters V and VIII.
Rinse the metal (the gold) with some hot water, filtering the
washings in with the main solution. Now add a little more nitric
acid to the metal, as before, so as to be sure you have dissolved all
the base metal and silver that you can.
With a little experience you will learn how much acid to use. Let
each lot of acid be thoroughly killed before you pour it off and add a new
lot. You will get better results in less time by using two or three small
lots of acid than one large lot.
It is not easy to calculate exactly how much acid will be needed, partly
because the process never acts with 100% efficiency, partly because you
rarely know exactly the composition of the metals you are putting into
solution, and partly because you rarely carry the acid to complete exhaus
tion. Experience shows, however, that about 4 to 6 lbs. of concentrated
acid will be used for each pound of metal that dissolves. As was pointed
out before, the acid works with less waste when diluted, but more slowly.
Finally when you are sure that everything has dissolved that will
dissolve, rinse the metal well, filtering the wash water in with
the acid.
* • * *
Under some conditions the metal remaining in the dish is nearly

pure gold. If your original metal ran 6-k or less, and if you rolled
it thin and boiled it well in nitric acid, the remaining gold may be
pretty nearly fine gold, in the form of a brown powder. But under
most conditions it still contains considerable base metal, and often
is in the form of flakes or lumps.
As we said at the beginning of this chapter, some workers desire
merely to get their gold into a clean button of good working quality,
without going through the final process of obtaining fine gold free
from all alloy. If such is your purpose, wash the remaining metal
thoroughly, to remove the dissolved copper, etc., then dry it and
melt it.
But to obtain high quality fine gold, you must carry your metal,
in most cases, through this next process.
(4) DISSOLVING GOLD IN AQUA REGIA, IF DESIRED
For every Troy ounce of metal in the dish (more or less), mix
up at least
I fluid ounce nitric acid,
4 fluid ounces hydrochloric acid.
Mix the acids in a graduate glass. Use "Technical" or C.P. Acids.
Use enough to cover the metal well, but add it slowly, a little at a
time, as it soon boils vigorously and might boil over.
Poisonous fumes appear again, different from the nitric acid fumes
but just as bad. In cold weather you may have to warm the acid to
start it to work. Stir now and then with a glass rod or porcelain
spoon, so that the acids can reach all parts of the metal mass.
You want to use just enough aqua regia to dissolve your gold,
but no more. It is better to use three or four small lots of mixed
acid than to put on one large lot in the beginning. Let each lot be
well killed before adding the next lot.
As each lot of acid is killed, pour it off into a big pitcher or crock.
Do not filter it yet.
Toward the last you may want to heat your acids. If your metal
is in small or thin pieces this will not be necessary. A big funnel,
inverted over the dish, will prevent spattering.
Whatever silver is present in the metal is changed by aqua regia
into the mud or scum of silver chloride. It dissolves to only a minute
degree. If your metal is in thick pieces, the scum will cover the
metal and keep the acids from reaching it, so stir often, and crush
the lumps with the rod or spoon, to loosen this scum. The more you
stir, and the hotter the acids, the sooner the work will be done.
Finally there will remain a fine mud of silver chloride in the bottom
of your casserole, and all the metal will be dissolved.
Rinse this mud to free it of dissolved gold, adding the washings
to your main solution. If the mud looks clean, put it into your Silver
Jar; but if it is dirty, with cinders, emery, etc., it may be better to put
it into your sweeps.
(5) REMOVING EXCESS NITRIC ACID
You were told to use no more aqua regia than was required. Be-
ginners always use too much. Experienced workers often do the
same thing, especially when in a hurry, because fresh acids work
more rapidly than those that are partly exhausted.
Sooner or later you must get rid of the unused nitric acid. Most
workers think it pays to do so right away, as follows:
Put the aqua regia solution into a big evaporating dish. It will
contain some sediment, mostly silver chloride; this does no harm at
this point, so do not filter it yet. Now add a little sulphuric acid,
stirring all the time, and working slowly, as it might spatter. Use
about an ounce or less of the sulphuric acid to each quart of liquid.
Now boil the solution down gently, to a syrup, being careful not to
spatter. The purpose of the sulphuric acid will be explained later.
This evaporation is a tiresome process. It goes rapidly at first, but
soon the liquid begins to spatter and you must lower the flame. Do
not cover the dish, as that merely impedes evaporation. A sand bath
or a steam bath is useful; these have already been described.
When using a sand bath, start work by pushing the sand to the
sides of the pan, so that the heat can quickly reach the evaporating
dish; later, as evaporation has progressed, scrape the sand down

underneath the dish, to lessen the tendency to spatter.
The advantage of the steam hath is that it does not cause spatter-
ing, and a dish can be left on it without supervision, without danger
of overheating.
Let the evaporation continue until the liquid becomes syrupy;
then slowly add a little full-strength hydrochloric acid. Brown
fumes will bubble off; this is the unused nitric acid being expelled.
Usually this evaporation must be repeated at least once, maybe twice
evaporating to a syrup and adding a little hydrochloric acid in
order to get rid of all the unused nitric acid.
Do not carry the evaporation far enough to form a hard crust.
One reason for the sulphuric acid is that it hastens the expulsion of
the nitric acid, and when it is present there is less tendency for a
hard crust to be formed. The sulphuric acid also serves another pur-
pose to get rid of lead which is described fully in a later chapter.
When you are sure that the excess nitric acid has been driven off,
let the dish cool, wash down its sides with a little water (use the
wash bottle) and get the solution into a jar or pitcher, preferably
a tall one.
Now dilute the solution with three or four times its bulk of water,
preferably hot, and let it stand (for some days if possible) for the
silver chloride to settle. The solution must be perfectly clear and
separated from the sediment before you go to the next process.
Usually it settles well in a couple of days, and you can pour off the
bulk of the solution, filtering only the last part.
The sediment of silver chloride should be collected and washed,
and put in the Silver Jar (or in the sweeps if very dirty). If there
was any lead in your original material, the silver chloride will be
contaminated with lead salts.
The clear solution contains dissolved gold, copper, zinc, etc. If

there is much copper or nickel in it, it will be greenish brown.
When gold is the main element it is yellowish brown. If it is not
perfectly clear and free from sediment, let it stand another day or
two for the sediment to settle; then pour off again into a clean wide-
mouthed vessel of generous size.
(6) RECOVERING THE DISSOLVED GOLD WITH COPPERAS
If possible, have the solution in a transparent vessel a glass jar,

beaker, or pitcher. If it is transparent you can see what goes on and
understand the process better. Later, when you handle large quan
tities, you will carry out this process in a chemical stoneware crock,
but while learning the processes, use transparent vessels whenever
possible.
In another vessel, dissolve in warm water about a pound of green
copperas for each 50 pennyweight of fine gold that you expect to
recover. Use fresh green copperas, with about a quart of water to a
pound. Add a little hydrochloric acid, enough to make the copperas
solution a clear green.
Pour the copperas solution slowly into the gold solution, and stir.
Metallic gold will appear as a dark cloud. Let this settle some hours,
or overnight, and you should find your gold in the bottom of the
dish, as a heavy brown powder.
Sometimes an impatient worker will cut short the evaporation described
in Section 5, and will fail to remove all the excess nitric acid. When this
has happened, bubbles will arise on the addition of the copperas to the
gold solution, sometimes so freely as to make the mixture boil over. So
add the copperas slowly at first. These bubbles are the same brown
fumes that you notice when nitric acid interacts with anything. In this
case the nitric acid is reacting with the copperas. Gold will not come
down until all the nitric acid has been disposed of. The solution will turn
dark, but no gold will precipitate. When this happens you must do
one of two things to get rid of this nitric acid.
One way is to add an extra quantity of copperas and hot water, stirring
constantly. Keep on adding copperas until the mixture stops fuming and
the gold comes down. Before trying this method, however, take out
about on ounce of the solution and get a rough idea of how much cop-
peras is needed. The other way to get rid of nitric acid is to evaporate
the whole solution down to a syrup again, and again add hydrochloric
acid, as in Section 5.
After a while, as we said, you will find the fine gold in the bot-
tom of the dish as a brown powder. Test the liquid to see if you
used enough copperas to throw down all the gold that was in solu-
tion; take a few drops of it in the porcelain spoon, and drop in a
clean green crystal of copperas. Watch the crystal closely; if nothing
happens in two minutes, you used enough copperas. But if a dark
cloud forms around the crystal, add more copperas and hot water to
your main solution, and let it settle again. You may have to do
this more than once.
(If your original material had contained platinum or palladium,
these metals would now be in this solution. If there is any possi-
bility of their being present, test the solution for these metals. Full
directions are given in Chapter IX.)
When, after satisfactory tests, you are sure that all the gold has
been precipitated, and that the liquid now contains no dissolved
precious metal, pour or siphon it off and throw it away. You will
have to filter the last part of it. Finally wash the mass of gold
powder into the casserole, as it must now be cleaned of the copperas
solution. In this operation the wash-bottle is helpful.
(7) WASHING THE FINE GOLD
Fill the casserole twice with hot water, letting it settle, and pour-
ing off (or filtering) the liquid, which is thrown away. Now and
then a thin film of gold floats on the surface; this is so light that it
does not amount to much. You can avoid this by having your
Using the wash-bottle to direct a Washing a powder, such as the

thin stream against a certain spot. gold powder, while it is in the funnel.
Hold your left thumb against the Wash down the sides of the funnel,
stopper. This also shows how to so as to get the powder into a com-
wash heavy materials, such as fine pact clean lump.
gold powder, out of a dish.
casserole quite clean before using it. Wash it with soap, rinse well,
and do not touch the inside with your fingers. It is the grease from
your fingers that floats this film of gold.
Next, cover the powder with hydrochloric acid and heat it to
boiling. This dissolves the light yellow mud that is mixed with it;
(it comes from the copperas), which must be washed out well.
Stir. Pour off this acid and throw it away. Add a little more acid,
to be sure your gold powder is clean.
Then wash off the acid with plenty of hot water.
Finally get the gold powder into a clean filter paper, washing it
down into a lump. See illustrations.
Let the gold powder dry in the funnel. After it has dried partly
you can lift the paper out of the funnel, spread it out on a plate in
a warm place, and it will dry quickly. You will note that clean gold
powder dries more quickly than powder containing dirt of any kind.
(8) MELTING THE FINE GOLD
Place the gold when dry in a clean sand crucible that has been
rubbed well inside with borax. Put borax on top, and melt as usual
in a gas furnace, or with a Hoke Phoenix oxy-gas torch. When
molten, add a pinch of nitre or saltpetre from time to time. This
burns out impurities. Let the gold cool in the crucible. Then break
the crucible and remove the button.
If the refining is properly done, this button will be of high quality
gold. If it does not roll perfectly, remelt it in a clean crucible with
borax and more nitre, as before. Or, add some ammonium chloride,
or a little bichloride of mercury. Do not inhale the fumes.
Pour into an ingot if desired, or let cool in the crucible.
Your gold should now be ready to be used again, or to be sold as

fine gold. Sometimes, either because of faulty refining, or because
of the presence of certain impurities that this method does not re-
move (these will be discussed later), the gold button is not as pure
as you wish. If this happens, turn to Chapter XV on the special
repurification of gold.
(9) SAVING THE SILVER
Going back to the Silver Jar, which contains silver, copper, etc.,
dissolved in nitric acid: add some table salt, dissolved in a little
water, roughly 1/2 ounce or more of table salt for each ounce of
silver in solution. A white cloud of silver chloride will form. Let
it settle.
To see if you used enough salt, add a little more salt after waiting
until the cloud has settled some. If more white cloud forms, add
still more salt. Hydrochloric acid can be used instead of the table
salt. It forms silver chloride in the same manner.
After a satisfactory test, in which no new clouds form, let the
chloride settle well. Let it stand in the dark, preferably for sev-
eral days.
Then pour off the clear liquid and throw it away.
Leave the chloride in the crock, covered to keep out dust, and
next time you refine pour the nitric acid solution in on top of it.
Add hydrochloric acid or salt as before, let settle, test, pour off the
liquid and throw it away as before. Repeat this until you have a
pound or two of chloride. Keep it moist.
It usually pays to let this silver chloride accumulate for some
time before refining it. We shall therefore leave it now, returning
to it in Chapter VIII.
OLD FILTER PAPERS
These sometimes contain considerable value. Tear off and throw

away any paper that is plainly worthless, and spread out the rest to
dry. When you have several papers, burn them in a frying pan;
pour on gasoline and burn until they will burn no more. If the
remains are very dirty, put them in your sweeps or your paper
storage. If fairly clean, add them to your next lot of filings.
WHAT TO DO IF YOU SPILL ACIDS OR CAUSTIC ON HANDS OR CLOTHING
Wash them at once with much water, and little or no harm will be
done. With acids: after you have washed off all the acid possible, it
may help to put on a little very dilute ammonia water, or a little bicarbo-
nate of soda. With lye or caustic: after washing well with plenty of
water, neutralize with a little lemon juice or vinegar, or some extremely
dilute hydrochloric acid, say three drops of acid to a glassful of water.
However, if you use plenty of water at once, you will not need any neu-
tralizing treatment. In Chapter B of the Appendix, this subject is dis-
cussed in greater detail.
* • • #
As we said at the beginning of this chapter, there are several ways

to refine this kind of scrap. In the next chapter we shall discuss
some of the alternative methods.
CHAPTER VI
MORE ABOUT GOLD. ALTERNATIVE METHODS
(I) Alternative preliminary treatments. (2) Melting the material into

a button. Quartation. (3) Removing base metals with acid. (4) Dis-
solving the gold in aqua regia. (5) Evaporating off the excess nitric
acid. (6) Recovering the dissolved gold by alternative methods: sulphur
dioxide; oxalic acid; sodium nitrite; metallic precipitants copper, iron,
zinc. (7) Washing and (8) Melting the fine gold. (9) Saving the silver.
Summary. Questions and Answers.
* * # *
In Chapter V we described a standard way to handle high grade
gold scrap. Briefly the steps were as follows:
(1) Preliminary treatments as required.
(2) If necessary, melting the material into a button, and rolling it
thin or granulating it.
(3) Removing as much base metal as possible, using nitric acid.
(4) Dissolving the gold in aqua regia.
(5) Evaporating off the excess nitric acid.
(6) Recovering the dissolved gold with copperas.
(7) Washing the refined gold.
(8) Melting the refined gold.
(9) Saving the silver if worth while.
In this chapter we shall describe several alternative methods that
are frequently employed. The student should be thoroughly familiar
with Chapter V before attempting to use or even to understand the
present chapter, which indeed may well be skipped for the present.
He should carry through at least two small lots of metal according
to the procedures of Chapter V, then try those outlined below. With
that background of experience he can adopt for regular use the
method that best suits his material and purposes.
In this chapter also we shall handle only high grade gold scrap
or filings or dentures; no platinum-group metals, soft solder, rolled,
filled or plated gold, amalgams, green gold of 16-k or better, or
white-metal dental alloy.
50
MORE ABOUT GOLD. ALTERNATIVE METHODS 51
Since we are assuming that the reader has already acquired some
facility in handling acids and equipment, our descriptions will be
less detailed than in earlier chapters.
(I) ALTERNATIVE PRELIMINARY TREATMENTS
In Chapter V you were told to sieve your filings, then to remove

iron with a magnet, then either to burn out the trash or to boil in
caustic solution. Some workers burn them first, especially if they
are very linty, then sieve, and finally use the magnet. Thus the
order of procedure will be governed by the nature and proportions
of the dirt to be removed.
At this point many workers weigh the filings; since it would be mis-
leading to weigh them while wet, such operators will burn the trash
rather than use the caustic treatment, which leaves the filings damp.
Considerations such as these often determine the nature and order of
the treatments.
Before using the magnet, give your filings time to cool completely
after being burned; iron has no magnetism when hot. Remember
also that some types of white gold are attracted to a magnet. In
any case, there is always the probability that some particles of gold
are caught up by the magnet, mixed with the iron particles. Some
workers place all these particles in a jar of acid, usually dilute hydro-
chloric or sulphuric acid; the iron and steel dissolve, and at inter-
vals the remaining gold is collected and refined.
Some workers like to pan their filings, the way miners pan dirt when
looking for gold dust. If there is a chance that the stuff contains small
pearls or other objects that could be harmed by heat, pan it without burn-
Gold washing pan. Eimer

and Amend No. 29698.
ing it. Some workers boil in caustic, or burn, to get rid of grease, and
then pan. It takes only a little practice to get the trick of panning. Shift
the flat pan round and round, with water, letting the water and the
lighter particles slip over the edge, while the heavy bright metal collects
in the bottom.
When burning paper and trash out of filings, it hastens matters to add
a little gasoline or alcohol.
(2) MELTING THE MATERIAL INTO A BUTTON. QUARTATION
In Chapter V we said that this step may or may not he necessary.

But it is usually employed when much sand or emery is present.
There are two other cases where it is advised: when there is good
reason for reducing the metal to 6-k or lower, and when you have
green gold that is 16-k or higher. (Green golds will be discussed
again in Chapter VII.)
You will recall that when 6-k gold is treated with nitric acid, it
promptly goes to pieces, the silver and copper dissolving, and the
gold remaining as a dark powder. This gold powder, after thorough
treatment with nitric acid, is frequently of high purity. Many shops,
therefore, make it a practice to melt up all their gold scrap, adding
enough other metal to reduce the gold content to 25% or less—
that is, to 6-k or lower.
In that way they avoid steps 4, 5 and 6 dissolving in aqua regia,
evaporating off the excess nitric acid, and precipitating with copperas.
The disadvantages of this plan are that it uses up much more nitric
acid, and calls for large crocks, etc., for handling the bulky solutions that
are required. Also, for reasons that will be clear in Chapter XII it
is not suitable when platinum-group metals are present. But when
platinum-group metals are absent, it offers certain advantages, and the
student should become familiar with it.
It is necessary to decide what metal to add, and how much
of it. Many shops add silver. Since they recover it later on and
use it over and over again, this involves comparatively small expense.
Other shops use copper or brass; this does not cost much initially,
but since it is not feasible to recover the copper or brass for re-use,
the expense may be considerable in time. Rolled- or filled gold scrap,
or electroplated gold scrap, and old jewelry made of these materials
are often used for this purpose, and if you have such scrap on hand
you will want to use it. However, for this first refining, we ask you
to use silver, or clean copper, or clean brass. Later, after you have
studied Chapter VII, which covers rolled and filled stuff, we advise
you to make up a button with it. But for reasons that will then be
clear, we ask you now to use silver, or clean copper, or clean brass.
You must next decide how much metal to add.
You want the resulting button to be about 6-k (25% gold) or
less; the other 75% to be metal that will dissolve in nitric acid.
Suppose your original scrap ran about 15-k on the average and
weighed 100 dwt. Fifteen-karat gold is 15/24 or 5/8 gold of
100 dwt. is 62 1/2 dwt. The rest, or 37 1/2 dwt., is metal that will dis
solve in nitric acid. You want to make up a button that will be
one part gold to three (or more) parts metal that will dissolve in
acid; 3 times 62 1/2equals 187 1/2. Your metal already contains 37 1/2
dwt. soluble metal, so, if you add 150 dwt. of silver or copper or
brass, your button will be one-quarter gold.
(This practice of making a button that is one-quarter gold, is called
quartation or in quart ation and has been employed for many years in
assaying and refining.)
If you have doubts as to the fineness of the original material, add

more silver or copper, rather than less.
The resulting button should be rolled thin if possible, and cut up
into small pieces, which are then twisted and annealed. Usually
the metal is too brittle to be rolled, so it is melted again and
granulated.
( 3 ) REMOVING BASE METALS WITH ACID
In all cases, the object of this step is to remove as much as pos-

sible of the base metals and silver. Nitric acid is used in the majority
of cases, especially in the small jobs. When very large lots, running
100 ounces and over, are treated, sulphuric acid may be chosen; but
it is suitable only when the added metal is silver, when the copper
content of the resulting button is 10% or less, and when the lead
content is low. Sulphuric acid is cheaper than nitric, but more
unpleasant to handle. Since your purpose at this time is to learn
the various modifications of these processes, we suggest that you
now take a button in which silver is the added metal, which con-
tains little or no lead, and in which copper runs not over 10%, and
treat it with sulphuric acid instead of nitric acid.
A small lot would be treated in a casserole or flask. Large lots
are treated in cast iron pots; sulphuric acid, so long as it is concen-
trated, has surprisingly little effect on cast iron. Heavy fumes are
produced, differing chemically from those yielded by nitric acid,
but equally offensive and dangerous. The mixture should be
stirred often.
From two to three ounces of concentrated sulphuric acid are
needed for each ounce of metal (more or less), and the mixture is
heated cautiously until action begins.
When the action of the acid ceases or becomes very slow, let the
vessel cool. This takes some time, because boiling sulphuric acid
is much hotter than boiling nitric acid. Then cautiously ladle the
liquid into a barrel or big crock of water. Pour this acid into the
water, slowly. {Never pour water into sulphuric acid it would
instantly become steaming hot and spurt out. Instead, pour the acid
slowly into the water, giving the mixture time to cool.) Leave the
remaining metal in the vessel and treat it with a small lot of fresh
acid, breaking up the lumps. Finally scrape it into a porcelain dish.
Do not put water into the iron pot, because dilute sulphuric acid
attacks iron rapidly.
The solution in the barrel contains dissolved silver, copper, and other
base metals in the form of sulphates. Often when ladling it into the
barrel of water you will see a white cloud form, which may or may not
redissolve as you stir it through the water; this is silver sulphate. Silver
sulphate dissolves scantily in water. That is, if much water is present,
all the silver sulphate may dissolve. Since the quantity of silver present
may be considerable, your next task is to save it. As before, you add
either hydrochloric acid or a solution of table salt, and the insoluble
silver chloride will be precipitated as a white cloud, which does not
redissolve on stirring. Let it settle, then test the top liquor with another
drop of salt to see if all the silver is down. If not, add more salt, and
let it settle again.
After a satisfactory test, pour off the liquid and throw it away,
as it should contain nothing of value. Methods of recovering the
silver from the silver chloride are given in Chapter VIII. (If your
original material had contained platinum or palladium, this liquid
would be of value, but that is another story.)
Returning now to the gold in the porcelain dish: wash it well,
to remove all sulphates of copper or silver. As we said before, this
gold may be as pure as you require. The exact purity will depend
mainly upon the nature of the metal you started with, partly on
your skill and patience, and partly on the composition of the button
you obtained by inquartation. Thus, if the gold in your button is
too high, say 30% or so, the gold you recover will be of doubtful
purity. (That is why some workers make up their buttons with a
gold content of only 10% or 15%, although this means added ex-
pense in acid and time, and rarely pays.) Note also that if your
original material contains much lead, sulphuric acid may not remove
all of it.
When highest purity is required, this gold must be dissolved in aqua
regia and precipitated again. Chapter XV on the special repurification
of precious metals tells you more about this.
(4) DISSOLVING THE GOLD IN AQUA REGIA
When you handle large quantities of metal, you will use larger
vessels, but the processes will not differ in principle. Filtering a
big jar of solution is slow work, but it can be hastened by such
devices as the filter pump and the Blichner funnel. These are de
scribed fully in Chapter XX. Hot aqua regia works more rapidly
than cold, but you are advised not to let it boil, because boiling
drives off the acids and wastes them; gentle heating on the steam
bath is better.
(5) EVAPORATING OFF THE EXCESS NITRIC ACID
Here again the steam bath is advised, especially for large quan
tities of material. Sometimes this evaporation may go too far, and
brown metallic gold will bake out onto the dish. You can easily
dissolve this baked out gold (without adding any nitric acid) in
this way: cover it with a few drops of full-strength hydrochloric
acid, then slowly add a few drops not more of a solution of
sodium chlorate in warm water. A very small amount of this so
dium chlorate solution will do the work; add it a drop at a time,
tilting the dish to wash the dissolved gold into the syrup. Stir
well, as this mixture of chlorate and hydrochloric acid can form an
explosive gas, especially when cold, but if your dish is warm, and
your chlorate solution is warm, and if you work slowly, little if any
of the explosive gas will collect.
(6) RECOVERING THE DISSOLVED GOLD BY ALTERNATIVE METHODS
In Chapter V you were told to use copperas (ferrous sulphate) to

precipitate the gold that was dissolved in the aqua regia. Copperas
is not the only reagent that will accomplish this, though it is the
most popular one. It is cheap, easy to use, obtainable everywhere,
and offers no hazards; in time it will lose its strength if exposed to
air, but with ordinary care it lasts for years. It is the small refiner's
standby.
There are several things to be considered when choosing a precipitant:

its cost; the ease of handling; and the completeness and cleanness with
which it works. That is, you want it to bring down 100% of the dis
solved gold, and nothing else. This final consideration the cleanliness
with which it works is one that has been somewhat neglected, with the
result that some recovered gold is not pure enough to use.
Reagents that we shall describe here are: sulphur dioxide gas;

oxalic acid; sodium nitrite; and the metallic precipitants such as
copper wire, sheet iron and zinc. This does not complete the list
of possible precipitants, but covers those in general use. Each has
its advantages and its disadvantages.
SULPHUR DIOXIDE GAS
A new way of precipitating gold from aqua regia solution employs

sulphur dioxide gas. This method is destined to become very popu-
lar, especially among large users.
Sulphur dioxide gas is dispensed in steel cylinders, something like

oxygen tanks; several sizes are on the market. It is one of the gases used
in refrigerators such as Frigidaire. In the past it has not been generally
available, but its sale is becoming common, and it is now available in
most parts of the country at a price within the reach of even the small
user. In most cases the dealers sell the tank with its contents, and then
will refill the tank or exchange it when empty for a full one, at a rea-
sonable charge.
When bubbled slowly through your regia aqua solution, it throws
down the gold as that same brown powder that you obtain with
copperas.
It is normally much cheaper in operation than are other methods
equally suitable. Even with beginners the gas is cheaper, and time and
labor costs are less. It is also less bulky than copperas or other materials.
Thus, suppose you have two gallons of solution containing gold; if using
copperas you would dissolve the copperas in two or more gallons of
water; that means four or more gallons to handle and filter. Other
reagents, such as the oxalic acid that is shortly to be described, also involve
bulky solutions, which sometimes require heating, an operation which
involves further time and expense.
Sulphur dioxide gas is simply passed into the original two gallons
of solution. This saves floor space, equipment, and time especially
when filtering. The gas keeps indefinitely until used.
Most workers report that they get purer gold with sulphur dioxide than
with any other precipitant. Even when soft solder is present, the fine
gold obtained is apt to be of high quality. In other words, sulphur diox-
ide is not likely to "carry down" other substances with the gold, and the
chemist speaks of it as giving a "good separation."
Sulphur dioxide has two disadvantages: first, the initial cost is high,
since you have to buy the first cylinder, as well as its contents. After
that, the expense is low, as you return the tank to be refilled, or exchange
it for a full tank. The average medium sized shop will find that a tank
will last for a year or more, so the bother of changing tanks does not
come often. The second great disadvantage is that the gas has an
extremely evil smell the smell of burning sulphur. If it escapes from
the tank it is stifling and dangerous, also it tarnishes shop machinery,
jewelry and other metal objects. When using it you must work near a
fan or under a hood. However, you have already had some experience
in handling dangerous fumes, and this is no worse than those you are
already familiar with. Store the tank in a safe place and handle it intel-
ligently, and it will be no more dangerous than your illuminating gas.
HOW TO RECOVER DISSOLVED GOLD WITH SULPHUR DIOXIDE
As in Chapter V, you must remove the excess nitric acid by evap-

orating the solution to a syrup, adding hydrochloric acid, and boil-
ing down again, finally diluting with considerable water and getting
rid of all sediment. This procedure is necessary when you use cop-
Precipitating gold with sul-

phur dioxide gas.
peras, or sulphur dioxide, or any other precipitant. Be sure that

the resulting solution is perfectly clear and free from sediment or
cloudiness.
The solution should now be in a jar or crock, preferably one that
is tall rather than wide. A glass vessel is pleasant to work with,
since it permits you to watch the operation. Simply bubble the gas
through the solution, a bubble at a time, as shown in the sketch.
Use tubing of at least 1/4 inch inside diameter; smaller tubing might
become clogged by the gold. Fasten the tank in an upright posi-
tion, so that it cannot tip over; the gas within is really in liquefied
form, and if the tank were on its side, the liquid would run out,
which would be wasteful and very unpleasant.
At first the bubbles of gas will all be absorbed. Let them pass in
slowly, say sixty to a hundred bubbles a minute. After five or ten
minutes let them run more slowly. The solution will soon become
dark with the precipitated gold. Stir gently, to bring all parts of
the liquid in contact with the bubbles.
The length of time required will depend upon the quantity of solution
you have, the amount of gold it contains, also whether you removed all
excess nitric acid or not. Normally it takes about 45 minutes to throw
down 10 ounces of gold contained in two gallons of solution, assuming
that you had removed practically all the excess nitric acid.
The beginner sometimes finds that the gold is very slow to come down.
Usually the trouble is that the liquid contains too high a concentration of
acid; that is, maybe he failed to evaporate off all his excess nitric acid;
or maybe he used too much sulphuric or hydrochloric acid in previous
steps. One remedy is to add water to dilute the liquid and reduce the
concentration of acid; another is to add pinches of sodium carbonate, a
very small quantity at a time, stirring well. When the acid concentra-
tion is right, the dark cloud of precipitated gold will appear promptly.
Let the powder settle as usual; take out a few drops of the liquid
and test it to be sure that you have used enough of the sulphur
dioxide to throw down all of the gold; place a few drops in a test
tube, heat it gently to boiling, and then add a clean green crystal
of copperas and watch to see if any more gold powder appears.
When you are sure that all the gold has been precipitated, let it
settle, pour off the solution as usual, and collect the gold powder,
and wash it with plenty of hot water, then with hot strong hydro-
chloric acid (to remove dissolved copper, etc.), then rinse well, dry,
and melt as usual.
If your material had contained no platinum metals, the solution
will now be worthless and ready to throw away.
PRECIPITATING GOLD WITH OXALIC ACID
Oxalic acid is often used to precipitate gold from aqua regia solu-
tions, but since it is relatively expensive, it is usually saved for occasions
where the purpose is to obtain specially pure gold; that is, when the gold
you start with is already nearly pure enough for your purpose. Thus,
suppose you had your gold content down to 6-k or less, had dissolved
out the base metals with nitric acid, and had obtained a gold residue that
was almost pure enough to be used again. This in turn you had dis-
solved in aqua regia, and you are now ready to re-precipitate it. This
aqua regia solution therefore contains only small amounts of other metals.
This is a case where oxalic acid might well be chosen as the most suitable
reagent.
You will observe that copperas and sulphur dioxide were employed on
solutions that contained (in addition to dissolved gold) much dissolved
copper, zinc, and possibly nickel and other metals.
Oxalic acid has this advantage over copperas: it does not introduce any
iron or other material which, if melted in with the fine gold, could hurt
its quality. The disadvantages are its cost, and the fact that when using
it you must heat your solutions, both the aqua regia solution and the dis-
solved oxalic acid that is added to it, preferably to the boiling point. Hot
solutions, especially when bulky, are awkward to handle, and may be
dangerous. For these reasons, as we said, oxalic acid is usually reserved
for jobs where the quantity of solution is relatively small, or when you
are working with a solution that contains little or no dissolved base
metals.
The method of using it, therefore, will be described in Chapter XV
on the special purification of gold. You should become acquainted with
this method, even though you do not plan to use it regularly.
PRECIPITATING GOLD WITH SODIUM NITRITE
This method is mentioned in British Patent 380,324, February 12,

1931. One of its advantages is that this reagent, if melted in with
the fine gold, will cause no trouble; sodium nitrite is chemically
related to sodium nitrate, and the latter, often called Chile saltpetre,
is used as a flux and purifier when gold is melted. Sodium nitrite
is not expensive, but it may weaken on long standing and eventually
lose its power to react with gold solutions.
The method of using it is simple practically the same as for
copperas. The aqua regia solution is evaporated to a syrup as
usual, taken up with water and a little hydrochloric acid as usual,
and filtered clear as usual. Do not use too much hydrochloric acid;
6o REFINING PRECIOUS METAL WASTES
the final clear and filtered solution should contain not more than
about 5% of total acids. This includes the sulphuric acid you added
to your aqua regia. If too much acid is present, the acid will react
with the nitrite and waste it, producing reddish fumes, but no gold
will come down. The nitrite may be added in solid form, a little
at a time, or may first be dissolved in water, as convenient. The
gold appears as an attractive yellow or brownish powder, which
settles promptly under normal conditions that is, when the solu
tion is not too concentrated. If it seems slow to settle, add water.
Finally, after proper tests, pour or filter off the solution, discard
it, and wash the gold as usual.
METALLIC PRECIPITANTS COPPER, IRON, ZINC
One of the oldest ways to get gold out of solution is to hang bunches
of copper wire in the jar. The copper slowly dissolves and grains of
metallic gold appear in its place, clinging to the remaining copper. (The
process is often spoken of as "cementing the gold out," as the same
method can be used to take silver out of solution, and the resulting grains
of silver bear a fancied resemblance to cement and are called "cement
silver." The phrase was then extended to cover the similar process when
applied to gold.) At intervals the grains of gold should be knocked off,
to expose a fresh copper surface.
The process is slow, and you must test the liquid now and then to see
if the copper has had time to remove all the gold. When you are sure
that this has taken place, the liquid (which by now is deep green with
dissolved copper chloride) should be dipped or filtered off and thrown
away. The remaining metal consists of gold and probably bits of unused
copper wire. It should be washed, boiled in nitric acid to remove all the
copper, and finally washed, dried and melted as usual.
This method has the disadvantage of being slow, and also more expen
sive than other methods except in those rare cases where you can use
the dissolved copper in some manner or other. Its less obvious disad
vantage is this: if any platinum or palladium should be present, they will
come down with the gold, in metallic form also. This means that your
"refined" gold will simply have to be refined over again in order to save
these two metals and provide you with pure gold. In this respect the
method differs conspicuously from copperas, sulphur dioxide, sodium
nitrite, or oxalic acid methods; you will recall that when these latter are
used, any platinum and palladium will remain in solution while the
gold alone is thrown down. This difference is important.
Iron and zinc may be used in the same way; wire, foil, shavings, or
granulated metals are used. The reaction is similar to that between

copper wire and dissolved gold; the iron or zinc dissolves and the metallic
gold is cemented out in its place. Iron and zinc work somewhat more
rapidly than copper. The disadvantage here is the same as with copper;
if platinum or palladium be present, they also will come down with the
gold. In addition, whatever dissolved copper is present will also come
down. The result is a mass of granular particles of gold, copper, plati-
num, and palladium, mixed with any iron or zinc that may not have
been used up in the reaction.
These methods, therefore, are not true refining methods, since the
gold they bring down is mixed with quantities of other metallic par-
ticles. Accordingly "cementing out" is done only when the purpose is
simply to recover precious metal, without regard to its purity. Instances
of this will come up in later chapters.
(7) WASHING THE FINE GOLD
In Chapter V we washed the fine gold powder first with water,

then with hydrochloric acid, then with water again. Some people
use two acid washes. Sometimes sulphuric acid is used instead of
hydrochloric, thus: cover the powder with cold water and add a
few drops of acid, slowly, and stir well. (Do not add the water to
the acid.) Work slowly or it may spatter. Heat to boiling, then
wash well with plain water to remove the acid, following the routine
suggested in Chapter V.
(8) MELTING THE FINE GOLD
(9) SAVING THE SILVER
Melting the fine gold and saving the silver will be discussed fully
in chapters of their own Chapter XV, Chapter A of the Appendix,
and Chapter VIII so no more will be said about them now.
* # * #
To summarize:
High grade gold scrap is treated as follows:
(1) Preliminary treatments: sieving, burning, panning, treatment
with caustic and with the magnet, as required by circumstances.
(2) Melting into a button if necessary. In some cases a metal
such as silver or copper or brass is added in sufficient quantity to
reduce the gold content to 25% or less. Rolling the button thin,
or granulating it.
(3) Treatment with nitric or sulphuric acid, to remove base
metals and silver.
(4) Dissolving the remaining gold in aqua regia, if desirable.
(5) Evaporating off the excess nitric acid.
(6) Recovering the dissolved gold, using copperas, or sulphur
dioxide, or, in certain cases, oxalic acid, sodium nitrite, or a me-
tallic precipitant.
(7) Washing the fine gold.
(8) Melting the fine gold.
(9) Saving the silver when advisable.
In Chapters XXI and XXII of this book, other alternative methods,
suitable primarily to the large plant, are described.
# # # #
* # • *
In this chapter the following equipment and chemicals are mentioned
for the first time:
Flat pan for "panning" filings; metallic silver, copper, or brass, to serve
as added metal; cast iron pot; sodium chlorate; sulphur dioxide; oxalic
acid; sodium nitrite; metallic precipitants such as copper, iron or zinc;
rubber and glass tubing for use with sulphur dioxide.
Note that all of these materials are employed in alternative methods;
none are absolutely essential, and many workers will not find them
desirable.
material had been, for example, dirty low-grade scrap containing much
soft solder.
* # # *
RECOGNIZING QUALITY. Q. HOW can I tell whether my fine gold is
really of good quality and fineness or not? Do I have to assay every lot?
A. The usual procedure is to melt the gold powder into a button, then
roll or hammer it. With a little experience you can tell by the appear-
ance of the button whether it is good or bad, closely enough for most
commercial purposes. A good button has a characteristic yellow color,
which rarely appears when impurities are present even in small amounts.
It also shows a characteristic depression or "pipe," a sort of dimple on its
surface where the metal contracted on cooling.
When placed under the hammer or rolls, a good button flattens
smoothly and easily. As the melters say, "It rolls like butter." Most of
the very large bullion dealers, whose business is to manufacture gold of
definite purity, do assay each batch, but the jewelry shop whose business
is simply to get gold for re-use, does not assay its refined gold. The
beginner would do well to obtain some fine gold of unimpeachable
quality, such as that sold by the U. S. Mint, and melt it and roll it, in
order to see the characteristic color and "pipe" and to appreciate its
characteristic softness under the rolls.
COPPERAS. Q. What is copperas? Does it contain any copper?

A. Copperas does not contain any copper. The name is definitely
misleading; it is an iron salt. One way to make it is to dissolve iron in
sulphuric acid, then crystallize out the salt that forms. Its chemical name
is ferrous sulphate, and it is also called green vitriol. When exposed to
air the surface turns yellow. Only the green salt is useful in precipitat-
ing gold, the yellow stuff being worthless for this purpose.
CHAPTER VII
SOME SPECIAL CASES
Green gold. White golds. Rolled, filled, and electroplated gold, and
gold scrap, containing soft solder. Procedure to be followed when tin
and lead are present.
* # # #
It is now time to take up some of the materials that were ex-
cluded from study in Chapters V and VI. .
GREEN GOLD
Acquaintance Experiments: Cover a piece of 16-k or 18-k green

gold with nitric acid, and warm it. There will be no action; this is
not surprising, since it contains so much gold. Now add a little
hydrochloric acid to the nitric acid in the vessel, thus forming aqua
regia. . . You expect aqua regia to dissolve gold, but in this case
it acts for a few minutes, and then stops. Warm the aqua regia,
and watch carefully. . . A firm coating of a grayish substance will
collect on the green gold, and even prolonged heating in aqua regia
will have surprisingly little effect.
The explanation is that green gold contains considerable silver,
and, as you know, aqua regia does not dissolve silver but merely
converts it into this whitish silver chloride. Eighteen karat green
gold is usually 25% silver.
These facts leave the beginner in a quandary. He says to him
self: it contains so much gold that nitric acid will not dissolve it;
and so much silver that aqua regia will not dissolve it; what is
there to do?
One answer is to melt it with enough copper, silver, brass, or
zinc to reduce it to 6-k or less; then it will yield to nitric acid and
crumble to a powder. A second answer is to melt it up with enough
gold (or other metal) to make it yield to aqua regia; this means
reducing the silver percentage to 10% or less. And there is a third
way out, as will be described in Chapter XVI, that is especially
65
suitable when green gold is soldered to, or mixed with, platinum,

as in the case of platinum topped jewelry with a green-gold base.
You are already familiar with the idea of inquartation, or adding
enough silver, copper, or brass, to reduce the gold content to 25%
or less. This is the usual plan when green gold is refined.
The other alternative, of melting it up with some metal that will
make it soluble in aqua regia, is not so popular, but the student
should understand it. A button that contains more than 12% silver
is very slow to dissolve in aqua regia. With silver as low as 8%,
aqua regia works readily enough, especially if hot. (We are as
suming that the other constituents are metals that are themselves
soluble in aqua regia, such as gold, copper, zinc, or nickel.) We
can easily imagine a situation wherein a jeweler might have on
hand a little green gold, and considerable yellow gold, red gold, or
fine gold. By melting all these together he would obtain a button
of less than 12% silver, that would dissolve in aqua regia, and that
could be refined by familiar methods.
WHITE GOLDS
White golds, to the refiner, are of two classes: those that are
whitened with nickel and those that are whitened with palladium.
The nickel golds are the cheaper and commoner class. Palladium
golds are used frequently in the dental laboratory, sometimes in
high grade jewelry, and sometimes in chemical apparatus; it is not
unusual for platinum to be another constituent in these special
dental golds. The refiner's first task, therefore, is to establish which
class his white gold belongs to. If it contains palladium, with or
without platinum, it must be handled according to the plan de
scribed in Chapters XI, XII and XIII, in order that the valuable
platinum-group metals may be saved.
Chapter IX describes methods of determining whether your gold
belongs to the nickel class or the palladium class. If after reading
Chapter IX and utilizing the tests described, you are still in doubt,
you are advised to assume that your white gold may contain pal
ladium, and treat it accordingly.
And if you are sure that no palladium or platinum can be present
and if the gold is low karat, or of low price, you can be reasonably
sure they are absent your white gold should be refined exactly as
described in Chapters V and VI, with which you are now familiar.
SOME SPECIAL CASES 67
Nickel, which is such an important element in the commoner

white golds, has a rather high melting point; you notice this when
melting the scrap into a button. The button is apt to be brittle and
difficult to roll; therefore it is customary to granulate it at once.
Some nickel golds are attracted by the magnet and may be con-
fused with steel.
With these facts in mind, the student is now ready to handle any
of the nickel white golds.
ROLLED, FILLED AND ELECTROPLATED GOLD, AND GOLD SCRAP,

CONTAINING SOFT SOLDER
Rolled and filled gold consist of thin shells of a gold alloy

usually low-karat gold soldered onto a heavier core of brass or
nickel silver.
Even in the better grades of filled goods, the quantity of gold is
small. Thus, goods stamped 1/10 12-k" at best will assay one
twentieth fine gold, the other 95% being base metal. Old jewelry
Billet of base metal with thin layer of

karat gold soldered to top and bottom,
ready to be rolled.
whose surfaces are worn down will contain even less gold than the
karat-mark will indicate. Electroplated gold contains an even lower
percentage of precious metal.
For our convenience we shall group these three kinds of scrap
together as "cheap" scrap. They have two things in common
they contain very little gold, and usually they contain soft solder.
The scantiness of the gold, and the presence of the soft solder, com
bine to make this type of scrap relatively unprofitable and trouble
some to handle.
Soft solder contains tin and lead, both of which are difficult to
remove. Moreover, if small amounts are left to contaminate the
recovered gold, they will make it brittle and useless.
When the depression of the 1930's forced many jewelers to go
into the business of buying old gold, many shops that had pre
viously handled only high grade gold bought old cheap jewelry,
old spectacle frames, old jewelry that had been repaired with soft
solder, and so on. They quickly found themselves in difficulty with
their refining; many a refiner with years of good work behind him
began to turn out brittle buttons, pale and crumbling "fine" gold,
and unworkable alloys. In most cases the trouble was due to the
tin and lead, which he had failed to remove completely.
The fact that some of this cheap jewelry is hollow, and is weighted
with sealing wax, shellac, or lead, did not help matters.
It is the purpose of this chapter to clear up these various problems.
At certain times the United States Assay Office has not bought
cheap scrap rolled, filled, or plated stuff. However, the regulations
change from time to time, and at certain times it may be wise to
sell all this material to Uncle Sam. The reader is advised to apply
for new rulings from season to season.
There are several ways in which this cheap scrap may be disposed
of. The larger handlers sell it to a copper refinery. The stuff will
then be treated as though it were impure copper, and the gold and
silver will become by-products. When several hundred pounds are
available, this method is normally chosen.
A second plan utilized by large shops is to melt it with low-grade
wastes such as floor sweeps or polishings. Low grade waste must be
melted up with considerable clean metal in order to collect its
precious metal particles into a button. If there is some rolled-gold
scrap on hand, for example, it can be melted in with the sweeps;
that saves the expense of buying clean metal for this purpose, and
all the precious metals are collected and refined together. This is
referred to again in Chapter XVIII.
A third plan, suitable to the average shop, is to use cheap scrap
as the added material in the inquartation of high grade scrap, as
described in Chapter VI. This gives the worker a button similar
to that described in Chapter VI except that it probably contains
tin and lead.
In a shop that handles gold and silver only no platinum-group
metals this is an excellent scheme so long as the worker understands
the effects of the tin and lead, and knows how to get rid of them.
We urge the student to become familiar with this plan, as he will
have many opportunities to use it.
Two other plans, less advisable, are mentioned now simply for the
sake of the record:
(a) Stripping the gold layer off the base metal core, using a "reversed"
electric current and a cyanide solution. The idea is to dissolve the gold
layer only. This is hard to do economically, and is dangerous to the
workmen. People who have equipment for electroplating or gilding,

and who have had experience in handling cyanide, should experiment
with this process, but others are advised not to. More will be said on
this subject in Chapter XVI under the heading of "Green Gold with
Platinum."
(b) Stripping the gold off by dipping the articles into aqua regia, and
fishing them out as soon as all the gold is dissolved. This also is hard
to do economically; sometimes the acid works in and acts more rapidly
on the core than on the outside gold layer. We do not recommend this
method.
• • • •
Sometimes the worker has on hand only cheap scrap, such as old
spectacle frames, and wants to refine them as they are. That is, he
does not wish to melt them up with sweeps, nor with high grade
gold. Perhaps he has no facilities for melting them up at all. In
that case he will use an acid method similar in principle to that
described in Chapter V, making only such changes as are demanded
by the tin and lead.
For your first refining of cheap scrap rolled gold, filled gold,
electroplated goods, or goods that have been repaired with soft
solder we suggest that you employ this modified acid method,
which will now be described fully, in order to learn to recognize
the effects of tin and lead, and to get rid of them. Having learned
these important things, you can then use the method of inquarta
tion when desired, as it employs the same general principles.
Acquaintance Tests: Obtain from your supply house small samples

of pure tin and pure lead. A pennyweight of each will be enough.
Do not use metals of whose purity you have doubts. The metals
may be in pellets, foil, grains, or any other convenient form.
First take a bit of tin about the size of the head of a pin and cover
it with a few drops of hydrochloric acid, plus a couple of drops of
water. Heat to boiling and watch the results. Now do the same
thing with a little lead; use a clean test tube or beaker for each test.
Note that hot hydrochloric acid dissolves tin promptly. It is
slower in its action on lead. Now, let the lead-hydrochloric acid
mixture cool. . . Note that a heavy precipitate of white crystals
appears. This is lead chloride. Add some water to make up for
evaporation, then alternately heat and cool the mixture, observing
that considerable lead chloride will be held in solution by a little
hot water, only to crystallize out when it cools.
You will note that the tin chloride, formed when the tin is acted
on by hot hydrochloric acid, is not so markedly affected by heat.
Because of the above facts, we use hot hydrochloric acid to dis-
solve solder (tin plus lead) in cases where the hot acid can reach
the solder. So, if your old jewelry shows lumps of soft solder, boil
it in hydrochloric acid until the solder loosens and can be scraped
off. But if the solder is inside the article, the hot acid will not
reach it.
Second: Try the effect of hot strong nitric acid, first on a scrap of
tin, then (in a clean test tube or beaker) on lead. . . The effect is
quite different. In both cases the acid goes to work at once, giving
off the usual brown fumes. . . The tin goes to pieces, and in its
place appears a white pasty or slimy substance. We call this tin
paste. The chemists call it meta stannic acid. If you add some
water the tin paste will thin out, like jelly, but it will not dissolve.
This tin paste, which forms when nitric acid acts on tin (and at
other times as well) is the main reason why tin is troublesome in
refining.
The lead, as we saw at first, reacts promptly with the nitric acid,
but in a few minutes it becomes coated over with a white salt,
which soon becomes so thick that action ceases. When this happens,
add a little water. . . The white salt will dissolve.
This means that lead, treated with strong nitric acid, forms a
white salt of lead nitrate, which may coat over the lead so closely
that the acid cannot reach it. But this lead nitrate will dissolve if
water is added. In other words, lead nitrate is fairly soluble in
water, less soluble in concentrated nitric acid.
Because of these facts, we dissolve lead metal in diluted nitric
acid. But when it comes to tin we see that, while nitric acid acts
on it, it does not really dissolve it, but merely converts it into that
troublesome tin paste.
Third: Repeat these tests, using clean test tubes and fresh samples
of tin and lead, with sulphuric acid. Have the test tubes dry when
making the first of these tests; just cover the small samples with a
few drops of the strong acid, and warm cautiously.
Strong sulphuric acid, powerful as it is in some respects, acts
rather oddly on metals. If diluted with an equal volume of water,
or more, it acts much more like the other acids. Try this: take a
clean cup or beaker, place in it an ounce of water, and then add
slowly a quarter ounce or less of sulphuric acid. Add the acid

slowly, and stir it between additions. Note how hot it gets.
(Never pour water into sulphuric acid. If you should it would at
once boil and spurt out of the dish. Always pour the acid, slowly, into
the water, stirring all the time.)
Now try the effect of the diluted acid on samples of tin and
lead. . . We find that, if heated, the acid will dissolve tin promptly,
but it does not dissolve lead at all; merely converts it into a white
salt (lead sulphate) which does not dissolve in more water. If you
still have some tin paste, left over from an earlier experiment, try
the effect of sulphuric acid upon it. . . It dissolves it, whether dilute
or concentrated.
It is valuable to repeat these tests using aqua regia, first strong,
then diluted, to see what it does to these two metals. Use small new
samples, and write down the results of each test.
By this time you will see that tin and lead are quite different in
their reactions; also different from copper or silver. This means
that each one demands attention in the refining process.
You are now familiar with the regular acid refining of high grade
gold the method described in Chapter V. By making a few
changes you can adapt this method to take care of tin and lead.
PROCEDURE WHEN TIN AND LEAD ARE PRESENT ACID METHOD
For your first refining of rolled, filled, or plated scrap, take clean
stuff such as old spectacle frames, or old jewelry or collar buttons.
Sometimes the thin outer layer of gold is firm enough to keep the
acids from reaching the base metal core; therefore roll this scrap
out thin, cut up the pieces and twist them to keep them from lying
flat in the acid. In many cases the gold layer is so thin that this
precaution is not necessary.
If there are visible lumps of soft solder, cover the articles with
full strength hydrochloric acid, and heat to boiling until the solder
loosens and can be scraped off. Then pour off the acid and throw
it away, and wash the metal thoroughly to get rid of all traces of
this acid. Wash till the washings have no more sour taste, or fail
to turn blue litmus paper red. (See Chapter IX.)
Next, treat the metal with nitric acid and a little water, as usual.
Since low grade scrap is mostly base metal, you will need much
more acid than in Chapters V and VI, and there is much more
chance that it will boil over, so add the acid slowly. This dissolves
all the copper, silver, brass, zinc, iron, cadmium, and nickel, that
the acid can reach. Tin is converted to the gelatinous tin paste.
Lead turns into the scantily soluble lead nitrate; add a little more
water, and all the lead nitrate will dissolve.
Pour or filter off this solution, looking out for flakes of gold.
Usually the liquid contains so little silver, if any, that it should be
thrown away. Repeat the nitric treatment if necessary, then wash
the residue well, but do not let it dry. The residue is gold, plus tin
paste, plus such base metals as were not reached by the nitric acid.
Next, mix up a little sulphuric acid and water, half and half (pour
the acid into the water, slowly, and stir all the time) and cover
your residue with this solution. Warm gently. This dissolves the
tin paste, dissolves any remaining tin metal that the acid can reach,
and changes any remaining lead into white insoluble lead sulphate.
The next task is to pour off and get rid of this sulphuric acid
solution. This is not as easy as it might appear; it is so strongly
acid that it would destroy a filter paper, so it must first be diluted.
The best way to do this is to pour everything acid, residue, and
all into another larger vessel which contains considerable water.
Let it settle somewhat, and you can then easily pour off and throw
away the bulk of the liquid. Then get the washed residue back
into your casserole for the next treatment.
From here on, follow one or another of the methods with which
you are already familiar. Some workers will melt this gold residue
(it usually is in the form of thin flakes) into a button first, but
usually the most economical plan is to dissolve it at once in aqua
regia. Its value will depend upon the type of stuff you started with,
but it rarely runs as high as 14-k. It may still contain a little tin
and lead, which must be taken into consideration in later stages.
Thus, when making up aqua regia, add some sulphuric acid, say
an ounce of the concentrated acid to each pint of aqua regia, more
or less. As we said in Chapter V, this sulphuric acid serves two
purposes: it hastens the expulsion of the excess nitric acid, and it
also helps to get rid of lead. You will now see how it does this:
by converting the lead into the insoluble white lead sulphate, which
will remain with the mud of silver chloride that is left after the
other metals dissolve in aqua regia, and which is filtered out at
the same time.
Tin dissolves in aqua regia, and will go along with the dissolved
gold. Small traces of lead sometimes do the same. If you precipi
tate your gold with copperas, some tin paste may be carried down
mechanically with the gold, and some lead may be precipitated with
it as lead sulphate. (With sulphur dioxide there is less tendency
for tin to be dragged down.) But if you use special care when
washing the gold powder you can expect good results. Thus, with
out letting the gold powder get dry, wash it first with much hot
water; then (second) boil it in hydrochloric acid to remove com
pounds of tin, iron, and copper; wash (third) in hot water to re
move the acid; then (fourth) boil it in a strong solution of caustic
soda to dissolve lead sulphate; and finally (fifth) rinse again in
much hot water.
But the cautious worker does not melt up all his recovered gold
at once; he takes about a pennyweight and makes a small test
button. If it is good, he melts the whole quantity. If it is brittle
or of! color, he turns to Chapter XV and follows the suggestions
given there for the further repurification of gold that is already of
fairly high quality.
* # • #
NOTE. As we go to press, we learn that the Western Electric Co.

is working on a process for stripping the gold from gold-plated
articles. Details are not yet available.
CHAPTER VIII
SILVER
Forms of silver encountered by the refiner. Characteristics and properties

of silver; acquaintance experiments. Metallic silver wastes scrap, filings,
polishings. Silver chloride, and its conversion into metallic silver. Silver
in solution: silver nitrate; silver sulphate; stripping and pickling solu-
tions. Cyanide solutions of silver or gold; plating baths. Mirror solu-
tions. Alternative methods of reducing silver chloride. Photographers'
wastes.
• * * *
The reader is already familiar with many of the more important
characteristics of silver. He knows that it is a white metal popu-
larly included in the group known as "precious" metals. This
inclusion has sometimes had unfortunate consequences, because
silver, in spite of its beauty and usefulness, is much less expensive
than the other precious metals. Thus, during the last few years
the highest ratio, in terms of gold, was reached in November, 1919,
when silver was quoted at 1/15 the price of gold. In December,
1932, it fell to 1/8 and in July, 1939, "foreign" silver was exactly
1/100 the price of gold. This relative cheapness must not be lost
sight of by the refiner, especially the beginner, when deciding how
to handle a given lot of material.
FORMS OF SILVER ENCOUNTERED
The jeweler, dental mechanic, and refiner will encounter silver in

several forms. These may be divided into two kinds metallic and
non metallic. Of the metallic group the most valuable form is
clean metallic scrap such as old anodes, old Sterling silver tableware,
old silver jewelry, or the clippings from the silversmith's bench.
Good quality silver filings are nearly as valuable. These will be
discussed in this chapter. The fine metallic particles found in pol
ishings, floor sweeps, and wash-barrel deposits will be discussed in
Chapter XVIII on Low Grade Wastes.
74
SILVER 75
Silver is also present as an alloying element in most gold alloys.

Even the higher grade dental golds usually contain silver, and some
of the green golds used in jewelry contain as much as 25% silver.
In the processes of refining, this silver is separated from the gold
and other elements, usually in the form of silver chloride.
Non-metallic wastes include silver salts such as the silver chloride
and silver sulphate which the reader has already met in Chapters V
and VI; and the solutions used in pickling, stripping, and electro-
plating. Most of these solutions are acid in reaction; a few, such as
the cyanide plating solutions, are alkaline.
Photographers' wastes, and the solutions left over from the making of
mirrors, do not come strictly within the scope of this book, but so many
people are interested in their treatment that a brief discussion of them will
be included.
This chapter will try to cover the ways in which these forms of
silver metallic and non-metallic may be converted into clean
pure metallic silver, ready to be used again or sold.
METALLIC SILVER WASTES
Clean scraps of silver metal, such as clippings of Sterling silver,

old Sterling tableware, wire, sheet, and the like, if perfectly free
from solder or Britannia metal, are usually melted, cast into ingot
form and used again. Impurities are apt to get in, however, and
bits of solder, binding wire, iron from the shears, etc., will cause
brittleness if melted into the bar, and a good proportion of what
seems to be clean scrap will yield a bad ingot. The process of
melting often changes the fineness of the Sterling silver, and in
large plants these ingots are assayed to establish their quality before
they are re-used.
Good quality silver filings should be sieved, gone through with
a magnet, and the organic matter should be burned out before they
are melted and cast into ingot form. The result, however, is fre-
quently a brittle bar, which may be sold as is, or perhaps used as
the "added metal" in the inquartation of gold.
Obviously these metallic wastes could be refined by acid methods; the
process is simple and the reader probably has guessed the steps that
might be used. But with prices at present levels it would not pay; much
acid and labor would be required, and it is better to return brittle bars
to the professional refiner, who will refine them in large quantities ac

cording to methods described in Chapter XXII.
Old suspension wires used in silver-plating become coated with
silver and finally break. These may be melted up with filings or
scrap that is to be sold.
NON-METALLIC WASTES
These include various salts of silver, either in solid or liquid form.

Of these the most important is silver chloride, and its recovery will
be discussed fully in this chapter.
Before going further, it would be well for the reader to recall the
several acquaintance experiments in previous chapters, in which he
handled silver. Thus in Chapter V he observed that silver dissolved
readily in nitric acid, and he prepared some silver nitrate, laying
some aside for later examination. When freshly made from fine
silver, the silver nitrate solution is colorless, like water. If made
from coin silver it will be greenish, the green color being due to
dissolved copper nitrate. If exposed to light, silver nitrate (as well
as most other silver salts) darkens and finally becomes black.
Silver also will dissolve in hot concentrated sulphuric acid, to
form silver sulphate. (See Chapter VI.) Aqua regia does not dis
solve silver, but attacks the surface layer and converts it into a
coating of silver chloride. Straight hydrochloric acid if hot and con
centrated does the same thing; cold dilute hydrochloric acid has
almost no action on silver.
One of the Acquaintance Tests of Chapter V was the addition of
a little hydrochloric acid or table salt to a solution of silver nitrate.
The resulting precipitate of silver chloride was seen to form as a
white cloud which gradually settled to the bottom of the vessel.
For the last five chapters silver has been accumulating in our
Silver Jar, in which we originally placed the nitric acid with which
we had dissolved as much as possible of the base metals that con
taminated our gold scrap. This liquid contained dissolved silver
nitrate, copper nitrate, probably zinc and nickel nitrates as well.
We added some hydrochloric acid (or perhaps some table salt) and
we obtained the white precipitate. This silver chloride was allowed
to settle, and when we were sure that the liquid over it had been
freed from all dissolved silver, we threw the spent liquid away,
leaving the deposit of silver chloride in the jar. At our next refining
SILVER 77
we put the new nitric solution in on top of the old white chloride,
and again we added hydrochloric acid or table salt, and again let
the heavy chloride settle, and again, after proper tests, threw away
the spent liquid.
Silver chloride obtained as above, or secured in other ways that
will soon be mentioned, is on hand in almost every shop that handles
precious metals.
Many jewelers, especially those who are cramped for room, do not
refine it; they wash and dry it and sell it as silver chloride because
of the difficulty in obtaining recovered silver of sufficient purity for

the manufacture of new alloys. However this book would be in-
complete without full instructions for carrying out this step.
It is almost as easy to refine a large lot of chloride as a small lot,
and in general you are advised to refine silver salts only in fairly
large quantities.
RECOVERING SILVER FROM SILVER CHLORIDE
There are several ways to reduce silver chloride to metallic silver.

We shall first describe a standard method; you are advised to learn
later the alternative methods described at the end of this chapter,
then to choose the one best suited to your requirements.
In all methods you begin by collecting the deposit of silver chlo-
ride and washing it well, with much hot water, breaking up the
lumps and washing until all green copper stains are gone. You will
recall that lead chloride, which may be present and which looks like
silver chloride, will dissolve in hot water. Lead sulphate, unfor-

tunately, will not dissolve.
(Sometimes, when breaking up lumps of silver chloride obtained from

dissolving gold alloys in aqua regia, you will find that some gold re
mains undissolved. Beginners often have this experience. If such bits
of residual gold do appear, pick them out and return them for further
treatment. This mistake is one that is easily recognized and should not
happen to the careful worker. In later chapters you will learn that when
alloys containing gold plus silver and platinum-group metals are dis-
solved in aqua regia, there are occasional complications and the silver
chloride thus obtained may require special treatment, which will be dis-
cussed in the proper place. For the present we are considering silver
chloride in which gold and platinum-group metals are absent.)
After the chloride has been thoroughly washed, cover it with

diluted sulphuric acid, using about one part acid to fifteen or
twenty parts water. (Pour the acid into the water, slowly, stirring
all the time; it may spatter.) Use enough to cover the chloride
well. Next add zinc, preferably in the form of shavings or thin
foil. Zinc plates or shot work more slowly. The zinc will begin
to dissolve at once, giving off many bubbles; as the zinc dissolves,
the silver chloride that is in contact with it is changed into grains
of metallic silver. Stir often. The bubbles consist of hydrogen gas;
this gas will burn, and if confined may become a fire hazard; there-
fore it should be permitted to escape freely into the outside air.
For each pound of chloride, you will use about one-quarter to
one-half pound of zinc (more or less) and about one-half pound
of concentrated sulphuric acid.
The chloride changes to granules of metallic silver, called "cement
silver." Dip some out with a porcelain spoon and look at it, and
you will see it looks quite different from the chloride.
Stir the mass often to permit all parts of the chloride to come in
contact with the zinc as it dissolves. Stirring hastens matters, and
also reduces the loss of zinc that dissolves while not in contact
with chloride.
When all the silver chloride has changed over into cement silver,
pour off the liquid and throw it away. The length of time required
will depend upon the rapidity with which you bring zinc and chlo-
ride into contact with each other; that is why zinc shavings or thin
foil work faster than thick plates; they present more surface.
SILVER 79
The cement silver must now be freed of the unused zinc. This
can be done by picking out any pieces large enough to be seen, then
by dissolving the rest in weak acid. Mix up some weak sulphuric
or hydrochloric acid say one part sulphuric acid to twenty parts
of water, more or less; or one part hydrochloric acid to fifteen parts
water and cover the cement silver with either mixture and watch
for bubbles. When there are no more bubbles on stirring, you may
conclude that all unused zinc has dissolved. Wash well with plenty
of clean hot water to remove all salts and acid, and let the cement
silver dry.
This process is entirely comparable to the precipitation of dissolved
gold by means of copper wire or other metallic precipitants, as described
in Chapter VI.
Usually the next step is to melt the cement silver into a bar or
button; use a flux containing borax with charcoal or sugar, in a
graphite crucible. The silver thus obtained is rarely pure enough
for new alloys, and most jewelers sell it as is. However, if you
use the inquartation process described in Chapter VI, this silver is
precisely what you want for reducing your scrap gold to 6-k or less.
(Alternative methods of reducing silver chloride will be given later in
this chapter.)
RECOVERING SILVER FROM SILVER NITRATE AND SILVER SULPHATE
These salts are formed by dissolving silver in nitric acid or sul-

phuric acid. Examples will be found in Chapters V and VI, in
which we removed the silver from gold alloys. Nitric acid acts
quickly on silver, dissolving it rapidly even when cold. Moreover,
a small amount of acid will hold in solution a relatively large
amount of silver, which is another way of saying that silver nitrate
is a very soluble salt 100 grams of water being able to dissolve
213.4 grams of silver nitrate at room temperature. Sulphuric acid
acts less readily, and the silver sulphate thus formed is only scantily
soluble in water, only .55 gram being held in solution by 100 grams
of water. If any larger quantity of silver sulphate should find itself
in a scanty supply of water it would appear in solid form, that is,
as a precipitate. This precipitate will dissolve on the addition of
more water, as you will recall from Chapter VI.
The first step toward the recovery of dissolved silver from silver
nitrate or from silver sulphate is, as you have already learned, to
add hydrochloric acid or table salt, to bring down the insoluble

white chloride. A consideration of the figures just given will show
you why a nitrate solution may require more salt than a sulphate
solution.
In general, a better precipitate is obtained if the solutions are fairly
dilute; also if warm rather than cold. To explain: a "good" precipitate
is one that settles quickly and completely, and that does not carry other
substances down with it. Here is an interesting experiment to illustrate
this point: divide a small amount of concentrated silver nitrate solution
into halves, placing each in a beaker or glass. Make up a concen-
trated solution of table salt, and divide it also into halves. Now pour
the strong salt solution into the strong nitrate solution, and set aside.
Next, dilute the remaining salt solution with considerable hot water, and
dilute the remaining silver nitrate solution with considerable hot water,
finally adding these two hot dilute liquids together, and set aside as before.
In both cases you have a precipitate of silver chloride; but watch the
two vessels and compare the way in which the precipitates settle. . . .
You will find that when silver chloride precipitates in the presence of
considerable hot water, it settles much more promptly than when formed
in cold concentrated solution. Analogous experiments would show that
when other substances (for example dissolved lead or palladium) are
present, the precipitate obtained from a dilute solution carries down much
less of the other substance.
Accordingly, in most cases not all when recovering dissolved silver
we add hot water, and we dissolve our table salt or hydrochloric acid in
hot water before adding it. The exception to this rule appears when a
bulky or dilute solution is objectionable for one reason or another, such
as lack of room, or when the solution remaining must undergo further
treatment that would be made difficult by the presence of much water.
The second step in the recovery of silver from solutions of silver
nitrate or silver sulphate is to collect the precipitated silver chloride,
to wash it well, and then to treat it with zinc, as described in the
previous section of this chapter.
ACID STRIPPING AND PICKLING SOLUTIONS
In this classification come old sulphuric acid stripping solutions

used to remove deposited silver from electroplated articles before
replating. Such solutions contain strong sulphuric acid to which
has been added some nitric acid, or perhaps potassium nitrate. To
recover the silver in as pure form as possible, dilute such solutions
with considerable hot water, and let any deposit or sediment settle
SILVER 81
overnight. Pour or filter off the clear liquid, and add to it a

solution of table salt, or some hydrochloric acid diluted with warm
water. Treat the precipitated silver chloride as usual.
The sediment that was filtered out of the solution above should
be examined, as it may contain considerable silver or silver com
pounds, mixed with material from the walls of the vessels, organic
trash, and so on. If a lead vessel had been used, more or less lead
sulphate, also particles of metallic silver deposited out on the lead
walls, may be included. The deposit, therefore, may well be worth
salvaging; usually it is melted into a button and sold with other
impure silver.
The pickling solutions used to strip firecoat from silver articles
belong to the same group, except that they may not require dilution.
Otherwise they are treated the same as above: the sediment is fil
tered out and saved if worth while, and the silver is precipitated as
silver chloride, as usual.
CYANIDE SOLUTIONS OF SILVER OR GOLD; PLATING BATHS
Old silver cyanide solutions, such as foul electroplating baths,

sometimes contain enough silver to justify treatment. All cyanides
are poisonous, and when mixed with acid they are doubly so. The
fumes are deadly if breathed, and the solution should not be per
mitted to touch the skin if the latter is broken or chapped.
It is possible possible but not advisable to recover silver from cyanide
solutions by the same method we used on acid solutions, namely to add
plenty of hydrochloric acid. This method is not advised, however, for two
reasons: the fumes are extremely dangerous to the worker, and in addi
tion they may cause the solution to froth over and be lost.
Another way, safer and better, is to add zinc to the cyanide solu
tion, and let it stand for some hours or days, stirring often. It is
not necessary to add any acid; some workers do add a little acid,
however (sulphuric or hydrochloric), because it hastens matters.
We do not advise this because it produces the poisonous hydro
cyanic acid gas, which may cause the liquid to froth over. The
zinc may be in the form of turnings or foil, or zinc sheets may be
suspended so as to hang free in the solution. Stir often. An ounce
of zinc will reduce about a gallon of solution if agitated well, the
exact quantity depending upon the amount of metal present and
the thoroughness with which it is brought in contact with the zinc.
fl REFINING PRECIOUS METAL WASTES
This process is essentially the same as the precipitation of gold by

metallic precipitants (Chapter VI) and the reduction of silver chloride
with zinc (this chapter) Silver appears in fine grains, called cement
silver Most old plating baths contain impurities such as dissolved copper,
and the treatment with zinc will carry down the copper too, in grain
form, and as a result the recovered silver will be impure The usual
procedure is to wash it well and melt it, the impure silver button being
used in inquartation or sold to a refiner.
A second way to recover silver is to evaporate the old solution to

a crust. This takes time, but if your shop has facilities for the
task, the method has advantages. Be sure the fumes escape into
the outer air. Heat the dried crust in a crucible; the salts that are
present serve as a flux and on stronger heating a silver button will
collect. The button is apt to contain copper and other base metals,
and usually is sold with other silver-bearing wastes.
A third way to get silver out of cyanide solutions is to plate it

out electrically, using steel or platinum anodes and a silver cathode.
This is slow and the last few grains of silver are hard to recover,
so in general the method is not recommended.
Gold can be recovered from cyanide solutions in the same ways.

Thus old gilding baths are dried to a crust and melted down; or
they may be treated with zinc (without acid) exactly as above, the
cement gold being washed extremely well, then refined as in Chap
ter V or XV. Sometimes it pays to plate the gold out electro
lytically, as described in Chapter XVI in the section on 18-k green
gold with platinum, to which you are referred.
In an article entitled RECOVERY OF GOLD AND SILVER FROM
PLATING SOLUTIONS l F r a n k K Savage describes the methods used
in the C G Conn plant at Elkhart, Indiana The silver rinse waters
pass over glass baffles that force it to travel some twenty-five feet before
reaching the sewer The space between the baffles is packed with zinc
shavings, held in place with excelsior In this particular plant the gold
is plated out electrolytically from the drag out solutions, work from the
various gilding operations being rinsed into a tank where the free sodium
cyanide concentration is kept at 1/4 to 1/2 ounce per gallon. Carbon
anodes and brass cathodes are used, the cathodes having cloth bags to
catch any gold that may drop off after deposition.
1 The Metal Industry, New York Page 159, April, 1939
SILVER 83
Further information on the recovery of these metals from cyanide solu-

tions will be found in S. B. Christy's 171-page pamphlet entitled
ELECTRO-DEPOSITION OF GOLD AND SILVER FROM
CYANIDE SOLUTIONS.2
SOLUTIONS USED IN MAKING MIRRORS
These solutions are made up with silver nitrate and ammonia, to

which is added a reducing agent such as sugar, or Rochelle salt, or
formaldehyde, or the like. The compounds formed are apt to be
highly explosive, and care must be taken to prevent any residues
becoming dry. This hazard is emphasized in a pamphlet 3 Pu 1S e
by the U. S. Bureau of Standards, and the reader is told that all
vessels containing silver solutions should be carefully cleaned at
once, and all residues disposed of without delay. The proper
method of disposal is to dilute with cold water, then add enough
hydrochloric acid to neutralize the ammonia (in which process the
solution becomes warm) with sufficient excess to throw down the
silver as silver chloride.
From here on the process is familiar, the silver chloride being
collected, washed, and reduced to metallic silver as usual.
ALTERNATIVE METHODS OF REDUCING SILVER CHLORIDE TO SILVER
You are already familiar with the standard method, in which the
washed silver chloride is covered with acid and left in contact with zinc
to form cement silver, which is washed and melted into a button.
There are however several other ways of reducing silver chloride to
metallic silver. The standard method is susceptible of variations. Thus
aluminum or iron may be used instead of zinc. Aluminum, as powder,
foil, or sheet, works more rapidly than zinc, and is much more apt to
make the jar boil over. Also aluminum is more expensive than zinc.
It is chosen only when rapid action is required, and the worker is warned
to add it very slowly, with frequent stirring, because its action with acid
produces heat, sufficient to crack the jar if produced suddenly.
Iron is cheaper than zinc and works more slowly. Flat iron plates
may be used, the silver chloride being spread over them in a thin
layer, the whole being covered with dilute sulphuric acid as before.
2 Bulletin 150 of the Bureau of Mines. Obtainable from the Superintendent of
DEPOSITION OF METAL ON GLASS.

Circular No. 389 of the Bureau of Standards. Price $0.05 January 6, 1931.
Rake the chloride back and forth at intervals, to bring all parts of
it in contact with the iron. Finally lift out the iron plates, which
possibly can be used again.
The liquid which covers the cement silver now contains iron sul
phate, also called ferrous sulphate or copperas; this was formed by
the interaction of iron and dilute sulphuric acid. It is impure, and
contains ferrous chloride and possibly other salts as well. However,
you will recall that you used ferrous sulphate (copperas) to pre-
cipitate your dissolved gold. Some refiners, therefore, collect this
ferrous sulphate solution and use it for this purpose; they let the
cement silver settle, pour off the supernatant liquid, and sometimes
evaporate it down and get the green crystals of copperas. In most
parts of the country, however, fresh pure copperas can be obtained
so cheaply that this expedient does not pay.
The cement silver obtained as above should be washed in dilute
acid as before, to remove all unused aluminum or iron, and dried
and melted as usual. If any copper salts were present in the original
silver chloride, metallic copper will be present now, mixed with
the cement silver.
REDUCING SILVER CHLORIDE THE DRY WAY
Here is another method which involves a different principle, and

is quicker than the first method, but requires a furnace with a
good draft.
After the chloride has been washed well with plenty of hot water,
spread it out and let it dry. Mix up a flux of equal parts dry borax
glass and soda ash; or use a prepared flux made for this purpose.
When the silver chloride is perfectly dry, mix it well with an equal
weight (or more) of flux.
Put the mixture into a large graphite crucible, filling it less than
half full. Melt in a gas furnace, slowly. Unpleasant fumes come
off and the crucible tends to boil over, so watch it and reduce the
heat if necessary. As it boils down, add more chloride-flux mixture.
Continue until everything becomes quiet, then let the mass cool in
the crucible. The button will be metallic silver, the chlorine having
gone off as fumes.
Here again the metal obtained is of doubtful purity, but con
venient to use in the inquartation process described in Chapter VI.
SILVER 85
PHOTOGRAPHERS' SOLUTIONS
With silver at present prices, there is small profit in these wastes,

except when handled on a large scale. Photography produces silver
wastes of several kinds: old films and prints, and the solutions used
in developing and printing, the latter being the stuff that commands
most interest. Large quantities of silver go into photography, at
least one hundred tons of silver salts being dissolved annually by
America alone, and the jeweler or small refiner may have occasion
to handle hypo solutions to his advantage.
Several ways of recovering silver from old hypo (thiosulphate)
solutions are in use, and by an unfortunate coincidence, at least
three essentially different methods are commonly referred to as
"electrolytic." This has caused confusion both in the literature and
among users.
When silver is to be recovered from ordinary quantities of old
hypo baths, such as accumulate in the average photo shop, the
commonest procedure is to add sodium sulphide, which immediately
precipitates the black silver sulphide; the liquid is discarded and
the sulphide is washed and melted with a flux (litharge and soda
ash perhaps) which gives a lead-silver button ready for further
purification.
A second method, equally applicable to small or medium-sized
lots, is the familiar one of adding metallic zinc, which gradually
deposits cement silver. This method is cleaner than the sulphide
method, but takes longer. The cement silver is melted into a button,
which as before is often of doubtful purity. Since the reaction be
tween the zinc and the dissolved silver depends upon the fact that
zinc is higher than silver in the electromotive series of metals, the
method is loosely referred to as an "electrolytic" method, even
though no generator is involved.
Indeed, the name "Electrolytic Unit" has been given to a prepared
device, which when immersed in an old hypo bath will collect silver
upon itself, the solution being saved for re-use. Such a device is
being provided by refiners for the use of their clients; it utilizes in
a convenient form the principle that is employed when zinc dis
solves to displace a heavier metal.
In both the above methods, the metallic silver obtained requires
further purification. When large quantities of such silver are avail
able, the most popular methods of purifying it are the Thum and
Moebius processes, described in Chapters XXI and XXII. These

are true electrolytic methods, employing electric generators, an
electrolyte, a heavy current of electricity, etc. The impure metal is
made the anode in an acid bath, it is carried into solution by an
electric current, and is then deposited out again on a cathode as
pure silver, the impurities being collected in the electrolyte.
The Eastman Kodak Co. has worked out another and quite different
method. It is truly electrolytic, in that the bath is traversed by an electric
current provided by a generator, and silver is deposited on the cathode.
The equipment required is extensive, and rigid control of the process
is imperative; hence the method is one that has found successful applica
tion only where large quantities of film are being treated.
The Eastman Kodak Co. controls several patents covering both the
principles and details of this method, which is described in an article
entitled ELECTROLYSIS OF SILVER-BEARING THIOSULPHATE
SOLUTIONS 4 w r i t t e n by engineers of its staff.
The process is carried on in large cells containing 100 square feet of
cathode surface, through which a current of 300 amperes is passed at
1 to 1.5 volts. At the anode, thiosulphate (hypo) is oxidized to tetra
thionate and trithionate sulphate; at the cathode silver is deposited with
small inclusions of silver sulphide and gelatin from the film; some of
the tetrathionate is reduced to thiosulphate, which is then available for
re-use.
The electric efficiency varies between 65% and 80% in large installa
tions, and the yield per million feet of film is about 1200 ounces of silver.
By utilizing the solutions from which silver has been removed, the con
sumption of fixing baths is reduced to 35% of that previously demanded.
When a simple silver-bearing hypo solution is electrolysed under ordi
nary conditions with fixed electrodes, for example as soon as the cur
rent is applied a stream of brown colloidal silver sulphide falls away
from the cathode, yielding a dirty unfilterable liquid. For this reason
earlier workers on this problem dismissed as impracticable the electrolysis
of hypo solutions.
However, by strictly regulating several factors, electrolysis can be
accomplished with the deposition of high-purity, mirror-bright silver.
Vigorous agitation, together with proper concentration of acid, sulphite,
and certain promoting agents, is essential.
In general, cells should be as large as will pass through a doorway, and
the cathode assemblies, plus their load of deposited silver, should not be
too heavy to be lifted by hand. The choice of building material for the
4 By K. Hickman, W. Weyerts, and O. E. Goehler. Journal of Industrial and
Engineering Chemistry, pp. 202-212, vol. 25, February, 1933.
SILVER 87
cells was found to be difficult, partly because of the corrosive action of

acid hypo solutions, and also because, as the action proceeds, silver tends
to deposit and spread from any spot where the cathodes touch the wall,
eventually causing short circuits.
Carbon or graphite anodes are used; the cathode may be of Monel
metal or Alleghany metal. The silver complex that deposits out is melted
in 100 lb. lots in an iron crucible, with a little nitre. In the first six
months of operation, the Metro Goldwyn Mayer plant at Culver City,
California, with one tailing and four working cells, recovered 60,000
ounces of silver. There was a saving of 65% in fixing and precipitating
chemicals.
OTHER PHOTOGRAPHIC WASTES
Old print paper containing silver can be burned, and the ashes collected
and melted. Old film contains two valuable ingredients, the silver-
bearing emulsion and the celluloid base. Some large establishments are
salvaging both of these. They remove the emulsion with caustics, and
the clean celluloid is sold for use in making campaign buttons and the
like. At present silver prices, the profit on this kind of waste is small,
even when relatively large lots are handled.
Several other metals, including gold, platinum, iron, palladium, ura-
nium and selenium, find employment in photography, in printing papers
and toning solutions. Occasionally it pays to burn such papers and to
add sodium sulphide or zinc to such solutions. Chapter IX gives tests
for the identification of such precious metals as these residues may
contain. Photographic solutions normally contain complex organic
chemicals, some of which inhibit the reactions of the heavy metals.
Chapters XII, XIII and XIV describe recovery methods, some of which
can be adapted to residues of this kind. The salvaging of these wastes
by the non-chemist is not advised, partly because of their small intrinsic
value, partly because of the complications occasioned by these organic
chemicals.
CHAPTER IX
IDENTIFYING METALS IN SOLUTION
The need for identification. Definition of "in solution." Equipment for

making tests. Testing for gold in acid solution, with copperas. Testing
for gold in alkaline or cyanide solution. Testing for silver. Stannous
chloride testing solution, for all precious metals. Standard solutions of
gold, platinum, and palladium. Using stannous chloride with platinum.
With gold. With palladium. With silver. Usefulness of stannous
chloride. If the tests are not conclusive. . . . Two or more precious
metals. Dimethyl glyoxime for palladium. DMG and nickel. Other
chemical tests—The glow test. Copper. Iron. Acid and alkaline solu-
tion: litmus paper. Other tests. A warning. Identifying metals in
solid form.
An important part of the refiner's job is to know whether solu-

tions do or do not contain dissolved metals, and, if so, roughly in
what quantity.
Thus you have frequently been told to throw away a solution
which had once contained precious metal. For example, in Chap-
ters V, VI, VII and VIII you are told to discard solutions from
which gold or silver has been precipitated. The beginner is always
reluctant to do this, and as a result his shop is soon cluttered with
jars and bottles of worthless solutions. Before long his troubles
are further complicated by his forgetting what the various liquids
originally contained. Finally in desperation he packs them up ex-
pensively and ships them off to some refiner, who naturally gives
him nothing in return.
It is therefore advisable to test all spent solutions at once, and to
throw them away promptly if they are worthless. Wash the sink out
well afterwards with plenty of water, to protect the plumbing. In
general it is well to send worthless solutions directly down the drain,
not into the settling tanks; being acid they may dissolve precious
metals that are present in the settlings and cause their loss.
But before throwing away any solution, you must assure yourself
by proper tests that it is indeed worthless. Fortunately there are
88
IDENTIFYING METALS IN SOLUTION
89
several such tests, all easy to perform and so dependable that the
most self-distrustful worker can soon use them with confidence.
Two of the most useful ones are already familiar to you.
DEFINITION OF "IN SOLUTION"
But before describing these tests, we must first make clear what
we mean by "dissolved metals" or metals "in solution." We mean
that they are distributed throughout the fluid in the sense that sugar
may be dissolved in water. Sometimes we may see particles in the
solid form as a sediment in the bottom of a dish, like undissolved
sugar in a teacup; this sediment is a different thing from a metal
"in solution" do not confuse the two. Some workers, especially
those to whom English is an acquired language, make this mistake,
and when they are told that there is no metal in a given solution,
they will say: "But I see little pieces of it there in the bottom of
the jar!" When metal is present in solid form, no matter how
small the particles, we call it a sediment, or a residue, or a precipi-
tate; it is not "in solution." Substance "in solution" will pass
through a filter paper; if a substance will not pass through a filter
paper, it is not "in solution."
(When you wish to test metal in solid form, your first step, as
will be described on page 102, is to get it into solution.)
All of these tests consist of taking a small quantity of the sus-
pected solution which we shall call the "unknown" and adding to
it a substance that will show the presence of a small amount of the
precious metal in question. That is the whole story in one sentence.
Certain ways of handling the solutions tend to make the tests
more delicate or more sure, and these will be described in turn.
EQUIPMENT FOR MAKING TESTS
Chemists generally use test tubes for making tests of this kind.
Small glass beakers are almost as good. Porcelain spoons and the
porcelain spot plates that will be described shortly are excellent in
many cases. The porcelain spoon and spot plate are especially
useful when the test involves a change of color, but they cannot be
used when the solution must be boiled.
All testing equipment must be kept scrupulously clean.
TESTING FOR GOLD IN ACID SOLUTIONS, WITH COPPERAS
You will recall that you recovered dissolved gold by means of

copperas. After the gold powder had settled, you tested the re-
maining liquid to see if you had thrown down all the gold, by
taking a few drops of the liquid in a porcelain spoon and adding a
clean green crystal of copperas. If after a few minutes a dark cloud
formed around the crystal, you realized that your "unknown" still
contained some dissolved gold. This is one of the most useful tests
for gold, especially as it can be used even when other heavy metals
such as the platinum-group metals are present in quantity.
Note that your unknown was originally an aqua regia solution,
and you had driven off the excess nitric acid by repeated evapora-
tions. You will also recall that copperas will not throw down the
gold until all the unused nitric acid has been disposed of; it will
produce brown bubbles but no brown powder. This fact must be
remembered when using this same reaction as a test; that is, if on
adding copperas you observe a brownness or darkening remember
that this may be gold or it may be merely brown bubbles. To settle
this doubt, repeat the test using a test tube or beaker, and heat the
solution to boiling; brown bubbles will be driven off but brown
powder (gold) will remain. Add enough copperas to be sure that
you have "killed" any nitric acid that may be present, and if the
gold is indeed present, you will soon observe the brown powder.
In order to gain assurance in making this test, practise first with solu-
tions whose compositions you know, noticing for example how a con-
centrated solution behaves how instantly the dark cloud forms; then
take a few drops of this same solution and dilute it with three or four
ounces of water and test it again; notice the difference in the appear
ance of the dark cloud. Repeat a third 'time, taking a single drop of
your rich solution and diluting it with a pint of water, then testing the
result; as your skill increases you will be able to detect smaller and
smaller quantities of gold. This skill is especially useful to the man
who buys old solutions for their metal value.
TESTING FOR GOLD IN ALKALINE OR CYANIDE SOLUTION
The test with copperas works only in acid solutions; it will not
work at all in alkaline or cyanide solutions until the alkalinity is
destroyed.
IDENTIFYING METALS IN SOLUTION 91
Thus, suppose you have an old cyanide gilding solution, and

wonder if it might contain dissolved gold. Take about five drops
of the suspected liquid in a small test tube or tiny beaker, stand
near a window or an exhaust fan, and add four or five drops of
hydrochloric acid. The fumes evolved are extremely poisonous
hence the need for working near an exhaust. Bring the mixture
cautiously to a boil; this drives off the cyanide. Let it cool, and
you can now test with copperas as usual.
TESTING FOR SILVER IN SOLUTION
This test also is already familiar. On several occasions in the

past you have added hydrochloric acid (or table salt) to solutions
that contain silver, and you have seen silver chloride form and
settle as a white precipitate. You observed that a small amount of
silver produced a surprisingly large amount of chloride.
Your purpose when examining spent solutions, to see whether
they are worthless or not, is to recognize small quantities of precious
metals dissolved in a relatively large quantity of liquid. Accord
ingly you must give your eye some practice in recognizing small
precipitates and thin clouds. Thus, take one drop of silver nitrate
solution (left over from the acquaintance experiments of Chap
ter V, for instance) and dilute it with an ounce or so of plain
water; then add a few drops of hydrochloric acid. You will see
the same white cloud of silver chloride, but much less dense than
before.
It is interesting to repeat this, diluting one drop of silver nitrate with
a pint of water, stirring well. This will convince you that the test,
properly conducted, is indeed delicate. Do not let this fact lose its
economic significance that is, if you find that a certain solution con
tains so little silver that the test gives only a thin cloud, you must realize
that it is not worth refining.
This test for silver can be used on either nitric acid or sulphuric
acid solutions; also on pickling solutions (which contain both nitric
and sulphuric acid). Cyanide or ammonia solutions require slightly
different handling; use a test tube or small beaker and take only a
few drops of the unknown solution, and cautiously add five or six
drops of hydrochloric acid; the ammonia mixture will become hot;
with cyanide solutions there will be bubbles of the deadly hydro
cyanic acid gas, which should be driven off by gentle heating. As
Purchase the following from any chemical supply house:

1 oz. stannous chloride crystals;
1 oz. or less pure tin metal—mossy, granular, or foil;
Porcelain spot plate with cavities—Eimer and Amend No. 29844
or equivalent; about $0.70
Dropping bottle size 30 to 50 cc; Eimer and Amend No. 18824
or equivalent. (Dropping bottles have grooves in the stopper, and
with them it is easy to get one drop, or as many drops as needed,
without delay or waste.) Price$.30to $0.50
You will also need hydrochloric acid and water, and a small glass
rod. Total cost, not over $2.00; you can hardly make a better in-
vestment. These quantities will provide hundreds of tests.
Make up only a small amount of Testing Solution A at a time, as
it does not keep indefinitely. Take about a pennyweight of the
stannous chloride crystals (also called tin salts) in the dropping
bottle, add a pennyweight or less of tin metal, and fill the bottle
three-fourths full of water. Ordinary tap water will do. Now add
a little hydrochloric acid, about 25 or 30 drops to a 30 cc. bottle.
This gives a milky liquid which is ready to use. The tin metal will
dissolve very slowly, and it serves to keep the solution in good con-
dition. Label the bottle "Testing Solution A."
As we said, Testing Solution A when properly used shows the
presence of precious metals in solution. In order to get acquainted
with the color changes involved, you should first make up some
solutions containing these metals. You should have a solution con-
taining gold, one containing platinum, and another containing pal-
ladium. This method is so useful and fascinating that most users
wind up with a whole series of standard solutions, so perhaps you
might as well get a half dozen dropping bottles in the first place.
STANDARD SOLUTIONS OF GOLD, PLATINUM, AND PALLADIUM
To make a standard solution, dissolve a weighed piece of metal

in a definite quantity of solution. Thus, dissolve two grains of
fine gold in a few drops of aqua regia, wash it into a two-ounce
bottle, and fill the bottle up to the mark with water; label it "Two
Grains Gold in Two Fluid Ounces Water."
When your only object is to become acquainted with the various
solutions, it is not necessary to use exact measurements. But later on,
when trying to estimate precious metal in a solution, it will be extremely
helpful to have standard solutions made up with a definite weight of

metal in a definite volume of liquid. Therefore it saves time to make
up your solutions in the beginning according to a definite plan. Some
people use 100 milligrams to 100 cc. of solution. It makes no differ
ence which unit you use, but use the same unit every time.
Use pure metals; not alloys. The gold and the palladium will dissolve
readily in cold aqua regia, but it will be necessary to heat the platinum.
Use as little aqua regia in each case as will do the work, then evaporate
the solutions gently to get rid of excess acid, finally washing them into
their respective bottles with a little water.
USING STANNOUS CHLORIDE WITH PLATINUM
Take the spot plate and drop one drop of the standard platinum
solution into a cavity. Notice the pale yellow color. Add a drop
or more of Testing Solution A. If properly prepared the two will
react instantly to give a deep yellow or brown color. If too con
centrated, the color will be almost black; in that case, dilute the
platinum solution with an equal volume of water. This deep yellow
color with Solution A is a characteristic of platinum and iridium.
WITH GOLD
In another cavity, place one drop of gold solution, and add a drop
of Solution A. After several moments add several more drops of
Solution A. Note the first intense dark color, deep purple or black.
This is characteristic of gold. After it stands a few minutes, notice
the purple stain on the white porcelain.
Now, in another cavity, take just one drop of your gold solution,
and dilute it with five drops of plain water. Take one drop of this
dilute gold, in another cavity, and add a drop of Solution A. Note
that the color is still definite. Dilute with five more drops of plain
water, and try again. See how dilute this gold solution must be
before it becomes so weak that you cannot detect a change with
Testing Solution A. If you figure this out, you will find that Solu
tion A constitutes a delicate test, and will reveal the presence of a
very small percentage of gold.
WITH PALLADIUM
In the same way, learn the color changes shown when mixing
standard palladium solution with Testing Solution A. This color
change is even more interesting than the others. When the two
drops are first admixed, you see a deep yellow, not unlike the effect
produced by platinum. After some minutes the yellow turns blue-
green. This blue-green color is characteristic of palladium.
WITH SILVER
Silver solutions, such as silver nitrate, do not give any color reac-
tion with Testing Solution A. What you will see when the two are
mixed is the white cheesy precipitate of silver chloride, similar to
that obtained when table salt is added to a silver nitrate solution.
USEFULNESS OF STANNOUS CHLORIDE
By the time you are familiar with the color changes produced by
gold, platinum, and palladium in solution, you will also be able to
gauge the approximate quantities of metal present. This knowl-
edge is especially valuable when you are examining spent solutions
to decide whether they should be discarded or not. You will also
have occasion in later chapters to use Testing Solution A to decide
upon what methods of recovery to use. If you are a professional
refiner who buys old solutions, you will find this solution invaluable
when making quick appraisals of your purchases.
IF THE TESTS ARE NOT CONCLUSIVE ....
Of course the beginner sometimes gets confusing results. Some-

times the colors fail to appear. This may mean that the testing
solution has lost its strength, or, more likely, too much strong acid
is present.
(Testing Solution A must be made up freshly from time to time.

It loses its virtue completely in a few weeks. Therefore, begin the
day's work by checking your Solution A against a drop of standard
gold solution. If it fails to respond, throw it away at once.)
Solutions containing only such base metals as iron, copper, zinc,

nickel and cadmium give no color change with stannous chloride.
Lead may give a white precipitate of lead chloride, which may be
confused with silver chloride see page 92 for a way to distinguish
between them. Another base metal that is apt to confuse the
beginner is mercury, which under some conditions gives a heavy
grey precipitate with stannous chloride. If you suspect mercury,
make up a standard mercury solution and make a few acquaintance

tests in order to recognize this reaction.
Here is another complication: solutions of the platinum metals,
especially iridium, to which much ammonia has been added at cer-
tain stages of the procedure, sometimes fail to respond to the usual
tests. The ammonia can form a sort of complex which ties up the
metal and the usual reactions may not occur. This will be spoken
of again in Chapter XIII. Organic substances sometimes do the
same sort of thing. In this book we go to some pains to avoid the
formation of these complexes, hence this difficulty is not apt to occur
when following these routines. If it should occur, however, do this:
evaporate a sample of the solution to a crust, and heat the crust to
redness. This breaks up the complexes and reduces the precious
metals to metallic form. Dissolve the residue in a little aqua regia
and the usual tests can be made.
If your unknown solution contains a very large concentration of

iron salts (as from copperas) or of copper salts, the test with stan
nous chloride may seem slow to appear, and you will need more
testing solution than usual. This is because of a by-reaction that
consumes some of the stannous salt before it reacts with the precious
metal.
Remember that Testing Solution A spoils on standing. The
standard solutions do not spoil; but when you make them up, do
not use an excessive amount of acid to dissolve your bits of metal,
or the tests will be weakened or even destroyed.
TWO OR MORE PRECIOUS METALS
If your unknown contains two or more precious metals say gold

and platinum you may at first be confused, especially if one or
the other is in large excess. The answer is to practice with mix
tures of the standard solutions, varying the proportions and concen
trations, and in a short time your eye will become skilled.
Remember also that gold reacts with copperas, and platinum does
not. Therefore, if your stannous chloride test is confused, it may
be helpful to repeat the test using copperas instead.
Instead of the spot plate, some people use white blotting paper or white
filter paper. Place on it a drop of the unknown solution, then drop
one drop of Testing Solution A on top of the first drop. The colors
will spread through the paper, and if two or more metals are present the
characteristic colors of both will appear.
Another situation that may confuse a beginner is to find an unknown
containing much platinum and very little palladium; or much gold and
very little palladium. He finds that the palladium color is obscured by
the intense reactions of the platinum or the gold. As his eye becomes
skilled he can detect smaller and smaller proportions; however, he will
be glad to know that there is another solution that is especially valuable
in detecting small amounts of palladium.
DIMETHYL GLYOXIME SOLUTION AND PALLADIUM
This solution has the added virtue of showing up nickel even in

small amounts. It will show up nickel in a platinum alloy; in a
white gold alloy; in a dental alloy. It will show up palladium and
nickel when both are present in small amounts in an alloy that is
largely platinum or gold.
One gram of dimethyl glyoxime will be enough for several hun-
dred spot plate tests. It is a white or pale yellow powder. The
name is pronounced "dye-methyl glyoxeem," but no one will blame
us if we refer to it as DMG.
Dissolve this gram of DMG by bringing it to a boil in about
100 cc. of water about 4 fluid ounces. It dissolves rather slowly.
If possible let it stand overnight; then filter. It is important that
the solution be free from sediment or crystals. It is now ready to
place in a dropping bottle, which should be labeled DMG. It
keeps quite well for years, except that you may have to filter it again.
Let us get acquainted with DMG. Its most interesting charac-
teristic, as we said, is to show up palladium and nickel, in the pres-
ence of other metals and in the presence of each other.
You can see this more clearly by using a few small test tubes. In
No. 1 place a few drops of standard platinum solution; in No. 2 a
little of your standard palladium solution; in No. 3 some standard
gold solution. All three have a pale yellow color. Be sure that all
the solutions are free from sediment. Now add a few drops of
DMG to each test tube, and note the change in the palladium tube.
Note the precipitate that forms, and its color. The pale yellow
colors do not change, but in the test tube containing palladium you
will see a precipitate. Note carefully that while Testing Solution A
gives prompt change in color, DMG distinguishes between pal-
ladium and the other metals by forming a precipitate.
In a fourth test tube take several drops of water and a very small
amount of palladium solution; add DMG and see how small a quantity
of palladium is the smallest that you can detect. With practice in look-
ing at precipitates, your eye will detect smaller and smaller percentages.
Be sure that all your solutions are clear and free from precipitates
or flocculence; otherwise you will be deceived.
DMG AND NICKEL
Rub a piece of nickel a five-cent piece will do vigorously on

the unglazed bottom of your spot plate, so as to obtain a "streak"
of metal particles. Dissolve the streak with a drop of nitric acid or
aqua regia. Add a drop or two of DMG. Nothing special will
happen.
Now repeat, but with this difference: after dissolving the nickel
streak in acid, add a big drop of ammonia. Then add the DMG.
If you used enough ammonia to kill the acid, you will see a very
beautiful and characteristic color change.
From one point of view it is a waste of time to describe these
color changes, since they become of value only through being seen
by each observer. However, for the sake of the record we may say
that this color, produced by DMG, nickel, and ammonia, is a strik-
ing rose-red. And the precipitate produced by palladium and DMG
in acid solution is canary yellow.
Note this particularly. The test for palladium appears only in

acid solution; the test for nickel must be made in ammoniacal
solution.
This is very handy. Thus, suppose you have a piece of white
dental alloy as an unknown, and you suspect that it may contain
nickel or palladium. First get some of it into solution with aqua
regia. Then add DMG. If palladium is present you will see the
copious precipitate, canary yellow in color. Now add enough am-
monia to make the mixture smell faintly. If nickel is there, you
will at once see the beautiful rose-red. All in one test tube. Con-
tinue to use various scraps of metal, such as cheap white gold (which
is almost sure to contain nickel) and high grade dental alloys (which
are almost sure to contain palladium). These tests are so fascinating
that it will be no hardship to practice until your eye is quite thor-
oughly trained.
OTHER CHEMICAL TESTS THE GLOW TEST
There are other chemicals, in addition to Testing Solution A and

DMG, that can be used in identifying precious metals.
Mercuric cyanide (poison), which dissolves readily in water, can
serve as a test for palladium. It gives a dense precipitate when
added to mildly acid palladium solutions.
One interesting peculiarity of platinum and palladium is their
ability to soak up gases. This is utilized in several technical appli
cations, including a popular kind of cigarette lighter; a bit of finely
divided platinum or palladium is exposed to the fumes of naphtha;
it soaks up the fumes, and in so doing becomes hot, and finally the
fumes ignite.
It is essential that the metal be in a finely-divided form; preferably
as a fine powder. This phenomenon can be used as a test for
platinum-group metals in some cases. To make the test, dissolve
your metal in aqua regia, dip into the solution a bit of thin asbestos
paper; heat this for a few seconds in a flame to drive off the aqua
regia and dry the residue (which by now has been converted into
a metallic powder) and let it cool somewhat. Now hold the asbestos
over an unlit jet of gas; soon the powder will begin to glow, and
finally the gas will be ignited.
This glow test works successfully with platinum, palladium and
rhodium. However, if your solution contains considerable quan
tities of other metals, this test should not be used, as some of them
inhibit it, and others, such as copper, may sometimes produce a
false glow.
COPPER
The green color of dissolved copper sometimes identifies it. There

are several other metals, however, that give green solutions, includ
ing nickel, chromium, and sometimes iron. For most purposes, the
addition of strong ammonia to a green solution will settle the ques
tion. Take a few drops in a test tube or beaker, and add a few
drops of ammonia; the mixture may spatter, so work slowly. First
there will be a curdlike precipitate, but on the addition of more
ammonia, if copper is present, the precipitate will dissolve to yield
a clear solution of a beautiful sapphire blue color. This blue color
is characteristic of copper, and will distinguish it from chromium or
iron. Unfortunately, nickel also gives a blue color with ammonia
though not so deep a blue which means that you will then have
TOO REFINING PRECIOUS METAL WASTES
to add a little DMG to settle your doubts; thus nickel with DMG,
in the presence of ammonia, gives a rose-pink flaky precipitate
while copper does not.
IRON
Iron in solution has sometimes a pale green color the color of

copperas but more frequently a brown, yellow, or rust color. The
rust color is so much like the color of dissolved gold or platinum,
that the worker may hesitate to throw away a brown solution be
cause it looks so much like something of value. Here is a quick
test that reveals the presence of iron: dissolve a gram of ammonium
thiocyanate (potassium thiocyanate works the same way) in three
or four ounces of water; place one drop of your unknown solution
in a cavity of the porcelain spot-plate, and add one drop of ammo
nium thiocyanate. If dissolved iron is present a deep blood red
color will appear, usually at once, perhaps only after a few minutes'
exposure to the air.
This deep red color constitutes a very delicate test, as it will
appear when even a minute trace of iron is present. Indeed, if
you drop some thiocyanate on your hand, the chances are that
there will be enough iron on your skin, picked up from handling
iron and steel articles, to give a red stain. Do not make this test
in solutions containing much nitric acid, because the nitric acid
itself gives a red color with the thiocyanate. Mercury, oxalic acid,
and some other substances also interfere with this test. If copper is
present there may be a brown precipitate, which may or may not
obscure the red color produced by iron. A few acquaintance tests
made with solutions whose compositions are known will be help
ful here.
Here is a second test for iron in solution: add a few drops of po
tassium ferro cyanide, and if iron is present an intense blue color
will appear, either immediately or after exposure to the air.
RECOGNIZING ACID AND ALKALINE SOLUTION; LITMUS PAPER
One test is to take a few drops in a beaker and add a little

ordinary bicarbonate of soda; nearly all acids will produce an
effervescence. Chemists, however, usually use litmus paper. Lit
mus paper can be bought at any drug store for a few cents. Litmus
itself is a dyestuff extracted from a seaweed; it has the valuable
property of being red when in an acid solution, blue when in an
alkali solution; when in a solution that is exactly neutral, such as

pure distilled water, it is lavender-colored.
Get some litmus paper paper impregnated with this dyestuff
and learn its characteristics. Tear off a small piece, and drop it
into a beaker of water and add a drop of sulphuric, or nitric, or
hydrochloric, or oxalic acid. (Dozens of other acids will behave
the same way.) Litmus paper as bought will probably be of the
lavender color; when immersed in the acid it turns red. Now drop
a second scrap of litmus paper into a second beaker containing
water to which you have added a drop of ammonia, or a very small
amount of caustic soda, or a little potassium cyanide or sodium
cyanide. (There are dozens of other substances with alkaline or
caustic reactions that will behave the same way.) Notice that the
paper turns blue.
It is valuable to repeat these tests with many variations. Thus,
with a glass rod fish the red scrap out of the first beaker and drop
it into the second beaker. Also fish the blue scrap out of the second
beaker and drop it into the acid solution. If, as advised, your solu
tions are fairly dilute, you can repeat this many times, making the
dyestuff change from blue to red, from red to blue, over and over
again. (Concentrated acids or alkalies will destroy the paper.)
Finally, suspend a piece of damp red litmus paper over the open
mouth of the ammonia bottle; note that even the fumes will turn
it blue. Suspend a piece of damp blue litmus paper over the open
mouth of the hydrochloric acid bottle; and it will soon turn red.
Remember in acid, litmus is red; in alkaline solution it is blue.
College boys have a little rhyme that helps them remember this they
say "Alkaloo turns it blue." You will have occasion to use these tests
often in later chapters.
Another characteristic common to all soluble acids is their sour
taste. A single drop of hydrochloric acid will confer a sour taste
on a pint of water. However, so many acids are poisonous, espe
cially if concentrated, that it is unwise to taste them.
OTHER TESTS. A WARNING
The foregoing does not exhaust the list of tests that can be made.
There are dozens of others, but some of them are so difficult, or
so complicated by the presence or absence of this or that other sub
stance, that the non-chemist should not attempt them.
The non-chemist would also do well to confine himself, at least

in the beginning, to solutions that originate in his own shop.
Strange mixtures may be offered him by itinerant prospectors or
old-gold buyers. The non-chemist (or the chemist who is inex-
perienced in this field) would do well to exercise extreme caution
before purchasing such things, no matter what color tests they may
produce.
IDENTIFYING METALS IN SOLID FORM
The refiner has frequent occasion to examine buttons, old den-

tures, sediments, residues, and the like, to learn which precious
metals are present. If he is familiar with the reactions described
in this chapter, this will be no problem. As we say, the usual pro-
cedure is to get some of the metal into solution by warming it in
a test tube with nitric acid, or sulphuric acid, or hydrochloric acid,
or aqua regia—whatever is found necessary. An excess of acid should
be avoided. The solution is then tested by the methods just de-
scribed.
The familiar touchstone tests, used by jewelers when buying old
gold, utilize this principle on a small scale. These are described in
the book TESTING PRECIOUS METALS, mentioned on page
92. The metal is rubbed against a stone to leave a streak of metallic
particles which are brought into solution with a drop or two of acid.
The porcelain spot plate can be used instead of the stone, and re-
agents such as stannous chloride, copperas, or DMG will be added,
as the ingenuity of the operator may suggest.
CHAPTER X
GETTING ACQUAINTED WITH THE PLATINUM GROUP

Especially Platinum, Palladium and Iridium.
The six metals. Their employment in art and industry. Books to read.
Familiar characteristics. Platinum and ammonium chloride. Palladium
and ammonium chloride. Palladium and acids. Possible experiments.
A little about iridium. Questions and answers.
* # • •
Platinum, palladium and iridium are the commoner members of
this group, and the ones of greatest general interest. The other three,
osmium, ruthenium, and rhodium, will be discussed in Chapter
XVII.
These six sister metals share many characteristics, especially rarity,
white color, unusual resistance to corrosion, high melting point,
and many chemical peculiarities; but like human sisters, each has
an individuality of its own.
The reader undoubtedly has had some previous acquaintance
with the group. Thus if he is a jeweler he knows that pure platinum
is almost as soft as fine gold, and must be hardened for most indus-
trial purposes. He knows that the addition of a little indium will
produce an alloy of beauty and working qualities that make it
the supreme jewelry material. The alloy 10% iridium with 90%
platinum, called in the trade "10% iridio-platinum," is the favorite,
though lower percentages of iridium are often used.
Pure iridium is rarely seen, except by jewelers who make up
their own alloys. So popular are these iridium-platinum alloys that
the word "platinum" is now commonly applied not only to the pure
soft metal, but to these alloys as well.
The jeweler will also be acquainted with "platinum solders,"
which normally contain no platinum at all, the best ones consisting
of gold and palladium; the cheaper ones, of gold and palladium plus
other metals such as brass and silver. Perhaps the jeweler will
have met palladium in other forms as well; in England during the
103
first World War much handsome jewelry was made of it, and it is
used today as an ingredient of high grade white golds.
The dentist reader knows platinum and lndio-platinum, and
probably several of the alloys of platinum with gold and palladium.
Dental alloys, of which more will be said in Chapter XII, are of
many kinds, the precious metals being combined in various propor-
tions, oftimes with the addition of base metals.
The desirable chemical and physical properties of the platinum-
group metals have insured their use in many other fields as well.
Chemical ware is made of platinum because of its resistance to most
chemical reagents. Hypodermic needles are made of 20% lndio-
platinum because of its strength and the ease with which it can be
sterilized. Immense quantities of platinum and its alloys go into
radio and electrical equipment. An alloy of osmium and iridium
(called osmindium) is used to tip fountain pen points, because of its
hardness and resistance to corrosion. Most of these uses are familiar
to the layman.
The word '"platinum" is derived from the Spanish "platina,"
meaning "little silver." Iridium gets its name from its iridescent or
many-colored compounds. Palladium was named from the planet
Pallas, which happened to be discovered at about the same time,
Pallas in turn being named for the Greek goddess of Wisdom.
The history of the platinum metals, and the story of their oc-
currence in nature, are interesting and romantic, but have no place
in this book. Nevertheless, a knowledge of that history will have
a dollar-and-cents value to every man who refines, works, buys or
sells these metals, because it gives him that familiarity that is the
basis of successful business.
BOOKS TO READ
Chapter E at the end of this volume, on "Books to Read," might

be consulted at this point. If you are in a business that employs
the precious metals, you can without cost obtain several publications
of high merit, simply by writing for them on your letterhead.
While they are primarily advertisements, they contain general in-
formation of especial interest to the refiner. The following are
notably useful:
TECHNICAL AND INDUSTRIAL PLATINUM, put out by
J. Bishop and Co., Malvern, Pa.
GETTING ACQUAINTED WITH PLATINUM 105
PLATINUM, A THUMBNAIL HISTORY, by Sigmund Cohn,

44 Gold St., New York City.
DATA CONCERNING PLATINUM, by Baker & Co., Newark,
New Jersey.
PLATINUM-PALLADIUM-GOLD IN DENTISTRY, by the
International Nickel Co., Inc., 67 Wall St., New York City.
LABORATORY APPARATUS OF PLATINUM, GOLD,
SILVER, by the American Platinum Works, Newark, N. J.
THE HANDY BOOK FOR JEWELERS, by Handy & Harman,
Bridgeport, Conn. Price, $1.00.
FIFTY YEARS OF PROGRESS, by Thomas J. Dee & Co.,
Chicago, Illinios.
FAMILIAR CHEMICAL CHARACTERISTICS
The reader who has studied previous chapters of this book will
already be familiar with several characteristics of interest to the
platinum refiner. Thus in Chapter III, in which he removed dirt
and base metal fiom platinum filings, he observed the following:
(1) Platinum may be boiled in nitric acid without dissolving or
suffering other change.
(2) Platinum is not attracted noticeably by the magnet.
(3) Platinum does not dissolve in caustic soda or potash.
(4) Platinum shows no tarnish after being heated to redness.
(5) If he has done electro-plating he has observed that platinum
does not dissolve in cyanide solutions, even under the electric current.
(6) In Chapter VI he learned about the use of metallic precipitants
such as copper wire, or zinc shavings, to get gold out of solution.
In the discussion on this method he read "One disadvantage is this:
if any platinum or palladium should be present in the solution, they
will come down with the gold, in metallic form also. . . . In this
respect the metallic precipitants differ from copperas, sulphur
dioxide, sodium nitrite or oxalic acid; when these latter are used,
platinum or palladium will remain in solution while the gold alone
is thrown down. This difference is important." Indeed, as he
will shortly learn, differences such as these are the basis of most
of our refining procedures.
(7) In Chapter IX, when making up his standard solutions, he
dissolved weighed samples of pure platinum, palladium and gold;
he observed that gold and palladium dissolved readily in cold aqua
regia, but that platinum reacted so slowly that he heated the dish
to hasten matters. This difference in speed with which gold and
platinum are attacked by cold aqua regia is important and will be
referred to again later.
There is another respect in which platinum differs from gold.
You will recall from Chapter VI (See Quartation) that gold can be
separated from silver by treating a silver-rich alloy with nitric or
sulphuric acid. That is, if an alloy containing three times as much
silver as gold is boiled with either of these acids, the silver will
dissolve and all the gold will remain as a brown powder, thus
effecting a separation. This method cannot be used to separate
platinum from silver, as platinum-silver alloys behave quite dif-
ferently from gold-silver alloys. Thus, if you should expose an
alloy of 95% silver and 5% platinum to boiling nitric acid, every-
thing will dissolve, the platinum forming a soluble platinum nitrate,
from which it is not easy to recover the platinum. If you should
expose an alloy of 75% silver and 25% platinum to nitric acid, most
of the silver and some of the platinum will dissolve, and there will
be a considerable residue which contains both metals. In other
words, you will not effect a separation.
Silver-rich platinum alloys react to boiling concentrated sulphuric
acid in about the same way no separation is effected. With sul-
phuric acid that is slightly diluted, a fair separation can be obtained.
This point will be mentioned again in the latter part of Chapter XII.
Before the reader becomes an accomplished refiner he will come
to know several other characteristics of platinum, but for the
moment (since we are anxious to begin actual work) we shall
content ourselves with only one or two further observations.
(Several new chemicals and pieces of equipment will be needed in
Chapters XI, XII and XIII. These are described in Chapter XI. Of
these, the following will be employed now: the porcelain crucible and
the clay triangle that supports it; sodium chlorate; ammonium chloride.)
PLATINUM AND AMMONIUM CHLORIDE
Place in a small beaker a half ounce or less of the platinum solu-

tion you prepared in Chapter IX. This is a solution of platinum in
hot aqua regia, made up with as little aqua regia as would do the
work, plus water; its chemical name is "Platinum Chloride" or
"Chloro-platinic acid." Notice the yellow color. In a second vessel
dissolve a tablespoonful of good quality ammonium chloride in a

little water, to make a clear concentrated solution. (Cheap am-
monium chloride gives a muddy solution that must be filtered before
use.) Add one-third of the ammonium chloride solution to your
platinum chloride solution, stir, and let settle.
The orange-colored powder that forms is called platinum am-
monium chloride or ammonium chloro-platinate. These names are
long, and it is usual to refer to it simply as the "orange powder."
Next, let us separate the orange powder from the liquid, using
a small filter paper, and collecting the clear solution in a clean
beaker. Test a drop of the clear solution in a cavity of the spot
plate with stannous chloride solution. . . . You will find that it
still contains a little dissolved platinum.
In other words, the ammonium chloride precipitated most of
the platinum, but not all. Naturally you will add a little more
ammonium chloride with the idea of completing the precipitation,
after which you will test the clear liquid again with stannous
chloride. . . . You will find that it still contains a little platinum—
enough to be detectable.
In this respect ammonium chloride differs from reagents like the cop-
peras with which you can precipitate gold so completely that no ordinary
test will show its presence.
There are at least two good ways to reduce the orange powder
back into metallic platinum: by heating it strongly in a crucible,
and by boiling it in formic acid; and even the amateur refiner
should know both methods.
The first way is the older method, and the one we shall use now.
(Later, in Chapter XIV, we shall use the formic acid method.) Let
the powder dry, then brush it into a porcelain crucible and heat
it, gently at first, then more strongly. Use a Bunsen flame, or a
small gas flame such as is used in cooking, or a gas-fired muffle fur-
nace. Do not use a gold-melting furnace.
Observe the effect of heat. The powder shrinks and changes
color, and fumes appear which are hard on machinery as well as
on human lungs, so work under the hood. If the fumes are white
you are working at proper speed; black or yellow fumes mean you
are heating too rapidly. Soon you will have left a lump of gray
metallic powder, called platinum sponge, all the ammonium

chloride having gone off in fumes.
This sponge is (if these directions have been followed) quite
pure, and should be saved. Residues such as the liquid above, which
contain a trace of dissolved platinum, should be saved too, though
of small value. Place this liquid aside for the moment, and later add
it to the "Stock Pot" which will be described in Chapter XI.
PALLADIUM AND AMMONIUM CHLORIDE
Place in a small beaker a half ounce or less of the palladium solu-

tion you prepared in Chapter IX. This was made by dissolving pure
palladium in a little aqua regia, evaporating off the excess acid, and
adding a little water; its chemical name is "palladium chloride."
In appearance it resembles both gold chloride and platinum chloride.
To this add a little of the ammonium chloride solution you pre-
pared for the previous test a quarter ounce, more or less. Does
anything happen ?
If your palladium is pure, and if your standard solution was pre-
pared as directed, nothing will happen. This experiment shows
that palladium differs from platinum in that it is not precipitated
with ammonium chloride from an aqua regia solution from which
excess nitric acid has been boiled off.
Warm the mixture gently, and add a single crystal of sodium
chlorate. Or, dissolve a pennyweight of chlorate in a little warm
water, and add a drop at a time. There will be considerable fizzing,
and as it subsides, add another small crystal or another drop. The
solution should be just short of boiling. After the addition of a little
more of the chlorate, a change in color will be noted. Stir the
mixture with a glass stirring rod; soon a dark red precipitate will be
seen, floating on the surface and then sinking. Continue to add
chlorate, a little at a time until the liquid is almost colorless, and
then let cool. This red powder is palladium ammonium chloride
chemically much like the orange powder above.
If you heat this red powder in a crucible in the same way that
you treated the orange powder, the results will be practically the
same, as would be expected. However, the palladium sponge that
remains after heating will rarely show a clean gray color; usually
there are bluish tones suggesting peacock feathers, due to palladium
oxides. This is one difference between platinum and palladium;
palladium is apt to form oxides, and the oxides show rainbow colors.
Another and very important difference was hinted at in Chapter
III, where you were removing base metals from platinum filings,
using nitric acid. You were told not to use the nitric acid treat-
ment if your material contained any pure palladium, or platinum-
palladium alloys in which the latter metal ran as high as 40%. The
reason for that is that palladium dissolves in nitric acid.
Palladium is the only metal of the platinum-group that shows this
characteristic. It also dissolves in boiling sulphuric acid. Hydro-
chloric acid has little effect upon it.
In the following chapter we shall expose platinum, iridio-platinum, and
palladium to a variety of reagents. A full discussion of all the chemical
changes that might go on would consume many pages, and the list of
acquaintance tests that might be performed, if one had unlimited time,
is almost endless. Thus, the reader might profitably take samples of
his standard platinum and palladium solutions and expose each in turn
to copperas, sodium nitrite, oxalic acid, sulphur dioxide gas, zinc shav-
ings, copper wire, table salt, and many other things.
We have assumed all along that the reader has never formally studied
chemistry in school or college; we can assure him that the formal study
of this science consists mostly of just such simple acquaintance tests, the
addition of one substance to another under varying conditions of concen-
tration, temperature, and so on, with careful observations of the effects.
If the reader has performed all the experiments thus far described he
has learned a method of approach which if followed will carry him well
along toward an understanding of the precious metals.
A complete understanding would involve much study and many

experiments, as can be judged from a glance at the list of books that
have been written about these metals. Indeed, if you should ask any one
of the experts on this subject for example the chemists at the National
Bureau of Standards they would assure you that one lifetime is far too
short for a complete understanding of this fascinating group. But that
does not mean that an intelligent layman cannot learn enough to refine
profitably the materials used in his own shop.
Experiments, therefore, will suggest themselves to him from time to

time. If he uses small samples, and works slowly, he will run little risk
of accident and be sure to learn much. Beginners always use too large
samples, which waste time and money, and often by their sheer bulk,
obscure the reaction.
Moreover, it is possible also to learn while doing; that is, to proceed

with the actual work of refining some jewelers' filings or some dental
scrap, accomplishing at the same time the recovery of precious metal
and the attainment of precious knowledge. This we shall do, pausing
now only to mention a few more facts.
A LITTLE ABOUT IRIDIUM
Pure iridium is a very hard metal, insoluble in any single acid,

and attacked even by aqua regia with the greatest difficulty, and
only under special conditions. Platinum that is alloyed with 30%
iridium can be boiled for hours in aqua regia with only minor
loss. In its chemical functions it resembles platinum in many re-
spects; for example, if ammonium chloride is added to iridium
chloride, there will soon appear a precipitate of dark iridium am-
monium chloride, chemically so much like the orange powder that it
is difficult to separate the two. Other characteristics will be dis-
cussed in later chapters, as the process of refining reveals them.
* * * *
QUESTIONS AND ANSWERS
Q. I saw a display of platinum-group metals at the World's Fair a

series of buttons, bars, crucibles, articles of jewelry, etc. To my eye all
six metals were exactly the same white color. But in a book I read, one
metal is described as "steel white," another as "silvery," another as "tin-
white," etc. How do they differ?
A. To most eyes, rhodium is the whitest, with iridium and palladium
tied for second place, and platinum close behind. The color depends
partly on the form of the metal. Thus the platinum sponge obtained
by heating the orange powder is gray; melt this sponge into a button
and it will be tin-white.
The color of electrodeposited metal will vary with the conditions under
which it was deposited, from the mirror-like brilliance of good rhodium
plating to a dull leaden hue in badly done work.
A well-polished surface seems whiter than one bearing many small
scratches. The usual cause of darkening, however, is an impurity; thus
small amounts of other metals, of impurities soaked up from the cru-
cible, or gases absorbed from the torch while the metal was molten, can
all cause darkening. But it is generally agreed now that all six metals
when pure and properly melted, are so nearly the same tin-white color
that experts cannot certainly distinguish them by the eye alone.
CHAPTER XI
JEWELERS' FILINGS AND CLIPPINGS CONTAINING

MIXTURES OF GOLD WITH PLATINUM-GROUP
METALS
Kind of materials handled. Equipment and chemicals. Beginning work,

(1) General cleaning. (2) Removing base metals with nitric acid. (3)
Removing gold with aqua regia. (4) Removing silver chloride.
(5) Melting the platinum filings. (6) Recovering dissolved platinum.
(7) Burning the orange powder. (8) Melting the platinum sponge.
(9) Recovering the dissolved gold. (10) Washing and melting the fine
gold. The Stock Pot. (11) The Silver Jar. Summary. Questions
and answers.
* * * *
KIND OF MATERIALS HANDLED
This chapter covers the filings and clippings ordinarily produced

in a jewelry shop where both gold and platinum articles are made.
Material of this type is rarely found in dental laboratories, but it
is one of the most important encountered in the jewelry industry,
and usually offers more profit to the small-scale refiner than any
other. Even the smallest jewelry shop will find that these gold and
platinum mixtures can be refined easily and with profit.
So important is this type of waste that some readers may turn
immediately to this chapter and skip the preceding ones. If so,
they should be warned that a knowledge of much that has gone
before is essential for the proper understanding of what follows.
We shall exclude from this chapter all buttons or alloys in which
gold and platinum-group metals are MELTED together. (In
Chapter XII we shall discuss such buttons or alloys fully; they are
an important part of dental wastes.) This distinction between a
mixture of particles of gold and platinum, and an alloy in which
these same metals are melted together, is important; the reason will
be clear shortly. We shall for the present, exclude amalgams also,
Jewelers' filings of this type contain the usual dirt, such as iron
from the files, emery and bits of paper that we found in Chapters III
and V.
The platinum used in ordinary high grade jewelry, included in this
chapter, normally contains a little iridium to stiffen it. White golds,
and the solders used in making platinum jewelry may contain pal-
ladium. This iridium and palladium are taken care of in this
chapter, and recovered in due time. Included also is old jewelry in
which platinum and gold are soldered together; for example,
platinum-topped gold jewelry.
We shall include in this chapter all the kinds of gold normally
used in jewelry, except one. Thus, we include yellow gold, red
gold, rolled and filled gold, white gold, and low-karat green gold.
We shall exclude only 18-k green gold. You will recall from Chap-
ter VII, that 18-k green gold behaves differently from most gold
alloys in that it does not dissolve in aqua regia. For that reason we
exclude it from this chapter, reserving it for Chapter XVI.
* * * *
The procedure consists essentially of the removal of first one kind
of unwanted matter, then another, from the particles of platinum.
That is, we shall remove, in turn, the shellac and grease, the iron
from the files, the base metals, and the gold, our effort being to leave
the platinum particles unchanged as far as is possible.
With some mixtures this is quite easy; with others it is impossible.
That is why we have spent so much time in listing the mixtures that
can be handled in this way. Our reasons for including some things
and excluding others will be clear to the reader after he has been
through the procedure once or twice.
This chapter, then, can be used only when the composition of
the mixture is known, in the sense that a jeweler working on his
own wastes knows what sort of alloys and materials have been
employed. The professional refiner who buys filings, old dental
scrap, and the like, usually does not have sufficient knowledge of his
purchases to justify him in using this chapter.
To summarize: This chapter treats of the filings and clippings

produced in a jewelry factory where platinum and ordinary gold jew-
elry are being made, and in which particles of gold and platinum
are mixed—not melted—together.
JEWELERS' FILINGS AND CLIPPINGS 113
We shall use all the equipment already employed, and a few other
things beside. That includes table, burner, exhaust fan or hood,
wash-bottle, funnels, funnel support, evaporating dish, graduate
glass, glass stirring rods, spoon-spatula, casserole, asbestos to go
between casserole and flame, glass beakers, test tubes, the spot plate
and dropping bottle used in testing; steam bath, magnet, frying
pan for burning out dirt; a crucible for melting the refined gold,
and at least one stoneware crock of generous size.
Wash-bottle. Mortar and pestle. Pipe-clay triangle.
Watch glass.
Porcelain crucible. Tongs for handling crucible.
In addition you will want the following for use in this chapter
and in Chapters XII and XIII: a second wash-bottle, smaller than
the first (Eimer & Amend No. 18952, size 500 cc, at$0.60);a mortar
and pestle, No. 28688,5 1/2 inches across, at $0.95. A porcelain crucible
No. 21241 or 21243, size 4, at about$0.50,will be big enough for the
small shop. Sand crucibles used in melting gold can be used instead,
but are not nearly so satisfactory. (Graphite crucibles should not
be used at all with platinum.) A pipe-clay triangle No. 32646,
size 7 1/2 cm., is needed to support your porcelain crucible while it is

being heated; price $0.13. Crucible tongs No. 32553, 9 inches, at
$1.10, will be used for picking up hot articles. Watch glasses, to
cover dishes and beakers, make for cleanliness; No. 32850, assorted
sizes, say one each of sizes 3, 4, 5, 6, and 7 inches, cost about $0.62
You will use the following familiar chemicals: nitric acid, hydro-
chloric acid, copperas, sulphuric acid, stannous chloride testing solu-
tion, granulated zinc or zinc shavings, and fluxes for melting gold.
In addition the following new chemicals: 5 lb. ammonium chloride,
U. S. P. or C.P., at about$0.25per pound; 1 pound cyanide sodium
cyanide, or cyanide-chloride mixture, or "Cyanegg"; (ordinary
commercial purity is good enough) price about $0.40; a pound of
sodium chlorate, C.P., about $0.80 and a pound of pure ammonia
(ammonium hydroxide) at $0.50 or less.
BEGINNING WORK
For your first refining, handle only a small quantity of material,

not over three or four ounces. As you become familiar with the
procedures, you will handle larger and larger batches with ease.
Briefly, the material will be treated in the following steps:
(1) Removing bits of iron with a magnet; removing shellac and
grease by caustic or by burning in a frying pan. These steps were
discussed fully in Chapters III, V, and VI.
(2) Removing base metals with nitric acid; this is the same as in
Chapters III and V.
(3) Removing gold by means of cold aqua regia; at the same time
a little platinum is dissolved. This differs only slightly from the
procedure described in Chapter V.
(4) Removing silver chloride. This normally leaves your plat-
inum clean.
(5) Melting the clean platinum filings.
(6) Recovering the platinum that is dissolved in aqua regia.
(7) Melting the platinum sponge.
(8) Recovering the gold that is dissolved in aqua regia, using
methods already learned in Chapters V and VI.
(9) Recovering the silver, if worth while.
(10) Laying aside the remaining solution—called the Stock Pot—
which contains a little platinum and probably some palladium and
iridium, for treatment in Chapter XIII.
Clippings and old jewelry, etc., must be rolled thin; the thinner
the better. Hammer them flat; then feed them into the rolls with
a little scoop whose sides have been cut out so that it will fit against
the rolls; or put them on a piece of paper and feed in paper and all.
Roll very thin; a few minutes spent at the rolls may save several
hours later. Twist the pieces so that they will not lie flat in the acid,
then anneal.
Do NOT melt the filings into a button no matter how dirty
they may be if you expect to refine them according to the procedure
of this chapter.
(I) GENERAL CLEANING
This is in every way similar to the methods used in Chapters III,

V and VI. Thus:
A. Take out bits of iron with a magnet. Put these with your
sweeps, as they may contain some value. (Some kinds of white gold
are attracted to a magnet; so if your white gold is of this kind, do
not use the magnet.)
B. Most workers burn out organic matter such as paper, lint, etc.,
in a frying pan. A few little cinders will be produced; they will
be washed out later.
C. But if your filings contain much shellac or grease, instead of
burning them, place them in a casserole and boil in lye or caustic, as
described in previous instructions, after which wash well, with plenty
of hot water. Let the wash-water settle overnight, then throw it
away, being careful not to pour over any filings. The mud that has
settled out contains some platinum, so put it with your polishings,
with old filter papers, or in the sweeps.
( 2 ) REMOVING BASE METALS WITH NITRIC ACID
Base metals, including some of the alloy in the gold, will now be
dissolved in nitric acid. Have the filings or clips in the casserole, and
cover them well with nitric acid. Use C.P. acid, and add a little
water, exactly as described in Chapters III or V. If there is much
base metal add the acid slowly, as it may boil over. The familiar
heavy brown fumes appear, so you must work near the fan. Dis
solve as much as possible of the base metals. The more base metals,
the more acid needed. When there are no more brown fumes on
stirring, add a good quantity of hot water to dilute the acid, and
pour it off through a clean filter, leaving the filings in the dish. Let
the liquid run through into a jar with a wide mouth; it contains
dissolved silver, copper, and base metals, and is the same as the
"Silver Jar" of Chapter V. Wash the filings clean with plenty of
hot water, as usual. Filter this wash water in with the acid. The
cinders produced when you burned out the organic matter will
gradually be flushed out by this and subsequent washings.
Sometimes there is so little silver in your material that it does not
pay you to save this nitric acid solution. But if there is enough
silver in it to be worth recovering, follow the procedure given in
Chapter VIII.
If your original material contained any tin, you will now see
more or less of that jelly-like tin paste noticed in Chapter VII in the
discussion on soft solder. You can dissolve this tin paste by means
of a little sulphuric acid; then wash this off and throw it away.
( 3 ) REMOVING GOLD WITH AQUA REGIA
All the gold is now to be dissolved. A little platinum will dis-

solve unavoidably at the same time. For every Troy ounce of gold,
more or less, add
1 fluid ounce nitric acid,
1 fluid ounce water,
4 fluid ounces hydrochloric acid.
Do not add sulphuric acid yet. Use enough aqua regia to cover
the filings well. Do not heat this (unless you are in a great hurry)
because the hotter it is the more platinum it dissolves, and we do
not want to dissolve any more platinum than necessary. If your
material is mainly gold, add the aqua regia slowly, as it may bubble
over.
Let this stand till there is no more bubbling when you stir.
Usually 24 hours are enough, but the thicker the lumps of gold the
longer it takes. In hot weather it works more rapidly. You can
hasten action by stirring frequently, by heating, and by having all
pieces rolled very thin. In some cases two hours are long enough
to dissolve all the gold. If you have much gold, put on a little aqua
regia at first, let it act all it will, then pour it off and add more. It is
better to do this than to add all the aqua regia at once.
Try not to use any more aqua regia than is necessary to dissolve all
the gold. If you use more than necessary, you not only waste acids and
time, but also make trouble for yourself later. With a little experience
you will learn how to dissolve all the gold, and at the same time have all
the acids killed when you are through.
When there is no further bubbling on stirring, pour off the aqua
regia solution (do not filter it yet) into a pitcher or jar, leaving the
filings in the casserole. This liquid contains dissolved gold, a little
dissolved platinum, perhaps some palladium and iridium, and such
base metals as were not reached by the nitric acid. This includes a
little lead in some cases.
To the liquid now add a little sulphuric acid, to throw down the
lead, if any is present, as white lead sulphate. Use say an ounce or
less of concentrated sulphuric acid to each quart of solution, add it
slowly, stirring all the while. The lead sulphate settles slowly.
Returning now to the platinum filings in the casserole: they are

mixed with a greyish mud of silver chloride, formed from the silver
that was alloyed with the gold. It forms a tight scum over the metal
and may keep the aqua regia from acting on it.
Dump the mass of filings into the mortar, and rub them hard
with the pestle, to loosen this scum of silver chloride. Then dump
them back into the casserole, washing them in with the wash-bottle.
Pour the water and loosened chloride into the pitcher of aqua
regia solution, leaving the filings in the casserole. The idea is
to wash off as much of this loose chloride as possible, but try not to
use more water than is necessary. At the same time you will be
flushing off cinders and other light-weight dirt.
Look through your material carefully, and if any large lumps of
gold are left, you will see them. Test to see if there are any small
pieces by adding a little fresh aqua regia, and watching the effect; if
in five minutes you see no bubbles on stirring, then the gold is prob-
ably all dissolved. If it is not all dissolved, leave the metal covered
with aqua regia for a few more hours, or until all the gold is dis-
solved.
Wash this aqua regia in with the other aqua regia in the pitcher,
washing well but using as little water as will do the work; you can
get your filings cleaner with several portions of water, of a few
spoonfuls each, than with one large portion, if you stir well and
pour off each portion completely before putting on the next. Wash
until there is no more color in the wash waters.
(4) REMOVING SILVER CHLORIDE FROM THE FILINGS WITH CYANIDE

(OR AMMONIA)
When you are sure that your filings are free from acid, fill up the
casserole with hot water, and add a piece of cyanide. For 100 dwt of
filings a piece as big as a walnut may be enough. Stir well, and heat
to boiling if possible. This treatment dissolves any remaining silver
chloride mud, and leaves the platinum clean and bright. Use either
sodium cyanide or potassium cyanide, or the chloride-cyanide
mixture.
Cyanide is a deadly poison. When mixed with acid it gives off its
deadly fumes more rapidly. That is why you must wash off the acid
so thoroughly.
If you do not like to use cyanide, you can use ammonia instead, but
it is slower. Use it exactly as you use cyanide wash the filings WELL,
then add a strong solution of ammonia, chemically pure, in hot water.
Stir often. Do not boil the ammonia, but set it in a warm place.
If there was much gold in your filings originally, there will be so
much silver chloride that you may have to repeat this treatment.
Finally pour or filter off the cyanide water (or ammonia) and wash
the filings well with much hot water. Be careful not to pour off
bits of platinum. Washings containing cyanide should not be put
in the Silver Jar; if the amount of silver is small you are advised
to throw them away at once. But if you want to keep and refine
them, keep them in a separate container, and follow the suggestions
in Chapter VIII.
As soon as the filings are clean and dry, they are ready to be
melted with the oxygen-gas flame. Pick them over well; you may
find bits of gold. Remember that small pieces dissolve before the
big ones, so if you cannot find any big ones there probably are no
small ones. Pick out any tiny squares of metal with a white look;
they are a certain kind of platinum solder.
If you have any 18-k green gold, or any gold-platinum dental
alloy, in your material, you will find them now. Pick them out, as
they require methods to be described in later chapters. Run through
the filings with a magnet to be sure all steel is removed. If the
filings contain much sand or enamel, see later instruction. A little
emery is often seen but it does no harm.
(5) MELTING THE PLATINUM FILINGS
It is always wise to melt about a pennyweight of metal first, to

make a test button If good, melt the whole lot. Remember that
it is easy to get dirt out of filings, but very hard to purify an impure
bar. So if your sample is bad, now is the time to clean your filings
further. Sometimes it is enough to wash them once more with
cyanide and water. Sometimes you must soak them again in aqua
regia. Clean filings will melt easily and perfectly, without smell or
smoke, and without loss. See Chapter A of the Appendix.
(6) RECOVERING DISSOLVED PLATINUM
Return now to the aqua regia solution which contains dissolved

gold, also some dissolved platinum; (also dissolved base metals and
probably some dissolved palladium and iridium). If the copper or
nickel content is high it will be greenish; otherwise it will be brown.
Probably by this time the sediment of silver chloride and lead
sulphate has had time to settle. This sediment must be removed
sooner or later, and now is one good time to get rid of it, so let it
settle well, pour off the clear part, and filter the remainder. The
sediment should be washed free of solution and placed in the Silver
Jar.
The next step is to drive off excess nitric acid. You were told to
use no more aqua regia than was needed, but there is always a
tendency to use too much, because fresh acids work much more
rapidly than acids that are almost exhausted. Put the solution in an
evaporating dish and heat it slowly, preferably on a steam bath, until
syrupy. Do not let it spatter. A maximum temperature of 140 to
150 degrees Centigrade is advised. Now slowly add a little full
strength hydrochloric acid to the syrup, and you will see the brown
nitric acid fumes being expelled. Evaporate to a syrup a second
time, possibly a third time, adding hydrochloric acid each time.
This is the same procedure that you used in Chapters V and VI.
If you carry the evaporation too far, you will form a hard cake
and gold may appear as a brown metallic film on the sides of the
dish. Try to avoid this. If metallic gold should bake out, dissolve
a crystal of sodium chlorate in a spoonful of hot water, and pour the
solution drop by drop where the gold has collected; this dissolves it.
Work slowly, drop by drop, as this combination (chlorate and acid)
tends to spatter violently.
120 R E F I N I N G PRECIOUS M E T A L W A S T E S
You may have to evaporate a third or fourth time, each time add-
ing hydrochloric acid, to get rid of all the brown nitric acid fumes.
One reason why we added sulphuric acid to the aqua regia solution
is because with sulphuric acid the unused nitric acid can be driven
off with fewer evaporations. Finally thin the syrup with several
volumes of water, and wash it into a large vessel (not a bottle)
which should not be more than two-thirds full. There should not
be more than about three ounces of precious metal to each quart of
liquid. Normally there is still some sediment present silver
chloride and lead sulphate that have managed to escape previous
efforts for their removal.
Get rid of this sediment, either by letting it settle and pouring off
the clear solution, or by filtering. The solution must be perfectly
free from cloudiness before you go to the next step.
The next step is to precipitate platinum by means of ammonium

chloride. For each quart of solution you will need about half a
pound of ammonium chloride, dissolved in hot water to make a
concentrated solution. Good quality (C. P. or T. P.) chloride is
best; cheaper stuff contains trash which must be filtered out.
Add the strong ammonium chloride solution to your aqua regia
solution, and stir well. Use plenty; for each pennyweight of plat-
inum that is in solution, you will need a little over 1/2 a pennyweight
of ammonium chloride, and a definite excess is advisable.
Immediately the orange powder of platinum ammonium chloride
will appear. About 95 per cent, more or less, of the platinum that
was in solution will be precipitated in this form.
The color will range from canary yellow through orange to a deep
maroon. Do not be surprised if the shade is different every time; it
depends upon the temperature at which it forms, the impurities present,
and other factors as well.
Let the orange powder settle. If your solution contains much

IRIDIUM, some iridium-ammonium-chloride may come down,
mixed with the platinum salt. The longer it stands, the more of
this dark iridium compound will come down.
(Moral: if you want your recovered platinum to be free from iridium,

filter off the orange powder within an hour or so. But sometimes you
want the iridium to come down with the platinum; in that case let the
jar stand overnight in a cool place.)
The next step is to get the orange powder into a filter paper; pour
the clear liquid through first, and then dump the orange powder in,
rinsing it in with a little of the solution that has already run
through. Let the liquid (which contains dissolved gold, etc.) run
into a big vessel such as a crock not a bottle.
Some of the powder will stick to the vessel. So dissolve a handful
of pure ammonium chloride in a pint of water, making up roughly
a 15% solution, and put it in your small wash-bottle. Label this
little wash-bottle, so as not to confuse it with the big one containing
plain water. Wash the orange powder into the filter with this am-
monium-chloride-water, using as little as will do the work. Chemists
often use a glass rod tipped with a bit of rubber to wipe off the
powder that sticks to the glass; these rubber tips are called "police
men," and can be bought at your supply store.
When all the orange powder is in the filter, and all the liquid has
run through, and before the powder dries, wash down the sides of
the filter with some of this ammonium-chloride-water, so as to get
the orange powder into a lump. We use ammonium-chloride-
water because some of the powder would dissolve in plain water,
but very little dissolves in ammonium-chloride-water. But use as
little as will do the work.
It is now necessary to wash the orange powder, because it is wet
with dissolved gold, dissolved copper, etc. A quick way is to pierce
the bottom of the filter paper with a glass rod and permit the powder
to run through into a clean dish below. Wash it down well, using
the ammonium-chloride-water. If you have time, wash the powder
into a wide flat vessel, cover it with plain water, and heat it gently
on a hot plate or small flame for several hours. Use a large jar so
that the layer of powder will be relatively thin. Let cool before
you filter again. These schemes serve to rinse and clean the orange
powder.
Finally get the powder into a new filter paper, washing it into a
lump as before. The liquid that runs through is of value, so add it to
the aqua regia solution. Spread the paper out on a plate to dry.
(7) BURNING THE ORANGE POWDER
When the orange powder is dry, dump it into a clean shallow

porcelain or sand crucible, filling it not quite full. Knock all the
orange off the paper that you can (if you washed it well this will be
easy; if not, very difficult), and put the paper with your other old
filter papers. Do not burn the paper with the powder. Porcelain
crucibles cost more than sand and are more fragile, but are preferred
because the powder does not cling to the smooth surface. If you
can afford it, get a fused quartz evaporating dish for this job. It is
expensive but should last indefinitely.
Heat the powder, as in the acquaintance test, for some time. A
large lot will take an hour or more. The ideal method is to use a
gas-fired muffle furnace with a good flue; small shops use a gas
burner such as is used for cooking, or a Bunsen burner, supporting
the crucible on a clay triangle. Do not use a gold-melting furnace,
as the draft may carry the powder up the flue. Do not use an
electric furnace unless the wiring is carefully protected against the
fumes. The fumes are dangerous to the lungs and should be carried
off, preferably through a duct that can be cleaned out periodically to
salvage any metal that might collect in it. Refiners handling much
metal sometimes play a stream of burning hydrogen gas on the
powder as it is being reduced. This method can be dangerous if
not performed with the utmost precision and is not advised for the
small shop.
Heat the powder gently at first, and hold a cold object, such as a
pair of tongs, in the fumes. If a white crust appears on it, all right;
but if yellow or black stains appear, you are heating too rapidly and
will lose platinum. If you are using a pyrometer, proceed slowly
till a temperature of 370° C. is reached; after that the heat can be
increased rapidly to 900°, a t w h i c h a l l v o l a t i l e m a t t e r i s b u r n e d off .
You now have a mass of grey platinum sponge metallic platinum,
finely divided; sometimes it clings together in a sort of clinker.
Sponge reduced in this way by heat is sometimes spoken of as
having been ignited or calcined. There is an alternative method of
reducing the orange powder, preferred by many workers because it
involves less chance of loss; it employs formic acid, and is described
fully in Chapter XIV.
(8) MELTING THE PLATINUM SPONGE
Sponge is melted in practically the same way as filings; this process

will be described more fully in the Appendix. The result, assuming
that the refining was properly carried out, should be a button of
almost pure platinum rather soft because it can contain at most

only a little iridium.
Beginners sometimes make a bad refining the first time, so it is
wise to make a small test button first, before melting the whole lot.
(More trouble and loss in refining is due to neglect of this last para-
graph, than all other mistakes put together. The wise refiner takes this
precaution every time.)
If you get a bad test button, boil the sponge for ten minutes in
strong cyanide water. This generally dissolves the dirt, which is
probably silver chloride. Wash well with much hot water, dry, and
make another test melt.
Generally this gives good metal; or sometimes one or two re-
meltings will give a good button. (Another practice, which we do
not recommend, is to melt up a small lot of bad metal with a large
amount of good new metal, gradually getting rid of bad metal in
that way.)
But perhaps you made some serious mistake in your refining, and
cannot get good metal. In that case, turn to Chapter XV on special
repurification, and follow those directions. It is much easier to
purify bad sponge than to purify bad buttons.
(9) RECOVERING THE GOLD FROM THE AQUA REGIA SOLUTION
Return now to the big vessel of aqua regia solution, which con-
tains all the gold. (It also contains base metals, palladium, perhaps
some iridium, and a little platinum.)
The dissolved gold is now to be recovered. Use copperas, or
sodium nitrite, or sulphur dioxide, as preferred, as described in
Chapters V and VI.
When using copperas, be sure to add a little hydrochloric acid
to the copperas solution, enough to make it a clear green. Among
other things, this precaution tends to prevent palladium from being
carried down with the gold.
If, in procedure (6) you did not succeed in driving off all the
unused nitric acid in your aqua regia, you will have difficulty here;
brown bubbles will appear and gold will refuse to come down.
When this happens, you must get rid of this nitric acid by methods
described in Chapter V.
(10) WASHING AND MELTING THE FINE GOLD—THE STOCK POT
Let the fine gold powder settle for some hours; then test a few
drops of the liquid to see if all the gold has been precipitated, using
a clean green crystal of copperas as described in Chapter IX. Watch
it for a couple of minutes at least. If nothing happens you have used
enough copperas. But if a dark cloud forms around the crystal,
add more copperas (or sulphur dioxide) to the main solution and let
the gold settle again.
Or, if the solution does not contain too much platinum or palladium,
use the stannous chloride test. (Large quantities of platinum or pal-
ladium in the solution will mask the gold reaction.)
After a satisfactory test, dip or siphon off the liquid into the
largest crock you have, called the Stock Pot. You will have to
filter the last part of the liquid.
Dump the mass of gold into the casserole, and wash it as de-
scribed in Chapter V, first with hot water, then with hydrochloric
acid, then with water again. These washings are all filtered and
added to the Stock Pot, as they contain dissolved platinum-group
metals.
Do not let any gold get into the Stock Pot if you can help it.
The liquid in the Stock Pot always contains a small amount of
platinum as well as palladium and iridium. These will be recovered
in Chapter XIII. It also contains considerable copper, nickel, etc.
If you used copperas as your precipitant, the Stock Pot will contain
much ferric iron in solution.
If your original material contained much palladium, some pal-
ladium may come down with your gold. If so, the gold powder will
look dark, even black, and when melted will give a whitish button.
When this happens, turn to Chapter XV on Repurification.
After thorough washing, the gold powder is usually ready to be
dried and melted. But here again it pays to make a small test melt,
before melting the whole batch. Normally this method gives gold
pure enough for most jewelers' requirements; but if the original
material contained much tin and lead, or much palladium, for
example, it may be advisable to go through the repurification.
(The purity of recovered gold depends, first, upon the kind and quan-
tity of impurities that you started with, and second, upon your skill and
patience.)
IIl) T H E SILVER JAR
You will recall that in some instances you were told to throw away
silver residues, especially the cyanide residues that are so dangerous to
handle. This advice is meant primarily for the small user.
The larger user might do well to save these, if he has good storage
facilities, and can keep the cyanide fumes from endangering his health.
But he should recall that even at best, silver residues have small value,
and usually are not worth refining as such, but should be put in with
the sweeps, wash-barrel wastes, or other low-grade stuff.
The Stock Pot the solution from which all the gold and most
of the dissolved platinum have been recovered is set aside for
further treatment in Chapter XIII.
SUMMARY
The following diagram, designed for the use of the man to whom
the various steps are already somewhat familiar, will be helpful as
a reminder and summary. ("Ppt." means "precipitate.")
JFWELERS' MIXTURES particles of PLATINUM; particles of GOLD alloyed with SILVER and BASE
METALS; organic TRASH such as paper, grease, shellac; IRON from the files; SAND or emery
(The platinum may be a loyed with palladium or iridium, or both )
Remove organic matter either by burning or by means of a caustic solution
Remove iron particles with a magnet
Dissolve with nitric acid all base metals that can be reached Filter
RESIDUE: PLATINUM; GOLD alloyed with BASE METALS and SILVER; FILTRATE: Nitrates of
SAND. SILVER andBASF METALS
Treat with cold aqua reg ia until all gold particles have been dissolved. Add hydrochloric acid or
Filter. table Salt. Let settle
RESIDUE: Bulk of FILTRATE: Dissolved GOLD, some dissolved PPT: FILTRATE
PLATINUM in almost its PLATINUM, possibly some dissolved PALLA- SILVER BASEMFTAL
original form, mixed DIUM and IRIDIUM, with dissolved BASE CHLORIDE salts
with SILVER CHLORIDE METALS. See Discard
(and SAND). ChapVIII
Expel excess nitric acid, take up with water
Treat with cyanide or and hydrochloric acid, filter and discard any
ammonia. residue, add ammonium chloride.
RESIDUE: FILTRATE: RATE: FILTRATE contains dissolved
PLATINUM Cyanide or Oruige GOLD, BASE METALS, and
(and ammonia powder. traces of dissolved PLATINUM,
sand). salt of Reduce PALLADIUM, and IRIDIUM.
SILVER. tQ metallic
See PLATnrou Add copperas. Let settle and
Chap. VIII. filter.
PPT: is FILTRATE:BASEMET-
FINE ALS, traces of PLATI-
GOLD. NUM-GROUP metals.
To Stock Pot Chap
ter XIII.
Q. Is ammonium chloride poisonous? In using the little wash-bottle

I got some of the solution into my mouth, and it had a bitter taste.
A. Chemically this salt is related to table salt (sodium chloride), and
small quantities will do you no harm. In fact, it is used in some kinds
of cough medicine. However, you are advised not to swallow the solu-
tion, as it can upset the stomach. Ammonium chloride is also called
"sal ammoniac," especially by older chemists.
* • • *
Q. Is there any way of knowing exactly how much platinum will go

into solution in aqua regia in a given time?
A. The quantity will depend upon the following factors: (1) the area
of platinum the surface that is reached by the acids; fine filings expose
much area, thick lumps expose relatively little; (2) the temperature of the
acids; the warmer the aqua regia the more rapidly it acts; (3) how
often you stir; frequent stirring permits fresh acid to reach the metal
and hastens action; (4) the composition of the platinum; pure soft plati-
num dissolves more rapidly than the iridio-platinum that is normally used
in jewelry, and the higher the percentage of iridium, the less of the metal
will dissolve. Also, some "platinum" is alloyed with more or less pal-
ladium, with or without gold and perhaps with nickel and other metals.
These alloys dissolve more rapidly than pure platinum; as the percentage
of these metals goes up, the rapidity of solution goes up.
* • * *
Q. We are told not to burn the paper with the orange powder. This
paper has considerable powder sticking to it. What should be done
with it?
A. Discard any part of the paper that shows no stains, and put the
rest in with your "Paper Storage," which will be taken care of in
Chapter XVIII. We do not burn the paper with the orange powder
because the carbon in the paper can combine with platinum, under cer-
tain conditions, to form a compound (a carbonyl) that is volatile.
* # # *
Q. Why is it necessary to remove all the excess nitric acid? Could we
not add an extra quantity of copperas and get rid of it that way?
A. Several things are accomplished by the removal of the nitric acid,
the first being the familiar one of facilitating the precipitation of gold.
Also, nitric acid hinders the precipitation of the orange powder, and at
the same time causes part of the palladium to come down with the
platinum.
Q. I have a friend who is doing this work, and he throws down his
gold before throwing down the platinum. Is this all right?
A. Under certain conditions it is advisable. There are several possible
modifications of this procedure which will be discussed in Chapter XIV.
Q. What are the fumes that appear when the orange powder is burned?
A. Several different substances are present: ammonia, hydrochloric
acid, chlorine gas, and probably water and other things. This particular
reaction has been studied by chemists for a long time, but they are dis-
satisfied and at least one laboratory has been investigating the matter
recently. Undoubtedly some platinum can be carried off mechanically,
and apparently a volatile platinum compound can be formed, especially
if carbon is present.
* # # *
Q. When I heated the orange powder it darkened and then seemed
to creep up the sides of the crucible. Later it clung so tightly to the
crucible that I could not get it off. What causes this?
A. This happens when the powder is damp when being burned, also
when the ammonium chloride wash water is too concentrated. This
excess ammonium chloride is what causes the stuff to cling to the cru-
cible. Impurities such as copper salts will stain the crucible.
Q. You tell us to clean the platinum filings with cyanide, to get rid of
the mud of silver chloride. Now, I have heard that cyanide will dis-
solve gold. Isn't it possible that I should lose some gold this way,
especially if I do as you say and throw away the cyanide afterwards?
A. Yes, there is a possibility of a small loss of gold here. That is one
reason why some people use ammonia for this step, in spite of its slow-
ness. However, cyanide dissolves gold very slowly, even when helped
along by aeration, which is not done here. Moreover you were not ad-
vised to use the cyanide until after you had reason to believe that the
gold was already removed, either wholly or in large part.
Q. After evaporating to a syrup to get rid of excess nitric acid, I am

told to dilute the syrup with "several" volumes of water. What does
this mean?
A. With a little experience you will learn the quantity best suited to
your material. Some workers use only three or four volumes of water,
others use as much as ten volumes. If you use too much water you
waste time in filtering the bulky solutions, and the proportion of plati-
num recovered as orange powder will be relatively low. On the other
hand, if you use too little water, your nickel or copper salts may deposit
out as a green crystalline crust mixed with the orange powder. Watch
out for this crust, and use enough water to prevent its formation.
• * * *
Q. I have a friend who does this work, and he always adds a little
ammonia when he precipitates the orange powder. Does this do any
good?
A. It does no good, and it can do harm. For example, if too much
ammonia is added (enough to neutralize the acid), all your metals will
precipitate together gold, iron, tin, platinum, zinc as a mass of pasty
compounds, difficult to redissolve and separate. Even a little ammonia
can do harm; you will recall in Chapter IX we said that ammonia can
form complex substances in which the platinum metals are so tied up
that they may not give the usual reactions. If these so-called amino
compounds do form, they may make your work in Chapter XIII more
difficult.
# * * •
Q. I have a friend who is doing this work, and when he precipitates
the orange powder he adds alcohol and a big piece of ice. Does this do
any good?
A. Ice and alcohol, either together or separately, are sometimes used.
Ice is added because more orange powder comes down in a cold solution
than in a warm one. Some people put the vessel in the icebox over-
night for this reason. However, this chilling may cause nickel or copper
salts to crystallize out with the orange powder; accordingly, the method
is not generally recommended. Alcohol can also increase the proportion
of platinum that comes down. However it reacts with the iridium and
palladium (like ammonia) to form compounds that may not be desirable.
Accordingly the addition of alcohol is not recommended except in a
few rare cases where this reaction with iridium is desirable.
Q. I was reading an old book about platinum, and it said that platinum
should be precipitated by sulphuretted hydrogen. Is that a good method?
A. Sulphuretted hydrogen, whose modern name is hydrogen sulphide,
is a poisonous gas with an unpleasant odor the odor of rotten eggs.
It will precipitate platinum if the acid concentration is right, but it will
also precipitate gold, copper, lead, mercury, and many other metals as
well, especially the heavy metals. Because of this latter fact it is rarely
used in the recoverv of nrecious metals.
CHAPTER XII
ALLOYS AND BUTTONS CONTAINING GOLD AND

PLATINUM-GROUP METALS
Material handled. Jewelers' buttons. Dental alloys. Equipment and

chemicals. (A) Preliminary treatments. (B) Melting into a button if
necessary. (C) Granulating the button or rolling it thin. (D) Dissolv-
ing the metal in aqua regia. (E) Recovering iridium from the silver
chloride. (F) Recovering dissolved platinum. (G) Recovering gold.
(H) Recovering silver if worth while. (I) The Stock Pot. Summary
and diagram. Silver-rich buttons. Questions and answers.
* # # *
MATERIAL HANDLED
In Chapter XI we refined mixtures of gold and platinum par-

ticles, the process consisting of the removal of first one, then another
contaminating substance, the platinum particles finally being left
clean and in practically their original form.
In this chapter we assume that our gold and platinum metals are
already melted and alloyed together, as in the case of certain dental
alloys. Such material must be dissolved completely, all the gold and
all the platinum going into solution and then being precipitated
separately. Both jewelers and dental technicians will handle ma-
terials of this kind.
It is assumed that the reader is familiar with earlier chapters of
this book, has had experience in dissolving and recovering gold, and
has performed the acquaintance experiments in Chapter X. If he
has carried through a refining according to Chapter XI, so much
the better.
JEWELERS' BUTTONS AND ALLOYS
The jeweler occasionally has materials suitable for refining ac-

cording to this chapter. Thus, after a few weeks of refining the
mixtures described in Chapter XI he will have accumulated a
number of old filter papers his paper storage containing particles
129
of precious metals and considerable dirt. He will reduce this to a

button, as described in Chapter XVIII on Low Grade Wastes, which
button should be refined according to the procedure shortly to be
described.
In past years, when indium and platinum were much more ex-
pensive than they are now, platinum alloys containing gold, pal-
ladium, nickel, silver, and even other metals, were sometimes made
into jewelry. Scrap and old jewelry made of this stock should be
handled according to this chapter. White gold of the highest class
(see Chapter VII) contains palladium; scrap and old jewelry made of
such white gold should also be handled as described here. Chapter
IX on "Identifying" tells how to recognize these complex platinum
alloys and how to determine whether a white gold does or does not
contain palladium.
DENTAL ALLOYS
Precious metals are employed in dentistry in several forms, some

of which have already been discussed. Pure gold in the form of foil
or adhesive sponge, is used in fillings; the scrap, which is nearly 24-k
gold, should be refined according to Chapter V or VI. There are
several kinds of dental gold solders and alloys in which gold and
silver are the only precious metals present; these also should be
refined according to Chapter V or VI. Amalgams and amalgam
alloys contain silver, and sometimes (by accident) other precious
metals as well; these should be laid aside and treated according to
Chapter XVI. Platinum and iridio-platinum pins also are discussed
in Chapter XVI, and generally should be treated separately, but if
they are included by accident in the material described below, no
harm will be done. Finally, we reach the very important modern
alloys that are the main concern of this chapter the wrought gold
alloys and the casting gold alloys in which gold and platinum-group
metals are present.
Base metals also find large employment in dentistry, and nat-
urally become mixed with the precious metal wastes. Skinner's
SCIENCE OF DENTAL MATERIALS, listed in the Appendix,
describes their application. Thus aluminum is used in making den-
tal plates; so are stainless steel and some of the chromium alloys.
Nickel and tungsten are used in dental pins. Steel is employed for
dental instruments, drills, etc. "Technic" alloys, which are not used
in the mouth, but only for practice, demonstration, etc., are of two
ALLOYS AND BUTTONS 131
types first the fusible alloys such as the bismuth-lead-tin alloys

which are used in dies and counterdies, some of which melt at
temperatures as low as that of boiling water, and second the yellow
silver-copper technic alloys that look like gold. The workman
should try to exclude these base metals as far as possible from his
valuable wastes, as they not only consume time and chemicals, but
some of them, like lead and tungsten, complicate the refining even
when present in small quantities.
Modern dental gold alloys always contain silver and copper, fre-
quently platinum, palladium, zinc and tin as well. Sometimes
iridium, rhodium, nickel, and indium are present in small amounts,
and the base metals aluminum, magnesium, manganese and iron
have been found.
Much interesting research has been done on dental alloys during
the last few years, both by the firms that sell them and by the Na-
tional Bureau of Standards, with the cooperation of the American
Dental Association. If the reader is not already familiar with these
investigations, he should at once obtain some of the publications
described in Chapter E of the Appendix. A familiarity with these
recent researches will be of great usefulness both to the user and the
refiner of these alloys.
The wrought gold alloys and casting alloys which are the main interest
of this chapter, are used for a variety of prosthetic purposes, and many
formulas have been proposed. Gold is the principal constituent of most
of them, especially the yellow ones. The silver content of the wrought
gold alloys ranges from about 5% to as much as 15%; this point will be
brought up again later. Platinum and palladium together may run
from 10% to as much as 25%. Normally iridium is present in small
proportions less than 1% and probably got in by accident rather than
by design. In the white alloys there is considerable nickel as much as
18%; copper runs from 9% to 15%, more or less, while zinc rarely runs
over 1%. The American Dental Association Specifications2 r e q uire
1 Indium is a rare metal, of the aluminum group, sometimes added in small quan-
tities. See U. S. Patents 1,987,451 and 1,987,452, N. O. Taylor, January 8, 1935.
Because of its rarity indium is not taken into consideration in this book. If you know
that your metal contains indium in sufficient quantity to justify its recovery, sec
NEW PROCEDURE FOR THE ANALYSIS OF DENTAL GOLD ALLOYS, by
Raleigh Gilchrist; Research paper RP 1103, Journal of Research of the Bureau of
Standards, vol. 20, pp. 745-771, June, 1938.
2 These specifications were prepared by the National Bureau of Standards, in
cooperation with the Dental Association and the manufacturers of dental gold ma-
terials. See Chapter E of this volume.
that dental wrought gold alloys should contain at least 7 5 % gold-and-

platinum-group metals.
The casting gold alloys are also made up in a wide variety of formulas.
They differ from the wrought golds primarily in that the gold content
is relatively higher up to 90% and over and the platinum-group metals
are lower. The Association Specifications require for soft casting alloys
a minimum of 83% gold and platinum group, and for medium and hard
casting alloys a 78% minimum.
The dental technician will have a variety of wastes containing these
alloys, including scrap, clippings, old dentures, and the stuff swept from
the bench where they are fabricated.
For convenience we shall divide these materials into two grades:
first grade, consisting of clean buttons, sprues, dentures, clean scrap,
and clean old jewelry; while second grade scrap consists of filings
and stuff swept from the bench, which contain considerable dirt
such as emery, rouge, carborundum, wax, rubber, plaster of paris, in
vestment compound, charcoal, and what not.
For the present we shall exclude all amalgams and mercury; these
will be taken care of in Chapter XVI.
For the present we shall handle material that contains compara-
tively little silver. Later in this chapter we shall consider the button
in which silver is the major element.
These are the same as were employed in Chapter XI. Since in this
chapter we dissolve all our metal, we may need some more large
vessels, such as a large evaporating dish and one or more large jars
or crocks, in addition to those already used.
BEGINNING WORK
For your first refining take a very small quantity, say 40 or 50 dwt.,
in order to gauge the size of vessels that this method calls for.
Briefly, the procedures are as follows, most of them being already
familiar.
A. Any preliminary treatment that may be useful.

B. Melting into a button if necessary.
C. Granulating this button, or rolling it thin, as preferred.
D. Dissolving the metal in aqua regia.
E. Recovering the iridium from the silver chloride.
F. Recovering platinum from the aqua regia solution, as usual.

G. Recovering gold from the aqua regia solution, as usual.
H. Recovering silver if worth while.
I. Laying aside the remaining solution the Stock Pot for treat
ment in Chapter XIII.
A. PRELIMINARY TREATMENTS
First grade clean scrap rarely needs any preliminary treatment.

But second grade stuff, in which there is much assorted dirt, should
be sieved, and probably burned in the frying pan or boiled in caustic,
and it may help to take out iron particles with a magnet. Remember
that if appreciable quantities of amalgam are present, the material
should be treated by methods to be described later Chapter XVI.
Nitric acid, which we used in Chapters III, V, VI and XI to get
rid of base metals, is sometimes employed here, but not always, for
a variety of reasons. Thus, high grade wrought or casting golds
are not attacked by it; therefore nothing would be accomplished by
using it. Sometimes it pays to use it on the second grade (dirty) ma-
terials, depending upon the kind of dirt present, but usually it is
easier to flux the dirt off in the next procedure. A further considera-
tion is the possible presence of palladium. You will recall that pal-
ladium, when present in pure form or in certain alloys, is soluble
in nitric acid; also a few rare alloys containing platinum will go
completely into solution when boiled in nitric acid.
Accordingly, take a sample of your material, and boil it in nitric
acid for a few minutes, then test the liquid with stannous chloride
testing solution. If it shows that platinum or palladium has gone
into solution, omit the treatment with nitric acid.
If you find that nitric acid dissolves only base metals, then treat
the whole quantity of material with nitric acid as usual. If the
solution contains worth while quantities of silver, put it in your
Silver Jar and refine according to Chapter VIII.
If you are in doubt, omit the nitric acid treatment.
B. MELTING THE MATERIAL INTO A BUTTON, IF NECESSARY
Second-grade stuff, especially if it contains much carborundum or

emery, is generally melted into a button.
Mix it well with a flux such as crushed borax glass and soda ash,
using equal weights of flux and material, and melt in a gold-melting
furnace, using a sand crucible. The flux serves to liquefy the dirt,
while the metallic particles collect into a button at the bottom of the
crucible. If the proportion of emery is high, use more flux. The
chapter on Low Grade Wastes gives formulas of other fluxes. Do
not use litharge. Let the melt cool in the crucible, which is then
broken and the button carefully separated from the slag.
First-grade clean stuff, such as old dentures, may or may not

need to be melted into a button. It does no harm in any case, and
after a little experience you can tell whether it does any good or
not; different kinds of metal call for different treatments.
Thus, if your metal either first-grade or second-grade material

runs more than half platinum, whose melting point is high, ordinary
melting will not get it into a homogeneous button; you will want
to lower the melting point of the whole mass by adding some brass,
or perhaps some copper or zinc. If you have some rolled or filled
gold scrap to get rid of, here is a good place to utilize it by melting
it in with your second-grade dental wastes.
When breaking the crucible to remove the button, look out for beads
or shot metal embedded in the slag; if there are many of these it means
that you did not heat long enough, or did not use enough flux, or that
your platinum percentage was too high. If properly done, the melting
will consolidate all the metallic particles into a button, and all the
emery, rouge, plaster of paris, etc. will go into the slag which then can
be thrown away.
C. GRANULATING THE METAL, OR ROLLING IT THIN
The next task is to get the metal into such form that acids can
work upon it easily. Sometimes the lumps or buttons are so soft
that they can be rolled thin as paper. Cut the strips into short
lengths and twist them so they will not lie flat in the acid, then
anneal. Roll very thin; a few minutes spent at the rolls will save
hours in the acid.
But often these metals refuse to roll, and it is easier to granulate
them. This is done by remelting the metal and pouring it while
molten into a big vessel of water. (See Chapter V.)
If your metal is as much as 12% silver, the silver will cause
trouble when you try to dissolve the button in aqua regia. In
Chapter VII on green gold we discussed this subject and the reader
would do well to refer to this chapter now. So, if you know your
metal to contain as much as 12% silver, melt some other metal in
with it gold, or copper, or zinc or brass sufficient to get the silver
content down to say 10% or less.
If you do not know what the percentage of silver is, drop a piece
of your metal into a little aqua regia, heat to a boil, and see what
happens. If the acids are very slow to act the chances are that it
would pay you to melt your metal up with brass, gold, or other
metal, in proper quantity to reduce the silver to 10% or less.
Usually it is best to add this extra metal when you are about to
granulate the material. But some workers like to add it when
making the first melt, just described.
D. DISSOLVING THE METAL IN AQUA REGIA
Place the metal, which by now is rolled thin or granulated, in a

casserole and cover it with aqua regia. Mix as follows:
Nitric acid, full strength 1 part
Hydrochloric acid, full strength 4 parts
Water . 2 parts
Do not add any sulphuric acid yet.
Use 6 or 8 fluid ounces of the mixture to each Troy ounce of
metal; with much platinum you may need more.
Sometimes the acids go to work at once; usually you will have to
heat. If possible, set the casserole on a steam bath as that does not
cause spattering or require continuous attention. Stir often to dis
lodge the scum of silver chloride. Place an inverted funnel over

the dish, to prevent spattering, save acid, and keep out dirt.
Some people like to use an Erlenmeyer flask for this step, because it
permits them to see what is going on. The thin Pyrex glass models
withstand heat well. The frontispiece shows several Erlenmeyer flasks
in operation, and on page 171 the reader will find other illustrations,
with catalogue numbers and prices. The refiner may have occasion to
use both the narrow-mouthed form and the wide-mouthed; the latter
when there is some danger of boiling over. The narrow-mouthed form
comes in many sizes, from the tiny 10 cc. size to the 6000 cc. model.
The wide-mouthed form is made in three sizes 250, 500, and 1000 cc.
Another advantage of these flasks is that the material therein can be
stirred easily merely by rotating the flask not too briskly, of course,
or it might froth over.
Do not use an evaporating dish for this procedure, as it permits too
much acid to evaporate off wastefully.
The length of time required to dissolve the metals depends upon
several factors. You can hasten matters by having the pieces very small
and thin, by heating, and by stirring often to loosen the coating of silver
chloride. Also, the more platinum and the more silver in the metal,
the slower the action of the acids.
Silver, especially, makes the process slow. If it runs as high as 12%
the slowness is noticeable; if as high as 25% the metal is practically
insoluble.
Sometimes it helps matters to pour the solution off into a pitcher,

then dump the whole mass of metal into the mortar and grind it
hard with the pestle, to loosen the chloride. Then dump it back
into the casserole, washing it back with a little water; then cover it
again with the aqua regia solution, which will now go to work
again with vigor.
Sometimes individual lumps or strips of metal will be much slower
to dissolve than the bulk of the material. It may pay to pick out
such lumps and lay them aside until they can be melted up and the
silver content reduced to 10% or less.
Finally all the metal will seem to be dissolved, and you will have
left only the powdery residue of silver chloride. Here, however, we
meet a new consideration a point that has not come up before. If
your original material contained iridium, the iridium will now re
main undissolved as a metallic dar\ powder, mixed with the silver
chloride.
This fact that when these alloys are dissolved in aqua regia the
iridium remains undissolved was not appreciated by most early
workers, but careful research at the National Bureau of Standards 3
and elsewhere has established the fact that iridium does have this
peculiarity. And if the platinum content is high, sometimes some
of the platinum also will remain with the silver chloride. This may
or may not complicate your task, however, first because dental alloys
normally contain little if any iridium, and second because the plati-
num content is rarely high. The iridium residue has a dark color,
and as you become more experienced you can tell by its appearance
whether the silver chloride does or does not contain enough value
to justify its further treatment. Assuming for the moment that our
material might contain a good amount of iridium, we would pro-
ceed as follows:
E. RECOVERING IRIDIUM FROM THE SILVER CHLORIDE RESIDUE
When the aqua regia has dissolved all the metal that will dissolve,
dilute it with its own volume of hot water, let it stand a little while;
then pour or siphon off the solution do not filter it yet leaving
the residue in the dish. Wash the residue and add the washings to
the main solution. This main solution contains the gold, platinum,
and palladium, and the base metals. Add a little sulphuric acid to
it at this point, an ounce or less to each quart of solution, stirring
well and adding the acid slowly. This precipitates any lead that
might be present, and also facilitates the removal later of the excess
nitric acid. Set the solution aside to settle, while we turn our atten-
tion to the residue in the dish.
Take a small sample, say as much as can be taken on the tip of

the porcelain spoon, place it in a small beaker or tiny porcelain cas-
serole, and add chemically pure ammonia diluted with hot water.
Ammonia dissolves silver chloride, leaving iridium or platinum un
dissolved as a fine heavy dark residue, so fine that it may be almost
3 See William H. Swanger's paper on ANALYSIS OF DENTAL GOLD ALLOYS,
Scientific Paper of the Bureau of Standards No. 532, of August 11, 1926. Also see
Raleigh Gilchrist's NEW PROCEDURE FOR THE ANALYSIS OF DENTAL GOLD
ALLOYS, Journal of Research of the National Bureau of Standards, Vol. 20, June
1938, pages 745-771. These papers, which are devoted primarily to the analysis
of these alloys, rather than to their refining, are mentioned again in Chapter E of
the Appendix.
invisible. Rotate the little vessel and whirl all the powder into the
center, and it will be easier to see.
Frequently there is a light-weight light-colored residue, insoluble
in ammonia, that is merely flakes of glass or porcelain scaled off
the dishes and acid bottles. This is plainly of no value, and should
not deceive a careful observer.
You may find that no precious-metal residue remains; if so your
silver chloride is of small value and may be placed in the Silver Jar
or with other silver residues. But if the ammonia test reveals ap-
preciable quantities of iridium, then the whole quantity of silver
chloride must be dissolved out.
This procedure is similar to procedure 4 in Chapter XI, where you
dissolved the silver chloride from the platinum filings, using either
cyanide or ammonia. If you are not already familiar with this pro-
cedure, go over it again now. Simply cover the material with water to
which cyanide or ammonia is added, and heat. Most workers prefer
ammonia for this task, rather than cyanide, because it is safer and be-
cause the silver may be recovered therefrom more conveniently.
Sometimes this treatment reveals pieces of metal that have not

been acted upon sufficiently by the aqua regia, and which must be
returned for further treatment, but after you have had a little ex-
perience this will not occur.
The residue that remains after all the silver chloride is dissolved
will consist of the fine dark iridium powder, which may contain
some platinum, and almost always there will be the light flakes of
porcelain or glass from the vessels. Wash it well and catch it in a
filter paper. The disposal of this powder will depend upon the
needs of the worker. If the iridium is to be used in making up
new alloys, the flakes of porcelain or glass should be removed with
hydrofluoric acid according to the method given in Chapter XVI.
If not, let the residue dry and then sell it to a professional refiner
who gives credit for iridium.
F. RECOVERING PLATINUM FROM THE AQUA REGIA
Returning now to the aqua regia solution that contains dissolved

gold, platinum, palladium, and base metals: we shall recover first
the platinum, then the gold, according to methods that were used
in Chapter XI. Turn back now to Chapter XI, procedure 6, and
proceed exactly as there described.
You will note that the first step is to consider any sediment that
may be present; sometimes there is much sediment, sometimes very
little, and it can be removed now or later, as desired. The next step
is to drive off excess nitric acid, dilute with water, filter if not per-
fectly clear, and recover platinum with ammonium chloride, exactly
as in procedures 6, 7 and 8 in Chapter XI.
G. RECOVERING GOLD
This is accomplished precisely as in procedures 9 and 10, Chap-

ter XL
H. RECOVERING SILVER IF WORTH WHILE
This is accomplished as in Chapter VIII.
I. THE STOCK POT
Here again we follow the procedure of Chapter XL The solution

from which you recovered the gold and most of the platinum still
contains all the palladium, as well as a little platinum. This consti-

tutes what is called the Stock Pot, and it is laid aside for further
treatment in Chapter XIII.
• * # *
SUMMARY AND DIAGRAM
The following diagram will summarize this chapter, and will be

useful to the worker who is already familiar with the various pro-
cedures. It is helpful to compare it with the scheme given at the
end of Chapter XI.
REFINING PRECIOUS METAL WASTES
DENTAL ALLOYS alloys containing GOLD with PLATINUM GROUP metals and BASE METALS
possibly mixed with organic TRASH such as paper, wax, vulcanite, as well as sand, emery, abrasives,
iron from files, etc
Remove iron with a magnet

Remove organic trash by burning, or by means of caustic, or by melting the material into a button
Heat with aqua regia until all metallic particles are dissolved
RESIDUF SILVER CHLORIDE, mixed FILTRATE GOLD, PLATINUM GROUP metals (except
with dark powder of IRIDIUM, also iridium) and BASE METALS, in solution
SAND
Expel excess nitric, take up with water and hydro-
Treat with cyanide or ammonia chloric acid, filter and discard residue, and to the clear
solution add ammonium chloride Let settle and filter
RESIDUE: IRIDIUM, FILTRATE Cyan
SAND. See ide or ammonia PPT Orange FILTRATE: GOLD, traces of PLATINUM-
Chap. XVI. salt of SILVER See powder Re GROUP metals, and BASE METALS, in
Chap VIII.or dis duce to me solution.
card talhc plati
num Add copperas. Let settle and filter.
PPT: FILTRATE contains traces of

FINE PLATINUM-GROUPmetals,espe
GOLD. cially PALLADIUM, with base
metals. To Stock Pot Chap
ter X I I I .
SILVER-RICH BUTTONS
As we intimated earlier in this chapter, if an alloy containing both

gold and platinum-group metals should also contain much silver, it
will require somewhat different treatment. If the silver content is
as high as 15%, the button will be very difficult to dissolve in aqua
regia. The best plan, in such cases, is to add enough gold (or copper
or brass or zinc) to reduce the silver content to 10% or less, then to
proceed according to the method just given.
However, it occasionally happens that an alloy contains so much
silver, and so little gold and platinum-group metals, that the above
expedient will be unsuitable. If you can so manipulate your ma-
terials as to avoid the formation of buttons of this type, do so, as
they are not easy to refine.
You will recall in Chapter X, that when a button containing much
silver and only a little platinum is treated with nitric acid, more or
less platinum goes into solution. If the platinum content is low, it
will all dissolve; if the platinum content is high, part of it will dis-
solve, and part of the silver will remain in the residue. In other
words, nitric acid does not effect a separation. And it is not easy to
recover the platinum from the nitric acid solution.
The way out is to dissolve such buttons in sulphuric acid not

full strength acid, but acid containing about one part water to three
parts acid. (Pour the acid into the water, slowly, and stir well.)
This can be done when the gold-platinum content is not over 25%.
The process is similar in principle to the inquartation described in
Chapter VI, where a silver-gold alloy was refined by means of sul-
phuric acid; and the same precautions should be observed.
After the sulphuric acid has acted all it will, pour it off into a
large jar of water, leaving the residue in the vessel. The liquid in
the large jar will contain sulphate of silver (and sulphates of base
metals), and the silver can be recovered as in Chapter VIII. The
residue will contain gold, platinum, and possibly some silver that
was not reached by the acid. Collect it and wash well, then treat it
with aqua regia exactly as you treated the gold-platinum alloys that
constitute the main topic of this chapter.
If you should, by inadvertence, dissolve a silver-platinum alloy in nitric
acid, and find that a worthwhile quantity of platinum has gone into
solution as platinum nitrate, do this: to the nitric solution, which con-
tains platinum, silver, and perhaps other metals, add enough hydrochloric
acid to precipitate all the silver as silver chloride. Filter, and handle
the silver chloride as described in Chapter VIII. The solution will con-
tain platinum dissolved in a sort of aqua regia in which the proportion
of nitric acid is unusually high. Evaporate as usual to expel the nitric
acid, then recover the platinum with ammonium chloride, as usual.
Q. Is it essential to add sulphuric acid to the aqua regia solution? I

have a friend who does this work, and he says he never uses it.
A. No, the sulphuric acid is not necessary. Many expert workers
never use it at all, even when lead is present, maintaining that it tends
to make the solution so strongly acid that the precipitation of gold is
hindered. On the other hand, one of the largest and best refiners of
dental wastes uses considerably more sulphuric acid than is suggested
in this book. See REFINING AND MELTING SOME PLATINUM
METALS, by J. O. Whiteley and C. Dietz, of the Dentists' Supply Co.
Laboratories, published by Mining and Metallurgy, March, 1928.
* * * *
Q. After precipitating the orange powder, I set the jar aside overnight
and next morning there was a crust of green granular stuff all over the
orange powder, and some was floating on the liquid. What shall I do?
A. The green crystals are probably a nickel or copper salt, crystallized

out when the solution cooled. You may be able to get rid of some of
the crystals by fishing them out with the porcelain spoon. Dissolve the
rest in a little hot water, stirring well; then before they have time to
form again capture your orange powder in a filter paper, and proceed
as usual. This happens often when your material contains much nickel
or copper.
• * • •
Q. My material contains considerable palladium, but very little plati-
num, and no indium to my knowledge. When I add the ammonium
chloride in Procedure F, I get hardly enough orange powder to show
on a filter paper. Can I omit this treatment?
A. Yes. If you know that your solution contains very little platinum
you can omit the ammonium chloride, and proceed immediately to take
down the gold. Whatever platinum is present will be recovered when
you refine the Stock Pot in the next chapter.
* # # •
If you have not already done so, we suggest that you read the group
of Questions and Answers at the end of Chapter XI.
CHAPTER
THE STOCK POT
Material handled. Equipment and chemicals. Collecting the precious

metals with zinc. Refining the black powder. The iridium residue.
The dilute aqua regia solution platinum, palladium, gold. Burning
the red palladium powder. Further treatment. Disposal of the final
tailings. Diagram. Questions and answers.
* * * *
MATERIAL HANDLED
This chapter covers the recovery of the precious metals, usually

present in small amounts, that remain in the solutions that were
left over at the ends of Chapters X, XI and XII.
These solutions should be stored in a large crock of good-quality
stoneware. Cheap earthenware sometimes absorbs solution and in
time may crumble.
These solutions normally contain only a little platinum. Palla-
dium and indium will be present in either large or small amount,
depending upon the original material; dental scrap usually contains
considerable palladium and very little iridium; jewelers' scrap will
differ from batch to batch.
If you used enough copperas in Chapters XI and XII there will
be no dissolved gold in the Stock Pot; if some should remain
through accident, it will be recovered in time. If you filtered prop-
erly there will be no silver chloride and no lead sulphate.
Other base metals will be present in considerable quantities. Thus
if you used copperas (iron sulphate) to precipitate your gold, there
will be much dissolved iron in the Stock Pot; this gives it a brown
color. Copper, nickel, zinc, tin, cadmium, and possibly other base
metals will be present as well.
The other metals of the platinum group rhodium, osmium, and
ruthenium are present here only under exceptional circumstances.
If present, they will be recovered in time. They will be discussed
more fully in Chapter XVII.
143
The Stock Pot should contain only such solutions as are definitely
described as suitable to be added to it. In particular, never let any
cyanide get into the Stock Pot, nor the silver nitrate solutions that
belong in the Silver Jar, nor any of the other solutions you were
told explicitly to throw away.
The Stock Pot should be kept lightly covered to exclude dust. If
possible, set it on top of a radiator or steam pipes, that it may slowly
concentrate by evaporation.
At the beginning of Chapter XI we mentioned the equipment

and chemicals that will be used in this chapter.
COLLECTING THE PRECIOUS METALS WITH ZINC
Some workers wait until they have accumulated several gallons of

this solution before going to the next treatment. The man who has
very little room will probably want to refine each lot at once, instead
of waiting. For your first refining, handle not more than three
gallons.
When you are ready to clean out the Stock Pot, add a good quan-
tity of zinc shavings and a little hydrochloric acid. Zinc shavings
are much better than granulated zinc or mossy zinc, but these latter
forms can be used if you cannot get the shavings. The kind de-
scribed as "zinc shavings for cyaniding" is the best, and well worth
the slight extra cost.
There should be some bubbling when it goes in; if not, slowly add
more acid. But be careful, as sometimes it boils over, or it might
get hot enough to crack the jar.
You cannot tell beforehand how much acid or how much zinc
will be needed; it all depends upon how much precious metal is
present, and how much acid. A pound of zinc shavings may be
enough for a three gallon lot of solution; or you may need much
more.
Let this stand for at least three or four hours. If you stir it often
it may be ready sooner.
The zinc reduces or throws down the precious metals (also cop-
per) as a fine black powder. If you have used enough acid and
enough zinc, in a few hours your solution will be a pale green, almost
colorless.
THE STOCK POT 145
Sometimes the solution suddenly becomes thick and muddy, with

a scum on top. In that case, add hydrochloric acid and stir, until it
clears up. Try to avoid this thick mud, by having plenty of acid in
your solution.
When the solution becomes a clear pale green, almost colorless,
you are usually safe in concluding that all the precious metal has
been thrown down, and that the solution contains nothing of value.
Test it with stannous chloride (see Chapter IX), which will show
if any precious metal is still in solution. (The only exception is
iridium, which reacts slowly with zinc and is the last to come down.)
After the black powder is once thrown down, do not leave it for
long with the pale green solution standing over it, or it will re-dis-
solve. In that case you must add more zinc and more hydrochloric
acid, just as before, until the black powder comes down again and
the liquid again becomes pale green and is proven to be of no value.
Let the black powder settle, dip off the clear solution and throw
it away. If the powder is slow to settle, pour in a good quantity of
hot water, and it will settle more promptly. Or, if necessary, filter.
Next get the black powder out of the big crock into a casserole or
big beaker, washing it out with the big wash-bottle.
Some people like to refine each batch of black powder at once.
Some merely wash each batch, and wait until they have accumulated
several batches before refining it. But for your first refining of the
black powder use a small quantity, not more than enough to fill a
300 cc. beaker.
If you prefer to put it away for a while, wash it into a vessel that
can be closed air-tight; a Mason jar is good. Use a wide-mouthed
vessel because it is hard to get powder out of a vessel with a narrow
mouth. Wash the powder down, so as not to leave any sticking to
the sides of the jar, and cover it with water, to stay wet until refined.
Stopper the jar firmly.
REFINING THE BLACK POWDER
There are several ways of refining this powder. The following is

a general method, and other methods will be given in Chapter XIV.
If your black powder has been left standing and if some of it has
re-dissolved and turned the liquid yellow, add a little hydrochloric
acid and zinc, until the liquid is colorless again. Next, wash the
black powder well to get rid of acid and zinc salts, using plenty of
water; then dry it. The usual method is to get it into a clean filter
paper, and spread it out to dry in a warm place.
The next step, if possible, is to ignite this black powder to heat

it strongly, but not to melt it. The ideal arrangement is to place it
in a quartz evaporating dish and heat it in a muffle furnace. This
step is not absolutely necessary, and may be omitted; its purpose is
to convert the iridium into a form that is not readily soluble in aqua
regia, so if you are not anxious to obtain your iridium separate from
the platinum, this ignition may be omitted. If the ignition causes
the powder to stick together in clumps, break these up by crushing
with a glass rod or grinding in a mortar. Part of the palladium will
oxidize, and you will see the blue and green colors that you observed
in the acquaintance tests.
The next step is to dissolve platinum, palladium, palladium oxides,

gold and base metals, and to leave the iridium (and the other plati
num group metals, if any) undissolved. To do this we first treat
the powder with plain hydrochloric acid; then without filtering we
add nitric acid and some water to form dilute aqua regia. (Palla
dium oxides dissolve readily in plain hydrochloric acid, but not in
aqua regia; that is why we add the hydrochloric acid first). Since
we do not know how much palladium oxide is present, we begin by
using only enough hydrochloric acid to cover the powder, then heat
gently and watch the effect. Perhaps there will be very little action;
perhaps you will have so much palladium oxide that more hydro
chloric acid will be needed; the intention is to dissolve everything
that will dissolve in hydrochloric acid.
Then, as we said, without filtering, we add dilute aqua regia to

dissolve platinum, palladium, and whatever gold and base metals
are there. By "dilute" aqua regia we mean a mixture of 1 part
nitric acid to 4 parts hydrochloric acid and about 6 parts water.
This acts readily upon the finely divided metal particles; heat gently
for a few minutes; it may not be necessary to bring it to a boil.
Let the dilute aqua regia act all it will, stirring well so that it can
reach all parts of the material. Do not leave the acids on any longer
than is necessary to dissolve everything that will dissolve under this
gentle treatment. Then filter off the liquid and set it aside for fur
ther treatment, leaving the residue in the dish.
THE STOCK POT 147
THE IRIDIUM RESIDUE
This residue should be washed, the washings being added to

the main solution. Possibly there was no iridium, rhodium, or
ruthenium in your original material; in such case this residue will
be merely light-colored flakes of glass or porcelain that have scaled
off the vessels and acid bottles, which naturally may be discarded
at once. Iridium, if present, will be in a fine dark powder, very
heavy. Normally your residue will consist of both the light flakes
and the heavy iridium particles. Wash this into a filter, using plain
water at first. Then as you wash add an occasional drop no more
of hydrochloric acid. When washed in plain water these fine
metallic powders sometimes go through the paper; a little acid, or
an occasional crystal of table salt, in some odd way lessens this ten-
dency. Add the washings to your main solution.
If your original Stock Pot solution contained any rhodium,
ruthenium or osmium, these metals will now be found as heavy
metallic powders, mixed with the iridium powder, presenting the
same appearance.
The disposal of this residue will depend upon individual circum-
stances. Often the best thing is to sell it to a professional refiner who
gives credit for indium. Sometimes you can use it to make up new
alloys. Unfortunately this separation is never a perfect one, no matter
how skilled you may be; that is, some iridium dissolves in the dilute aqua
regia, and usually the iridium residue contains a little platinum, and
possibly other things. Hence the advisability of subjecting the residue
to further purification as in Chapter XV, or of selling it as it is. The
flakes of glass and porcelain can usually be dissolved out with hydro-
fluoric acid, as in Chapter XVI, and with a little experience you can
tell by its looks whether the residue is of good quality or not.
THE DILUTE AQUA REGIA SOLUTION PLATINUM, PALLADIUM, GOLD
This contains all the palladium, a little platinum, any gold that
might have gotten into the Stock Pot, and base metals; also consid-
erable water, which we now get rid of by evaporating to a syrup.
Use a big evaporating dish and heat on the steam bath. At the
same time the excess nitric acid will be driven off. It does no harm
here to add a few drops of sulphuric acid, to throw down any lead
that might be present and at the same time to facilitate the expulsion
of the excess nitric acid.
This is in every way comparable to the evaporations in previous

chapters. Take up the syrup with a little hydrochloric acid, as usual,
repeating if necessary, to get rid of all nitric acid; then dilute the
syrup with several volumes of water, all just as usual. If gold bakes
out, dissolve it with a few drops of hot water containing a little
sodium chlorate. (This combination of chlorate and acid produces
a gas which can be explosive under certain conditions, so work
slowly and keep everything hot; if all your solutions are hot there
is little danger of this gas accumulating in sufficient quantity to
cause damage.)
If any sediment or residue remains, filter it out and put it in
your paper storage. It is probably lead sulphate, but may contain
something valuable.
PLATINUM
To the solution now add ammonium chloride either the dry,

pure salt, or a concentrated solution. This will precipitate platinum
as the familiar orange powder, exactly the same as in Chapter XI
or XII. Let it settle, pour or siphon off the solution, catch the
orange powder in a filter, wash it well as usual, then dry and burn
it, all just as usual. If your original material contained no platinum,
nothing will happen here, so proceed at once to the next step. (But
even if you should know that platinum is absent, add the ammonium
chloride, as it is required for the next step.)
PALLADIUM
The solution from which you have removed the orange powder
should now be in a vessel that will stand heating; a tall glass beaker
is best, or a casserole. Heat it almost to boiling, and slowly add a
few small crystals, one or two at a time, of sodium chlorate. (See
Acquaintance Test with Palladium, Chapter X.)
Turn the flame down so that the liquid just boils. It is very apt
to boil over. Soon you will see bubbles of chlorine gas, so do this
work near the exhaust fan or flue.
For a quart of liquid a teaspoonful of chlorate may be enough;
sometimes much less, sometimes more. The amount of chlorate de-
pends upon the amount of palladium present, and upon other factors
not easily regulated.
(Potassium chlorate is not as good for this work as sodium chlo-
rate, but it can be used if you cannot get the sodium chlorate.)
THE STOCK POT 149
Soon you will see a red powder forming on the surface of the
liquid and on the sides of the dish. If no red powder forms in five
minutes, add some more hydrochloric acid and more chlorate, and
scratch the inside of the vessel with the glass rod.
This red powder is palladium-ammonium-chloride.
Use only enough chlorate to bring down all the red powder pos-
sible. Keep the solution hot almost boiling. With a little practice
you can tell when you have used enough chlorate, because as the
palladium goes into the form of red powder, and settles to the bottom
of the beaker, the liquid becomes paler and paler in color.
If you are interrupted while doing this, you may find on your re-
turn that some of the red powder has re-dissolved. If so, heat the
solution again and add a little more chlorate.
Let this stand on the little flame for perhaps half an hour, adding
an occasional crystal of chlorate, and filter within 24 hours. Wash
it into the filter as usual, washing the powder down into a lump
with a thin stream of ammonium-chloride-water. Drop a crystal of
chlorate into the funnel as it runs through. Wash well, the same as
you washed the orange powder in Chapter X. Then let it dry.
BURNING THE RED PALLADIUM POWDER
Dump the dry red powder out of the paper into a small porcelain
crucible, and ignite it exactly the way that you burned the orange
powder of platinum-ammonium-chloride. It shrinks on being
burned, so add more powder from time to time. Sand crucibles
have been used for this work, but the palladium sticks to them so
tightly that there is always loss in getting it out.
The red powder melts and burns down to a crust of palladium
sponge. It may be grey, or tinted with red, blue and green oxides.
Loosen the crust with a pen-knife and drop it out of the crucible
into a casserole.
It pays to reduce the palladium oxides before melting the sponge,
thus: moisten the sponge with alcohol and heat it gently until the
colors go and the sponge is a clean grey or black. The alcohol will
take fire, so work carefully.
Most people make up new alloys with the sponge, without melting
it into a button. With careful work you may get sponge pure
enough for commercial purposes at the first precipitation, but most
workers like to purify it further. (See Chapter XV.) The red
powder is more easily purified than the sponge, accordingly, if you

intend to repurify, do not ignite the powder, but proceed at once
to Chapter XV.
Another method of reducing the red powder, in which formic
acid is used, will be described in Chapter XIV.
To melt the palladium sponge, use an oxy-gas torch, as described
in Chapter A of the Appendix. The button may swell up and act
queerly, because it absorbs large amounts of gases when hot, then
gives them off when cooling. After the metal is all molten, move
the flame away very slowly, turning the oxygen off slowly so that
the button may cool gradually in a yellow flame. Then turn the
button over and heat the other side until it just runs smooth. You
can get a whiter button by dropping it into water while still very hot.
FURTHER TREATMENT OF THE SOLUTION FROM WHICH PLATINUM

AND PALLADIUM HAVE BEEN REMOVED
Returning to the solution from which we filtered off the red powder:
it will contain traces of both platinum and palladium, and possibly some
gold. (You tried to recover all your gold in Chapters XI and XII, but
now and then some finds its way into the Stock Pot.) Test it with the
stannous chloride testing solution, as your next treatment will depend
upon what it contains.
Normally it contains only traces of platinum and palladium, and
therefore, will be placed in the Stock Pot again. (Note that this com-
pletes a sort of circle, the residues from the Stock Pot going back to
become a new Stock Pot.) But sometimes it contains considerable
palladium so much palladium that you feel that further treatment is
necessary in which case you may follow the procedure below. If
palladium is present only in small amount, however, proceed to the later
section in which gold, if any, is recovered.
Further treatment to recover palladium, to be used when the chlorate
treatment was inadequate and when considerable palladium is present,
follows.
First, add ammonia, slowly, and stir all the time. It gets hot and
fumes. With the first few drops you will get pasty precipitates of dif-
ferent colors, depending upon what base metals are present. As you
add more ammonia, the precipitates may get heavier. Add still more,
and most of them will re-dissolve.
The amount of ammonia needed depends largely upon the amount
of acids and base metals present. You want to add so much that there
is no further change in color, and no more precipitate dissolves. The
chemist would say that this is an "excess of ammonia."
THE STOCK POT 151
If much copper or nickel is present, the liquid will now be a beautiful

sapphire blue. If much iron is present, the precipitate will be a brown
jelly in little flakes. Lead and tin give pale precipitates.
After a few minutes filter off this precipitate. Generally it can be
Adding ammonia to a large quantity of palladium solution.
thrown away, as it consists of base metal compounds. But if very bulky,

put it in the paper storage or the sweeps, as bulky precipitates sometimes
take some palladium down with them. Wash the precipitate in the
funnel with a little ammonia.
Now take the clear ammonia-palladium solution, and very slowly add
just enough hydrochloric acid to it to make it change color again. Make

it distinctly acid with the hydrochloric acid. Stir well with a glass rod.
The mixture becomes hot and bubbles appear, of a gas whose odor
suggests mice, and a bright yellow powder is soon seen. Let it stand
half an hour or so; over-night if desired. This yellow powder is called
palladosamine chloride, as well as by other names. It is much like the
red palladium powder in its general nature.
Filter off the yellow powder, washing it down into a lump with the
ammonium-chloride-water as usual. Wash and dry as usual, and burn
it the same as you burned the red powder. The two may be mixed
together if desired, or reduced by means of the formic acid method that
will be described in the next chapter, or repurified as described in
Chapter XV.
(Note: This "Further Treatment" should be avoided when your
solution contains very large quantities of copper. The sapphire-blue
copper-ammonia compound, observed above, has an unexpected effect
upon paper, and when present in large amount will weaken the filter
paper and render filtration difficult.)
DISPOSAL OF THE FINAL TAILINGS
The solution that remains should now contain only a trace of precious
metal. It is usually referred to as the final tailings, to distinguish it from
the solutions left over from Chapters XI and XII. Frequently it is of
such low value that you should throw it away without further ado.
Take out small samples and test them, first with copperas to see if gold
is present, then with stannous chloride to see if there is enough platinum
or palladium to justify further effort. At this point you find use for a
thorough familiarity with the color changes described in Chapter IX,
especially in comparison with standard solutions containing a known
weight of metal.
Normally gold is absent. But if present in paying quantities, recover
it as usual with copperas.
You will recall that in Chapter IX, you were told that ammonia added
to these solutions, can form amino compounds from which it is some-
times not always difficult to recover platinum and iridium, even by
treatment with zinc. When you recall this fact you see why we hesitate
to add this particular solution with its high concentration of ammonia
salts to the Stock Pot. One alternative is to evaporate it to a crust and
put the crust in the sweeps. Another alternative is to keep it in a small
jar, separate from your big Stock Pot, but to treat it with zinc in the
same way, using extra time and care to insure the recovery of all
precious metal.
THE STOCK POT 153
You will recall, also, that you have been told not to pour acid solutions
into the settling barrels or tanks through which your factory wash-
waters pass, because they might dissolve solid particles of precious metal.
You can get around this dilemma sometimes, however, by neutralizing
the acid in your tailings. (This may or may not be feasible.) Take
out a pint or so of the tailing solution, and add small lumps of air-slaked
lime, for example, until the acid is killed. This point can be recognized
by the litmus paper test described in Chapter IX; also by color changes in
the solution. Work slowly, as the lime and liquid become hot, and
might break the crock. Usually the mixture becomes pasty and many-
colored, depending upon the variety of base metals present. This pasty
stuff may clog some types of wash-barrel systems so consider this pos-
sibility before using this plan on the whole solution.
Other suggestions will be found in Chapter XVIII on Low Grade
Wastes. Most jewelers, it might be noted at this point, are inclined to
waste time, chemicals and floor space on these low-grade solutions, and
workers have been known to spend fifteen or twenty dollars packing up
and shipping off a solution that contained about two dollars' worth of
metal. The way to avoid such mistakes is to utilize the tests and the
Standard Solutions that were described in Chapter IX.
# # # #
The following scheme will be helpful in following the routine of

the Stock Pot refining, but should not be used until after one has
become familiar with the individual steps.
BLACK POWDER obtained from concentrating the Stock Pot solution with zinc contains PLATI
NUM GROUP m e t a l s , BASE METALS, and possibly some GOLD
Wash ignite if possible to convert iridium to a less soluble form, treat with dilute aqua regia,
applying the hydrochloric acid first to dissolve the palladium oxides Filter
FILTRATE contains PLATINUM, PALLADIUM, GOLD if any, and BASE METALS RESIDUE IS IRIDIUM
with flakes of glass
Expel excess nitric acid, dilute, add hydrochloric acid, discard residue, add and porcelain from
ammontum chloride Filter the vessels Possi
•—• ——• — bly r u t h e n i u m and
PPT IS orange FILTRATE contains PALLADIUM, GOLD if any, BASE other platinum
powder of PLAT METALS, and traces of the other platinum-group metals group metals
INUM AMMONIUM
CHLORIDE Add SODIUM CHLORATE slowly to hot solution. Filter.
F P T , IS red FILTRATE contains traces of PLATINUU-

pawder of GROUP tnetals. GOLD, if any, and BASK
PAILADIUM METALS. Consider "Further Treat-
AMMONIUK m e n t " with ammonia, etc.
CHLORIDE
When you read this over for the first time you will think that the
refining of the Stock Pot is a long and complicated process. But re-
member that the amount of metal handled is only a small percentage
of the original material, and that you may do only one Stock Pot
refining to several regular re finings. We might add also that the
Stock Pot refining is the most interesting part of the work.
Q. I have run through four Stock Pot refinings, and have recovered
very little metal. Doubtless that is because my material contains no
palladium, and very little iridium, and I am always careful to recover
all my gold the first time. The Stock Pot seems to be a lot of work
for a very small return. I did get some platinum out, so I hesitate to
throw away the next Stock Pot. Have you any suggestions? My
material contains considerable copper and nickel.
A. In Chapter XIV you will find several other methods of refining the
Stock Pot solutions, some of which are especially suitable when the copper
concentration is high, others of which are advised when palladium is
absent, and so on. We suggest that you become familiar with more than
one method, then use the one best suited to your needs and equipment.
Q. When we refined rolled and filled gold we took special precautions

to get rid of tin and lead. Do these metals cause any trouble in the
Stock Pot?
A. Occasionally. Small amounts of lead sometimes get into the Stock
Pot, in spite of the fact that you tried to get rid of it when, in mixing
up your original aqua regia, you added a little sulphuric acid. This
sulphuric acid precipitated most of the lead as lead sulphate, which was
discarded. Tin is rather more apt to get into the Stock Pot. For one
thing, it gets in from the stannous chloride solution that you use in
testing stannous chloride being another name for tin chloride.
You recall that the first treatment of the Stock Pot is to add zinc and
hydrochloric acid. This throws down the heavy metals and may carry
down some tin. But if plenty of hydrochloric acid is present at the end
of this process, the amount of tin that comes down will be very small.
That is, if you have more than enough hydrochloric acid to dissolve all
the unused zinc, you probably will at the same time dissolve any tin that
may have been deposited.
But if your black powder does contain tin or lead, they may show up
in one or several places. They may be carried down with the colored
THE STOCK POT 155
powder of palladium or platinum; tin oxide may contaminate the iridium.

If so, the repurification methods described in Chapter XV will remove
them.
* * • #
Q. It is slow work to filter these big jars of solution. How can I

hasten the process?
A. In many cases you can save labor by letting the solution settle over-
night, then syphoning off the clear part and filtering only the last frac-
tion. A solution goes through a filter more rapidly when hot, and some-
times this fact can be utilized. Sometimes you can use three or four
funnels on the same job. This question comes up again in Chapter XX,
in which the Buchner funnel is described, that being a device that is
commonly used when handling large quantities of solutions.
CHAPTER XIV
ALTERNATIVE METHODS FOR CHAPTERS XI, XII

AND XIII
Part I. An alternative procedure for the aqua regia solution in Chapters

XI or XII. Part II. Formic acid method of reducing the colored powders.
Part III. An alternative method of precipitating palladium, using dimethyl
glyoxime. Part IV. Alternative methods of refining the Stock Pot.
Alternative acid method for the black powder (Stock Pot concentrates);
recovering palladium from nitric acid solution; the second black powder,
platinum and iridium. Diagram of second scheme for Stock Pot con-
centrates. Questions and answers.
• * • •
In this chapter we shall describe some alternative methods of han-
dling gold and platinum mixtures, gold and platinum alloys, and the
Stock Pot solution. It is assumed that the reader is already thor-
oughly familiar with Chapters XI, XII and XIII.
PART I. AN ALTERNATIVE PROCEDURE FOR THE AQUA REGIA SOLUTION

IN CHAPTER XI OR CHAPTER XII
In Chapters XI and XII we precipitated first the platinum, then

the gold, from the aqua regia solution. You will recall that platinum
comes down more completely in a concentrated solution, while gold
comes down more cleanly in a dilute solution. This is one of the
considerations that impel us to precipitate the platinum first, the
washings that follow the first precipitation serving to dilute the solu-
tion and prepare it for the subsequent treatment with copperas.
There are other considerations as well, some of which are not easily
explained to the non-chemist, and the beginner would do well to
follow the regular procedure unless he has excellent reason for doing
otherwise.
It is true that some workers, including Whiteley and Dietz,1 r e g u
i REFINING AND MELTING SOME PLATINUM METALS, by J. O. Whiteley
and C. Dietz, Technical Publication No. 84 of the American Institute of Mining and
Metallurgical Engineers. Published with Mining and Metallurgy, March, 1928. This
paper describes the methods employed in the commercial refining of jewelry and
dental scrap.
156
ALTERNATIVE METHODS 157
larly precipitate the gold first, using sulphur dioxide gas. They
free the aqua regia solution from excess nitric acid as usual, filter
it clear of all sediment or cloudiness, then saturate with sulphur
dioxide gas until all the gold is precipitated. The gold powder is
filtered off, washed thoroughly as usual, first with water, then with
hot hydrochloric acid, then with water again, to remove dissolved
platinum and base metals. The filtrate from the gold precipitation
contains base metals and the platinum-group metals. It is now thor-
oughly saturated with chlorine gas for some time, in order to coun-
teract the reducing effect of the sulphur dioxide upon the platinum.
(This effect is one of the considerations mentioned above, which is
not easily explained to the non-chemist and about which we shall
say something in the "Questions and Answers" at the end of this
chapter.) After that, the solution is placed in an evaporating dish
and concentrated somewhat, until there is no longer an odor of
chlorine.
If ammonium chloride should be added at this point, it will pre-

cipitate platinum as the orange powder, and most of the palladium
and iridium will come down with it. (The color of the powder
will run from pale yellow to deep maroon, depending upon the
proportions of the various metals and the temperature of the react-
ing substances.)
In some cases this is a satisfactory method that is, some workers

are willing to obtain all their platinum-group metals together in
this way, and then repurify the powder as required. Other workers,
including Whitely and Dietz, vary the procedure thus: just before
adding ammonium chloride, they stir some alcohol into the hot
solution from which the chlorine gas has just been expelled. This
reduces the palladium and iridium compounds to the "ous" condi
tion, in which they stay in solution while the platinum salt comes
down. The palladium and iridium are recovered later from the
Stock Pot, as usual.
As we said above, to the non-chemist this is confusing. We have

mentioned certain "considerations" and "effects" that are difficult to
explain, and this is one of them. In the Questions and Answers at
the end of this chapter we shall discuss these more fully under the
heading of "Oxidizing and reducing effects changes of valence
'ous' and"ic"conditions."
PART II. ALTERNATIVE METHODS OF REDUCING THE COLORED POWDERS OF

PLATINUM-AMMONIUM-CHLORIDE, PALLADIUM-AMMONIUM-CHLORIDE, AND
PALLADOSAMINE CHLORIDE TO METALLIC FORM
In previous chapters we reduced the orange, red, and yellow pow-

ders to platinum or palladium metal by heating them strongly until
they decomposed and the metallic sponge remained. In this chapter
we shall refer to these salts simply as the "colored powders," since
the method is the same with all three, also with the related dark
iridium salt. There are certain disadvantages to that strong heat
ing the danger of loss of precious metal in the fumes, and the fact
that the method does not remove any impurities. In spite of this,
the method is widely used. Large plants cut down the losses in two
ways; they pass the fumes through special baffles to capture any vola
tilized metal, and sometimes they play a stream of burning hydrogen
or illuminating gas on the colored powder as it is being decomposed.
This lessens the tendency to volatilize, but hydrogen and illuminat
ing gas are highly flammable, and can be explosive if not properly
handled.
The other method of reducing the colored powders is to boil them
with formic acid and ammonia as described below. This formic
acid method has several advantages: there is little danger of loss of
metal; many impurities are automatically removed during the re
duction; no muffle or furnace is needed; the method is quick; and
the fumes produced are not highly objectionable. The expense is
low. The precious metal is obtained as a finely divided powder,
called "platinum black" or "palladium black," not so easily handled
as the grey sponge but equally suitable for making up new alloys,
and easily susceptible to further purification if such is desired. The
method carries no personal hazards, and is destined to become
popular.
EQUIPMENT NEEDED: Formic acid, which is a colorless liquid with
a strong odor; ammonia; litmus paper; one or more Pyrex beakers,
preferably with a lip; the 2-liter size is good. A casserole can be
used, but the glass vessel is better since you can see through it and
watch the color changes. For large quantities of material, use sev
eral 2-liter beakers.
For each ounce of platinum expected, you will need approximately
the same weight of formic acid. It is easy to measure out a definite
weight of the acid by placing a tared beaker on the scales and pour
ing in acid until the proper weight is reached. Formic acid loses
strength on standing, so you may need more; do not be afraid of
using too much, as a moderate excess does no harm. Palladium
requires a little more formic acid than platinum.
When using this method the first time, use only a few pennyweight
of the colored powder.
Mix the formic acid solution thus: place the proper quantity of
formic acid in a big beaker, dilute it with two or three volumes of
water, and drop in a piece of litmus paper. Now very slowly add
ammonia, stirring constantly. The solution gets hot and will spatter
unless you work slowly. The litmus paper will at first be red; use
enough ammonia so that, on stirring, the litmus just turns blue.
Now, add more formic acid, so that the litmus just turns red again.
(In other words, you want a mixture of ammonia and formic acid
that is slightly acid, and just turns litmus paper red. Fish out the
litmus paper with a glass rod and throw it away.)
Next take one of your 2-liter beakers and fill it about one-quarter
full with this formic-ammonia solution. Place your colored powder
in a casserole or beaker, cover it well with water and bring it to a
boil. At the same time the formic-ammonia mixture, in its beaker,
is also being brought to a boil.
Now slowly pour the precious metal mixture into the formic-
ammonia solution, and if everything is quite hot the reaction will
take place at once, the mixture foaming up and turning black.
(That is why you were told to use a small quantity and to have
your beaker not over one-fourth full so that your foaming mixture
would not overflow the beaker.)
After the reaction once starts, the heat of the reaction will keep
the liquid hot, and there will be considerable bubbling and fuming.
Sometimes rarely you need to bring the mixture to a boil a second
time before the reaction will begin.
If your mixture is still yellow or brown after being boiled a few
minutes, try adding a little more ammonia. If that does not help,
add some more formic acid, slowly, and again bring to a boil.
When the reaction is complete your precious metal will all be in
the form of a black powder, which soon settles, leaving the liquid
colorless (or possibly pale green from nickel or copper). Let the
beaker stand in a warm place for about an hour. If your metal is
pure, the heavy black powder settles quickly; if impure, it settles
slowly and the mixture looks like ink.
After the black powder has settled, it is easy to wash it by pouring

off the clear liquid (filter if necessary), adding hot water and pour-
ing off as often as seems necessary.
Work with clean vessels, free from fingerprints inside. Otherwise,
precious metal will deposit as a mirror on the inside of the glass.
For very large quantities, do this: place the colored powder in a big
crock, cover it with water and run in steam. While it gets hot, mix the
formic acid, water, and ammonia, and pour the mixture into a number
of 2-liter beakers, about two inches to a beaker. Place these beakers in
a row on a steam bath or hot plate, and start the fan to draw off fumes.
When everything is hot, dip up the mixture of colored powder and water,
and pour it into the beakers; with a little practice in judging quantities
you will be able to use up all the formic acid and still not add too much
colored powder.
Let the beakers stand in a warm place till the black powder has
"digested," then wash as usual, and dry.
If your original powder is contaminated with small amounts of
iron, zinc, or other light metals, this method will separate your
precious metal from them. That is, the light metals will remain
dissolved in the liquid after the black powder has been thrown
down, but if your colored powders are contaminated with heavy
metals, such as gold, mercury or iridium, this method will not re-
move them. They will come down with the black powder, indis-
tinguishable from it.
If nickel, copper, or silver are present in very large amounts, they
may come down to some extent, especially if the acid concentration
is too low. In such case a single repurification is normally sufficient.
When properly performed, this method converts all precious metals
into the metallic powder, and the liquid that remains, and the wash
waters, can be discarded as worthless.
• * • *
A final method for reducing the colored powders to metallic form
is the familiar one of cementing them out with zinc in the presence
of a dilute acid. Weak hydrochloric acid is commonly used.
PART III. AN ALTERNATIVE METHOD OF PRECIPITATING PALLADIUM

WITH DMG
It will be recalled from Chapter IX, on the identifying of precious

metals, that palladium reacts with dimethyl glyoxime to produce a
bright yellow precipitate. The reaction constitutes a delicate test, a
very small amount of palladium in a large amount of solution pro

ducing a very conspicuous precipitate.
Dimethyl glyoxime called DMG for convenience can also be
used to recover palladium from solutions in which the other plati
num group metals are present. In the assay of ores and bullion
containing these metals, it is standard practice to precipitate pal
ladium with DMG.
However, the method is rarely used in refining these metals; only
in their assay. The reasons are: first, the DMG is relatively ex
pensive; second, the precipitate produced is so voluminous as to be
difficult to handle. Even a few pennyweight of palladium will pro
duce so copious a precipitate as to demand large vessels much larger
than needed for the formation of the red palladium-ammonium
chloride. The same peculiarity that makes DMG useful in identi
fying small quantities makes it unhandy in recovering large ones.
There may be occasions however when the method may be useful;
for example, to recover palladium from an aqua regia solution con
taining platinum-group metals, the usual base metals, but no gold,
when you are anxious to recover the palladium at once, and are
willing to postpone the recovery of the other platinum-group metals
to a later day.
For each pennyweight of palladium expected, at least 21/2 penny
weight of DMG are required, and a slight excess is advisable. Dis
solve the DMG in hot water a pennyweight of DMG to about
5 fluid ounces hot water (roughly the same as 1 gram of DMG to
100 cc. water). The powder dissolves slowly; if necessary filter out
any cloudiness, and let cool. The aqua regia solution, which may
contain platinum-group metals, and base metals, but no gold, should
contain about 3% to 5% acid; if it is stronger than that, dilute it
with water.
Add the DMG solution slowly; do not heat because heating will
bring down platinum as well. The voluminous yellow precipitate
ofpalladiumdimethylglyoximedimethylglyoximeappearsatonce.Ifyour solutions
are so concentrated that the mixture becomes thick, dilute it with
considerable cool water, and let the yellow precipitate settle, then
filter. Wash well with much hot water. The solution and washings
are treated with zinc and hydrochloric acid, to cement out the other
platinum-group metals.
The yellow precipitate can be reduced to palladium metal by strong
heating, or by means of formic acid, as in the case of the other col
ored powders.
PART IV. ALTERNATIVE METHODS OF REFINING THE STOCK POT
In Chapter XIII the precious metals in the Stock Pot were recov-
ered by a standard method, one that the student should understand
thoroughly before attempting other methods. The diagrammatic
scheme at the end of Chapter XIII will serve as a reminder of the
various steps.
• # • •
Sometimes the Stock Pot is of such slight value that you find it
does not pay to refine it. This is especially the case where your
original material contained little or no palladium or iridium, and
where you succeeded in removing all your gold at the first precipi-
tation with copperas. In such a case the Stock Pot will contain
merely a small amount of dissolved platinum, plus much base metal.
If your time is valuable, and if your plant is crowded, it may pay
you, instead of refining this solution, to follow one or another of
these suggestions: (a) evaporate the Stock Pot solution to a crust,
and put the crust in with the low grade wastes that will be sold to
a professional refiner who gives credit for platinum-group metals;
(b) neutralize the acid in the solution by adding lime, then put the
thick mixture in with the wash barrel residues; (c) seal it up in a
tight barrel and sell it as is.
But generally the Stock Pot solution is well worth the time and
effort to refine it. Assuming such to be the case, the following are
several of the possible methods.
(I) Instead of precipitating the precious metals with zinc or other
precipitants, evaporate the whole Stock Pot solution down to a crust
on a sand bath or steam bath. This takes some time, but can go on
without attention from the worker, and if you are equipped with a
good steam bath (and you should be) it may pay to use this method.
Melt the crust down with a reducing flux such as soda ash with
charcoal, using a sand or graphite crucible and a gold-melting fur-
nace. This gives a metallic button that must be refined by methods
to be determined by an examination of the button probably by the
methods given in Chapter XII.
Normally this button contains much iron (from the copperas) plus
considerable other base metals. The main disadvantage of this
method is the presence of this large amount of iron.
(2) Cement out the heavy metals by means of zinc (or other me-
tallic precipitant) as described in Chapter XIII, then melt the black
powder into a button, using a reducing flux such as soda ash and
charcoal. This button may be refined as it is, or melted in with
other metallic residues. The main objection to this method is the
large quantity of copper that is usually present. Also, if the black
powder contains much platinum and iridium, with their high melt-
ing points, the melting point of the button may be so high as to be
troublesome.
(3) Cement out the heavy metals by means of zinc (or other me-
tallic precipitants), then melt the black powder into a button, using
an oxidizing flux, such as sodium nitrate. When properly done, as
described below, this method converts the copper into copper matte,
while the precious metals collect as a small and relatively valuable
button. Most of the iron and other base metals go into the copper
matte. Here are the details of this method, as described by the
refiner of one of our finest Eastern bullion plants:
"I take down my Stock Pot solutions with zinc, then wash the
precipitate well by decantation, filling the crock up at least four
times with hot water. This removes most of the acid. Then I cover
the metal with a little water and add considerable sal soda, stir well,
and leave overnight. I use about three pounds of sal soda to a five
gallon jar of our solutions, which are fairly rich.
"In the morning, add hot water again and decant, and by this
time the washings should be alkaline. If not, add more sal soda and
let stand again. Finally wash and catch the residue and dry it. I
use a suction funnel and wash it well with hot water in the funnel,
then dry it slowly on a hot plate.
"We melt about a 600 cc. beakerful of this dry powder at a time.
To the 600 cc. of powder, add 600 cc. of nitrate of soda and a table-
spoonful of ground glass. This goes nicely into a size M Dixon
sand crucible.
"If we know that there is much palladium present, or very much
platinum, we add a half ounce of cement silver to collect them. We
have an iron mould, cup shaped, and when melted we pour into
this iron mould; in that way we can use the sand pot several times.
When the melt cools, drop it out and you will see that the precious
metals form a button in the bottom. This comes off readily with a
knock of a hammer. The copper matte contains most of the other
base metals. If properly handled, this matte does not contain appre
ciable quantities of precious metal." The button will then be re-

fined according to the method of Chapter XII.
(4) The rousting method. Cement out the precious metals as
usual, wash well, and dry. Then roast the powder, preferably in a
muffle furnace where air can reach it, at red heat, and rake the
powder about so that air can reach all parts of it. This converts
the copper into copper oxide, a form that will dissolve easily in
dilute sulphuric acid. At the same time the platinum, palladium,
iridium (and tin) are converted into forms that resist dilute sulphuric
acid. Let cool in the presence of air, and break up the lumps.
Treat the roasted material with warm dilute sulphuric acid, made
by pouring one quart of acid into five or six quarts of water. (Add
the acid to the water, slowly, stirring all the time.) The copper-
oxide (or most of it) will dissolve readily, as will most of the other
base metals that may be present. The residue will consist of precious
metals, tin oxide, and traces of other base metals. Wash it well,
then treat it according to one or another of the methods already
familiar, preferably according to the scheme given in Chapter XIII,
where you dissolve the black powder in aqua regia, adding the hydro-
chloric acid first to dissolve palladium oxide, then adding nitric acid
and water. From that point, follow Chapter XIII.
The following methods differ only in detail from the scheme out-
lined in Chapter XIII.
(5) Instead of concentrating the heavy metals by means of zinc, use
some other metal precipitant. Thus it is possible to use aluminum, iron
or copper in the first step. Aluminum costs a little more, but works
much more rapidly. This rapidity may become a disadvantage unless
the aluminum is added very carefully, as it reacts quickly, the liquid be-
comes hot and may boil out of the vessel, or may cause sudden local
heating sufficient to crack an earthenware jar and permit the solution to
be lost. If you elect to use aluminum, use foil or wire or sheet, and add
them slowly; aluminum powder reacts too quickly for safety.
The black powder obtained should be treated exactly the same as if
it had been precipitated by zinc.
Iron is the cheapest of the metals that will serve this purpose. It may
be used in the form of rods or plates. It is noticeably slower than zinc,
but so cheap as to be very popular. Copper plates and copper wire are
also slow, but the black powder obtained is clean and easy to handle.
The heavy metals deposit upon the wire, sometimes forming a coating
on it. Some workers then take the coated wire and melt it all together
into a button, then refine the button. Others refine the coated wire
without melting it down, using one or another of the plans with which
you are now familiar.
(6) In Chapter XIII when adding zinc, we added a little hydrochloric
acid. Sulphuric acid can be used instead of hydrochloric. It costs less,
and the action is practically the same. Add it slowly, a little at a time,
stirring well to prevent local overheating.
* # # *
Assuming that we have concentrated our heavy metals as a black
powder by any one of the methods just mentioned, we are now ready
to consider another acid method by which this black powder can
be refined. It differs in only a few steps from the scheme used in
Chapter XIII.
ALTERNATIVE ACID METHOD FOR THE BLACK POWDER (STOCK POT
CONCENTRATE)
First wash the black powder with unusual care to get rid of every
trace of hydrochloric acid. Wash until the washings have no effect
upon blue litmus paper. Do not let the powder dry. The washings
should be of no value and should be discarded. (If possible, after you
have once started this washing, try to proceed without interruption until
you have completed the next process and the nitric acid solution described
below is filtered off and the filter washed down well.)
The washed black powder should now be in a casserole or beaker.
Cover it with nitric acid diluted with about two volumes of water. It
begins to act in a few seconds, dissolving the palladium and base metals,
and leaving undissolved the platinum, iridium, and any gold that might
be present. There will be the usual brown fumes. Warm gently, and
stir well. Do not leave the acid on any longer than is necessary less
than half an hour is generally enough even for a large lot.
As soon as all is dissolved that will dissolve under this gentle treat
ment, filter the liquid through a small filter, washing the remaining
powder well so as to free it of solution. Sometimes this remaining
powder is so heavy that it will stay in the vessel while the clear solution
is being poured off; but usually it is so fine that the use of filter paper
is essential. While washing the powder into the filter paper, drop in
one or two crystals of sodium nitrate. (If washed with plain water,
this very fine powder will run through the paper; we add the sodium
nitrate and in some way it coagulates the fine powder and keeps it from
running through.)
Your palladium is now in the nitric acid solution. This can be left
standing as long as you wish. Methods of recovery are given below.
If your original material contained no palladium, this nitric acid solu-

tion should be of no value, and should be thrown away. (Do not put it
in the Stock Pot.) If it contains much nickel or copper, it will be green.
Palladium in solution is brown.
Your platinum and iridium are now in the form of this second black
powder. For its further treatment, see later paragraphs in this chapter.
If any gold, rhodium, ruthenium, or osmium were present in the Stock
Pot, they also will be concentrated in this second black powder.
(Note: The six metals of the platinum-group are like each other in
many respects. As a result, these separations are never perfectly com-
plete. That is, a little platinum may dissolve with the palladium even
under the gentle treatment just described. Also, some of your palladium
may remain with the residue of platinum and indium. This is un-
avoidable.)
RECOVERING PALLADIUM FROM THE NITRIC ACID SOLUTION
This solution contains the bulk of the palladium, perhaps all of it,
with base metals. Place it in an evaporating dish and evaporate to a
syrup, preferably on a steam bath. Add a little hydrochloric acid slowly,
and again evaporate to a syrup. This is similar to previous evaporations,
and the purpose is the same to get rid of the unused nitric acid, and
replace it with hydrochloric acid.
Finally thin with water, and if any sediment appears, filter it out
and put it in the paper storage.
The solution should now be in a beaker or casserole, not more than
two-thirds full.
From here on, the recovery of the palladium is exactly the same process
as that used in Chapter XIII. You add ammonium chloride, as much as
will dissolve. Sometimes a slight orange precipitate of platinum-am
monium chloride appears here; if so, filter it out and add to your other
platinum powders. Next, heat the solution to boiling and slowly add
sodium chlorate, a crystal at a time, stirring well and scratching the sides
of the beaker with the glass rod. The familiar red powder of palladium
ammonium-chloride will form and settle as before.
This red powder will be collected, washed, and reduced to metallic
form exactly as before either by igniting as in Chapter XIII, or by
means of formic acid and ammonia, as described in this chapter.
The solution from which you filtered off the red powder should be
tested with stannous chloride Testing Solution A. Usually it contains
only a trace of precious metal, mixed with a considerable quantity of
base metals, and may often be discarded immediately. Sometimes it is
of sufficient value to be put into the Stock Pot. Sometimes, however,
you find that your treatment with the sodium chlorate was so unsuccess
ful that considerable palladium is still present; if so you may wish to

go through with the so-called "further treatment" in Chapter XIII, in
which you obtained the yellow powder.
THE SECOND BLACK POWDER PLATINUM AND IRIDIUM
Let us return now to the second black powder, insoluble in nitric acid,
which we left in a filter on page 166 of this chapter. The metals in this
form are called "platinum black" and "iridium black." Under ideal
working conditions this residue will consist solely of the platinum and
iridium that were not previously recovered in Chapters XI and XII,
plus little flakes of glass and porcelain from the vessels and reagent
bottles. (These flakes can be removed by hydrofluoric acid see Chap
ter XVI.)
Jewelers usually do not need to separate the platinum from the iridium
indeed, they prefer to keep these two metals together so their next
step is to melt this black powder, using the oxygen-gas flame as de
scribed in the Appendix. Usually one can tell by its appearance whether
this powder is clean enough to melt or not, or a small test melt can
be made first. If the original material contained much iridium, the
button will be rather hard, and some soft platinum must be added to
bring the alloy to the desired composition.
However, ideal working conditions do not occur every day, and this
black powder is often too impure to be melted up at this stage. For
instance, gold sometimes gets into the Stock Pot, and will show up in
this powder, giving it a brown color. Sometimes for other reasons the
powder is unattractive. Sometimes your original material contained
rhodium, ruthenium, or osmium, which would be accumulated here.
Accordingly this powder generally receives some further treatment
before being melted. Thus, suppose you suspect gold and base metals
to be present; you would proceed as follows: Dry the powder and heat
it to redness (do not melt) to convert the iridium to a less soluble form.
Then boil in concentrated aqua regia, adding the hydrochloric acid first.
This dissolves palladium, gold, platinum, and most base metals. Iridium
dissolves very slowly, if at all. Rhodium would remain with the iridium;
also undissolved are the flakes of glass and porcelain, and perhaps a little
silver chloride that had escaped previous treatments.
Dilute this aqua regia solution with two volumes of water and filter.
If the undissolved residue is whitish, it is of no value and can be thrown
away. If the residue is dark it probably is of value, and should be
treated according to the methods described in Chapter XV under the
repurification of iridium, or if known to contain ruthenium, osmium
or rhodium sold to a refiner who purchases such metals.
The aqua regia solution just obtained should be treated as usual for
the recovery of platinum and gold.
SECOND SCHEME FOR REFINING STOCK POT CONCENTRATES
BLACK POWDER obtained i n n Stock Pot solution Contains PLATINUM, PALLADIUM, IRIDIUM,
BASE METALS, possibly GOLD possibly other PLATINUM GROUP metals
Wash with great care to remove all hydrochloric acid Do not ignite Treat with warm dilute
nitric acid Filter
FILTRATEcontainsPALLADIUM,withbasemetals, RFSIDUE consists of PLATINUM, IRIDIUM, possi

possibly traces of other platinum group metals bly GOLD flakes of glass and porcelain possibly
other platinum group metals
Convert to chloride, add ammonium chloride
Filter if necessary Ignite Treat with aqua regia, applying the
• hydrochloric acid first Filter
PPT IS FILTRATE contains PALLADIUM and :
orange base metals FILTRATE contains PLATINUM, RESIDUE
PLATINUM GOLD, BASE METALS, possibly IS IRIDIUM,
salt Add sodium chlorate slowly to hot traces of other platinum-group glass,
solution Filter metals possibly
- —-— • ~—-— other
PPT IS FILTRATE may contain Expel nitric acid, add ammonium platinum-
red enough palladium to chloride Filter group
PALLADIUM justify further treat • metals
salt ment, to obtain the PPTIS F I L T R A T L contains
yellow salt Or, place orange GOLD, base metals,
in Stock Pot or sweeps PLATINUM traces of others Add
salt copperas
Ppris FILTRATE to
GOLD Stock Pot
* * # #
In Chapter XXI still more alternative methods methods suitable

primarily for the very large plant will be described. And in Chap
ter E of the Appendix the searcher for still further alternative meth
ods will be shown the way to find them.
# * * •
To the beginner, this wealth of alternative methods will be con
fusing, possibly baffling. The only safeguard is to confine oneself
to a single, well-recommended method until it is truly familiar, re
membering that the other man's method, like the grass on the other
side of the fence that always looks so green, may really be no better
than one's own.
Q. Please tell us more about the use of chlorine gas.

A. Chlorine gas is dispensed in steel cylinders, similar to those used
with sulphur dioxide or ammonia. Chlorine under pressure is a liquid;
when released it forms a gas. The gas is most unpleasant and corrosive,
attacking metals and fabrics as well as human lungs and eyes. It is
widely used in industry as a bleaching agent, and in sanitation to purify
water supplies and sewage. In the refining of precious metals it has
several uses, notably in the Miller process for refining gold see Chap
ters XXI and XXII in one method of extracting gold from low grade
ore, and for oxidizing platinum and palladium salts in solution to their
higher valences.
* * • *
Q. Please explain the following phrases, which I have found in articles
on the platinum metals: "Oxidizing with chlorine" "reducing with
alcohol" "changing the valence" "higher and lower valences" "the
ous condition" "the ic condition."
A. A full explanation of these phrases, all of which refer to the same
group of chemical relationships, is beyond the range of this book. It is
well known that metals exist in various forms; thus we know iron as a
metal, also as an element in a pale green compound called copperas,
whose chemical name is ferrous sulphate. This ferrous sulphate, after
reacting with dissolved gold (or being exposed a long time to air)
turns yellow, and is then known as ferric sulphate. Thus there are two
kinds of iron salts, the ferrous and the ferric.
There are also two kinds of platinum salts the platinous and the
platinic. In fact, most metals occur in both the ous and the ic condi
tions. Chemists have another way of saying the same thing; they say
that most metals have at least two valences, the salts of the higher
valence being called the ic form, and the lower valence being called the
ous form. The change from a lower to a higher valence is described as
an oxidation, though frequently no oxygen is present. The change
from the higher to the lower form is called a reduction. These changes
can be brought about in many ways. Some oxidations are accomplished
by mere exposure to oxygen, some reductions by heating the substance
in the presence of charcoal. The electric current used in a gilding bath
is said to reduce the dissolved gold when it deposits it upon the article
that is being plated, while at the same time it oxidizes the anode and
carries it into solution. An electric current can also change a salt in solu
tion from the ous to the ic condition, or vice versa.
This discussion is in no sense a satisfactory explanation. However,
this book attempts to describe the procedures of refining so that the
reader may perform them, even though unaware of their underlying
chemical principles. For an adequate discussion of oxidation and re
duction phenomena, valences, the differences between ous and ic salts,
the reader is referred to any text book of general chemistry.
CHAPTER XV
REPURIFICATION OF GOLD, PLATINUM, PALLADIUM,

SILVER, IRIDIUM
Part I. Repurifying gold: by reprecipitation; the use of oxalic acid; by

fire; by chlorine gas; by electrolysis. Part II. Repurifying platinum.
Part III. Repurifying palladium: the colored powders; buttons, bars, and
palladium black; palladium sponge. Part IV. Silver. Part V. Iridium.
# # # #
The precious metals that are recovered in the earlier chapters of
this book are not always as pure as your needs require. Small
amounts of tin, lead, iron or other metals may contaminate your
"fine" gold. Your platinum may contain some silver chloride or
palladium; your palladium may contain platinum; and so on. The
amount and nature of these impurities will depend partly upon your
skill and patience, but primarily upon the character of the original
material, and in most cases it is no reflection upon you. or your
method that repurification should be necessary.1 I n d e e d , in m a n y
plants at least one repurification is routine, except where the original
material was already of high quality.
You are advised to make a test melt of every batch of recovered
metal. Melt up a few grains of the gold powder, or the cleaned
platinum filings of Chapter XI, and if the sample button is unsat-
isfactory, do not melt up the bulk of the metal it is easier to repurify
powders or filings than to repurify buttons or bars.
With a little experience you can quickly judge the quality of your
metal, especially by its appearance when molten and its behavior when
hammered or drawn. As we suggested at the end of Chapter VI, it is
wise to obtain samples of gold, platinum, palladium, and silver of un
impeachable purity, then melt, roll and hammer them, noting their
characteristics carefully. Pure platinum, under the oxy-gas flame, melts
and rounds up into a thick almost globular button, with a mirror-like
i The serious student is referred to the 29-page paper PURIFICATION OF THE
SIX PLATINUM METALS, by Edward Wichers, Raleigh Gilchrist and William H.
Swanger, published in Mining and Metallurgy, March, 1928.
170
REPURIFICATION 171
surface, while impure platinum flattens out into a thin button with a
skin or film on the surface. Pure platinum or gold can be drawn into
extremely fine wire, while impure metals cannot. And so on the
"feel" of the metal under the saw, file, or hammer, conveys much in
formation to the experienced user.
The commonest process of repurification, and the one to be de
scribed first, is simply to dissolve the metal in acid and to precipi
tate it again. With the colored powders of palladium, a slightly
different plan is possible.
Inallcasesgold,platinum,orpalladiumdimethylglyoximeifyourmetalisalready
in the form of a button or bar, roll it out as thin as possible, and cut
Erlenmeyer flasks, two styles.
the strip into short lengths, twisting them a little. If too brittle to
be rolled, the metal may be granulated, or broken up with a hammer.
When redissolving the metal, use a casserole or flask. A Pyrex
glass flask, Erlenmeyer style, is the most popular vessel. Eimer &
Amend No. 23382 with the narrow mouth, and No. 23386 with the
wide mouth, are good. The liter size, thin glass, either model,
costs $0.38. (See also page 136.)
PART I. REPURIFYING GOLD
By Reprecipitation
If your gold is in the form of a powder, and if you believe it to be
contaminated with tin, begin by heating it gently almost to redness
in an open shallow crucible. Do not let it melt. This converts any
tin into tin oxide, a substance that does not easily dissolve in acid.
If your gold is in a button or bar, roll it out thin, or break it up, as
mentioned above, and cover it with aqua regia.
For each Troy ounce of gold use about one-half fluid ounce of
nitric acid, two ounces hydrochloric acid, a little water, and several
drops of sulphuric acid. Add slowly, as it may bubble over, espe
daily if your gold is in the form of a powder. Warm gently if

necessary.
When all the gold is dissolved, place the solution in an evaporating
dish and expel nitric acid, as usual, taking up the syrup in hydro-
chloric acid, and evaporating again, as usual. When you are sure
all the brown fumes have been expelled, add considerable water say
ten volumes or more and set aside for several days to settle.
Silver chloride, lead sulphate and other salts will deposit out
slowly as a fine mud. So will the tin oxide that was formed when
the powder was heated almost to redness. The liquid must be per
fectly clear and free from sediment before you go to the next step,
so use particular care in getting rid of this deposit. If time permits,
let it settle well, then siphon off the clear liquid, filtering only the
last part. This filtering is not always easy, as the fine powder some
times goes through the paper. One plan is to use filter paper spe
cially made for fine precipitates. Another plan is as follows: tear
up a filter paper into small scraps and drop them into the liquid to be
filtered, stir until the paper has disintegrated into shreds, then filter
as usual into a clean filter paper. The tiny particles of tin oxide,
silver chloride, and lead salts will cling to these shreds and the fil
trate will run clear.
You are now ready to precipitate the gold again. As in Chap-
ter VI, you have a choice of several precipitants, but sulphur dioxide,
oxalic acid, and sodium nitrite are the best. Sulphur dioxide is
especially suitable when tin is present. Use it exactly as described
in Chapter VI, then collect and wash the gold powder as usual
first in hot water, then in hydrochloric acid, then in hot water again
after which it is ready to be dried and melted. Test a few drops
of the remaining solution with stannous chloride Testing Solution A,
and if it contains appreciable quantities of precious metals, put it in
the Stock Pot.
If you have no sulphur dioxide gas, use sodium nitrite as described
in Chapter VI, or oxalic acid, as described below. Some people use
pure copperas for this work, but it is not as good as oxalic acid or
sodium nitrite.
In general, do not use the same precipitant for your repurification
that you used in the first recovery. Thus: sulphur dioxide tends to
carry down traces of the platinum group metals with the gold; oxalic
acid does not. But oxalic acid does tend to carry down tin and cop
per; sulphur dioxide does not. Therefore, if you use sulphur dioxide
REPURIFICATION 173
for your first precipitation, use oxalic acid for the repurification, and
vice versa.
The Use of Oxalic Acid
This method was mentioned briefly in Chapter VI as being espe-
cially suitable for the recovery of gold from solutions that contain
relatively few impurities. Its disadvantage is that it will work only
when all the solutions are hot preferably boiling which means
that the precipitation must be carried out in vessels that will with
stand considerable heat. It costs more than copperas or sulphur
dioxide, about the same as sodium nitrite. Under most conditions
the gold recovered is of high purity.
For each Troy ounce of gold you will need at least a Troy ounce
of oxalic acid crystals. If you failed to expel all the excess nitric
acid, you will need more.
Dissolve the oxalic acid in hot water. Add a little ammonia, say
a teaspoonful or less to an ounce of oxalic acid; add slowly to avoid
spattering.
Do not use too much ammonia the mixture should not smell of
ammonia after being stirred. If it does, add enough additional
oxalic acid to kill the ammonia smell. If you have litmus paper,
use it here, and let the mixture show a slight acid reaction. Have
the mixture in a big beaker or jar that will withstand heat.
Next, heat your aqua regia solution to boiling. Also have the
oxalic solution hot boiling if possible. Pour the gold solution
slowly into the oxalic acid solution. Gold comes down at once with
hot oxalic acid, but takes hours or even days if it is cold.
Some people pour the oxalic into the gold; we think it is better
to pour the gold into the oxalic.
There will be bubbles and fumes, so work slowly or the liquids
may boil over when they come together. If you failed to expel all
the excess nitric acid, brown fumes will appear here; if this happens
you may have to add much more oxalic acid in order to kill all the
nitric acid.
Let the mixture stand for some time, keeping it hot if possible.
The gold comes down as bright clean crystals, often very beautiful,
which usually settle promptly.
Finally pour or filter off the liquid, after testing some of it with
the stannous chloride testing solution to be sure you used enough
oxalic acid. If worthless throw it away at once. In rare cases it
may show enough palladium or platinum to be put in the Stock Pot.

The brown powder of gold should now be washed well with much
hot water, to remove all traces of hydrochloric acid, then boiled with
a little nitric acid, to remove traces of palladium or base metals.
Finally wash with plenty of hot water again, and dry and melt as
usual. This normally gives gold of high quality.
By Fire
Sometimes a batch of brittle gold can be purified by one or more
remeltings with the addition of suitable fluxes. The commonest
(and one of the oldest) method is to add a little nitre (potassium
nitrate) or sodium nitrate to burn out the impurities. The gold is
heated in a clay crucible, and the nitre is added cautiously just before
the gold melts. Considerable bubbling occurs immediately, and the
oxygen in the nitre unites with the base metals. It is customary to
protect the crucible by a ring of bone ash, which absorbs the base
metal oxides, or to use a flux of borax glass. Iron and zinc can be
removed readily in this way. Lead is more difficult to remove.
Platinum-group metals are removed by prolonged nitre treatment
and will go off with the fumes and be lost.
Instead of nitre, ammonium chloride can be added to the molten
gold, and it is especially useful in removing lead. When consider-
able lead is present, alternate additions of nitre and ammonium
chloride are employed.
By Chlorine Gas
This method, called the Miller method, is described in detail in
Chapters XXI and XXII. It will not remove the platinum-group
metals, and has other disadvantages, including the formation of ex-
tremely dangerous fumes. It seems to be used only by large plants,
but there is no inherent reason why it should not be used by the
small plant as well, and the serious student is advised to consider
its merits.
By Electrolysis
Electrolytic methods, suitable to the large plant only, are described
in detail in Chapters XXI and XXII. Gold that is already about
860 fine or finer, and which contains platinum-group metals, is
carried to a high degree of purity by these electrolytic processes,
which are standard in the Mints, in the large copper and nickel
REPURIFICATION 175
refineries, and in several of the refineries that specialize in dental

gold. Electrolytic methods are not adaptable to the needs of the
average small plant.
* * * •
PART II. REPURIFYING PLATINUM
The commonest impurity in precipitated platinum is silver chlo-

ride, which makes it brittle and dark in color. Iridium and palladium,
which do not make it brittle, are also apt to be present, especially
if your original material is high in these metals, because they are so
similar to platinum in chemical characteristics that it is not easy to
separate them. Gold and tin appear often enough to be trouble-
some. Usually one precipitation is sufficient to remove most of
the silver chloride, gold and base metals. Complete freedom from
palladium and iridium is rarely required by the jeweler, which is
fortunate, for as we say, these metals are so much alike that the
task of separating them is difficult. The usual method of purifica-
tion is simply to dissolve the metal in aqua regia, and then to pre-
cipitate it again with ammonium chloride, and when greater purity
is desired, this may be repeated several times.
For each Troy ounce of platinum, mix up about 1 fluid ounce
of nitric acid and 4 of hydrochloric acid, with 1 ounce of water and
a few drops of sulphuric acid. You may need more aqua regia
than this, especially if the metal is in the form of shot or thick foil.
Sponge dissolves promptly. Stir often, by rotating the flask, and
heat gently, preferably on the steam bath, raising the temperature
slowly to avoid too vigorous action.
After the metal has all gone into solution, the next step is to drive
off the unused nitric acid, as usual. This may be done without
previous filtering. Heat almost to a syrup, and when a little water
is added there will be an evolution of brown fumes. Add some
hydrochloric acid also, and evaporate again, repeating as often as
necessary. After the last evaporation, add water only, adding suffi
cient to bring the platinum concentration to about three ounces to
each quart of liquid. Let settle overnight or longer.
Silver chloride, tin oxide and lead sulphate will settle out as a
whitish mud. After it has settled well, filter off the clear liquid
into a clean jar or big beaker, and catch the residue, using the
shredded filter paper that was described in the purification of gold,
and throw it away.
When your solution is perfectly free from cloudiness, reprecipitate

the orange powder, using about a Troy ounce of pure ammonium
chloride to each ounce of platinum. Some workers advise having
both solutions quite hot at the time of adding them together. The
bright orange powder will come down at once. If much iridium is
present, the dark iridium ammonium chloride will slowly come
down too. The longer it stands, the more platinum will be pre
cipitated out, but impurities will come down too, therefore for
highest purity, filter immediately.
Wash the orange powder before it dries, getting it into a beaker
and covering it with a good quantity of ammonium-chloride water
and stirring well. Let settle, and pour off the liquid. Repeat, stir
ring well to wash out all impurities. These washings always con
tain some platinum, and possibly gold and palladium as well, so
put them in the Stock Pot. Finally catch the washed orange powder
in a clean filter paper, reduce to metallic form, make a test button,
and if pure melt as usual. This precipitation can be repeated if
desired.
This method of repeated precipitations with ammonium chloride will
readily remove gold, silver chloride, lead, tin, and many other con
taminants. But it does not immediately remove the sister metals of
the platinum group. In their paper on the PURIFICATION OF THE
SIX PLATINUM METALS, Wichers and his co-workers say: "With
the possible exception of iridium, . . . one might expect to eliminate the
metals of the platinum group about as rapidly as gold and the base
metals, by the repeated precipitation of ammonium chloro platinate.
This is not the case. Rhodium, iridium, and palladium display re
markable persistence in contaminating the platinum salt." These writers
then give an alternative and quite different scheme, described as the
"collective precipitation of impurities," which however should not be
attempted by the non-chemist. As we said above, the jeweler for
tunately does not often demand that his platinum be completely free of
its sister metals, and accordingly the method of reprecipitation with
ammonium chloride normally suits his requirements.
Platinum, unlike gold, should not be repurified by fire. At

tempts to remove impurities by repeated remelts cause a loss of
metal, without compensating improvement in the quality of the
platinum.
REPURIFICATION 177
PART III. REPURIFYING PALLADIUM
Palladium recovered from dental and jewelry wastes is usually

contaminated with platinum, copper and other base metals, and
perhaps with iridium as well. The following cases will be con-
sidered: the red palladium ammonium chloride and the yellow
palladosamine chloride; the sponge obtained by burning either of
these colored salts; the palladium "black" obtained by reducing the
colored salts with formic acid; and palladium metal in the form of
button, bar, or foil.
The Colored Powders
This process utilizes the reactions employed in Chapter XIII under
the heading "Further Treatment for Palladium." You will recall
that you took a solution containing palladium, added ammonia to
form a flesh-colored precipitate which redissolved in an excess of
ammonia, filtered out base-metal impurities, then re-acidified to
obtain the yellow precipitate. The same chemical reactions will be
utilized again here, as follows:
Place the colored powder either the red or the yellow in a good
sized flask or beaker and add cold concentrated ammonia (am
monium hydroxide, C. P.) slowly, stirring well. The palladium salts
dissolve to give a solution that is almost colorless. Copper and
nickel salts dissolve to form sapphire-blue compounds. Iron forms
pasty rust-colored flakes, lead and tin form pale gelatinous flakes.
Platinum and iridium remain as orange or dark red powders. Use
just enough ammonia to dissolve everything that will dissolve, add
a volume or more of water and set aside an hour or two for the
pastry precipitates to coagulate. If too much ammonia was added
by mistake, some of it can be removed by warming the dish on the
steam bath. Let settle, then filter clear. The stuff in the filter will
consist of salts of platinum and iridium, plus the pasty hydroxides
of iron, tin, etc., and should be placed in the paper storage.
Next, add hydrochloric acid, a few drops at a time, to the strongly
ammoniacal solution. It will become hot and fume and bubble,
so work slowly and stir well. Add enough hydrochloric acid to
acidify the whole solution, as determined by the change in color,
and until no more yellow salt forms. The yellow salt settles readily,
and should be captured in a filter paper, washed in cold water,
dried, and a sample should be reduced to metal and melted into a
small test button. The solution that remains should be tested with
stannous chloride Testing Solution A, and disposed of as suggested

in Chapter XIII, according to its contents.
If the yellow powder just obtained is not yet of the desired purity,
repeat the above process.
Repurifying Palladium Buttons, Bars, and "Black

Palladium metal in the form of buttons or bars should be rolled
thin or granulated. Palladium "black," the fine powder obtained
by reducing the colored salts with formic acid, is already in a form
that will react almost instantly with acids.
The first step, in either case, is to get the palladium into solution.
You are already familiar with two ways of doing this. In Chap-
ter XIII you dissolved palladium in aqua regia, and in Chapter XIV
you dissolved it in nitric acid. Either solvent can be used here,
after which the solution is filtered well and the palladium repre-
cipitated exactly as in Chapter XIII or Chapter XIV.
If you decide to use nitric acid, dilute it with at least two parts
water to one of acid. Action is very rapid with "black" or thin foil,
but thick lumps may require some heat. Gold, tin oxide, silver
chloride, and a few other impurities will remain undissolved. If
platinum is present in the black powder, most of it will remain
undissolved. Platinum present as an alloy with the palladium in
massive form, however, will probably go into solution with it.
Filter the solution clear and put the residue in the paper storage,
as it may be of some value. You are now ready to precipitate the
red powder again, exactly as in Chapter XIV.
If you decide to use aqua regia as a solvent, dilute it also with
some water. Aqua regia will dissolve all the gold and probably all
the platinum group metals that are present, leaving as a residue
any silver, lead, or tin oxide. From here on, proceed exactly as in
Chapter XIII.
Repurifying Palladium Sponge
The sponge obtained by burning the colored powders usually
shows blue and green shades due to palladium oxides. These
oxides, as we remarked before, do not dissolve readily in nitric acid
nor in aqua regia. They do dissolve in hydrochloric acid, however.
This fact must be borne in mind. Sometimes the best plan is to re-
duce these oxides at once to metallic form, by heating them with
alcohol as in Chapter XIII. The alcohol usually catches fire, so work
REPURIFICATION 179
carefully to avoid burns on the hands or face. Another plan is to

boil the sponge with a little formic acid to which some ammonia
has been added, as in Chapter XIV.
A third way is the way mentioned in Chapter XIII when you
dissolved the black powder in aqua regia you applied plain hydro
chloric acid first, to dissolve the oxides, then without filtering added
the proper quantity of nitric acid to form aqua regia.
Whichever plan is used, the procedure from this step on will be
according to methods with which you are already familiar.
PART IV. REPURIFYING SILVER
In general it does not pay the small organization to rework silver.

It is better practice to sell it or exchange it for pure metal. However,
a worker occasionally wants to purify some silver for special reasons,
perhaps merely to satisfy his curiosity, and so a description of one method,
suitable for handling small lots, will be given.
Roll the metal thin or break it up, and dissolve it in hot concentrated
sulphuric acid. (See Chapter V I ) When the silver has dissolved, dilute
the solution by pouring it slowly into a considerable quantity of water,
and set it aside for the undissolved particles to settle. Capture these
as they may contain gold, possibly platinum group metals as well, and
put them in with your next batch of filings.
When the dilute sulphate solution is perfectly clear, add to it enough
hydrochloric acid to precipitate all the silver as silver chloride. Let this
settle, pour off the liquid, and wash the chloride well with much hot
water. Finally reduce it to metallic silver by one of the methods given
in Chapter VIII, using high grade reagents. (Thus, if you elect to
reduce it with zinc, use chemically pure zinc, free from arsenic or other
impurity.)
Test the liquid from which you recovered the chloride, and if it con
tains a worth while quantity of palladium or platinum, add it to the
Stock Pot, first evaporating off as much of the water as you conveniently
can.
Large plants handling silver in quantity, use the electrolytic methods
that will be described in Chapters XXI and XXII.
PART V. REPURIFYING IRIDIUM
In some cases the iridium black obtained from the Stock Pot in
Chapters XIII and XIV may contain ruthenium, rhodium, or osmium.
The non-chemist is advised to sell such powder to a professional refiner
who gives credit for these rarer elements. In Chapter XVII these metals
will be discussed again, and the difficulties attendant upon refining them
will be described.
It has been pointed out that this iridium black is often contaminated
with porcelain or glass that has flaked from the reagent bottles and
vessels. Earthenware is especially apt to scale under strong acids or
alkalies. A way to get rid of these flakes is given in the next chapter
Chapter XVI and the task is easy and inexpensive to perform.
The melting point of iridium is very high, and the attempt to melt
the clean black powder obtained from the Stock Pot may be discourag
ing, even when the powder is of good quality. Normally it contains
more or less platinum, and to most users this is not objectionable; in
deed, many workers try to keep the two metals together. Accordingly
the usual fate of this clean black powder is to be melted up with sufficient
soft platinum to give an alloy suitable to the manufacture of new goods.
If the refining has been carried on successfully, the result will be a
button of good quality and hardness. If too hard, remelt it with a little
more soft platinum. (Methods of melting platinum are described in
Chapter A of the Appendix.)
But sometimes this button is not of good quality, even when there
is no ruthenium, osmium, or rhodium, and even when the flakes of por
celain have been removed. For the beginner the best procedure may
be to sell such a button to a dealer who gives credit for iridium. But
the worker who has had a little experience, especially if he has the
inclination to study these metals, will sooner or later elect to repurify
such a button, in the same way that he has repurified his gold or
platinum. There are two or three ways of doing this, but none is really
satisfactory.
Suppose your purpose is not to separate the indium from the platinum,
but merely to obtain a good-quality iridio platinum alloy. In that case,
add enough soft platinum to reduce the iridium content to about five
percent, making an alloy that is soluble in aqua regia. Roll the metal
very thin, cut it up, and dissolve it in full strength aqua regia, as in
Part II of this chapter, and precipitate the orange powder of platinum-
ammonium-chloride as usual. Do not filter immediately let it stand
overnight and the dark iridio ammonium chloride will come down too.
Filter, wash, and burn as usual, and the resulting sponge should when
melted produce a button of good quality, containing most of the platinum
and a large part of the iridium. The precious metals remaining in the
solution should be precipitated by zinc, and recovered as described in
Chapter XIII.
Suppose, on the other hand, your purpose is to separate the platinum
from the iridium. One way is to proceed as above, obtaining the mixed
colored powders as above. Then (instead of burning them) reduce these
powders with formic acid as in Chapter XIV. This gives you a mixture
REPURIFICATION OF GOLD 181
of platinum black and iridium black, not unlike mixtures you have
already encountered in Chapters XIII and XIV. Wash well, dry, and
ignite.
Your iridium black is now in the less soluble form. Treat the ignited
powder with dilute aqua regia, which dissolves platinum, leaving the
iridium black undissolved. As noted before, this separation is not a
perfect one, but it serves many purposes.
For a more complete separation, see the section on Electric Contact
Points, in Chapter XVI, where the iridio-platinum alloy is melted up
with a quantity of zinc, then treated with acids as there described.
The classic separation of iridium and platinum, as used by assayers,
requires the alloy to be fused with six or more times its weight of lead,
at a temperature about that of melting gold. The iridio-platinum-lead
alloy is next treated with nitric acid to dispose of excess lead, the residue
then being boiled in aqua regia. This leaves the iridium unattacked,
in the form of small bright crystals. This method is laborious, and not
suited to the amateur refiner's abilities. Full details of this procedure
will be found in the publications of the National Bureau of Standards,
as listed at the end of Chapter XVII.
CHAPTER XVI
SOME MORE SPECIAL CASES
Sand, glass, porcelain and enamel. Emery and carborundum. Chromium

plate. The peculiar case of 18-k green gold with platinum; stripping
with the current; equipment; the current; recovering gold and silver
from cathode and cyanide bath. Mercury and amalgams. Dental pins.
Cold pen points with osmiridium tips. Gold leaf. Contact points.
Chemical ware.
SAND, GLASS, PORCELAIN AND ENAMEL
These substances, often found in jewelers' and dentists' bench

wastes, are not removed by ordinary solvents. They can be gotten
rid of by melting the metal into a button, using a flux, but sometimes
this is not desirable. Fortunately they dissolve readily in hydro-
fluoric acid.
Hydrofluoric acid is dispensed in wax bottles, and must not be used
with china or glass vessels, as it quickly attacks the surface and
would in time dissolve the whole vessel. Plastics such as Bakelite
are not attacked by it, nor is platinum or gold. Lead is attacked very
slowly.
To remove sand, enamel, porcelain, and glass from your platinum
filings or from the iridium powder recovered from the Stock Pot,
simply place them in a suitable dish and cover them with hydro-
fluoric acid, and warm gently. Do this after they have been through
all the other regular treatments with nitric acid, aqua regia, or the
like, and have been washed well.
Lead dishes are inexpensive, and can be bought from almost any
supply store, or shaped up out of a piece of sheet lead. Eimer and
Amend No. 22026, four inches in diameter, costs $0.28. Do not heat
it strongly or it will melt. The best vessel for this job is a dish
of platinum or gold, shaped about like a shallow evaporating dish.
It will last indefinitely.
182
SOME MORE SPECIAL CASES 183
The fumes are bad for the lungs and skin, and will frost your
window glass if allowed to touch it, so work carefully, outdoors if
possible.
Leave the filings in the acid until the foreign matter is dissolved,
stirring with a piece of platinum wire or strip of Bakelite or hard
rubber. If you use a platinum or gold dish, it is easy to apply heat,
Lead dish.
but a lead dish must be heated with caution, preferably on a sand

bath.
Sometimes you can use the same hydrofluoric acid on two or more
lots of filings, saving it in a wax bottle. Finally pour the spent acid
down the drain, washing it down with much water.
Wash the filings very clean, then dry and melt as usual.
EMERY AND CARBORUNDUM
These are not soluble in ordinary acids, nor in hydrofluoric acid.

Some kinds of emery used as a polishing material can be removed
from platinum filings by boiling them in water to which has been
added caustic soda.
Carborundum, and most other kinds of emery, are not readily
removed by any of the solvents in ordinary use, and when present
in sufficient quantity to be troublesome, are best gotten rid of by
melting the filings into a button, when possible, or by dissolving
all the metal in aqua regia.
CHROMIUM PLATE
White gold jewelry, especially the cheaper kind, is sometimes fin

ished with a thin coating of electro-deposited chromium. Chromium
differs from most base metals in that it does not dissolve easily in
nitric acid, and accordingly is sometimes mistaken for white gold
or platinum. It does dissolve readily in plain hydrochloric acid or
dilute sulphuric acid, if hot, and can be removed by either.
If chromium should get into your aqua regia solution, it prob
ably will cause no difficulty. It behaves a little like nickel in some
respects, like iron in other respects.
THE PECULIAR CASE OF l8-K GREEN GOLD WITH PLATINUM STRIPPING

WITH THE CURRENT
You will recall (see Chapter VII) that 18-k green gold, which con
sists of 75 parts gold and about 25 parts silver, is practically in
soluble in acids, as it contains so much gold that nitric acid will not
affect it, and so much silver that aqua regia will not affect it. When
mixed with platinum, as in jewelry with a green gold base and a
platinum top, it presents a real problem to the refiner. Sometimes it
is convenient to mix such scrap in with considerable other metal,
such as yellow gold or brass, to produce a button that will yield to
aqua regia; but here is another method, differing fundamentally
from those previously suggested, that is peculiarly suited to this
problem.
This method is being used occasionally with yellow or red gold
and platinum combinations as well. It consists of dissolving the gold
away from the platinum, using a cyanide solution and a "reversed"
electric current. If you have had experience in electroplating, this
method will be easy to understand and carry out.
The method is designed for mixtures of green gold and platinum,
not for buttons or alloys. This distinction is the same one that was
mentioned so often at the beginning of Chapter XI, and is based
on similar considerations, as will soon be clear.
Equipment. Use an earthenware jar or glass cylinder such as
is used in electroplating. It must be heated, preferably by a steam
bath. Use the same direct electric current that you use for gilding,
electro-stripping, or the like. Your clippings are wired directly to the
anode rod. Filings are placed in a platinum-lined dish, which is
wired to the anode rod. Take a china bowl and line it with thin
platinum foil, reaching not quite to the edge, using a dish large
enough to hold the filings comfortably, spread out thin, so they can
be stirred without spilling. All wires under the water line are of
platinum. This wire can be used for years without loss, so use wire
of generous size, as thin wire wastes current.
A sheet of lead or copper, called the cathode, is attached to the
work rod by means of stout copper wire. Use a small cathode for
small jobs, and a larger one for larger jobs. Thus for 300 or 600
dwt. of clippings, use a cathode 6 X 6 inches or larger. Filings can
take a larger cathode than the same weight of clippings, since they
present more surface to the current.
Make up a cyanide bath, with 4 to 8 ounces commercially pure

sodium cyanide or cyanide mixture to each gallon of water.
Large scrap should be rolled extremely thin. Filings should be
treated by the preliminary treatments mentioned in Chapters III, V
and XI burn out the grease, use a magnet, and remove as much
base metal as possible with the usual nitric acid treatment, and wash
with much hot water.
As soon as your metal is thoroughly washed, place it in the cyanide
bath, which should be heated to about 140° R ' a n d t u r n on t h e
current.
Gold, silver, and copper will dissolve, and finally will be deposited
(more or less completely) upon the cathode. Platinum will not be
affected at all. The gold dissolves slowly, the time depending upon
how thick the gold is, and how strong the current is. The hotter
the bath and the better the connections, the shorter the time, but
the more cyanide will be used up. Let it run all night if you wish.
You can tell when all the gold is gone by the looks of the remaining
platinum. Rinse it, and melt it as usual, with the oxy-gas torch.
The Current. Use a stronger current, if possible, than what
you are using for coloring or stripping. Have a generous "action,"
with perhaps 6 to 10 volts, and as much as 15 amperes or more. Use
as much current as you wish, as long as your wires and rods do not
get hot.
Watch out for clippings that sometimes drop off the wire, par
ticularly those on which the platinum is not continuous. Look out
for short circuits. But have the cathode and metal as close together
as convenient.
Remember also that the current will take the shortest path from the
anode rod to the work rod. You want it to go through your clip
pings. Some workers slip a small piece of rubber tubing over the
wires that are used for wiring on the clippings, so as to force the
current down into and through the clippings themselves.
It takes longer to refine filings than strips. If you can fix up
some method for stirring the bath all the time, the work will go
more rapidly, and a higher current can be employed.
Be sure that the current goes THROUGH your filings. That is
why you use a platinum-lined dish, instead of an all-platinum dish;
if the dish were all metal the current would go from the outside of
the dish to the cathode and skip the filings.
Another plan is to put the dish on the bottom of the earthenware
vessel, and hang the cathode over it, so that the current goes
straight up and down. In this case, you must do something to keep
flakes of gold from the cathode from dropping back into the dish,
where they will have to be dissolved again.
One way to keep these flakes from dropping back is this; buy at a
department store a pair of wooden rings used for stretching em
broidery. Stretch a piece of thin cotton cloth over these rings. Lay
this on your dish of filings to catch the loose gold.
If a dark cloud forms in the solution, around your metal, and
remains after you stir, add more cyanide. Some workers simply add
a little cyanide to the bath each time they use it. Now and then
scrape any loose flakes off the cathode, and save them for later puri
fication.
You can usually judge by the color of the metal when all the gold
has been removed from the platinum. Look for large pieces; the
small ones dissolve first. When all the gold has been removed,
wash the platinum well with hot water, adding the washings to the
cyanide solution. If you did not go through the filings with a
magnet at first, do so now, as the current does not remove iron.
Your platinum should now be clean and free from gold and base
metals, ready to be dried and melted.
When you turn off the current for any reason, lift the cathode out
of the bath, to keep it from dissolving in the cyanide.
Recovering Gold and Silver from Cathode and Bath. The cath
ode by this time is coated with a spongy layer of gold, silver, and
probably other metals. Scrape off the loose deposit, and use the
cathode over and over as often as desired, finally rolling it thin and
refining it as below.
When the cyanide bath becomes foul, cement out the gold and
silver by means of zinc see Chapter VIII on Cyanide Solutions.
Collect the deposited powder and wash it well.
All of these materials are refined by essentially the same process
the loose flakes that dropped or were scraped from the cathode, the
cathode itself after it has been rolled thin, and the cement gold and
silver that were deposited out by zinc. Remove silver and base
metals by means of nitric acid; the powder that remains will be
high quality gold, possibly fine enough for your needs. If not,
repurify it according to the methods of Chapter XV.
This plan of stripping gold from platinum by means of the re
versed current requires little labor, and its cost will depend upon
that of cyanide and electric energy. The consumption of cyanide

can be controlled by regulating the temperature of the bath, and by
occasional additions of caustic soda, sufficient to keep the concentra
tion of the latter at about 2 to 4 grams per liter (about 3 to 6 dwt.
per quart). The cyanide does not decompose appreciably below
113 deg. F. As the temperature rises the decomposition increases,
with the formation of ammonia and sodium formate. Above 149
deg. F. cyanide decomposes rapidly, and when nearly at the boiling
point the whole bath can be destroyed within a few hours. For most
economical results, keep the bath at about 140 deg. F. Unfor
tunately, the bath works faster when hot; the worker must there
fore make a compromise between speed and cyanide conservation.
This method has been described as an electrolytic method. How
ever, it differs in many ways from the better-known electrolytic
methods used in large-scale refining (see Chapters XXI and XXII)
and should not be confused with them.
• * • *
MERCURY AND AMALGAMS
In earlier chapters you were advised to keep mercury and its

alloys (known as amalgams) separate from your ordinary wastes.
In some ways mercury acts like silver, being soluble in nitric acid,
and being precipitated by hydrochloric acid as a white chloride.
It can cause confusion in other ways as well, contaminating the
gold and interfering with the stannous chloride tests.
The amalgams used in dentistry are made by mixing mercury with
alloys of tin, silver, zinc, and other base metals. Normally they are
of low value and inconvenient to refine. Now and then, accidental
bits of gold or even of gold-platinum alloys, get mixed in with a lot
of old amalgam, but this mixing should be avoided as far as possible.
In shops that handle gold leaf, gold amalgams are sometimes en
countered. They are normally of more value than dental amalgams.
The process of refining is essentially the same for both kinds of
amalgams.
The material is first heated strongly in an iron retort to drive
off the larger part of the mercury, which is caught in water and
should be redistilled before being used again.
A crust of metal remains in the retort; it always contains some
mercury. It is melted into a button. This second melting, of the
crust, is the most dangerous part of the work, as the heavy mercury
fumes can seep out into the building and make the workmen sick.
The button that is obtained from dental amalgams is difficult and
expensive to refine on a small scale, largely because of the large
proportion of tin it contains, and we do not encourage you to refine
it yourself.
The button obtained from gold amalgams may be of high quality
and easy to work with, and usually may be refined by methods that
are already familiar.
* * • *
Now for the details of the method of retorting off the mercury.
A very small amount of amalgam can be retorted in a glass tube,
sealed at one end and bent into a V. A tube of 3/4 inch diameter
is best. The amalgam is broken into small pieces and dropped into
the closed end, where it is heated, and the fumes condense in the
long open end of the tube.
In another method for small amounts of amalgam, a potato is
used as a condenser. This is a device popular with prospectors
because it is very simple, and saves a part of the mercury. A large
potato is cut smoothly in half, and in the flat surface of one half a
recess is hollowed which should be considerably larger than the
amount of amalgam to be treated. The amalgam is placed on a
clean sheet-iron surface (a frying pan or clean shovel), the half
potato is placed over it, and the whole is set over a hot fire. Some
mercury vapor will escape under the edges of the potato, and, as
before, these fumes must be avoided. After fifteen or twenty min
utes of strong heating the potato may be lifted off for inspection.
If all the mercury is gone from the gold the potato may be crushed
and panned, and a considerable part of the mercury will be re
covered.
However, it is much better to use a regular cast-iron retort or still.
Eimer and Amend list an iron retort for this purpose, their number
30574. Pint size $4.50; two-pint size $5.00.
The material is put in and the lid clamped down very carefully,
and a long iron pipe is connected with it. The retort should be
coated on the inside with chalk, or painted with a thin paste of
clay, and dried, to keep the gold from sticking to the iron. A lining
of paper serves the same purpose, but forms an objectionable deposit
in the condenser tube.
Do not fill the retort too full; pile the lumps of amalgam loosely
in the retort, to not over % of its volume. (When retorting liquid

mercury, 1/3 to 1/2full is enough.) Clamp the cover on tightly, being
sure that the condenser pipe is free from obstructions. The joints
Retort for mercury.
hould be luted with some sealing compound, such as a cement of

ground asbestos with litharge and glycerine.
Set the exit tube so that the mercury will run downward, and place
Cooling water
Alternative form of water-

_* sealed vapor trap
Apparatus for retorting amalgams A, set up of small retort with condenser,

B, water sealed vapor trap, C one style of retort, D, Nevada type retort, E, bullion
mold, F, graphite crucible
Reprinted from U.S. Bureau of Mines Information Circular 6787
a vessel of water under the open end. Wrap a piece of cloth around
the open end of the exit tube, like a cuff, and let it dip into the
water. The idea is to conduct the mercury into the water as it con
denses, the wet cloth helping to cool any fumes that have not al-
ready been condensed.
DO N O T let the IRON tube dip into the water only the bottom
edge of the cloth cuff. Remember that when the retort cools, water
may suck back into the hot iron pipe and cause it to explode. The
cloth cuff tends to prevent this sucking back, since it is not airtight.
But, after you once start to heat, keep the heat going steadily, and
remove the water before stopping the heat. This is important.
The retort may be heated by a Bunsen burner, blow torch, or other
means. Old-timers say a wood fire is better than a coal fire. Apply
a low heat at first, then after ten or fifteen minutes increase it grad
ually, just enough to maintain a small steady trickle of condensed
mercury. 1
Mercury boils at about 675 deg. F., which is halfway between the
boiling point of water and the first visible red heat of iron. How
ever, mercury begins to volatilize at the boiling point of water, and
gives off fumes that are dangerous to the worker even when heated
quite gently. When no more mercury appears, increase the tem
perature for a few minutes to red heat. Do not go much above this
or you will drive over the zinc, which may clog the pipe. Re
move the water, then let the retort cool. Some vapor may remain
in the retort, so be careful not to breathe the fumes when you take
off the cover.
A crust of metal remains in the retort. Sometimes it is sold or

refined as it is, but generally it is melted first, in a gold-melting fur
nace, with a flux such as borax glass. As we said above, this second
melting may be dangerous, as the fumes may not go up the flue
unless the draft is better than average. Aside from this hazard, this
step does not differ markedly from ordinary melting. The molten
metal may be cast into an ingot mold, or possibly granulated in
water, as circumstances will dictate.
The metal obtained from most dental amalgams will contain
much silver and tin, and is usually of low value. It is possible to
refine it by acid methods, of course, but the process is made tedious
by the tin, and the amount of acid and labor required is usually
enough to make the job unprofitable. In most cases the best dis
posal of such metal is to a silver buyer.
i Parts of this section are adapted from United States Bureau of Mines Informa-
tion Circular 6787, of October 1934, PLACER MINING IN THE WESTERN
UNITED STATES, Part II, by E. D. Gardner and C. H. Johnson.
As we said at the beginning, there is always a possibility that acci-

dental admixtures of gold, or gold and platinum, can get into collections
of low-grade dental amalgams. Accordingly a test for these elements is
always in order. Take a scrap of the metal and treat it first with nitric
acid to dissolve silver, then add hydrochloric acid (forming aqua regia)
to dissolve gold and platinum; add a little water and let the copious
white silver-chloride settle; if gold or platinum is present it will now be in
solution in the liquid, which should be tested with stannous chloride
Testing Solution A.
The final task is to repurify the distilled mercury. Sometimes a

single re-distillation is the best method. A clean retort and clean
condenser pipe, free from rust, should be used, and it is not necessary
to heat as strongly as before. This re-distillation is not as simple as it
would seem, because when the mercury boils it "bumps" violently,
and sometimes shakes the condenser pipe loose. Accordingly in
some plants that do this work on a large scale, the distillation is
carried out in a vacuum, at a relatively low temperature.
Or, the mercury can be purified by washing with chemicals.
Place very dilute nitric acid in a tall narrow vessel, such as a length
of wide glass tubing; let the mercury trickle down it a drop at a
time. One plan is to place the mercury in a bag of chamois and
squeeze it through, into a funnel that rests in the top of the wide
glass tubing. The acid dissolves tin, zinc, silver, etc. After that,
wash the mercury first with water to remove the acid; then with
sodium hydroxide to remove any residual traces of acid; then water
again to remove sodium hydroxide; then finally with ether to re-
move the water. The mercury should be bright and free from scum,
ready to be used again, or sold to a mercury dealer.
DENTAL PINS
High grade dental restorations, especially those made several years

ago, often contains substantial iridio-platinum pins to anchor the
porcelain teeth to a metallic plate. When the price of platinum be-
came high, many substitutes were devised. Some were of base metal
with platinum ends or coatings. Others were made of alloys of
platinum with gold, palladium, nickel, and whatnot. Others con-
tained no precious metals at all.
When handling dentures, first get rid of the porcelain and vul-
canite by hammering and pinching with pliers. Then heat the
M REFINING PRECIOUS METAL WASTES
metal parts with nitric acid. Base metals will be attacked and some
kinds will disappear in a few minutes. Test the nitric acid, as it
may contain palladium; if not, throw it away.
Platinum pins, and pins that consist of high grade platinum-gold
alloys, will not be attacked by the nitric acid. It may be worth while
to get pins of this type out of the metal in which they are embedded,
heating them in a gas flame until the solder softens and they can be
pulled out. This is slow work, but sometimes it pays. (The alterna
tive is to melt up everything and treat it according to Chapter XII.)
The metal from which the pins were pulled may consist of high
grade gold. Generally it contains neither platinum nor palladium
and should be refined as a simple gold job.
Next, place the pirn in nitric acid again, and heat, to remove any
remaining base metal, then treat with aqua regia. Iridio-platinum
pins will show almost no action, indeed they can be boiled in aqua
regia for hours before dissolving completely.
The cheaper pins will dissolve more readily in aqua regia, depend
ing upon their composition. If any silver is present it will form the
familiar grey scum of silver chloride.
Most people pick out the high-quality iridio-platinum pins and
melt them without further refining. The cheaper alloys should be
allowed to dissolve, and the precious elements will be recovered ac-
cording to methods with which you are already familiar.
• * * *
GOLD PEN POINTS
Some gold pen points are tipped with grains of native osmiridium
a naturally-occurring alloy of osmium and iridium. Synthetic
alloys of the platinum-group metals are also used, as well as a few
base-metal combinations. Good fountain-pen points must be very
hard, and most of them are also difficult to dissolve or melt.
The refiner's normal procedure is to clip off the points with shears,
and to handle the gold part of the pen as a simple high grade gold
job. The points should then be treated with aqua regia to remove
remaining gold. The grains that survive this treatment may or may
not be valuable.
Some of the inexpensive base-metal grains contain tungsten, and
will behave so much like osmiridium as to confuse the worker. Ex
posing the washed grains for a second or two to an oxy-gas flame
may reveal the base-metal particles, as tungsten forms colored oxides
and may take fire, while osmiridium remains unchanged. This test
should be used with care, however, as longer exposure to the flame
may volatilize osmium, as well as the ruthenium that is used in some
of the better-grade synthetic alloys, occasioning a loss of valuable
elements as well as danger to the worker, the fumes being poisonous.
When all these facts are considered, it is not surprising that many
professional refiners do not buy these grains, even when presented
in sufficient quantity to justify an assay. (The complete assay of
these grains can be a laborious task.)
In Chapter XVII more will be said about methods of handling the
rarer elements of the platinum group. Meanwhile, for most workers
the best program is to recover all the gold, and to accept whatever
price is offered for such grains as remain after the treatment outlined
above.
Now and then a worker makes the mistake of melting up a lot of
gold pens without first clipping off the points. The osmiridium does
not alloy with the gold, merely remains as lumps that appear when
the metal is rolled. To overcome this, remelt the gold in a tall nar
row crucible, and let it cool slowly in the crucible. The heavy os
miridium will settle to the lowest part, which can be sawed off and
treated separately. T. K. Rose 2 p o i n t s o u t t h a t osmiridium grains
settle better from an alloy that is mainly silver, so the rich bottom
portion should be remelted several times with silver, the lowest part
being clipped off each time. Finally granulate the metal, which by
that time is mainly silver, and part it with sulphuric acid (see Chap
ter VI) which dissolves silver and leaves a residue of gold and
osmiridium grains. Treat this with aqua regia to dissolve the gold,
and clean grains of osmiridium will remain.
GOLD LEAF
Gold leaf is thinly-beaten fine gold, or, more generally, an alloy

containing a very little copper or silver. The manufacture of gold
leaf is simple in theory sheets of high-karat gold are rolled and
hammered to extreme thinness but the actual operation is an art
that demands much deftness and close attention to many details, as
well as muscular prowess. (See Encyclopedia Britannica on Gold
beating.) Accordingly the user is not apt to make his own leaf,
and so has little incentive to refine the scrap. The usual procedure
2 THE METALLURGY OF GOLD. Seventh Edition, page 463.
is to return scrap to a professional refiner, who burns out the trash

and refines the residue as a simple gold job, such as would be
handled according to Chapter V or XV.
ELECTRICAL CONTACT POINTS
These are made in many sizes and shapes, of many materials, both
precious and non-precious. Silver is widely used because of its low
electrical resistance, but to the refiner the most interesting contact
points are those made of platinum and its alloys
For highest grade equipment, where price is no object, the
lndio-platinum alloys are preferred, from ten to twenty-five percent
iridium. The requirements 3 t o b e m e t ^ c o n t a c t p o i n t s differ
widely, and for many purposes cheaper materials are adequate.
Economy is effected in various ways, notably by making the bulk of
the article of brass or other base metal, and coating only the exposed
area with precious metal.
In the catalogue of a single manufacturer4 of c o n t a c t p o i n t s we
find the following metals and alloys mentioned fine silver; Sterling
silver; coin silver; silver overlaid or inlaid on monel metal, brass or
copper; silver welded to steel; platinum; lndio-platinum; platinum
group alloys welded to nickel, steel and bronze; platinum overlaid
on bronze; tungsten; tungsten carbide; special alloys of platinum,
iridium, gold, silver, palladium, and other precious metals. Other
sources mention gold-silver, platinum-ruthenium, gold-silver-nickel,
gold-nickel, silver-copper-tin, and many other alloys.
In the face of so extended a list, it is clear that the refining of old
contact points may reasonably be deferred until the worker has
acquired some familiarity with metals and their reactions.
Sometimes he is fortunate enough to know the composition of
the materials he is handling, and his task should not be difficult.
Other times his first, and perhaps his most difficult step, will be to
identify the materials. A few acquaintance tests on contact points
of known composition will save much time in later operations.
Points usually reach the refiner still attached to the brass, steel,
or other parts to which they were welded or screwed while in use
Tungsten and tungsten carbide should be gotten rid of promptly,
3 For a discussion of these requirements, see USE OF THE NOBLE METALS FOR
ELECTRICAL CONTACTS, by E F Kingsbury Mining and Metallurgy, New
4 The H A Wilson Company, Newark, N. J.
as they are of small intrinsic value and can cause trouble if melted
in with precious metals. A few seconds exposure to the oxy-gas
flame, as mentioned in the section on fountain pen points, may
reveal them, as tungsten forms colored oxides and presents a charac-
teristic reaction to the flame.
The magnet will locate steel parts, and hot dilute sulphuric or
hydrochloric acids will dissolve them away without affecting the
precious metals. If the articles are not too small, it helps to roll
them as thin as the rolls will take them, but this is hard on
the rolls. Nitric acid will remove brass, monel metal, bronze, and
other base metal parts, but will also dissolve silver and palladium,
so should be used only when provision is made for the recovery of
those metals.
After as much base metal has been disposed of as seems expedient
and the worker must exercise judgment as to the amount of effort
to expend on each step a treatment with aqua regia is indicated.
It may dissolve all the precious metal parts, and tests given in
Chapter IX will suggest proper methods of recovery.
Some of the higher grade contacts will not dissolve in aqua regia,
even after prolonged boiling. This includes the valuable iridio-
platinum alloys, the iridio-platinum-rhodium alloys, and the iridio
ruthenium complexes. Possibly the simple isolation of these high
grade bits of metal will serve the refiner's need and his work will
be done. But if he wants to separate the elements, the following
J 5 is suggested: Melt the contact points high grade
platinum-group alloys insoluble in aqua regia with considerable
zinc, in a gold melting furnace. Granulate the resulting alloy and
treat it with hydrochloric acid, which removes zinc and leaves the
heavy metals as a granular powder. Wash this residue and heat it
to redness to convert iridium to a less soluble form. Treat it with
aqua regia and the platinum will dissolve, the iridium and ruthenium
remaining as a dark residue. Recover the dissolved platinum as
usual, with ammonium chloride. (Do not use this plan on alloys
containing osmium.)
Professional refiners handling these contacts in quantities are not
apt to have the patience to examine each contact with care, and may
melt everything together, obtaining a button that would be treated
5 Adapted from A S S A Y I N G P L A T I N U M M E T A L S , by Martin Schwitter.
Metal and Mineral Markets, Jan. 2 1 , 1932, p p . 6 - 1 0 ; Feb. 18, p p . 6 - 7 ; March 17,
pp. 6- 1 1 .
according to the scheme of Chapter XII or some modification thereof.

The assay of these points is laborious, and unless the composition is
known the price paid is correspondingly low.
CHEMICAL WARE
The chemical industry is the second largest consumer of platinum,

using the metal as a catalyzer in the manufacture of sulphuric and
other acids, as electrode metal in a variety of electrochemical proc-
esses, and as laboratory equipment in the form of dishes, crucibles
and other apparatus. For most purposes pure platinum, or platinum
alloyed with a small amount of some sister metal, is preferred. Spe-
cial purposes may employ special alloys; the thermo-couple, for
example, consists of two wires, one a platinum-rhodium alloy, the
other pure platinum.
Manufacturers of platinum laboratory ware are generous in provid-
ing instructions on the care and handling of such equipment. They
also make reasonable allowances for the return of damaged articles.
Accordingly only a small amount of chemical platinum will find
its way to the jeweler or small refiner. In some cases, such as old
electrodes and dishes, the refiner's job is easy. The bulk of the metal
may be of high quality, ready to be remelted after a little treatment
with nitric acid to remove surface impurities, the remainder being
refined according to the procedures in Chapter XII. Other types of
waste such as old contact masses are much more difficult, and should
be sold to a professional refiner.
Chemical apparatus is made of other precious metals too. Pure
gold, pure silver, and gold-palladium alloys are popular. To the
student who has followed the suggestions in this volume, the identi-
fication and refining of such ware will be easy.
Platinum-clad nickel and platinum-clad copper are recent develop-
ments that may became of considerable commercial importance.
Sheets of platinum are welded to sheets of the base metal and rolled
down after the fashion of rolled-gold plate. The resulting platinum-
clad plate is then fabricated into vessels of all sizes and shapes,
highly suitable to many chemical and industrial processes. When
properly made the platinum layer is thick enough to protect the
inner layer from corrosion, but thin enough to reduce the cost of the
whole to an attractive figure.
The refining of such dual metals should not offer undue difficulty
to the student of this book.
CHAPTER XVII
RHODIUM, RUTHENIUM, OSMIUM
Their rarity. Rhodium. Ruthenium. Osmium. Refining rhodium,

ruthenium and osmium, the "mysterious" metals. Some schemes of
separation: I. The classic system. 2. The method of Gilchrist and
Wichers. A list of books and papers referring to the three rarer metals.
THEIR RARITY
According to the Minerals Yearbook, published annually by the

United States Bureau of Mines, platinum is by far the most widely
used metal of its group. In 1938 it accounted for 68.8 percent of the
total of platinum group metals sold by domestic refiners. The
jewelry industry was the largest purchaser, the chemical industry
second, dentistry third, and the electrical industry fourth. Next
to platinum, palladium is the most widely used metal, constituting
27.5 percent of the total sold by domestic refiners in 1938, of which
over half went into dentistry. Iridium sales came next, but amount
to only 2.6 percent of the total. Osmium, rhodium, and ruthenium
together amount to only 1.1 percent of the total consumption of the
group.
From these figures it will be seen that osmium, rhodium and
ruthenium, however interesting they may be from the chemical
viewpoint, are of secondary importance commercially. All six metals
of the group are normally found in crude platinum ores, but osmium
and ruthenium are (except in a few exceptional cases) not used in
either dentistry or jewelry. Rhodium is occasionally used in small
amount. Accordingly the jeweler or dental technician who refines
his own scrap has little occasion to consider these three less familiar
elements.
This is fortunate, for their separation from their sister metals is
tedious and unsatisfactory, even in the hands of experienced chemists.
In general, if a non-chemist believes that his residues contain appre
197
ciable quantities of these metals, his best procedure is to find a re-

finer equipped to separate them (many professional refiners are not
so equipped) and when selling such stuff, always to mention the
fact that these elements are present. Because of the difficulty of
refining and assaying the purchase, the financial return may be
disappointingly low.
Full descriptions of these metals, their histories, natural occur-
rence, physical characteristics such as melting points, electrical con-
ductivity, and so on, will be found in the publications mentioned at
the end of this chapter and in Chapter E of the Appendix. For our
immediate purposes, their most interesting characteristics are their
insolubility and the fact that two of them osmium and ruthenium
form volatile oxides when heated.
RHODIUM
This metal received its name from the characteristic rose color of
several of its compounds. When in compact form it does not dis
solve in any single acid, and is attacked very slowly by aqua regia
except when alloyed with considerable platinum or other metal;
rhodium "black" dissolves slowly in hot concentrated sulphuric
acid. Rhodium can be plated electrolytically to give a handsome
silvery finish to jewelry, silverware, or optical instruments. Since
it is almost totally insoluble in acids, the finish once applied is not
easily removed. Grinding is the method usually employed, and the
deposit is so hard that even this is difficult.1
Rhodium plate is perfectly non-tarnishing, and when properly
applied can protect the underlying metal against corrosion. The
amount of metal needed to give a serviceable deposit is minute, and
the refiner need not expect to recover much rhodium from his
wastes, even when handling stuff that is known to be rhodium
plated.
The solutions used for rhodium plating are sold by certain of the
precious metal dealers. When they become exhausted or foul, do
not add such solutions to your other platinum wastes their separa
tion would be unduly expensive and troublesome. Instead, evaporate
i Atkinson and Raper mention a method of stripping rhodium plate from silver,
the article being immersed in a molten mixture of cyanides of potassium and sodium;
if it is made anode and a current passed, the removal of rhodium is hastened. The
method should be operated only by experts. For details see British Patent No. 419,819.
There is obvious need for a simpler and more adaptable method.
RHODIUM, RUTHENIUM, OSMIUM 199
them down a bit to get rid of excess water, then return them for credit
to the dealer from whom they were purchased, shipping by express
(not mail) and using the same acid-proof container in which they
came. It is possible, of course, to precipitate the rhodium with zinc,
but the task of making it up into a new bath is beyond the skill
of most non-chemists.
A second use for rhodium is in scientific apparatus, especially the
so-called Le Chatelier thermo-couple, mentioned in Chapter XVI,
whose essential part is two wires, one pure platinum, the other an
alloy of 10 percent rhodium with 90 percent platinum. Rhodium-
platinum alloys resist high temperatures extremely well.
RUTHENIUM
Ruthenium is the lightest of the platinum group. Its name comes

from "Ruthenia," an old name for Russia, where it is found. Like
rhodium, it is insoluble in any single acid, and difficult to get into
solution by any means. It is occasionally used as a stiffener and
hardener for platinum, the alloy containing 8 percent ruthenium
being almost as hard as the one containing 20 percent iridium. This
fact has been utilized at times when the price of ruthenium was low
enough to compete with iridium.
Alloys of palladium stiffened with a little ruthenium and rhodium
have been recommended for jewelry. Their working qualities are
good, and because of their low specific gravity an ounce of such
alloys goes almost twice as far makes up almost twice as many
articles of a given size as an ounce of iridio-platinum. A typical
formula is 95 parts palladium, 2 to 4 parts ruthenium, and the re
mainder rhodium. The sale of these alloys for jewelry has been
hampered in this country for three reasons: first, our stamping laws
are so worded that it is difficult to stamp them in a way that will
appeal to the buyer; second, the refining of the mixed filings is com
plex; and third, the melting of the scrap metal calls for some extra
care.
Ruthenium, when heated to about 600 degrees C, forms a volatile
oxide. When alloyed with other metals this tendency is lessened,
but it can occur nonetheless. The fumes are poisonous and the loss
of metal may be appreciable, especially when the ruthenium content
is high.
OSMIUM
Osmium is the heaviest of the platinum group, and has the highest
melting point. Like ruthenium, it has been used as a stiffener and
hardener of platinum when its price compared favorably with that
of iridium. Its employment in fountain pen points is mentioned
in Chapter XVI, and one of its compounds is used as a stain in
microscopy. When heated in air or oxygen, osmium also forms
a volatile and very poisonous oxide, and is even more dangerous to
the operator than ruthenium, the fumes appearing at a lower tem-
perature. Partly because of this volatility its alloys have found
limited usefulness, in spite of the fact that it is more efficient than
iridium as a hardener of platinum.
Any operation involving the heating of osmium should be con-
ducted in a hood with a good draft, and a mask should be worn to
protect the face, eyes and throat of the operator. The fumes have a
garlicky odor and act as a poison to the lungs. The vapor is reduced
to the metallic state when in contact with the eye tissues, and if
deposited upon them in sufficient quantity can produce permanent
blindness. The word osmium is derived from a Greek word mean-
ing stench.
Osmium and ruthenium and most of their alloys are dissolved
in acids only with difficulty, even when in the finely divided form,
and in most systems of separation the volatility of these oxides is
utilized.
REFINING RHODIUM, RUTHENIUM, AND OSMIUM, THE

"MYSTERIOUS" METALS
As we said before, it is fortunate that jewelers and dental tech-

nicians have little occasion to accumulate these metals. Until the
last few years there was little in print on their refining, and an aura
of mystery, partly justified by the facts, surrounds them. The in-
telligent layman naturally resents this situation, and we shall now
mention the several reasons that have, in combination, brought
it about.
First, these three metals have so few industrial applications that
only a few people are interested in their technology, which means
that publishers are not tempted to print books about them. A second
reason is that some companies that sell these metals have deliberately
shrouded their laboratories in secrecy. During the last few years,
however, the free publication of research by other firms, and espe

cially by our National Bureau of Standards, has made much infor
mation available to all, and the tendency now is toward greater
frankness, as is proven by the list of publications at the end of this
chapter.
A third reason, and the main one, is directly related to their
chemical characteristics. With the discovery of new reactions and
better procedures, their separation will undoubtedly become easier
and more dependable, and the cloud of mystery will blow away.
Active research is going on in several famous laboratories, some of
it directed at precisely this problem. At present, however, the
refining of these rarer elements is not for the laymen, mainly for
chemical reasons such as will now be described.
The reader has met several procedures described as "good" separa
tions. For example, the precipitation of gold from acid solution by
copperas is normally a "good" or "clean" separation; when properly
conducted it brings down the gold all the gold and nothing but the
gold cleanly and completely. There are thousands of such separa
tions known to chemistry, many being used by non-chemists. But
when the reader precipitated platinum as the orange platinum-
ammonium-chloride he did not bring down all the platinum, and he
did bring down some of the sister metals, notably iridium and
rhodium if they were present. This precipitation therefore is not a
"clean" one, and affords only partial separation.
Again, the reader recalls that when he separated platinum from
iridium by boiling the mixed blacks in dilute aqua regia, some of
both metals dissolved, and some of both metals remained in the
residue. This is another incomplete separation, highly characteristic
of the group. It is chemical characteristics such as these which are
the real reason for the apparent mystery that surrounds these metals.
Nor is that all. The reader must have been struck by the number
of alternative methods that are possible in the refining of even the
simpler gold-silver mixtures; and when several elements are present,
as in the Stock Pot, the possible variations become even more numer
ous. Now, there are six platinum-group metals, and wastes contain
ing them normally have other metals in addition gold, silver,
copper, nickel, tin, iron, and what not. A listing of all the combina
tions and permutations in which this baker's dozen of elements
might associate themselves, with sometimes one, sometimes another
as the main constituent, would run into the thousands. And with
every change there might be a modification of the procedure.
It should now be clear why even the skilled chemist finds this
refining a laborious task, and why he must be prepared to vary his
procedure from batch to batch if necessary. Whiteley and Dietz be
gin their paper 2 w i t h t h e w o r d s : " I t is difficult to give a refining
outline that may be followed for any and all combinations of the
platinum metals; different combinations require different. methods
of attack." The same point is brought out in the paper 3 in w 1C
Johnson and Atkinson describe the processes used in the Acton
(England) refinery of the Mond Nickel Company, Ltd. They
say: "In precious metal refining, the order of operation depends
upon the ratio in which the precious metals are present. The order
described here is followed because it has been found to be the most
suitable for these by-product concentrates, although it is realized
that other sequences of operations might be more suitable for other
types of material."
We shall therefore make no attempt to give full directions for
the refining and separation of these three rarer metals. We shall
instead give a brief outline of the general principles used, and, for
the reader with a knowledge of chemistry, a list of publications
wherein he can find detailed procedures.
SOME SCHEMES OF SEPARATION
I. The Classic System

The classic system of separation of the platinum group was worked
out over sixty years ago. Sainte-Claire Deville andJ. S. Staspublished
4 in which they described the refining of the metals used to
r
make the standard bar of iridio-platinum on which the international
metric system is based. Even then the shortcomings of the ammonium
chloride separation were recognized. The following quotation from
the book THE PLATINUM METALS, by E. A. Smith, published in
London in 1924 by Sir Isaac Pitman & Sons, Ltd., is an outline of a
2 REFINING AND MELTING SOME PLATINUM METALS, by J. O. White
', by C. Johnson and R. II.
3YE PAR LA SECTION

FRANQAISE DE LA COMMISSION INTERNATIONALE DU MfeTRE A LA
CONFERENCES DES PROTOTYPES, Proces-Vcrbaux, Comite International des Poids
et Mesures (1878), Annexe No. II. By E. H. Sainte-Claire Deville and J. S. Stas.
modern adaptation of the classic system, as used in many modern re-

fineries for the treatment of crude platinum; it is summarized from
Louis Duparc's method as given in his LE PLATINE DU MONDE,
Geneva, 1920. This scheme is susceptible of dozens of variations as to
details and order of procedure. We quote:
"The process involves ten main operations, as follows:
" 1. The treatment of the mineral with aqua regia with the separation
of the insoluble osm-iridium and sand, etc.
"2. The precipitation of the platinum as the orange ammonium
chloroplatinate and subsequent calcination of the precipitate to produce
platinum sponge.
"3. The separation of the iridium as ammonium chloro-iridiate from
the mother liquor by long standing, and calcination of the precipitate to
obtain metallic iridium.
"4. The treatment of the mother liquor (after removal of the platinum
and indium as above) with iron or zinc, to separate in the metallic state
other metals of the platinum group (together with any small quantities
of platinum not previously separated) which are deposited as a fine
black precipitate called first blacks Drying and roasting of the blacks
and treatment with dilute sulphuric acid to remove copper and other
base metals.
"5. Treatment of the cleaned blacks with dilute aqua regia to obtain
a solution containing palladium (with traces of platinum, rhodium, and
iridium) and an insoluble residue containing rhodium.
"6. After separating the traces of platinum, rhodium and iridium
from the solution from 5 with ammonium chloride, the palladium is
separated by means of metallic iron as a black mass which is purified
and the metal sold as palladium sponge.
"7. The insoluble residue from 5 is fused with barium dioxide, and
the fused mass treated with aqua regia to effect solution of the rhodium
and indium. These metals are then precipitated with ammonium
chloride and finally converted into metallic sponge.
"8. The insoluble residue of osm-iridium remaining after the first
treatment of the ore with acid, is fused with zinc to effect its decom-
position. The fused product is treated with aqua regia, and the acid
liquor so obtained is submitted to distillation.
"9. The acid liquor is distilled in the presence of steam whereby
osmium distills over as osmic acid, and is collected in water, and the
osmium subsequently separated as metal.
"10. The acid liquor retains the iridium and ruthenium which are
separated with ammonium chloride, and finally converted into sponge."
The last four processes, involving fusion with barium dioxide, fusion
with zinc, and the distillation of the volatile oxides, have not been dis
cussed in this book. As we said above, they are subject to many modi
fications. Thus, potassium pyrosulphate or sodium peroxide may be
used instead of the barium peroxide. Sometimes the distillation is carried
out in an atmosphere of chlorine. And so on and so on.
2. The Method of Gilchrist and Wichers

Recognizing the importance of the platinum metals, our Bureau
of Standards has been studying them for some years. The first
task was to make sure that each metal could be determined, once it
had been separated from its sisters. The second was to find good
methods for the separation of any pair of the metals. The third was
to provide a method of analysis for the entire group, on which could
be based practical schemes of assaying and of refining.
On March 9, 1939, two of the Bureau's workers, Dr. Raleigh

Gilchrist and Dr. Edward Wichers, were awarded the Hillebrand
Prize of the Chemical Society of Washington, for their paper en
titled A NEW SYSTEM OF ANALYTICAL CHEMISTRY
FOR THE PLATINUM METALS, which had been presented
before the Ninth International Congress of Pure and Applied
Chemistry at Madrid, Spain, in 1934. In reporting this award the
magazine Science, in its issue of March 3, 1939, says:
"The work represents the first general advance in the analytical

chemistry of the platinum group since Deville and Stas devised
methods for the analysis of the platinum alloys used in fabricating
the international prototype meter and kilogram, over sixty years ago.
The new system enables the chemist to separate and determine these
six closely related metals with a degree of accuracy equal to that of
the more exact procedures used for the common metals. Related
procedures applied to refining have resulted in the preparation of
the platinum metals in an exceptionally high degree of purity/'
Gilchrist and Wichers have since prepared a second paper that

details the procedures of the prize paper. This latter paper, entitled
A PROCEDURE FOR THE SEPARATION OF THE SIX
PLATINUM METALS FROM ONE ANOTHER AND FOR
THEIR GRAVIMETRIC DETERMINATION, was published by
the Journal of the American Chemical Society (volume 57, 1935,
pages 2565 to 2573), a periodical that can be found in a majority of

American public libraries.
This new procedure is not suited to the needs of the non-chemist
refiner, but serious students of the platinum group are urged to
consult it, as it seems destined to serve as the foundation for im-
proved methods in many fields.
In this new system, in the words of the authors, "Precipitation with

ammonium chloride, fusion with pyrosulfate and extraction of metallic
residues with acids are avoided entirely. The separation of platinum
from palladium, rhodium and iridium is accomplished by controlled
hydrolytic precipitation. When a boiling solution containing these four
metals as chlorides, and in addition sodium bromate, is neutralized to
approximately pH 7, the hydrated dioxides of palladium, rhodium, and
iridium precipitate quantitatively, leaving platinum in solution. A second
precipitation of the dioxides suffices to effect complete removal of plati-
num. . . Palladium is separated from iridium and rhodium, in chloride
solution, by precipitation with dimethyl glyoxime. . . Rhodium is
separated from iridium by reducing it to metal with titanous chloride in
a boiling solution of the sulfates in diluted sulfuric acid. The metallic
rhodium is dissolved in hot sulfuric acid and precipitated a second time,
to remove iridium completely.
"When this new procedure is applied to the separation of all six of
the platinum metals from one another, osmium is the first metal to be
isolated. This is accomplished by distilling its volatile tetroxide from
a nitric acid solution. Ruthenium is next isolated by distilling its tetrox-
ide from a solution of the sulfates in diluted sulfuric acid to which sodium
bromate is added. A very suitable reagent for the absorption of osmium
tetroxide, and also of ruthenium tetroxide, is 6N hydrochloric acid
saturated with sulfur dioxide. . .
"In the recovery and determination of each of the six metals, the new
features of the procedure consist in the precipitation of osmium, of
ruthenium, and of iridium, as hydrated oxides, with final ignition to
metal. Rhodium, and also platinum, are recovered as sulfides, and
ignited to metal. Palladium is recovered as the glyoxime compound, in
which form it may be weighed, or it may be determined as metal after
ignition of the glyoxime precipitate. . . The procedure is designed for
the analysis of the platinum group, in the absence of other metals."
The last phrase of the above quotation is important "in the

absence of other metals/' That means that the procedure as there
outlined does not take care of gold or base metals. However, the
Bureau of Standards expects to give consideration to these elements
in subsequent work, and the student is urged to keep in touch with

current literature.
One piece of further research has already been published, A
NEW PROCEDURE FOR THE ANALYSIS OF DENTAL
GOLD ALLOYS, by Raleigh Gilchrist, printed as Research Paper
RP 1103, in the Journal of Research of the National Bureau of
Standards, Volume 20, June 1938. It provides for the separation
and gravimetric determination of silver, iridium, tin, gold, indium,
copper, zinc, nickel, palladium, rhodium, and platinum. Note that
osmium and ruthenium are not included. The procedure can also
be adapted to the refining of crude materials containing gold and
the platinum metals. This paper can be obtained from the Superin-
tendent of Documents, Washington, D. C, for ten cents. The non-
chemist will find it difficult to understand, but the chemist working
in this field will find it invaluable.
# * * *
(The frontispiece of this book shows Drs. Wichers and Gilchrist at
work. Dr. Gilchrist, on the right, is manipulating a series of flasks
arranged for the distillation of osmium and ruthenium. Dr. Wichers is
pouring something possibly a solution to which ammonium chloride
has been added to precipitate the orange powder from a big evaporating
dish into a Buchner filtering funnel of a type that will be described in
Chapter XX.)
A LIST OF BOOKS AND PAPERS REFERRING TO THE THREE RARER METALS
The reader should be familiar with the publications mentioned in

Chapter X, and should also refer to the more extended bibliography
in Chapter E of the Appendix. The Encyclopedia Britannica, and
all text and reference books of advanced chemistry, give interesting
and useful facts under the headings Platinum Group; Rhodium;
Ruthenium; Osmium.
• • • •
NEW PROCEDURE FOR THE ANALYSIS OF DENTAL
GOLD ALLOYS. By Raleigh Gilchrist. Research Paper RP
1103, part of the Journal of Research of the National Bureau of
Standards, vol. 20, pages 745-771, June 1938.
PLATINUM AND ALLIED METALS. SOME POINTS OF
INTEREST TO THE CHEMICAL ENGINEER. By C. John
son and R. H. Atkinson. Trans. Institute of Chemical Engineers,
London, vol. 15, pages 131-144, 1937. This describes the operations
at the Acton (England) refinery of the Mond Nickel Co., Ltd.,

where all six of the platinum metals are recovered. An abstract
of this article was printed in The Industrial Chemist (London),
pages 223-233, June 1937. A shorter abstract was printed in The
Chemical Age (London), pages 459-462, May, 1937. Extended
quotations will be found in Chapter XXII of this book.
METALS OF THE PLATINUM GROUP. ORES, RECOVERY
AND REFINING, FABRICATION AND USES, AND
PROPERTIES. By R. H. Atkinson and A. R. Raper. Journal
of the Institute of Metals (London), vol. 59, No. 2, pages 179-210,
1936. An abstract of this paper appears in The Metal Industry
(London), June 15, 1937.
A PROCEDURE FOR THE SEPARATION OF THE SIX
PLATINUM METALS FROM ONE ANOTHER AND FOR
THEIR GRAVIMETRIC DETERMINATION. By Raleigh
Gilchrist and Edward Wichers. Journal of the American Chemi-
cal Society, vol. 57, pages 2565-2573, 1935.
A NEW SYSTEM OF ANALYTICAL CHEMISTRY FOR THE
PLATINUM METALS. By Raleigh Gilchrist and Edward
Wichers. Publicaso en: Trabajos del IX Congreso Internacional
de Quimica Pura y Aplicado, Tomo VI, Quimica Analitica.
Madrid, Spain, 1934.
A METHOD FOR THE SEPARATION OF RUTHENIUM
FROM PLATINUM, PALLADIUM, RHODIUM AND
IRIDIUM. By Raleigh Gilchrist. Research Paper 654, Bureau
of Standards Journal of Research, vol. 12, pages 283-290, March
J934-
ASSAYING PLATINUM METALS. By Martin Schwitter. Metal

and Mineral Markets, Jan. 21, 1932, pages 6-10; Feb. 18, 1932, pages
6, 7; Mar. 17, 1932, pages 6-11. A reprint of this article, under
the heading REFINING PLATINUM METALS, will be found
in The Brass World, issues of Sept., Oct., Nov. and Dec, 1932. It
describes the methods used in a plant that refines jewelers' and
dentists' scrap.
A METHOD FOR THE SEPARATION AND GRAVIMETRIC
DETERMINATION OF OSMIUM. By Raleigh Gilchrist.
Research Paper 286, Bureau of Standards Journal of Research,
vol. 6, pages 421-448, March 1931.
FABRICATION OF THE PLATINUM METALS. By C. S.

Sivil. Transactions of the American Institute of Mining and
Metallurgical Engineers, Institute of Metals Division, pages 246-
261, 1931.
A GRAVIMETRIC METHOD FOR THE DETERMINATION
OF RUTHENIUM. By Raleigh Gilchrist. Research Paper 125,
Bureau of Standards Journal of Research, vol. 3, pages 993-1004,
December 1929.
MELTING, MECHANICAL WORKING AND SOME PHYSI-
CAL PROPERTIES OF RHODIUM. By Wm. H. Swanger.
Research Paper 127, Bureau of Standards Journal of Research,
vol. 3, pages 1029-1040, December 1929.
PURIFICATION OF THE SIX PLATINUM METALS. By
Edward Wichers, Raleigh Gilchrist and Wm. H. Swanger. Tech-
nical Publication No. 87 of the American Institute of Mining and
Metallurgical Engineers; 29 pages; issued with Mining and Metal-
lurgy, New York, March 1928.
THE PLATINUM METALS AND THEIR ALLOYS. By Fred-
erick E. Carter. This paper contains in tabular form many of the
physical characteristics of the six metals. Technical Publication
No. 70 of the American Institute of Mining and Metallurgical
Engineers; 24 pages; issued with Mining and Metallurgy, New
York, February 1928.
REFINING AND MELTING SOME PLATINUM METALS.
By J. O. Whiteley and C. Dietz. Osmium is not included. This
paper describes methods of refining dental and jewelry scrap and
laboratory ware, in 100 ounce lots. Technical Publication No. 84
of the American Institute of Mining and Metallurgical Engineers;
9 pages; issued with Mining and Metallurgy, New York, March
1928.
THE PLATINUM METALS. By Ernest A. Smith. A book,
published by Sir Isaac Pitman & Sons, London. 120 pages. 1924.
THE METALLURGY OF PLATINUM. By Louis duParc,
translated by Allen C. Redding and Geo. M. Nelson, with Frank
Sanborn. A translation of Chapter XV of duParc's "Le Platine
du Monde." Issued in "Mining in California," a publication of
the California State Mining Bureau, San Francisco. Vol. 18,
April 1922, pages 158-172.
DE L'ANALYSE DU PLATINE IRIDIE EMPLOYE PAR LA

SECTION FRANCAISE DE LA COMMISSION INTERNA-
TIONALE DU METRE A LA CONFERENCES DES PRO-
TOTYPES By E. H. Sainte-Claire DeVille and J. S. Stas.
Proces-verbeaux, Comite International des Poids et Mesures.
1878, Annexe No. II.
* * • •
For a discussion of the dangers connected with the heating of
osmium, see The Keystone, Philadelphia, issue of January, 1924,
page 92. Also Chemical and Metallurgical Engineering, New York,
November 10, 1924, page 743. Also The Brass World, vol. 20,
page 242, 1924.
CHAPTER XVIII
LOW GRADE WASTES
Materials discussed. Determining the method of disposal. Two extreme

cases. Part I—The shop of moderate size. Suggested procedure for
floor sweeps. Equipment for dressing floor sweeps. Procedure for wash-
barrel residues. For old crucibles. Flue dust. Paper storage old filter
papers. Polishings, grindings, old aprons, carpets, floors, etc. Old solu
tions. Fluxes for use with low grade wastes. Litharge. Lead buttons.
Cupellation. If platinum is the only precious metal. Part II The large
plant. Amalgamation not profitable. Disposal of the final fine dust.
Sampling. Questions and answers.
# # • #
In this chapter we shall discuss materials in which the proportion
of precious metal is low, such as the sweepings from the floor of a
jewelry factory.1
The literature contains frequent accounts of savings that have
been made by burning up old jewelry-shop floors, door mats, aprons,
and the like, and it is true that even lowlier materials, such as the
settlings of the water in which the jewelers wash their hands, will
yield up value under proper treatment.
However, there have been many occasions when these low-grade
materials have proven a loss, rather than a profit, through improper
handling. The literature does not go so fully into this aspect of
the case. This chapter, therefore, will attempt to help the owner to
decide whether a given material is worth salvaging or not, and
what methods to employ.
DETERMINING THE BEST METHOD OF DISPOSAL
It is not easy to draw the line between goods that it pays to

refine in your own shop, goods that should be sold to a professional
refiner and smelter, and goods which (like sea-water) should best
be let alone.
l Parts of this chapter arc based on the article JEWELERS* LOW1 GRADE
WASTES: AN ASSET OR A LIABILITY? by C. M. Hoke. The Metal Industry,
New York, November, 1934.
2IO
LOW GRADE WASTES 211
The bases of determination are: How much metal is present, and

how much will it cost to recover it? The latter depends in turn
partly upon the amount of dirt present, but mostly on the character
of the dirt, as well as upon the facilities for refining that your estab-
lishment maintains. All three factors must be considered.
TWO EXTREME CASES
Suppose one has a dollar's worth of gold, finely divided, and mixed
with a quantity of wood, as in the case of old flooring, with gold
Sweep reducer.
Courtesy of the American Gas Furnace Co.
embedded in the wood fibre, or in an old cocoa door-mat. Simple

burning (so long as care is taken to keep gold from going up
the flue) will almost solve the problem; pick out any nails with a.
magnet, then flux off the few ashes with a little borax, and your
metal is in almost salable form. Expense: perhaps $0.25 or so for the
crucible, three or four cents for fluxes and fuel, plus a little for labor.
On the other hand, suppose one starts with a dollar's worth of
gold, silver and platinum mixed with a similar bulk of old crucibles,
instead of wood. It would be necessary to (1) break up the cru
cibles, picking out and discarding the obviously worthless lumps;
then (2) grind and sieve the rest, recovering the good-sized metallic
shot; (3) mix the remaining fine dust with a large amount of flux
and treat it for a considerable time at a high heat; (4) refine the
resulting metal and the metallics from process 2 to separate the
gold, silver, and platinum from each other and from the base metals
present. The labor will be considerably more than in the first
instance, the fuel expense will be higher, more flux will be needed,
as well as a larger and more expensive crucible, and chemicals will
be needed for the separation of the pure metals. Possibly the process
will pay; but many a worker has found similar tasks unprofitable.
In other words, the presence of precious metal is only part of the
story.
These instances are extreme cases wood is easy to burn, while
old crucibles, whose materials were chosen precisely because they
do resist heat and fluxing, are very difficult to separate from the
precious metal particles.
The majority of jobs will fall somewhere between these two ex
treme cases. That is, part of the material can be removed by simple
burning, other parts will be less easy to handle.
The point to be remembered, then, is that the character of the
dirt with which your precious metal particles are mixed, as well as
its quantity, must be considered.
If daily care is taken in the collection of the low grade wastes, it
is possible to exercise control over both the character and the quan
tity of dirt that must be worked over before the precious metals can
be recovered. For example: the porter in a jewelry shop had been
told to collect and burn all papers. He obeyed literally, and col
lected and burned not only the tissues used in the factory, but also
the unopened magazines and newspapers from the office waste bas
kets the office in question being used only by clerical workers.
This meant an obvious waste of the porter's time, and in addition
the ashes from the coated magazine paper made the task of refining
difficult.
PART I. THE SHOP OF MODERATE SIZE
The reader of this book has observed that there are often several
alternative methods of refining a given type of waste. Low grade
wastes are no exception; hardly any two shops will follow exactly
the same routine, even though they may be governed by the same
principles. To describe all the possible alternatives would be ex-
hausting. The following paragraphs, then, describe typical pro-
cedures, but by no means all possible ones.
SUGGESTED PROCEDURE FOR FLOOR SWEEPS
It normally pays the average shop, having the average amount

of space and facilities, to burn, grind, and sieve its sweeps, and
to refine the metallic particles that are sieved out. It does not
pay to refine the low-grade fine dust that passes through the
sieve; this should be sold to a professional refiner and smelter.
This statement covers a large majority of cases, and every word
should be noted carefully.
The following routine is suggested: Do not let large amounts

of floor sweeps accumulate. Handle each day's lot as it comes in,
collecting it in a wire basket that rests in a metal box. This permits
fine stuff such as filings, small stones, etc., to sift through into
the box below.
Sometimes it pays to pan this stuff that goes through the
wire basket, the way miners pan earth when looking for gold dust.
If there is a possibility of the stuff containing small pearls or other
objects that can be harmed by heat, pan it without burning it;
otherwise burn it first to get rid of the grease. (See page 51.) Also
pass a magnet through it to take out iron and steel.
It takes only a little practice to get the trick of panning—shifting
the pan round and round, with water, letting the water and the
lighter particles slip over the edge. Finally the bright metal will
collect on the bottom. It may be possible to pan the stuff so well
that the metallic particles can be treated the same as filings. But
usually they will be melted up with a flux, as will be described later.
The papers, etc., in the wire basket, should be burned, and the
ashes ground and sifted to say 20 to 40 mesh. The grains that will
not pass the sieve should be examined; sometimes they should be
treated as clippings; in some cases they are plainly of no value and

should be discarded. Usually it is desirable to melt them down
into a button, using a flux.
In some cases the sifted material is then ground further, until
the bulk of it will pass the 8o-mesh screen, the metallics being again
collected and united with those previously obtained, and the re
sulting button being refined by acid methods with which you are
now familiar.
The fine dust is dampened lightly to prevent scattering, and is
then packed in barrels and sold to a refiner. The sampling process
by which its value is determined will be described shortly. The
purchaser should always be informed of the kind of material he
purchases. Tell him which precious metals are present, and so on.
This makes it easier for him to make an assay and appraisal, and
may make your net return higher.
There is some question as to the exact size or fineness to which
this material should be ground before being passed along to the
professional refiner. The process of grinding is noisy; also it calls
for considerable energy either hand power or electric current.
Therefore in a small shop the process is not apt to be carried far
because of the noise; on the other hand if power is readily available,
it is well to carry the process along further. The refiner's first task,
after receiving and weighing and drying the stuff, is to grind it
a process for which he makes a charge; therefore if the owner can
do this grinding cheaply himself, it is to his advantage to do it.
EQUIPMENT FOR DRESSING FLOOR SWEEPS
The choice of equipment is governed largely by the size of the

lots to be handled. Thus, the small shop will burn its sweepings
in a big frying pan, pouring on gasoline and burning until the
stuff will burn no more. It will grind the residue in a hand mortar,
and sieve by hand. A small gold-melting furnace will take care of
the melting of the metallics.
A somewhat larger shop will burn its sweeps in a stove made
for the purpose a simple affair, not unlike the old-fashioned wood-
burning heater. It will grind the ashes in a small tumbling barrel,
using iron balls, and will sieve by hand. The gold melting furnace
will take care of the metallics.
The larger plant, as will shortly be described, will have automatic
machinery for grinding, sieving, and mixing, and extensive equip-

ment for sampling and bagging.
The final smelting—the smelting down of the fine dust at the
plant of the professional refiner will take place in huge furnaces
of special design.
PROCEDURE FOR WASH-BARREL RESIDUES
In the past, a jewelry factory was inclined to estimate its activity

and importance by the number of barrels in which its wash-water
was settling. Now, however, many shops, especially where rents are
high, are using only one or two small barrels, plus a filter press.
The advantages and disadvantages of the two methods will be
discussed in Chapter XX.
In general wash-barrel residues are handled nearly the same as
floor sweeps; that is, dried, burned, sieved, ground, sieved again,
the metallics collected, and the fine dust that passes the last sieve
should be sold to a professional refiner. The grease that is present
tends to float off the fine particles of precious metals; for that
reason panning is not advised.
Whether you use a filter press or settling tanks, the residue col-
lected will contain soap, grease, washing compounds, rouge, pumice,
and sometimes lime, alum, excelsior or other substances added to
agglutinate and hold the flocculent particles.
PROCEDURE FOR OLD CRUCIBLES
As suggested above, these are difficult to handle, because crucible

materials are difficult to flux and resistant to heat treatment.
Dropping old crucibles into water while still hot, will make them
easy to crush, and sometimes permit the immediate recovery of
embedded lumps of metal. When breaking them up, if any lumps
are plainly of no value, they should be discarded promptly; a little
time spent at this task will pay for itself later. The remainder
should be crushed and passed through a series of sieves, in order
to recover the metallics, which sometimes are of good size and
appreciable value.
It probably will pay to pan the dust; it pans much more cleanly
than floor sweeps. If small lots are handled at regular intervals, the
task is not tiresome. After panning, the remaining low-grade dust
can be put in with other sifted dust from the floor sweeps or wash
barrel residue, to be sold.
The metallics, as usual, should be melted into a button; because
of the admixture with sand and other highly refractory particles,
this will require more flux than ordinary to obtain a clean button.
FLUE DUST
The flue or stack that carries off the fumes from the gold-melting
furnace should be cleaned out at intervals. The stuff collected is
quite fine, and may be sifted and sold with the fine dust obtained
from the final sifting of floor sweeps.
PAPER STORAGE OLD FILTER PAPERS
The paper towels used when handling precious metal solutions,

and the sawdust used to soak up solutions that have spilled, are gen
erally put in with the old filter papers. Many people do not like
to class these with the low grade wastes, as they sometimes are of
high value. It is difficult to say whether it pays to refine them
separately or not, since individual workers differ so greatly in the
amount of metal they permit to collect in filter papers or towels.
The following suggestions, however, may be helpful.
Tear off all paper that contains no valuable material, and spread
the rest out to dry. When you have several papers, burn them in
a frying pan, pouring on gasoline. Some workers divide filter
papers into three kinds: those whose main content is trash such
as the cinders flushed out of filings; those whose main content is
silver chloride; those that are high in platinum. Those containing
much trash should go to the sweeps. Those containing platinum
can sometimes, after burning, go in with the next lot of filings.
Those containing silver chloride should, after burning, be melted
down with soda ash to give a silver button see Chapter VIII to
be refined by methods already familiar.
POLISHINGS AND GRINDINGS; OLD APRONS, CARPETS, FLOORS, ETC.
The ashes obtained from this group frequently contain much more
precious metal per pound than those obtained from floor sweeps.
The cotton waste, old buffs, thread, rags, and so on, are sometimes
difficult to burn because the fluffy particles are carried up the flue.
The procedure is practically the same as for floor sweeps the
material is burned, sifted to about 20 mesh, iron particles may be
Ovens for eliminating

water from wet materials
or solutions
Courtesy of Handy &
Harman
taken out with a magnet, the metallics are collected, and if facilities
permit, the ashes are ground again to pass an 80 to 100 mesh.
The metallics are melted into a button as usual, using one or
another of the fluxes that will shortly be described, and refined
by acid methods. The final fine dust is, as was said above, normally
much richer than the final dust from floor sweeps, and some well
equipped shops find that it pays them to refine it, melting it down
in a sand crucible, as below, to obtain a button for acid treatment.
Most shops, however, add it to the collected sweeps and sell it to a
professional refiner.
OLD SOLUTIONS
Give attention to any residue or sediment. Sometimes it is the

only valuable part. Wash it and examine the particles with a hand
lens. Utilize the tests described in Chapter IX. If the residue
seems to be of value, capture it in a filter, dry and melt it into a
button, for treatment to be determined by its nature.
FLUXES FOR USE WITH LOW GRADE WASTES
In most of the procedures just described, the rnetallics that fail

to pass through the various sieves are collected and melted into a
button. Normally this is done in a sand crucible in a gold-melting
furnace, with the addition of a suitable flux. The button obtained
will be refined according to methods already familiar; if platinum
group metals are present, see Chapter XII. If gold and silver are
the only precious metals, see Chapters V, VI, and VII.
Slow burning to obtain a concentrated ash.

Courtesy of Handy & Harman.
The word "flux" means to flow. You want to make the sand, dirt,
emery, etc., become liquid and literally flow, so that the particles
of metal (which also have become liquid because of the heat applied)
can collect freely into a button. This is the most important duty
of a flux. Sometimes other purposes are accomplished at the
same time, as will be seen later. The choice of fluxing material
will depend upon the nature of the material that is being handled,
and the objects to be obtained. The quantity to be used will depend
upon the proportion of dirt that is present and may run from one-
half to three times the weight of the material.
Thus, if the metallics are fairly clean, consisting mostly of metal,

they can be melted down with a little calcined borax (also called
borax glass) without difficulty or great expense. The greater the
proportion of dirt, sand and the like, the more flux will be needed.
Many shops will make up a "stock flux" in quantities; then when
occasion demands they can add other ingredients to it. A popular
stock flux consists of:
Soda ash (or preferably a mixture of
soda ash and pearl ash) About 5 parts
Calcined borax, powdered 3 or 4 parts
Broken glass, crushed 1 or 2 parts
These should be mixed well and kept very dry. The broken glass
is not easy to prepare, and some people do not use it.
If the melt tends to boil up and foam unduly, add some common
salt.
If the material is almost wholly sand, as from old crucibles, use
considerable stock flux, possibly two pounds of flux for each pound
of material, and add an ounce of fluorite for each pound of material.
Fluorite is especially good for fluxing sand; that means it will be
correspondingly hard on the crucible in which the operation takes
place.
This does not in the least exhaust the list of possible fluxes.
Some workers add a little flour to the stock flux. Some workers
say that ordinary lye makes a good flux, the kind sold in cans being
mixed with materials such as we have been discussing. In addition
there are several prepared fluxes on the market, which if used
according to the maker's instructions, will give good results.
LITHARGE. LEAD BUTTONS
If much carbonaceous matter is present (that includes graphite)

use a good quantity of stock flux, perhaps two pounds to each pound
of material, and add about one to three ounces of litharge for each
pound of material. Litharge is a lead oxide; when heated it gives
off oxygen which helps to burn the carbon. The metallic lead that
is released will collect and wash down the metallic particles into a
compact button.
The charge is then poured into a mold. The lead-bearing metal
sinks to the bottom, and when cool can easily be broken from the
slag. Under proper working conditions this slag will show no
metallic particles. If it should, that may mean you did not use
enough flux, or perhaps did not leave the charge in the furnace long
enough.
Litharge is also advised for dust that contains very little metallic
content, or in which the particles are extremely small in size. For
that reason it is regularly used by the refiners and smelters when
running down the low-grade dust that has been ground and sieved
according to the procedure described in this article.
The formation of lead buttons is a very common procedure in
some refineries, as well as in the assay laboratory. This book has
generally recommended alternative procedures wherever possible,
because most small establishments are not equipped to handle lead
buttons by cupellation, and acid treatment is generally tedious.
CUPELLATION
Cupellation is the process of melting an alloy containing lead

with precious metals, in the presence of air. The lead is oxidized to
litharge, which liquefies and dissolves the oxides of the base metals
and carries them into a liquid layer than can be poured or bailed off,
leaving behind a precious metal alloy of much higher value than
that with which the cupellation started. This method of concentrat
ing the valuable metals will be mentioned again in later chapters.
The production of lead fumes involves unusual hazards to the
worker without proper equipment, and for that and other reasons,
cupellation is rarely used by those to whom this book is dedicated
the jeweler, the dental technician, and the small refiner.
IF PLATINUM IS YOUR ONLY PRECIOUS METAL
If the only precious metal present in the dust or metallics is

platinum (as in the case of the factory that handles platinum goods
only) some other metal must be added to the melt to collect the
platinum particles. The heat of a gas-air furnace will not melt
platinum; even though the heat be adequate to flux the sand into
a liquid, the platinum particles will not flow together unless some
other metal, of lower melting point, is added to dissolve and collect
them. Litharge is often used for this purpose. Or, if one has on
hand some rolled-gold or gold-plated scrap, or even brass or copper,
this can be used instead. This helps to avoid the use of lead, whose
fumes are dangerous, and offers an easy way to salvage the gold.
The resulting buttons will be refined by acid methods.
PART II. THE LARGE PLANT
In Part I the problem of the average establishment was discussed,

and a procedure was suggested for handling the ordinary quantity
of floor sweeps, wash barrel wastes, old crucibles and the like. As
a matter of interest we shall now describe the procedure followed
by one of the largest bullion houses in the United States, where gold
is handled literally by the ton, and platinum by the hundred
Pulverizing mills and screens.

Courtesy of Handy & Harman.
• .. 2 It is unusually well equipped, and has singularly excellent

facilities for cost accounting; for that reason alone it is interesting
in that it has checked the economics of various methods and has
reported on its conclusions.
Briefly, it burns, grinds, and sieves its floor sweeps, collects and
refines the metallics, and sells the remaining fine low-grade dust
to a smelter. This in principle is the procedure recommended for
the much smaller establishment, and differs from it only in details.
2 The United States Assay Office at New York City.
The sweepings are first burned in a large furnace. In this plant,

incidentally, there is a Cottrell precipitator for the fumes and dust
from the several flues. The ashes are reduced in a jaw crusher
to about 1/2 to 1 inch size, then go through crushing rolls onto a
vibrating screen, the larger particles being sent back for further
crushing, while bits of iron are caught by a magnet. The metallics
that fail to pass a 20-mesh screen are melted with a flux and refined.
AMALGAMATION NOT PROFITABLE
When this plant was first built, it used a mercury-amalgamation

process as the next step. The material was mixed with water and
passed over huge trays containing mercury; baffle-plates forced all
the mud to come in contact with the mercury, which picked up most
of the gold and silver, forming an amalgam.
This amalgam was collected and refined, the mercury being used
again and again. However, it was found that this mercury treat
ment did not pay, and this step has been abandoned. (Mercury does
not pick up platinum.)
DISPOSAL OF THE FINAL FINE DUST
After the material has passed the 20-mesh screen, the first step of
the sampling process is taken. Every third shovelful is placed to
one side and saved for sampling. The rest of the dust is placed in
bags to be sold. It is interesting to note that this plant, though
unusually rich and well equipped, does not attempt to refine this
fine dust, but sells it to the highest bidder.
(There are only a few firms in the whole world that can economi
cally refine this fine dust. The job calls for furnaces of huge dimen
sions, measuring ten to twenty feet or more on a side, and a single
melt will run two or three days and involve tons of material.)
SAMPLING
Sampling is a long and painstaking process. A large plant, such

as the above, will have the necessary equipment for this task; the
average shop probably cannot afford it. But there is another reason
why the small shop does not attempt to sample the big refiner
would not accept his figure. The refiner will in effect say this:
"We have perfect confidence in your good faith; however we know
from long experience that the job of sampling calls for skill, patience,
and suitable equipment, and we shall therefore courteously insist
upon doing the sampling ourselves."
As a result, the sampling is usually done in the presence of rep-
resentatives of both buyer and seller, and usually at the plant of
the buyer. The large plant we have been discussing is unusual in
having space and equipment under its own roof for this process;
therefore does its own sampling, but always has representatives of
one or several buyers present while the work is being done.
The material (every third shovelful of the ground and sifted
sweeps) is carefully mixed and piled in symmetrical cones, then
The dried material is piled in cones

and quartered.
quartered by means of a straight-edge. One quarter will be set

aside, the rest added to the stuff already bagged. The fourth cone
is combined with the fourth cone from other piles, and is again
ground to pass a finer sieve say 80-mesh. The process of piling
and quartering is repeated until the bulk has been greatly reduced
in this case to about forty pounds.
This is divided into four final lots; one goes to an outside chemist
for assay, another is set aside as a reserve to use in case of dispute;
the rest is distributed to the refiners who wish to bid on the lot.
The one offering the highest bid will receive the entire lot, though if
two bids are very close it is customary to divide the lot between
the two, handing alternate bags to each. If the stuff is to be
shipped any distance, it is packed in air-tight drums, to prevent loss.
This process of coning and quartering is only one of the methods
that may be used. Sometimes a borer, an instrument suggesting
an enormous apple corer, is plunged into the pile and samples ex
tracted. When very small quantities of material are to be refined,
this laborious sampling is sometimes omitted, the entire lot being
melted up and itself serving as the sample.3
3 For a description of methods of sampling this and other materials, see SAMPL
ING AND EVALUATING SECONDARY NON-FERROUS METALS, by T. A.
Wright. Mining and Metallurgy, New York, February, 1928.
REFINING THE FINAL SIFTED DUST
This material consists of a large number of finely divided particles

of worthless material, through which are scattered a few particles
of precious metals. The large refineries that do this work use the
same principles that have been outlined in this chapter, but handle
enormous quantities of material at a time, the mixture being charged
into a large furnace and the process of refining going on without
interruption for several days. The dust is mixed with a stock flux,
plus fluonte and litharge. The fluonte (also called fluorspar, or
calcium fluoride) liquefies the sandy stuff, the litharge (lead oxide)
furnishes oxygen to burn up combustible materials, and lead in
which the precious metals may dissolve. Base metal oxides, and the
fluxed sandy materials, form a slag that is bailed off at intervals.
The lead in which the precious metals are collected, is finally cast
into bars or pigs for further treatment.
The slag just mentioned still contains some valuable material, and
is smelted again in a cupola furnace, to give another metallic layer
and a final slag that may now be discarded.
This process of smelting with litharge will concentrate the precious
metals considerably, the metal in a ton of sweeps now being collected
into about a hundredweight of lead.
The next step is to get rid of the lead. This is done in a cupell
furnace, the molten metal being exposed to a blast of air which
changes the lead back into litharge, while the gold, silver and
platinum remain as an alloy. The molten litharge layer will dissolve
the oxides of copper and other base metals, and will be bailed off
and used again.
The alloy of precious metals, now practically free from lead and
other base metals, is cast into slabs or granulated for further puri
fication. Lfforts are made to arrange the materials in the original
charge so that this alloy will contain at least three parts of silver
to one part of combined gold and platinum-group metals. The
slabs or granular metal now obtained may be refined by several
different methods. Sometimes they are boiled with concentrated
sulphuric acid according to the routine suggested in Chapter XII
for silver-rich alloys. Usually, especially when the silver content
is high enough, they are cast into anodes and refined by the
electrolytic methods to be described in Chapter XXII. When the
final process is completed, and the purified metals are cast into
bars or bricks ready to return again to art or industry, the circle

will have been completed.
* * * *
The present United States Treasury regulations make it incon-
venient to export gold-bearing materials. But some years ago,
before these regulations came into effect, many jewelers, and indeed
many small refiners, sent their sweeps to England, and the returns
from certain of the English refineries were such as to justify the
extra shipping expense. This should not be interpreted as a
criticism of American skill; rather it reflects differences in labor
and monetary conditions.
* # * #
Q. It is well known that gold ores carrying only a few pennyweight

of gold per ton, are being worked at a profit. Why not use mining
methods, such as cyaniding, on floor sweeps and other low grade wastes,
which usually contain several ounces of gold per hundredweight?
A. The reason is that floor sweeps contain large quantities of materials
that use up or destroy cyanide. In the effort to dissolve out the gold
in the sweeps, enormous quantities of cyanide would be needed. It is
cheaper to get the gold out by smelting the sweeps with litharge and
other fluxes. United States Bureau of Mines Report of Investigations
Number 3275, dated June, 1935, entitled STUDIES IN THE RECOV-
ERY OF GOLD AND SILVER, says: "Certain substances in ores are
harmful to cyanidation; chief among these are those that combine with
cyanide, forming cyanides, and carbon, which precipitates gold from
solution." Floor sweeps contain large quantities of carbonaceous mate-
rials, also considerable copper, brass, and other substances that form
cyanides.
Certain other methods used in mining are being used in the refining
of secondary metals, for example the devices for crushing and sieving.
Amalgamation is another mining method occasionally utilized in the
recovery of secondary gold.
* * * #
Q. Why is it necessary to grind the sweeps so fine? If the stuff is
going to be smelted down anyway, why not leave it as it is after passing
the 20 mesh?
A. Fine grinding has been proven necessary for accurate sampling.
In cases where sampling is not important, there is no reason for fine
grinding.
CHAPTER XIX
LOSSES
Metal losses in the refining room; in the shop; in the furnace ,

How much loss is permissible' ^ g ^ e s on losses in melting gold; in
melting and reworking platinum; in polishing. Losses due to dishonesty.
Losses of money, rather than metal, due to mismanagement. Difficulty
of estimating the value of wastes. Assaying. Questions and answers.
# * # *
Losses in precious metal refining are of two kinds losses of
actual metal, and losses of money through mishandling or bad
management. This chapter will discuss both kinds. The first kind
is easier to trace, being revealed merely by the comparison of the
weight of new metal with that of the finished article plus recovered
scrap. Losses through mishandling are more difficult to recognize,
but should be easier to overcome.
METAL LOSSES IN THE REFINING ROOM
The obvious losses due to splashed or spilled solutions, or broken

vessels, are common only with the beginner. But the beginner
can expect them, which is another of the reasons why he is urged
to handle small quantities at first. If he has a good supply of paper
towels, and plenty of sawdust to soak up any spilled solution, the
loss will not be considerable.
A few hours after handling solutions containing gold, the begin
ner will notice deep purple stains on his fingers. The amount of
actual gold involved is small, but even so such stains represent a
loss of metal, and the more skilled the operator, the fewer the
stains. Silver and platinum in solution produce black stains on
skin and clothing.
Spattering of hot solutions is a cause of metal loss. Using dishes
of generous size, so that the liquid does not reach too near the top,
will reduce this. In many processes flasks should be used instead
of evaporating dishes or casseroles. A funnel inverted over a
226
LOSSES 227
casserole in which a solution is being boiled, will save much spatter-

ing. For evaporation, use a steam bath or sand bath as much as
possible, and maintain the solution just below the boiling point.
Readers often ask if the precious metals boil out when solutions
containing them are heated. They do not boil out in the way that
water or alcohol does, but if the solution spatters, the loss may be
considerable. The table on which the burner stands should be
easily cleaned; soapstone and heavy linoleum are better than rough
brick. The flue should not have too strong a draft, or droplets may
be carried away.
Clean platinum filings are very heavy, and when a mixture
containing them is poured from one vessel to another, they cling
with great tenacity, and because of their neutral color may escape
attention. Gold filings are just as heavy, but their yellow color
makes them less apt to be ignored.
Old filter papers often contain more metal than is evident; the
heavy valuable filings collect in the point of the filter cone and are
concealed by the lighter and less valuable stuff on top.
Waste of reagents is as expensive as any other waste. Thus, the
careless refiner when precipitating gold, will dissolve a large quantity
of copperas in considerable hot water and dump it in, even though
the amount of gold may be very small. This means more than
the waste of copperas; it also means loss of time in handling and
filtering the extra bulk of liquid. Frequent use of the tests de-
scribed in Chapter IX will minimize losses of this kind.
METAL LOSSES IN THE SHOP
These are of many kinds, some more easily controlled than others.
It has long been known that shop losses are greater in summer than
in winter, because the men's hands are damp and the filings cling
and are carried away, also because windows are open and fans are
blowing. Modern methods of air conditioning promise much to
the jewelry factory, as they circulate fresh air through windowless
rooms, filtering out not only the dust, but also particles of precious
metal.
Losses are greater with intricate jewelry designs than with simple
ones, and with hand work than with machine work.
Smooth hard-surfaced floors, free from cracks, reduce losses.
Gratings or cocoa mats, especially on concrete floors, will keep filings
from being ground into shoes and carried away. A vacuum cleaner
is a necessity. Good lighting pays for itself in preventing the
accumulation of sweeps in dark corners. Thought in the arrange-
ment of benches and heavy equipment, to permit the nozzle of the
vacuum cleaner to reach behind them, will be repaid.
So much has been said during the last generation regarding proper
suction devices for use in polishing, that it seems superfluous to
mention them again. Most health regulations require them.
Some factories employ a laundress to wash the workmen's aprons,

towels, scrub rags, etc., and all water must pass through the wash-
barrel system. Wash-barrels can be a big source of loss. Mere
settling is never enough an adequate filtering device (see Chapter
XX) is imperative, because soap scum will float filings through
a whole train of barrels. Keep oil and grease out of the wash waters,
as oil is a very efficient flotation agent. When discarding old acid
solutions, do not pass them through the wash-barrels as they might
dissolve metallic particles therein.
LOSSES IN THE FURNACE ROOM
Platinum filings and clippings must be thoroughly clean before

being remelted. Do not attempt to burn out the dirt. You may
succeed in burning out certain kinds of trash, but platinum will be
scattered, possibly volatilized, at the same time, and the expense of
gas and oxygen will be quadrupled. Clean platinum melts quickly,
and with little tendency to leave particles of metal embedded in the
crucible.
Flue loss in the melting of gold can be appreciable, and the
presence of volatile impurities such as zinc may increase it. It is
LOSSES 229
not due to true volatilization of metal, according to T. K. Rose,1

but rather to the spurting of tiny globules which are carried away by
the flue draft. Such globules are so small, some being even less
than 0.001 mm. in diameter, that they stay suspended like dust
motes even in a still atmosphere. This should be kept in mind
when the flue is being built. One plan is to pass the gases through
a chamber (lined with smooth tilework) where their velocity can be
so reduced that most of the metallic droplets will settle. Very
large plants such as Mints use the Cottrell system, described in
Chapter XX, and make a nearly complete recovery of this flue dust.
A similar loss by spurting may occur when the molten metal is
poured.
HOW MUCH LOSS IS PERMISSIBLE?
Exact figures on the percentage of loss in various factories are not

always helpful because of the many factors, differing from shop to
shop, that are involved. However, such figures are interesting, and
we shall accordingly provide a few.
FIGURES ON LOSSES IN MELTING GOLD
T. K. Rose, in the chapter just mentioned, reports that it is the

common experience in Mints that the gross loss in melting gold-
copper coinage alloys is usually from 0.2 to 0.25 part per 1,000, and
after allowing for the amounts recovered from condensation cham-
bers and flues, and from the ground-up crucibles, sweeps, etc., the
net loss amounts to 0.1 to 0.15 part per 1,000. This is with gas-
fired and coke-fired furnaces. With an electric resistance furnace
the crucible loss was only 0.03 part per 1,000, flue loss being
eliminated. Rose also reports (page 459) that with a Cottrell
precipitator, the flue losses occurring in the refining of bullion (which
are presumably greater than in the melting of clean alloys) can be
reduced by 97 to 99 per cent.
ON LOSSES IN MELTING AND REWORKING PLATINUM
The United States Bureau of Standards reports that in the

handling of $16,000 worth of platinum, involving various kinds of
reworking, repurifying, and remelting, there was a loss of only
!THE METALLURGY OF GOLD. Seventh edition, page 66,
$200 worth of metal. These figures would be more helpful if

accompanied by more information as to the nature of the reworking.
ON LOSSES IN POLISHING
Ernest A. Smith h a s m a d e an extensive study of the nature of

2
polishing processes, and quotes figures on several kinds of articles,

the effect of different abrasives, different designs, and so on. He
finds it difficult to state what percentage of loss can be called
legitimate. He says that in the Sterling silver trade it is usually con
sidered reasonable to have a net or unrecoverable loss from all
sources polishing, refining, smelting, etc., of about 1 or 2 per cent,
part of which is the refiner's fee. He concludes: "The net losses
sustained in working gold and platinum are so variable that no

general statement can be made regarding them."
The exhaust system for removing buffing dust, lint, and so on,
should be carefully designed to avoid pockets, and ample facilities
for cleaning should be provided. John S. Isdale 3 describes h i s
experience with a section of an exhaust system which, because of
its position, was left uncleaned for four years. The section in
question consisted of about sixty feet of 24" galvanized pipe. It
was taken down carefully and the dust was collected and refined.
The charges for cleaning, transportation and refining came to $33,
and the factory received a credit of $900. In other words, $933 worth
of metal accumulated in that pipe in four years.
2 JUST HOW MUCH PRECIOUS METAL IS LOST IN POLISHING IS AN
IMPORTANT QUESTION, by Ernest A. Smith. The Manufacturing Jeweler,
1
^ " T I I E ^ E ^ O V E S Y ^ F EWC?6US METALS FROM FACTORY WASTE, by
John S. Isdale. The Brass World, New York, pp. 138-9, July, 1922.
LOSSES 231
ON LOSSES IN THE REFINING OF CRUDE PLATINUM ORE CONCENTRATES
The Acton (England) refinery of the Mond Nickel Co., Ltd.,

subsidiary of the International Nickel Co. of Canada, Ltd., handles
concentrates from the company's mines and plants all over
the world. They report 4 t h a t n o t m o r e t h a n 0 . 2 5 p e r c e n t o f t h e
platinum metals contained in their concentrates is lost, another
1.25 per cent being tied up temporarily in furnace slags that go
back for retreatment. This figure is impressive, especially as the
concentrates are very complex, but it must be recalled that the
processes involve very little contact between metals and hands, as
many steps as possible being automatic. This plant will be re-
ferred to again in Chapter XXII.
LOSSES DUE TO DISHONESTY
So far nothing has been said regarding the obvious factor of

employee dishonesty. The practice of refining individual work-
men's filings in separate dishes is a potent incentive to honesty.
To carry through each man's return, from start to finish, as a
separate job, occasions considerable overhead, and there are ways
of reducing this without reducing the effectiveness of individual
treatment. In some shops, the following plan is used: All filings,
except those of one man, are thrown together; the refiner then has
two lots to handle, a large one and a small one, and he keeps the
two separate from start to finish. Next week, all filings are com-
bined except those of a second man. And so on, no man knowing
when his turn will come. Or if a shop uses much platinum and
very little gold, each lot may be kept separate until the platinum
filings are cleaned and weighed, while all the dissolved gold is
placed in one big jar and precipitated at the same operation.
In this connection an odd psychologic fact might be mentioned.
A supply house observed that on visitors' day the general losses
were slightly larger than usual. It was not suggested that the guests
had pilfered bits of metal; rather it seemed that the workmen,
diverted by visitors, and feeling subconsciously that their responsi-
bility was somehow subtly divided, were less scrupulous in making
4 PLATINUM AND ALLIED METALS, SOME POINTS OF INTEREST TO
THE CHEMICAL ENGINEER. The Chemical Age, (England.) Page 459-62,
May 22, 1937.
their returns. Whatever the explanation, the firm finally refused

to admit visitors at any time.
MONEY LOSSES DUE TO MISMANAGEMENT IN REFINING
We shall mention only a few instances of money loss, as dif

ferent from metal loss, our object being to suggest the principle
that should be followed in minimizing them. This principle may
perhaps be compressed into this one sentence: The truly skilled re
finer knows how to refine, what to refine, and when to refine it.
For example: one loss, more serious than apparent, comes from
letting scrap metal lie idle. Back in Chapter II it was pointed
out that when scrap is sold to a professional refiner, there may be
a long interval during which the metal is earning nothing, while
it accumulates in the shop or in transit. The operator who does his
own refining can work with a smaller capital investment because
he keeps substantially all his metal in use. Thus the platinum
filings in Chapter III need stay in the refining room for only an
hour or two. In Chapter XI the bulk of the platinum is back at
the bench within twenty-four hours, often less; the gold within two
days. The process given in Chapter XII takes longer, but it is
unusual for the bulk of the metal to be tied up for as long as a week.
On the other hand, too frequent refining, involving the handling
of many small lots, runs the labor costs up. A proper balancing
of these factors characterizes the refiner of good judgment the
refiner who knows when to refine.
Another common piece of mismanagement is to spend more in
labor and chemicals than a batch is worth to spend five dollars to
recover four dollars' worth of metal. The ability to make proper
decisions in borderline cases admittedly comes only with experi
ence. The higher the grade of waste, the easier it is to figure its
value and to estimate the cost of refining it, which is another reason
why the student is advised to begin on high grade materials. After
he has completed three or four refinings, and has a good conception
of the procedure, he must establish the habit of careful weighing.
He must weigh his filings as received, weigh them after the dif
ferent steps of the refining, making permanent records of the im
purities removed, the metals recovered, and so on.
As his skill increases, less and less metal will be tied up in old
LOSSES 233
filter papers, the percentage of metal recovered will be higher,

the time and chemicals consumed will be greatly reduced, and his
refining profits will grow accordingly. He will steadily become
able to handle lower grade or more difficult materials, in less time,
as well as the simpler high grade ones.
DIFFICULTY OF ESTIMATING THE VALUE OF MATERIALS
If there were some magic whereby we could know the value of

a dish of filings simply by looking at it, and could foretell the
exact cost of refining it, our work would be easier and our profits
much higher. Fortunately the man who refines only the wastes
generated in his own shop does not need precise figures, however
much he might enjoy having them. He knows what metals are
present, and he soon will know approximately how much dirt can
be expected in each workman's filings, how much metal to look
for in each pound of wash-barrel waste, and by using the tests in
Chapter IX he should make excellent appraisals of all solutions.
ASSAYING
The situation of the professional refiner who buys from many

sources is quite different. He needs an assay for every purchase.
An assay is a different thing from the estimates mentioned above,
being an exact determination of the precious metal content of a
very small, carefully segregated sample of the material. The task
of taking the sample is often difficult, and the determination of
the contents, while utilizing many of the chemical reactions with
which the refiner is familiar, requires a very different technique.
In the questions and answers at the end of this chapter, assaying
will be discussed again. As we said, the man to whom this book
is directed, who refines only the wastes generated in his own shop,
has no need for formal assays, since his task is not to buy metals,
but to take impure metals and convert them into pure ones.
Q. I want to learn to assay. I have had considerable experience in

refining. What equipment shall I need besides the things I already
have? How long will it take me to become a good assayer?
A. The ideal way to learn assaying is to attend a qualified institution

that teaches it. This takes time. There are some excellent assayers who
never attended a college, but learned through working in a good assay
laboratory under a good supervisor. This second method usually takes
much longer than the first one, and at the end the worker may be able
to handle only such materials as happened to come into the laboratory
during the period of his apprenticeship unless, of course, he has the
extraordinary luck to have a supervisor who is at the same time a
teacher, able and willing to impart the principles of the science as well
as its procedures. Such good fortune is rare, and many a young man
has spent months in a commercial laboratory, blindly following a
monotonous routine, and emerging almost as ignorant as when he
entered .
You, however, have had some experience in refining, which means
that you know the reactions of some of the metals and chemicals. Pre
sumably you also have acquired some finger skill. With this back
ground you can learn to assay in a shorter time than otherwise, whether
you travel the academic road or the route of apprenticeship.
The assayer needs more than a mere knowledge of the procedures,
which can be picked up from any textbook. He needs steady hands
that can manipulate minute quantities without fumbling; he needs a
knowledge of the chemistry of the materials and fluxes; and he also
needs a high degree of integrity, a rare mental honesty that will never
permit him to "doctor" a weighing no matter how unimportant, or to
color a result even when such hypocrisy will deceive only himself.
As to equipment: this is described fully in the catalogues of the
chemical supply houses, some of whom maintain special assay depart-
ments. A good assay balance is the first essential, and one of the most
expensive. The choice of furnace and other equipment will be gov
erned by the metals you plan to determine and the number of de
terminations you plan to run at the same time.
Assaying jewelers' and dentists' wastes for gold and silver is not dif
ficult. The presence of metals of the platinum group complicates the
assay, each additional element increasing the complexity sharply.
Your own immediate needs probably could best be solved (and most
economically), if you would engage an assayer to tutor you, preferably
in your own shop, in the technique of assaying your kind of materials.
Write to the head of the chemistry department of the college near you,
explaining your problem fully. With your experience you may well
find that a few hours of tutoring, spread over a few weeks, will give
you a valuable accomplishment at a moderate expense.
We suggest that you go to a college or university, rather than try to
hire some "practical" man from another jewelry shop, because the "prac
LOSSES 235
tical" man is often unable to impart his knowledge, while the university
man's business is to do just that. The university man may seem to take
longer in getting to the point, but he can teach you the underlying
principles that will enable you to go further. If possible, visit the uni-
versity laboratory and inspect the equipment. Also, go to the library
and look over the books on this subject.
CHAPTER XX
MORE ABOUT EQUIPMENT
Where to buy equipment. How to order. When handling large quan

tities. The Buchner funnel. Equipment used by the professional re
finery. New ideas in chemical apparatus. More about ventilation.
Crucibles. Cleanliness in the refinery. Wash barrels and filter-presses.
The Cottrell precipitator.
• • • #
Any effort to describe too strictly the equipment for refining
precious metal wastes is foredoomed to failure. First, because no
two establishments have exactly the same problem. Second, because
no two operators choose exactly the same apparatus, even when
handling similar materials. In earlier chapters, therefore, our lists
were suggestive, not restrictive. Fortunately the dealers provide
good catalogues, well illustrated, that are freely available to all users.
WHERE TO BUY EQUIPMENT
In Chapter D of the Appendix will be found a short list of dealers

in equipment, furnishings, and supplies. Other names will be
found in the telephone classified directories of all cities, as well as
in the trade registers such as the Chemical Engineering Catalogue,
Sweet's Catalogue, Thomas' Register, and MacRae's Blue Book,
which are found in nearly all public libraries. Your trade papers
also offer suggestions. There are several jewelry magazines, and
many magazines devoted to dentistry. The advertising managers
of these magazines will gladly assist you to get in touch with
manufacturers.
HOW TO ORDER
When asking for prices or catalogues, and especially when placing

orders, be as specific as possible. Thus, do not say "Send me price
list of your furnaces" to a dealer who sells several dozen models,
some burning oil, others using coke, coal, city gas, natural gas,
liquefied tank gas, or electricity; some being small enough to be
236
MORE ABOUT EQUIPMENT 237
lifted in one hand, others being as big as a bungalow. Instead, say

"Please send me descriptions and prices of furnaces that will melt
about 50 ounces of gold, and that will burn natural gas of 600
B.T.U." or whatever your needs are. When enquiring for equip
ment that uses electric current, say whether yours is A.C. or D.C.,
and mention the voltage and cycle. Or, if the equipment is some
thing with which you are not familiar, do not hesitate to tell the
dealer frankly that you are not sure which type of fan, or filter-pump,
or furnace (for example) will do your work, then describe your
situation as completely as possible, and ask his advice. This will
save you considerable time in the end, and probably money and
annoyance as well.
WHEN HANDLING LARGE QUANTITIES
The equipment described in earlier chapters was designed primarily

for the beginner. When large quantities of solution are handled,
larger vessels and more of them are required. Where a small shop
uses one or two casseroles, the big shop will use six or a dozen;
where the small shop has a 5-gallon Stock Pot, the big shop will
need a 20-gallon jar. As the quantity of solution increases, the task
of pouring and filtering it becomes more tedious. Syphons of rub
ber or of bent glass are helpful, but care must be used in starting
them. One way is to use the pump that will be described shortly.
Another way is to fill the syphon with plain water, closing one end
with the finger, the other end with a little clamp that can be re
leased when you are ready for the syphon to flow.
THE BUCHNER FUNNEL
For rapid filtering there is no more popular device than the

Buchner funnel. The sketch shows the arrangement. The Buchner
funnel contains a flat perforated plate, on which a filter paper is
laid. The paper should not be larger than the plate, and must lie
flat. The flask is of heavy glass, strong enough to withstand at
mospheric pressure when the pump has produced a partial vacuum
within it. Suction may be produced by a vacuum pump of any
type, but the little device that screws onto any faucet is the most
popular one. Most workers put in an empty "safety" bottle (shown
in dotted outline) to keep water from the faucet from being acci
dentally sucked back into the main solution.
The flask shown, Eimer & Amend No. 23072, costs $2.40 in the
2000 cc. size. The Buchner funnel, No. 23082, five inches across,
costs $3.60; the six-inch size is $4.20. A rubber stopper with a
large hole makes a tight seal between flask and funnel. The
pump, No. 30130, with coupling No. 30138, 1/4 inch pipe size,
together cost $1.55. Rubber tubing with extra heavy wall, Eimer &
Amend No. 30710, 1/4 ID., costs about $1.00.
Arrangement for
rapid filtering.
Longitudinal section of one type of vacuum 11
Coupling to attach pump to

faucet.
Filter papers are made in many qualities and thicknesses, as well

as in many sizes. Ask your dealer for samples of the different
qualities, telling him that you are precipitating gold, platinum am
monium-chlonde, silver chloride, and the like. He will help you
select a paper whose texture is close enough to capture your precipi
tates, without being so thick as to make filtration unduly slow.
Dampen the filter paper, start the pump and make sure it is
pulling well, and practice first with a little plain water containing
a handful of sand or sawdust. Notice that the mixture must be
poured steadily. You will soon acquire the knack of pouring at
the right speed. As the last drop of liquid goes through, disconnect
at point A. Too strong a suction may break the filter paper, espe
cially if it is not properly fitted.
The heavy-walled flask must not be heated strongly or it will

crack. When handled skilfully, this set-up is quick and satisfac
tory. When badly handled the paper breaks, or the water from
the faucet will gush back into the flask. That is why the beginner
is urged to practice first with some valueless mixture before working
with a valuable solution.
The frontispiece of this book shows a Buchner funnel and large
filter flask, in operation.
EQUIPMENT USED BY THE PROFESSIONAL REFINERY
In Chapter XXII of this book, the methods and the equipment

used by the very large professional refineries are described in detail.
NEW IDEAS IN CHEMICAL APPARATUS
The manufacturers of scientific supplies are just as progressive

as the makers of automobiles, jewelry, or any other line, and
new designs, new materials, and improvements of all kinds are
being developed every season. Most of the supply houses publish
periodical catalogues of new articles, and the wise refiner will adapt
whatever fits his needs and cuts down his operating expenses. For
example, within recent years the glass and ceramics industries to
mention only two have made great strides. Glass and stoneware
are now being fabricated into forms that were literally impossible
a generation ago. Glass cylinders of large size, so well annealed
as to be highly resistant to thermal extremes or mechanical shock,
are now available at low prices. Stoneware,1 v a t s , j a r s , c e l l s u s e d
in electrolytic work, and the like, can be bought in shapes weighing
up to a ton, and when special resistance to acids, or to alkalies,
or to thermal change, or to abrasion, is desired, it can be obtained.
1 NEW STONEWARE IN THE ELECTROCHEMICAL FIELD. By Percy C,
Kingsbury. Transactions of the Electrochemical Society, pages 303-310, 1939.
Fused quartz vessels can serve several purposes in the refinery.

Fused quartz differs from porcelain in several respects, notably in
its extraordinary resistance to thermal changes. That is, a quartz
dish can be heated to redness, then plunged into cold water, without
danger of cracking. Accordingly when valuable solutions are
being heated, quartz vessels are preferred in spite of their high cost
Quartz crucibles are used when burning the colored powders of
Purifying radium, the most precious of all metals. A battery of 12-inch Vitreosil
(fused quartz) evaporating dishes, in use at the Port Hope, Ontario, refinery of the
Eldorado Gold Mines, Ltd. Solutions containing radium-barium chloride are being
concentrated.
Courtesy of The Thermal Syndicate, Ltd.
platinum or palladium, and quartz dishes or bowls are advised

when aqua regia solutions are to be boiled or evaporated.
Within the last dozen years we have seen an impressive array of

new paints and paint-like materials, many of them of synthetic com-
pounds undreamed of in the past. Some of these are excellent for
finishing the refinery walls, tables, floors, hoods and fume ducts.
They come in several colors and can be applied in a variety of ways.
They are described in the chemical engineering magazines and
catalogues, and are sold by the chemical supply houses, as well as
by some paint stores. The lacquers and cements used in a chromium-

plating plant will resist strong acids and are useful to the metals
refiner; these are described in magazines that cater to the electro-
plating industry.
MORE ABOUT VENTILATION
With the development of the chemical industry during the last

two decades there has been parallel development of means of
handling chemical fumes. No matter how corrosive a gas may be
and the precious-metal refiner handles some of the worst ones
there is a ventilating system that will take care of it. Sometimes
the fan-blades are made of wood or laminated acid-resisting plastics,
the motor being so housed that fumes cannot reach it. Some
times a Venturi duct is used that is, a strong current of air is
directed into a duct with a narrowed throat, to produce a partial
Venturi or injector duct.

I
1
MOTOR
HEATING L-=~™,J
SURFACE
vacuum that carries off the fumes. Sometimes, indeed, the fumes
are salvaged and converted into by-products of value. The Chemical
Engineering Catalogue describes systems suitable for use by the
refiner.
The ducts must be made of an acid-resisting material, or coated
with an acid-proof paint. Wood coated with asphalt or rubber-
base paint is economical, light in weight, and lasts surprisingly well.
Ordinary sheet iron is worthless, but some of the new special irons,
made for the purpose, have shown excellent resistance. Lead, so
useful with some acids, is worthless with nitric fumes. Earthenware
pipe (sewer pipe) when properly luted at the joints is good.
Within recent years we have seen glass made up into tubing of
large diameter ideal material for fume ducts as soon as its price
reaches competitive levels.
One of the inconveniences of an exhaust system in winter is that,
while drawing out the fumes it also draws out the warm air in the
room, causing drafts and attendant discomfort to the workers.

This must be avoided by care in designing and by the maintenance
of a generous warm-air supply.
Chemical laboratories in which many Kjeldahl determinations

are being made face much the same problem as the refinery, which
might do well to adopt some of their ideas. (A Kjeldahl determina
tion pronounced Kelldoll is a method of determining the nitrogen
content of materials such as wheat, fertilizer, etc., and involves heavy
and dangerous sulphur trioxide fumes.) Sometimes several dozen
flasks will be boiling at a time, each flask emitting stifling fumes.
In one plan of fume disposal, manufactured by the Laboratory Con
struction Company of Kansas City, Mo., a stoneware duct of some
five or six inches diameter, with a series of small openings down its
side, is supported over the row of fuming flasks, each flask close to
its individual opening. The fumes are carried through the duct by
gentle suction. The same idea could be used with aqua regia or
other fumes.
In Chapter D of the Appendix will be found the names of sev

eral additional organizations that manufacture fume hoods and
will give helpful advice on the planning of a disposal system.
Details of design that seem unimportant to the layman may spell
the difference between an expensive failure and an inexpensive
success. Thus, ducts must be large enough but not too large, of
smooth bore and free from sharp bends; provision must be made for
condensed steam; the fan must be large enough but not too large;
and so on. A frank discussion with a firm of this type will cost
nothing and pay well.
CRUCIBLES
Reasonable care in the storing, handling, and use of crucibles is

essential, both for economy and safety. Believing that it would be
of advantage to their customers, the principal crucible manufacturers
of the country have pooled their information and ideas and published
a pamphlet that is available freely to all users. The pamphlet is
entitled HOW TO CUT CRUCIBLE COSTS, and may be ob
tained from the Crucible Manufacturers Association (a non-profit
group), at 90 West Street, New York City.
CLEANLINESS IN THE REFINERY
There are several reasons why the precious-metal refinery should

he kept "hospital" clean. When solutions are left to dry on a
vessel, there is first the obvious loss of metal, second the possible
contamination of the next batch of solution, and third (if the out-
side of the vessel is soiled) there is the danger that it will crack when
next exposed to strong heat. The well-managed refinery will show
a rack of flasks and dishes that sparkle as brightly as those in any
bacteriological laboratory. This is especially important with the
articles used in testing the spot-plate and test tubes used in identi-
fying the metals and establishing their presence or absence. (See
Chapter IX.) These tests will show the presence of extremely small
quantities of metal. Naturally the spot-plate and test tubes must
be kept scrupulously clean, or you will be getting positive tests in
everything you examine, no matter how worthless it may be in
truth. Vessels should be washed immediately after being used,
and stored in a dust-free cabinet. It will be recalled that finger-prints
inside a vessel will leave an oily film that will make precipitated
gold float on the surface of the liquid, in little shining flakes.
If you cannot spare time to wash the vessels immediately after
use, leave them completely immersed in plain water until oppor-
tunity permits thorough cleansing. The spot-plate is the article
that is most apt to show persistent stains, and one on which stains
are most objectionable. Gold and silver stains can be removed by
a little cyanide solution; platinum stains by aqua regia. Plain hot
water, with or without soap, is sufficient for most requirements. A
washing compound called "Laboratory Calgonite," sold by chemical
supply houses, is recommended for use with hard water.
WASH-BARRELS AND FILTER-PRESSES
Many schemes are being used for salvaging the fine bits of precious
metal from the water used in washing hands, tools, work, work-
aprons, scrubbing brushes, and so on. The older jewelry shop
uses wooden barrels, usually three or more in a train, through which
the water passes slowly while the metallic particles are settled out
by gravity. Wooden tanks, or concrete tanks with wooden or
concrete partitions, are also used, and frequently agglutinants (such
as alum or ferrous sulphate) are added; and usually the water passes
finally through a layer of excelsior or sawdust which later is dried

and burned.
One arrangement of wash-barrels.
But within recent years the filter-press a pumping device which

forces the liquid through filters has become increasingly popular.
There are several types on the market, and descriptions will be
One arrangement of settling tanks.
found in the Chemical Engineering Catalogue and in periodicals

that cater to chemical engineers. The set-up generally consists of
one barrel in which the wash-water collects until it reaches a certain
level, at which the pump automatically operates, forcing the liquid

through a firm cloth or paper-lined filtering medium. The liquid
escapes perfectly free from all sediment.
The filter-press has several advantages over the settling barrels.
For one thing, it occupies much less room. Second, it is more sani-
tary the water is filtered clear and discharged into the sewer within
A small filter-press, with motor-driven pump, for handling 50 to TOO gallons of

jewelers' wash waters per day. Larger and smaller units are available. The unit
shown is complete except for filter cloths and the pipe from the supply barrel and
the discharge line to the sewer.
Courtesy of D. R. Sperry & Co.
a few hours, instead of standing until odors may develop. The

filter-press yields a compact, almost dry cake, much less unpleasant
to handle than the slimy mud that was collected at intervals from
the bottom of the settling barrels.
The primary advantage of the filter-press, however, is that it
really removes all the metallic particles, a thing that the old settling
tanks rarely did. It has been proven repeatedly that particles of
platinum and gold will literally float through long series of settling
tanks, being carried over by soap scum and grease films. One
enterprising janitor in a New York jewelry building installed a trap
in the basement and captured sometimes as much as seventy-five
dollars' worth of metal a day, in spite of the fact that every jeweler
in the building had a settling-tank system.
When a filter-press is used, it is rarely necessary to add alum

or other agglutinants (except when metal is known to be in solution
in the wash water). The first cost of a small filter-press is only a
little more than that of a train of good settling barrels, and in most
localities the cost of power to run the pump is so low as to be
amply covered by the additional metal saving. Labor cost is low,
and as we said above, there is a saving in floor space.
THE COTTRELL PRECIPITATOR
It has long been known that when gold or silver is melted in

the ordinary furnace, fine particles are carried up the flue and
may be lost. Various schemes for salvaging this dust have been
employed, including the passage of the dust-laden air through long
chambers in which the particles were given an opportunity to
settle. Another device is a series of cloth bags. When large quanti
ties of gas are evolved, this becomes quite a problem.
The Cottrell precipitation process, widely used to capture smoke,
Portland cement dust, acid mists, etc., is also employed by most
of the Mints and larger refineries. The equipment is expensive and
bulky, and so far the method has not been adapted to the needs of
the small establishment. In Sir T. K. Rose's book THE METAL
LURGY OF GOLD, page 459, the process is described as follows:
"The fumes, preferably at a temperature of 90° C, are conducted
into a chamber in which are suspended a number of tubes, down the
centers of which are a corresponding number of wires with cross-
shaped sections or roughened surfaces, and weighted at the ends to
keep them taut. Plates may replace the tubes, in which case two
wires are placed opposite to each 4 or 6 inch plate. A high-voltage
unidirectional current passes across the gap between the wires (posi
tive electrodes) and the tubes or plates (negative electrodes). The
particles of gold and other materials in the intervening space are
thereby given a positive charge of electricity. They then flow to
the negative electrode where their charge is neutralized and they

may accumulate until shaken or washed off periodically.
"The voltage used is usually between 50,000 and 100,000, and
the intensity is augmented by the peculiarly roughened shapes of
the anodes. It is stated that 97 to 98 per cent of the gold and 98 to 99
per cent of the silver passing away as fume is thus recoverable."
For a description of the equipment and procedure used by one
large plant to recover the precious metals from furnace fumes, see
RECOVERY OF PRECIOUS METALS FROM ELEC-
TROLYTIC COPPER REFINING at the Canadian Copper Re-
fineries Plant, Montreal East, Quebec, by C. W. Clark and A. A.
Heimrod. This paper, published in the 1932 Transactions of the
Electrochemical Society (New York), tells how the fumes are
first led through large chambers to permit cooling and the settling
of the larger particles, then through towers in which they encounter
a shower of water, finally through the Cottrell electric precipitator.
Interesting statistics on the efficiency of the system, which is high,
and the power consumption, are included. Descriptions of some
other installations will be found in Chapter XXII of this book.
CHAPTER XXI
PROCESSES USED BY THE PROFESSIONAL

REFINERS
Materials handled by the professional refiners. Virgin metals. By

product metals. Secondary metals. The Miller chlorine process. The
various electrolytic methods. Possible variations. Advantages and dis
advantages of the electrolytic methods. Can professional methods be
adapted to small-scale operation ?
• • # #
In this chapter we shall describe the materials handled by the

professional refiners, and the processes used by them, some of which
have not been described or even mentioned in earlier parts of this
book. In Chapter XXII a few typical large establishments will be
described in detail, and literature in which the serious student can
find further information will be mentioned.
Occasionally a jeweler or dental mechanic who plans to refine the
wastes accumulated in his own shop will read up on the processes
used by the big refiners, with the idea of using them on his own
materials. This idea is only rarely feasible, for reasons that will
soon become clear, but the reader's interest in the big plant is
certainly legitimate. Hence these two chapters.
The materials refined by the big plants are of many kinds, and the
procedures will differ with the kind and quantity of metals and
impurities that are present, which are rarely alike in any two
plants. For our purposes it is convenient to divide these materials
roughly into three groups virgin metals, by-product metals, and
secondary metals, and to discuss them under those headings.
The various national Mints, though established originally to manu
facture coinage and maintain a market for the gold mines of the
nation, also have occasion to refine many sorts of materials, both
virgin and secondary, and operate well-conducted plants, using
processes similar to those of the commercial refineries.
248
PROCESSES USED BY PROFESSIONAL REFINERS 249
VIRGIN METALS
Gold, silver, and the platinum group metals are found native in
many forms, and contaminated with a wide variety of impurities,
both metallic and non-metallic. Nuggets and dust are found in
river beds, loosely mixed with sand or gravel from which they are
separated by panning, by gravity separation, or by other methods.
As men work out these alluvial deposits, they turn to the various
hard metal-bearing materials met with in veins, crushing the rock
and concentrating the metal particles by one technique or another.
The whole story of the extraction of precious metals from their
• 1r .__i and fascinating, but has no place in this book. For
ores is loiij^
tunately the story has been well told in the literature, and the stu
dent can find it in almost any reference library. The Encyclopedia
Bntannica in its articles on Gold, Silver, and Platinum, offers many
suggestions, and others will be found in the bibliography at the
end of this book.
It is the practice in most mines, after separating the precious
metal from the sand or rock, to melt it into bars and ship it to a
refinery, which sometimes is close by, but may be thousands of
miles away. It is not until then that our interest in it begins.
The gold bars, as well as the nuggets and dust from natural
deposits, are always impure, rarely containing as much as 90% gold.
Such gold is usually brittle, and in modern practice, whether brittle
or not, is always refined before being used. The ancients refined
gold by what they called "cementation" plates of gold were stacked
in an earthenware pot, and surrounded and separated by powdered
porous stone or brickdust, mixed with common salt and sulphate of
iron. The pot was co\ered and heated strongly, just short of the
melting point of the gold. Silver and most other impurities were
converted into chlorides, which oozed out and were absorbed by the
brickdust.
During the 16th century, nitric acid was used in refining gold.
The gold was melted with three times its weight of silver (inquarta
tion), granulated in water, and boiled in nitric acid, practically as
described in Chapter VI. The silver was recovered from solution
TK Roses authoritative book THE MFTALLURGY OF GOLD devotes 319 of
its 541 pages to the various general methods of extracting gold from its ores in
cluding panning, hydraulic separation, crushing, amalgarmtion, grinding, gravity
concentration, flotation, dry crushing, roasting, and the several cyaniding processes
and used again and again. In the 19th century, sulphuric acid,
being cheaper, largely replaced the nitric acid. All the processes
described in earlier chapters of this book are in use by the pro-
fessional refiners, at one stage or another of their routine, sometimes
on a large scale, sometimes on a small scale, depending upon cir
cumstances. In addition to the procedures already described, there
are two very important methods, used in large plants, that have not
as yet been discussed. These are the Miller chlorine process for
gold, and the several electrolytic processes for gold and silver. These
will be described briefly in this chapter, and in Chapter XXII some
typical plants that use them will be described in full detail.
BY-PRODUCT METALS
Even the highest grade native copper is never pure. Base metals
are always present, and often we find silver, gold, and platinum
group metals as well. Much copper is used in the electrical indus
tries because of its high conductivity. Conductivity is reduced by
impurities, even when present only as traces, and so the demand
is for copper of very high purity. The electrolytic refining of copper,
of which more will be said shortly, produces copper of high purity
and high conductivity, and at the same time permits the accumula
tion of the silver, gold, and platinum-group metals as by-products.
The story of electrolytic copper refining2 1S Mly as l o n g a n d f a s
cinating as that of the extraction of gold from its ores, but only a
small part of it has a place in this volume, namely the recovery of
the precious metal ingredients. The principles of the process will
be discussed briefly in this chapter, and in Chapter XXII a plant
that carries it on will be described in detail.
Native silver, native nickel, and native lead also are apt to con
tain precious metals as impurities, which can be separated out and
finally refined, either by familiar acid methods or by one or another
of the new methods to be described in this chapter. Considerable
quantities of platinum-group metals are obtained as by-products in
the refining of Canadian nickel.
2 See THE ELECTROPLATING and ELECTRO-REFINING OF METALS, by
Watt and Philip. 680 pages. D. Van Nostrand & Co.
SECONDARY METALS
The professional refiner with whom the jeweler or dentist is most

apt to be acquainted is the one who buys the old jewelry, dentures,
and sweeps that have been the subject of this book. These materials
are refined in many cases by precisely the methods that have been
described here. In other cases the methods that will be described
shortly the Miller chlorine method and the various electrolytic
methods are used.
THE MILLER CHLORINE PROCESS FOR GOLD
This method was first used in Australia, where acid was expen
sive, and is now employed in a number of large refineries. Small
operators such as jewelers have not as yet adopted it, though in the
opinion of this writer it would be suitable in many establishments,
both large and small. The gold, which should first be brought to
a fineness of about 700, is melted, and a stream of chlorine gas is
bubbled through it, converting silver and the base metals into
chlorides which come to the surface and can be bailed off. Platinum-
group metals cannot be removed by this process, so its usefulness is
confined to materials that are known to be free from this group.
The following description of the principles of the method is based
on T. K. Rose's THE METALLURGY OF GOLD, pages 466
et seq., of the seventh edition.
The method was first proposed by Lewis Thompson, and in 1867
F. B. Miller installed it in the Sydney Mint. Chlorine is passed
into molten gold covered with borax, and combines with all metals
present except those of the platinum group. Some chlorides volatil
ize and pass up the flue. When the gold approaches a fineness of
900, gold chloride begins to appear in the fumes that pass through
the slag. The end-point is reached when a peculiar stain, caused by
gold chloride, appears on a clay pipe-stem held in the fumes. The
remainder of the slag is then bailed off and the gold cast into ingots.
It is usually about 996 fine, the residue being mainly silver, with
0.5 or 0.6 parts per thousand of base metal, mostly copper.
A clay pot is used, as a graphite pot would reduce the silver
chloride back to silver.
The slag that is bailed off will contain about 2 percent of the
gold, mainly as small crystals; the slag is therefore melted again
with some soda ash, to reduce some of the silver chloride to silver,
which on settling carries the gold crystals down with it. This alloy
then passes through the process again. The remainder of the silver
chloride is freed of base-metal chlorides and reduced with iron.
A modification of the process consists of the addition of air to
the chlorine gas passed into the molten gold, so that part of the
metal is oxidized rather than chloridized. This modification was
introduced at the Perth (Australia) Mint in 1917.
The chlorine gas is usually bought compressed in iron cylinders,
but in isolated regions where tanked chlorine is not available it
must be generated on the spot from manganese chloride and hydro-
chloric acid an unhandy process which is avoided whenever pos-
sible.
The chlorine method is especially suitable for gold over 700 fine
with the remainder mostly silver. Gold of lower quality can be
treated, but the time of treatment is longer and the cost greater.
Bullion of suitable composition is obtained when possible by mix-
ing. If the silver is below 50 parts per 1000 there is some difficulty
in removing base metals.
Base metals differ considerably in their behavior. The chlorine
attacks the different metals and non-metals in an order believed
to be something like the following: zinc, iron, antimony, tin,
arsenic, copper, lead, bismuth, silver, tellurium, selenium, gold, with
the platinum metals somewhere near gold.
It is certain that dense fumes of the volatile chlorides of zinc, iron
and antimony are given off in the early stages. Lead is said to be
erratic in its behavior, the last traces being difficult to remove and
occasionally remaining in the finished product and making it so
brittle that it must be repurified, by one method or another. Tel-
lurium is the most difficult of the elements to eliminate. This is
common in some gold ores, but almost never present in secondary
gold, so would cause no difficulty to the jeweler-refiner. Platinum
metals, if present, remain with the gold.
The recovery of the silver from the silver chloride slag is simple
in principle, but has some practical difficulties; these are discussed
in Chapter XXII in the detailed description of the Homestake re
finery.
The one great problem is the fumes. Chlorine itself is one of
the most corrosive and unpleasant of gases, being used in chemical
warfare for its deadly effects. The volatile chlorides are of the same
order of unpleasantness, copper chloride being particularly distress-
ing Different kinds of condensing chambers are used water sprays,

baffles, screens, and so on. The Cottrell process appears to be far
more effective than other schemes.
One great advantage of the chlorine process is its rapidity, as it
brings 98 percent of the gold to a marketable form in a few hours.
The plant costs little, and can be expanded readily to meet growing
requirements, and the operating costs are low.
Its main disadvantage, as we said, is the fumes. Ordinary systems
are usually not able to cope with them. However, once the fume
problem is conquered, the method is easy, flexible, and economical.
A description of the equipment and procedure used in this process

by the gold refinery of the great Homestake Mine will be found in
Chapter XXII of this book.
THE VARIOUS ELECTROLYTIC METHODS
Every jeweler who has done gilding will understand the basic
principle of electrolytic refining. In all cases an anode of impure
metal is hung in a solution ("electrolyte") and is decomposed
by the electric current. As this goes on, pure metal deposits out on
a cathode. The impurities either drop down as a sludge or go
into solution in the electrolyte. There is the whole story in one
paragraph, but the practical applications of this principle have filled
many books and occupied many engineers and chemists for many
years. The method is simple in theory, but its successful application
is not easy.
There are several kinds of electrolytic refining. The reader should
understand at least three of these copper, silver, and gold, even
though his main interest may be with gold only.
(I) ELECTROLYTIC COPPER REFINING
Crude copper, after various appropriate treatments, is melted and

cast into huge anodes. Sometimes secondary metal such as scrap
copper from old dynamos, or scrap from the manufacture of cheap
gold filled or gold plated jewelry, is melted in with it. These
anodes usually run about 96 percent copper, the remainder being
precious metals with traces of many other elements zinc, nickel,
arsenic, tellurium, selenium, and often several others, depending
upon the source.
The sketch illustrates the principle only; the actual cells contain
many anodes and many cathodes, alternating. An electric current
of low voltage but high amperage is used. After a time the condi
tion suggested by the second sketch has come about, and the scrap
anode is melted in with the next batch of anodes, the cathode is
Crude copper anode,

about 96 per cent Pure copper cathode
copper. or starting sheet.
Copper sulphate electrolyte.
Copper cell at start of electrolytic refining.
ready to be melted into some marketable form, and the electrolyte

(now foul with dissolved zinc, iron, nickel, and other metals) is
treated for the recovery of such as have value. Our interest centers
upon the sludge or slime, in which are concentrated the precious
metals, mixed with more or less copper, selenium, tellurium, arsenic,
and other elements.
Pure copper depos-

Scrap anode.
ited on cathode.
Sludge containing
gold, silver, platinum Foul solution, con-
metals, copper, lead taining dissolved iron,
sulphate, selenium, tel- zinc, nickel, copper,
lurium, etc. etc.
Copper cell at end of electrolytic refining.
This sludge or mud is next washed, dried, roasted at about 700

degrees F., leached in dilute sulphuric acid to remove copper, and
then melted in a furnace with fluxes to remove selenium, tellurium,
and as much more of the base metals as possible. In the silver
refinery of the Raritan Copper Works at Perth • Amboy, New
Jersey (see Chapter XXII), the metal that remains after these treat-
ments runs about 980 parts silver, about 15 parts gold, and the re-
maining 5 parts are platinum-group metals and a few odds and ends.
This silver-gold alloy they call it dore silver is then cast into
anodes, somewhat similar to but smaller than the copper anodes,
and is ready for the electrolytic silver refinery.
(2) ELECTROLYTIC SILVER REFINING
The sketch shows the principle only; the actual cells may be
arranged quite differently. The impure silver may consist of
washed and purified slimes from a copper refinery, or may be sec-
ondary silver obtained in some way from old jewelry, old coinage,
Cathode.
Crude silver anode in
cloth bag.
Crude silver
anode in bag. Electrolyte. Cathode.
Silver cells at start of electrolytic refining. The Moebius or vertical arrangement is

shown on the left, and the horizontal Thum or Balbach-Thum arrangement is shown
on the right.
Cathode coated with

pure silver crystals. Bag containing sludge
o£ gold and platinum
metals.
Bag containing sludge Electrolyte, Cathode coated with

of gold and platinum. pure silver crystals.
Silver cells at end of electrolytic refining. The Moebius arrangement is to the left,
the Thum or Balbach-Thum arrangement is to the right.
old photographic wastes, or what not. For economical operation,

the composition of the anode must be carefully adjusted, the silver
content being kept as high as possible.
At the end of the process, pure silver has been deposited out on
the cathode, some impurities have gone into solution in the electro-
lyte, while gold and the platinum metals, if present, have dropped
away as a sludge.
Two cell arrangements are in comon use. In the Moebius method
the anodes and cathodes hang vertically, each anode being encased
in a cloth sack, and each cell containing several anodes and cathodes,
alternating. In the Thum process, also called Balbach-Thum, the
single anode is supported in a horizontal position by a wood frame,
cloth lined, the graphite cathode lying on the floor of the flat shallow
cell. Electrolyte of the same composition is used in both ar-
rangements.
The pure silver that collects at the cathode is washed well and
melted down for market. It usually runs about 999 fine. The cloth
sacks have captured the sludge of gold and platinum-group metals,
which is washed well and then treated by methods appropriate to
its composition. Thus, if platinum-group metals should be absent,
this gold may require only a boiling in sulphuric acid to make it
ready to be melted down for the market.
(3) ELECTROLYTIC GOLD REFINING
Here again the sketch shows only the principles, the actual cells
usually containing several anodes and several cathodes, alternating.
The anodes may be made up of the washed sludge from an electro-
Cathode or starting
impure gold anode. •• sheet of fine gold.
Electrolyte of gold chloride

with free hydrochloric acid.
Gold eel! at start of electrolytic refining by the Wohlwill process.
lytic silver cell such as that just described, or of virgin or secondary

gold from other sources. The anodes in this process called the
Wohlwill process frequently run as high as 980 parts gold, the
remainder being silver, platinum-group metals, and copper. Every

effort is made to keep the gold content as high as possible.
At the end of the process the cathode ("starting sheet") has become
heavy with pure deposited gold, the silver has formed a sludge of
silver chloride, while the copper and most of the platinum-group
P u r e cathode
Scrap anode. _ ~TW WYZ -
Foul electrolyte con-
tuning hydrochloric
Sludge of silver _ acid, dissolved gold,
chloride, osmium, etc, V •< i •. \7 copper, platinum, etc.
Gold cell at end of Wohlwill electrolytic refining.
metals have gone into solution in the electrolyte. The electrolyte,

which by this time is much the same thing as the aqua regia solu-
tions containing gold and platinum that were described in Chapters
XI to XVII of this book, is refined by acid methods with which
the reader is already familiar, the exact procedure depending upon
which metals are present and in what proportion.
POSSIBLE VARIATIONS
In the precious metal refinery of the Raritan Copper Works at

Perth Amboy, New Jersey, all three kinds of electrolytic cells are in
operation copper, silver, and gold. Another plant, handling con
siderable low-grade jewelry scrap, has copper cells and gold cells,
but no silver cells. It melts up the low grade gold scrap with
sufficient added copper to make it a "copper proposition," and re
fines it as described above. The sludge contains only a little silver,
but much gold and some platinum. After roasting and a treatment
with sulphuric acid to remove the copper and silver, this sludge is
cast into anodes and refined as a "gold proposition" by the Wohlwill
method.
In a third plant the plant at Acton, England the raw materials
contain much platinum and palladium, with only a little gold and
silver. Here, as will be described in Chapter XXII, the procedure
is essentially an acid process similar to that described in Chapters
XI to XVII, the gold being precipitated by copperas, and the

platinum by ammonium chloride. The gold is subsequently re-
purified in small Wohlwill cells.
Another plant, handling high grade stuff only, has no copper cells,
only silver cells and gold cells. It begins by making up silver anodes,
adding enough pure silver to run the fineness up to about 900. The
sludge from this electrolysis is mostly gold, and is treated by the
Wohlwill process. In this plant it is necessary to have considerable
fine silver on hand, and it is used over and over in making up the
first anodes.
This does not exhaust the list of possible variations. In Chap-
ter XXII other possibilities will be suggested, and it should be clear
to the reader that hardly any two establishments can be expected to
follow precisely the same routine.
ADVANTAGES AND DISADVANTAGES OF THE ELECTROLYTIC METHODS
The above summary gives principles only, and does not mention
the many precautions that must be observed in the actual practice of
electrolytic refining. When properly carried out, the methods yield
metal of high purity at a very low cost per ounce, and with rela-
tively few fumes. The installation cost is high, and the work must
be done under the supervision of highly trained personnel. How-
ever, since it is almost as easy to supervise a large plant as a small
one, the labor cost per ounce is usually low. The main disadvan-
tage of these methods is their slowness, and the fact that much
precious metal is permanently tied up in electrolyte, scrap anodes,
and starting sheets. (The Miller chlorine process, in contrast,
brings 98 percent of the gold to marketable form in a few hours.)
As compared with the acid methods given in Chapters I to XVII
of this book, the striking thing about the electrolytic methods
is that the "impure" silver and "impure" gold may run as high as
980 fine when the process begins. It is true that anodes of lower
quality are being worked, but as the quality goes down the costs
of operation go up.
If the reader is experienced in gilding or silver-plating he will
understand this necessity for a high-quality anode. He will recall
that even small impurities in his gilding bath can affect the color
of the deposited gold. The more impure the anode, the more
quickly the electrolyte becomes foul and the sooner the impurities
will contaminate the cathode. (This contamination may be either

an actual co-deposition or perhaps merely the mechanical inclusion
of small bits of metal on the rough cathode surface.)
Some impurities if present in sufficient concentration can actually
stop the process. Thus in a silver cell, an anode containing much
gold will soon coat over with an adherent layer of gold sludge that
will prevent further decomposition of the silver beneath. And if
the gold anode in a Wohlwill cell contains too much silver, a film
of silver chloride on its surface will in the same manner eventually
halt the electrolysis.
These difficulties are lessened by various expedients. The ad-
herent gold sludge can be scraped off mechanically, for instance.
And a very ingenious method is sometimes used in the Wohlwill
cell: a weak alternating current of electricity is sent through the sys-
tem, superposed upon the direct current that accomplishes the
electrolysis; this produces an electrical pulsation that dislodges the
silver chloride layer and permits the work to continue.
The electrolytic refining of gold and silver is spoken of as
involving relatively few fumes. This is true, but it most certainly
is not entirely free of fumes. Most routines call for various addi-
tional processes that involve fumes, and there is considerable fuming
in the Wohlwill process itself, enough to require an exhaust fan over
the cells. The foul electrolyte, which is a hydrochloric acid solu-
tion containing dissolved gold, platinum metals, and base metals,
must be refined by acid methods, and the platinum and palladium
sponge must be reduced (with the formation of fumes) by methods
with which the reader is already familiar.
The necessity for almost pure metal to start with is, of course,
one of the reasons why electrolytic methods are not generally suited
to the needs of the jewelry shop and dental laboratory. Thus, sup-
pose a jeweler has on hand some 14-k or 12-k scrap, mixed with
platinum; to refine it successfully by an electrolytic process he must
first raise its fineness considerably, either by melting the scrap in
with a generous quantity of fine gold, or by a preliminary treatment
by familiar acid methods, or by adding enough silver to make it a
silver proposition. All of these expedients involve so much expense
that they tend to undo the advantages attaching to the electrolytic
procedure.
In spite of its simple aspect, an electrolytic cell is not a fool-proof
device. It calls for close and intelligent supervision. For example,
if the bath becomes too highly contaminated, the deposited cathode

metal will be impure and the whole job will be to do over again.
Sometimes crystals form on the cathode and grow until they touch
the anode and cause a short circuit. If the bath is stirred too vigor
ously the sludge will churn up and be caught and included in the
cathode; if stirred too little the bath will become locally exhausted
and the operating costs will run up. And so on and so on. As
one experienced worker phrased it: "When you put in an electro
lytic system your troubles are not over; they have just begun."
CAN PROFESSIONAL METHODS BE USED IN THE SMALL SHOP?
The purpose of this book is to tell how some professional methods

those described in the first sixteen chapters can be learned and
utilized by any intelligent non-chemist who is willing to make, in
order, the acquaintance tests described herein, and to follow step by
step the procedures outlined. Twenty-five years of experience have
proven that this can be done.
In this writer's opinion, however, the electrolytic methods cannot
be operated economically by the non-chemist working alone. It is
believed, further, that present electrolytic methods are rarely suited
to small-scale operation, even under the supervision of a trained
electro-chemist. It is true that small systems have been set up
cells holding about a quart of gold chloride, for instance. In such
a cell a trained worker can produce cathode gold of as high quality
as he might desire, with mental satisfaction, no doubt, but probably
at unduly high cost, and the small cell will require almost as much
attention as an acid system handling many times the amount of
metal per day. And it is a matter of record that electrolytic systems
of considerable size have been installed and later abandoned, usually
for economic reasons.
Another "professional" method is the Miller chlorine method. In
this writer's opinion, this method might well be considered by some
of the smaller establishments. It has its hazards and its limitations,
true, but one can begin work on ,a small scale at slight expense, learn
the technique, and then expand if desired. The interested reader
will find further information on the method in Chapter XXII.
CHAPTER XXII
SOME LARGE PROFESSIONAL REFINERIES
The Irvington Smelting and Refining Works. The Miller Process at

the Homestake Mine. The Raritan Copper Works. The United States
Assay Office. The plant at Acton, England. Some other large plants
Copper Cliff; Montreal East; Old Hickory; the Royal Mint at Ottawa,
Ontario.
In this chapter we shall describe in detail a few typical large re-

fineries. The descriptions that follow are adapted or quoted from
papers that were, in most cases, written by men who are themselves
at work in these establishments. We are greatly indebted to the
publications in which the original papers appeared, for permission
to quote.
THE IRVINGTON SMELTING AND REFINING WORKS
Materials with which the reader is familiar, including floor

sweeps and jewelers' low grade wastes, are refined in this plant at
Irvington, New Jersey. The processes include treatments by blast
furnaces, converters, cupels, electrolysis, roasting, and leaching, as
well as by many of the acid processes mentioned in this volume. The
following paragraphs are adapted from an article by Adolph Breg
man entitled IRVINGTON SMELTING AND REFINING
WORKS, which appeared in The Metal Industry (New York),
vol. 25, pages 183-187, May, 1927.
The Irvington Works recover gold, silver and the platinum metals.
The copper with which these metals are normally associated is
turned out as bluestone (copper sulphate crystals).
Raw Materials
The raw materials treated are either primary, such as copper con-
centrates resulting from mining and milling operations where precious
metals are associated with the copper, or secondary, such as jewelers'
261
sweeps, copper bullions containing silver or gold, and a great variety of

waste materials that carry values.
Sampling
The sampling of fine materials is done by the coning and quartering
method. Bullions and other metallics are sampled by drilling or by
melting and granulating in water.
In the sampling room is installed an exhaust system leading to a small
Cottrell precipitator, which recovers the dust from the air exhausted.
The main purpose is to keep the room under suction, so that no dust-
laden air will find its way to the outside atmosphere or to other parts
of the plant.
Smelting Operations
Sweeps and other fine materials are charged into a pug mill in which
they are mixed with water before moving to the blast furnace. Slaked
lime is added to this mixture as a binder, when the materials are such
as would tend to cause excessive flue-dust during smelting. The water
used in the pug-mill is waste solutions from other parts of the plant,
and contains some platinum by-metals which otherwise might be lost,
or recovered only with difficulty.
Sweeps
The highest grade of sweeps, concentrates and similar materials, and
also rich copper bullions, are smelted directly in the cupels or "test"
furnaces. The lower grade bullions are charged directly into the con-
verter. The test furnaces handle also various rich products from other
operations in the plant, as noted later on.
There are three blast furnaces, two having a capacity of 30 tons each
per 24-hour day, and one of 50 tons. The latter is arranged for use in
conjunction with a small basic-lined converter for producing copper shot
from the matte delivered by the furnace.
Sweeps of medium grade and other materials adapted to blast furnace
treatment are usually smelted with lead as a carrier, in the blast furnaces,
the lead being supplied by including litharge (lead oxide) in the charge.
A typical blast furnace charge of this nature includes:—
Muddled (mixed with water) materials, such as sweeps, concentrates,
etc.
Leady matte from previous charges.
Fluxes (limestone, gypsum, pyrite cinder, or slag high in iron).
Litharge (if lead is to be used as carrier).
Return slag, as ballast.
SOME LARGE PROFESSIONAL REFINERIES 263
From the lead bullion resulting from such a charge, the precious
metals can be recovered promptly, and the litharge returned to the blast
furnace for re-use.
When running leady charges the lead bullion is tapped from the lower
of two tapping spouts in the crucible of the blast furnace, receiving the
lead and matte together in conical iron pots. The matte forms an upper
layer in the pot, solidifies quickly, and is removed with a chain link which
has been partly inserted while the matte was yet molten. The lead is
then ladled out into pigs.
Lead is not used in the blast furnace charge when the materials for
treatment would produce matte in their own right, merely by smelting
down. In such cases the furnace is run, not as a lead furnace, but as a
straight copper matting furnace; the precious metal values are collected
in the resulting copper matte, and eventually are found in the copper shot
produced therefrom in the converter. When such materials are being
smelted, it is found economical to include in the charge low grade sweeps
or other materials which of themselves would not produce matte, and
which otherwise would have to be smelted separately with lead. In this
way the expense of recovering these values from lead bullion is avoided,
and their cost of reduction is brought more nearly in range with the
low return that the smelter receives for their treatment.
When copper matte alone is produced, it is tapped from the 50-ton
furnace into an oil-heated settler or fore-hearth, from which it is tapped
at intervals and taken in a cast steel ladle to the converter, which has a
capacity of 7 tons of copper per 24-hour day. The leady matte likewise
reaches the converter eventually, after it has been used over and over
repeatedly until the quantity on hand has become too large to hold.
In the converter the sulphur and any iron and lead contained in the
matte are substantially removed, yielding a final product high in copper
and containing the gold, silver, etc., that had been in the matte. The
copper is then poured off through a launder into a pit of water, in which
it is granulated into shot form.
Of the converter slag, which is high in iron, the cleaner portions are
reserved for blast furnace flux, while the dirty or more highly metallic
portions are run down in one of the blast furnaces into the form of
200-lb. blocks which like the low grade bullions, are charged into the
converter along with molten matte.
The smelting plant includes also two oil-fired tilting furnaces, which
can be used to cast anodes or to remelt metal for producing shot.
Air is supplied to the blast furnaces and the cupel furnaces by a turbo
blower, and two positive-pressure blowers are held in reserve. The warm
water from the blast furnace jackets is held to maximum temperature and
minimum volume by thermo-static control, at the furnace, and is after
wards pumped to boilers and other parts of the plant for use where
required, thus utilizing both the water and the heat. Thermostatic con
trol and recorders of various kinds arc in liberal use throughout the entire
works.
Bluestone Plant
The copper shot made by the converter is charged into lead-lined
tanks, in which it is treated with steam, air, and sulphuric acid solution,
whereby the copper is taken up as sulphate in solution, leaving behind
the precious metal slime. The solution when heavily charged with
copper sulphate is passed through filter-presses to remove any sediment,
and then is run into crystallizers made of concrete with acid-proof brick
for lining, or wood with lead lining. In them are hung lead straps, and
when the crystals have deposited upon these straps and upon the sides
of the tank, the mother liquor is siphoned off and the crystals are removed
and placed in containers.
The bluestone is then washed and dried in a centrifuge, after which
it passes through a trommel screen for sizing into large and small crystals
before barrelling. A special granular bluestone can be made by running
either of the above through a pulverizer.
Cupel Furnaces
Lead pigs or lead bullion from the blast furnaces are charged into
oil-fired cupel or "test" furnaces where they are subjected to melting and
air oxidation. Litharge is formed as a liquid floating upon the molten
lead, and is allowed to overflow into pots or molds. This litharge, being
lean in precious metals, is returned to the blast furnace as before de-
scribed. The silver bullion remains behind in the hearth of the test
furnace and when enough has accumulated the feeding of lead bullion is
discontinued, the lead remaining in the bath is driven off and the silver,
then comparatively free of lead but carrying gold and the platinum
metals, is ladled out into molds.
If the silver bullion is low in gold and platinum metals, anode molds
are used and the anodes are sent to the electrolytic silver department.
If, however, there is a large quantity of contained gold and platinum
metals, the bullion is cast into thin slabs for the parting process, in order
to expedite the recovery of the gold and also because the presence of
excessive quantities of platinum metals renders the material poorly fitted
for the usual process of silver electrolysis.
Parting Plant
The parting or separation plant consists mainly of several cast iron
kettles heated by oil burners. Here the silver bullion is treated with
boiling sulphuric acid, the fumes from which are exhausted through
hoods above the kettles and pass to an acid Cottrell installation to pre-
vent pollution of the atmosphere. The resulting silver sulphate solution
is decanted off from the residue, which consists of gold and the platinum
metals. This residue then goes to the Wohlwill electrolytic refining
department. The silver is removed from the sulphate solution by cement-
ing out on copper sheets; the cement silver then is melted, cast into
anode form and electrolyzed. This further refining by electrolysis is
necessary because of the fact that platinum by-metals are present and
show some tendency to go over with the silver sulphate solution, partly
in solution and partly in suspension. The cement silver, therefore, con-
tains sufficient traces of these metals to require the additional refining
operation.
The Electrolytic Silver Refinery
The electrolytic silver refinery has been improved and remodeled
within the past few years and its capacity increased to about 700,000
ounces of silver per month, without enlarging the installation. Silver
nitrate is used for electrolyte, and the silver anodes are wrapped in muslin
sacks to hold the slimes. Mechanical rakes, constructed and operated in a
simple and effective manner, in accordance with a patent owned by the
Company, are used to brush or break off the longer silver crystals as they
form upon the cathodes. These fall to the bottom of the cell, and at
regular intervals each cell is emptietied and cleaned. The silver crystals
taken from the cell are washed, dried in a gas-fired rotating furnace, and
after melting in crucibles are poured into bars ready for the market.
The slimes left in the muslin sacks contain gold and other precious
metals and are sent to the Wohlwill department for further refining,
in the same manner as the gold and platinum metals obtained from the
parting.
Wohlwill Gold Refining Department
The raw material received consists of "sponge gold" residues from the
parting process and slimes from the silver refining process. These are
melted, cast into anodes, and electrolyzed with direct current at about
15 volts, in an electrolyte of gold chloride solution with some free hydro-
chloric acid. The tanks are of white porcelain and are set into a steam
bath under a hood. The resulting cathodes are fine gold and are melted
and cast into bars for the market, but the slimes which consist largely of
silver, gold, and some platinum by-metals, are reserved for further treat-
ment. Platinum and palladium enter the electrolyte, from which they
are subsequently recovered.
GeneralConsiderations
The main reason for the existence of a plant such as this is not the
recovery of copper or the manufacture of bluestone, but rather the recov
ery of precious metals, gold, silver, platinum, and the platinum by
metals. Copper as such can be recovered more economically by the large
copper refineries, while bluestone is essentially a by-product which ordi
narily would prove unprofitable unless, as here, its production is in con
junction with and secondary to some other process.
THE MILLER PROCESS AT THE HOMESTAKE MINE
The refinery of the Homestake Mine at Lead, South Dakota,

handles only the gold from its own mine. No platinum group
metals are present. In an illustrated paper entitled BULLION
PARTING AT THE HOMESTAKE MINE, in Mining and
Metallurgy (New York), pages 500 to 504, December, 1935, Na-
thaniel Herz describes the equipment and procedure of the Miller
chlorine process as employed there, and also explains why the
chlorine process was chosen rather than electrolytic methods or acid
parting. Parts of this paper are quoted below.
The bullion to be treated is of two kinds. That recovered from the

ore by amalgamation contains 800 to 820 parts gold with 175 to 190
parts silver, and makes up about 60 percent of the production. The
other 40 percent has been recovered from the ore by cyanidation, and
runs 730 to 755 parts gold with 205 to 230 parts silver. We quote:
Acid parting and the Moebius electrolytic process both require a bullion
with a silver to gold ratio of 2 or 2.5 to 1. This would necessitate keep-
ing a large stock of silver for alloying, and the parting of something
like 30,000 to 35,000 oz. of alloyed bullion weekly. The acid or power
needed for these processes and the large installation for such quantities
eliminated both of them from consideration. Similarly, the Wohlwill
process was eliminated because it requires a bullion over 900 fine in
gold; that would require a stock of gold for alloying and the parting of
between 30,000 and 40,000 oz. of alloyed bullion weekly. This left the
Miller process as the one adapted to the conditions, as it is suitable for
bullion of the class under consideration, can produce refined gold suit-
able for shipment in a few hours, and may be installed in a small space.
Therefore, plans were made for equipment for chlorine parting sufficient
to refine the weekly cyanide plant output in one day and the amalgama-
tion output in two.
Melting Furnaces
The melting furnaces are of local design, stationary, built directly
against the stack, with good covers and flue connections. This is advis-
able here, because there is always a remote possibility of mercury vapor
from an imperfectly retorted lot of amalgam; furnaces without a posi-
tive flue connection might allow mercury vapor to escape into the
room unless well hooded and provided with positive draft. The fur-
naces have a cylindrical iron shell, and a sloping cast-iron cover plate
over each pair. This cover plate supports the doors, and has a horizon-
tal ledge in front to hold moulds when pouring bullion bars, or a
saddle-shaped iron block to steady the crucible when pouring bullion
into the parting crucibles. These furnaces are 30 in. in diameter out-
side, and 18 inside. The inside depth varies from about 21 in. in front
to 28 in back, owing to the sloping cover.
Crucibles
The crucibles in use are generally No. 100 graphite melting pots;
sometimes No. 50 pots are used for small lots. The space between the
furnace lining and the crucible is much greater than is customary in
tilting furnaces, but is needed for the tongs used in pouring. The fur-
naces are lined with a monolithic rammed refractory, but are of such
a size that a standard cupola block may be substituted if desired.
Parting Furnaces
The parting furnaces are also of local design, and are made to accom-
modate No. 18 French clay crucibles, which are about 111/2 in. high
and 53/4in. outside diameter, within No. 18 graphite guard pots. The
furnaces are 18 in. outside diameter and 24 in. high, with the inside
diameter and depth 11 and 21 in. respectively. They are lined with a
rammed refractory, making repairs to the bottom easy. These fur-
naces, of which there are eight, are arranged along a horizontal flue
installed along one of the 40-ft. sides of the room, and are raised above
the floor so as to bring the tops to a convenient working height, about
30 in. They are spaced 3 ft. center to center, with a portable iron
working table about 16 in. square between each pair of furnaces. The
flue connections are about 5 by 6 in. and are well below the furnace
tops, being about at the level of the tops of the crucibles. The flue is
supported on a steel stand; the gas and air lines are carried below it with
feeders coming between pairs of furnaces, each furnace having individ-
ual control. There is a short vertical stack, which connects by means
of a short suspended horizontal flue to a tile tower leading into the
large brick stack at a point about 20 ft. above the floor. This tower
has a water spray in it to wash out excess chlorine before delivering

to the outside atmosphere and to recover some volatilized silver chloride.
All flues are of firebrick except the tile tower and its base, which are
of acidproof stoneware.
Chlorine Supply
Chlorine is obtained in 100 lb. cylinders, which are kept in a small
brick house adjoining the refinery. Connections are made by standard
chlorine fittings to flexible copper tubing leading to the pipe supplying
the furnaces. This pipe is of seamless steel, carried above the front of
the flue, with duplicate control valves for each furnace. The pressure
on this line is sometimes a little high when starting a run, but it soon
drops owing to the cooling of the liquid chlorine by evaporation. To
keep the pressure up to the desired point, it is usually necessary to
connect two or three cylinders to the line. In real cold weather a
small steam radiator in the chlorine house is required to keep the
temperature up to a point that prevents freezing of the chlorine in the
cylinders.
The hydrostatic pressure of molten gold is nearly 10 lb. per foot of
depth, and as the working depth is something like 6 in., good control is
obtained with ease when the line pressure is from 20 to 40 lb. The
valves in use are 1/2-in. high-pressure bronze valves with disk and seat
of nickel alloy, and packed with asbestos. From the control valve an
asbestos-covered rubber chlorine hose connects to the clay tube admit-
ting the chlorine to the bullion being treated. This hose is held in
place by clamps so as to put no strain on the clay tube, which is about
25 in. long and 1/2 in. in diameter outside; the bore is about 3/16 in.
In operation it reaches to the bottom of the crucible. Furnaces are
covered with two fire-clay tile, allowing a small opening for the chlorine
tube to enter and for observation.
Procedure
Parting is conducted in this manner: The clay crucibles are pre-
pared by filling with a hot borax solution, which is poured out after
about five minutes. Then the crucibles are dried for several days on a
warm flue. When put into service, they are placed in the graphite
guard pots in cold furnaces, and preheated slowly to avoid cracking.
While they are being preheated, the crude bullion is melted in one or
more of the large furnaces with the addition of a small amount of
borax to flux impurities; a little sand is added to the flux when melting
cyanide bullion. This slag is partly skimmed off, using a spiral iron
skimmer; it is unnecessary to complete the skimming at this stage.
Then the parting crucibles in their guard pots are carried to the melting
furnaces and the molten bullion poured into them. The charge is
about 650 oz. per crucible, although charges of over 700 oz. have been
handled occasionally. As soon as the crucibles are replaced in the part
ing furnaces, the clay tubes are lowered to a point just above the bullion
to preheat them for a few minutes. A few ounces of borax glass is
added to each crucible; if lead is present in the bullion a little fine sand
Flaw slwpt of Flomesltike parting plant
is also added. A small stream of chlorine is now turned on and the

tubes slowly lowered to the bottoms of the crucibles. The chlorine
flow is increased gradually and is all absorbed by the bullion.
The various metals present, with the exception of gold, combine with
chlorine at rates depending upon the heats of formation of the chlorides,
with a certain amount of overlap depending upon relative concentra
tions present. Lead is volatilized rapidly; soon the white fumes of lead,
if present, are replaced by darker fumes of cuprous chloride; copper is
only partly volatilized, and the formation of liquid silver chloride starts
soon after. The early silver chloride is more cupriferous than that
formed toward the end of treatment, but if much copper is present it is
found in all the chloride and some is found in the fine gold. The
silver chloride formed occupies about 2 1/2 times the volume of the
silver from which it was produced, so it must be removed from the
crucible at intervals. This is done by bailing it out with small triangu-
lar crucibles, pouring it into shallow moulds. The chloride is allowed
to rise almost to the top of the crucible before bailing; after skimming,
more borax glass is added and the chloride again allowed to build up.
The rate of formation of silver chloride is about 1 oz. per minute during
most of the treatment; with 20 percent silver, a charge is nearly parted
in a little over two hours. Sometimes additions of small pieces of crude
bullion are made at intervals during the parting as silver is removed.
The End Point

When the amount of silver present diminishes toward the end of
parting, the chlorine is incompletely absorbed, so the rate at which it is
supplied is diminished to reduce the splashing tendency. The crucible
is covered with a slotted cover, convex downward to hold back this
splashing. When practically no more chloride seems to be forming
the charge is skimmed as completely as possible and the gas allowed to
continue through the bullion at a low rate. The end of the parting
can be judged by experienced operators by observation, but it is difficult
to describe it. A test is sometimes used, consisting of turning on more
chlorine and holding a cold clay tube in the crucible above the gold.
A certain brown stain on the tube is an indication of the end point.
This must be used with some judgment, as under some conditions there
is a rather different tint of brown produced somewhat earlier by volatili-
zation of copper. The appearance at the end is rather distinctive.
When silver is present in any quantity, the surface seems to have streaks
that appear at a higher temperature than the rest; this may be actually
so, due to the heat of formation of silver chloride. When the parting is
finished, the surface of the gold has a different luster than previously;
this may be described as a high luster, almost adamantine, with some
semblance of transparency. This may not be true transparency, but
the gold beneath the surface has a greenish cast by artificial light which
gives it this appearance. When this stage is reached, the chlorine tube
is withdrawn, the valve closed, and the gold poured at once into bars.
With ordinary care, the gold bars as shipped are about 994 in fineness.
When there is as much silver to be removed, as in the Homestake
bullion, the preliminary treatment of the crucibles with borax solution

is of considerable advantage. The use of a little borax slag, as advised
in some references to chlorine parting in other places, would be suffi
cient if little silver is present; the longer time and greater depth of
silver chloride lead to more absorption by the crucible if it is not treated
to make it less pervious. When tried without the borax treatment, it was
found that chloride passes entirely through the clay wall and is reduced
to metal by the flame; sometimes enough of this reduced metal was
found in the guard pot to contaminate the gold unless great care was
taken in pouring.
This completes the gold treatment, which is rapid and comparatively
simple; about 98 percent or more of the gold is available for shipment
in a few hours.
Treatment of the Silver Chloride

Recovery of the gold in the silver chloride is the next step in the
process. Some is present in visible shots that adhered to the skimming
crucible; some may be present in such finely-divided form as to be
invisible. It is recovered by remelting the silver chloride under a borax
cover and reducing a portion of the silver by the gradual addition of a
little soda ash. This is done in graphite crucibles; it was tried in the
larger furnaces in lots of 100 lb., but there was some loss by volatiliza-
tion of silver, and no method for its recovery. Now the operation is
conducted in No. 18 crucibles in the parting furnaces so that the water
spray can recover the volatilized silver. Experiments are under way
with heavier slag covers; perhaps the loss may be eliminated in this way.
The charge used at present is about 20 lb. of chloride, 11/2to 2 lb. of
borax glass, and 1 lb. of soda ash added in small increments after
the charge is fused. The reaction forms sodium chloride and liberates
carbon dioxide and oxygen, as silver oxide is unstable above 300°
The reduced silver, in dropping through the melted chloride, carries
all the gold down with it. The temperature must be high enough to
melt the alloy into an ingot in the bottom of the crucible; then the heat
is turned off and the charge allowed to cool to a dull red heat. The
nearly solid slag is removed, and the purified silver chloride dipped out
and poured into shallow iron moulds making four cakes each 6 by 8 by
about 1/4 in. The last of the chloride is poured into these moulds and
the button of gold-silver alloy removed and added to a subsequent lot of
crude bullion for parting. Twenty pounds of silver chloride gives
about 3 lb. of alloy containing about 20 percent gold. Thus, something
like ten to fifteen per cent of the silver is reduced and returned for a
second chloridizing. The slabs of silver chloride now contain no im
purity of consequence except copper.
As now operated, the chloride cakes are packed in "soaking vats" of

acid-proof stoneware. These receptacles are rectangular trays of 12 gal.
capacity, 6 in. deep inside, and about 20 by 24 in. They hold about
75 lb. of chloride cakes, alternated with layers of iron punchings spaced
about 11/2 in. center to center. Dilute sulfuric acid is added to start
the reaction, which is allowed to continue for two or three days. The
liquid is then removed, as it is saturated with iron salts, stopping all
action. A fresh lot of very dilute acid is added, and in a day or two
the action is complete. The silver and copper are all reduced to metal,
but not alloyed. The metal is removed, picking out the remaining iron
slugs for further use. The reduced metal is now washed with water,
partly air-dried, and melted into bars averaging over 990 fine.
When bars of higher fineness were made the copper was removed
by leaching with ferric chloride. The reduced metal, after removing
the iron slugs, was placed in a soaking vat and treated with a ferric
chloride solution made by passing chlorine into some of the ferrous
solution produced by the reduction of the silver chloride with iron. The
chlorine is absorbed by this solution readily by feeding it through an
alundum diffuser. The reduced copper in the silver is finely divided,
and most of it is found on or near the outside of the cakes, as it tends
to travel to the iron; this condition makes its solution both rapid and
complete. That is fortunate, as the process is very sensitive owing to
action on the silver, tending to form chloride again. Silver is insoluble
in this solution, but both ferric and cupric chlorides react with finely
divided silver, producing a surface layer of chloride. Therefore, the
amount of ferric solution used must not be too great, and the contact
not too long. A day is enough to dissolve the copper.
Due to the reaction with silver, most of the copper is in this solution
as cuprous chloride and not the cupric salt as in leaching copper ores.
This necessitates a wash with acid before water, or the dilution would
cause the precipitation of cuprous chloride. The washing is done by
placing the silver in shallow canvas bags in the upper part of acidproof
stoneware suction filters, which are used as gravity filters because the
charges leach freely. After washing several times with acid of increas-
ing dilution, the last of the acid is washed out with water and the
silver removed and partly air-dried. It is melted with a borax slag; if
there seems to be an appreciable amount of chloride present, it may be
reduced by adding some soda ash to the borax. If much chloride appears
after melting, the bars are not poured, but the metal is allowed to
solidify in the crucible. The slag and chloride are removed and the
silver remelted and poured. Silver has been produced by this method
that is 999 to 9991/2 fine, the impurity being largely iron.
Protection for Workmen
Some observations on the operations may not be amiss at this point.
Shallower furnaces could be used for the parting operation, but the
effect would be to increase the fumes in the furnace room. With the
depth used and a good flue draft, the fumes are slight and the operators
suffer no discomfort from them. There is an exhaust fan in the wall
above the furnaces, producing a distinct draft away from the men. The
operators wear asbestos aprons, sleeves, and mittens, as well as dark
furnace spectacles. When pouring, they also wear asbestos leggings to
guard against burns in case of spills.
The tile tower with water spray removes most of the free chlorine
that escapes into the flue; also some silver chloride is recovered by filtering
the water.
THE RARITAN COPPER WORKS
This large plant, at Perth Amboy, New Jersey, is engaged pri-

marily in the large-scale refining of copper. In Chapter XXI we
showed how impurities in the crude copper accumulate as slimes or
sludge during the corrosion of the copper anodes. These slimes con-
tain not only the precious metals, but also many other materials,
and a complete by-products refinery such as this one will market six
or eight metals gold, silver, platinum, palladium, selenium, tel-
lurium, and in some cases arsenic, antimony and nickel, as trade
conditions dictate.
In a paper entitled RECOVERY OF PRECIOUS AND SEC-
ONDARY METALS FROM ELECTROLYTIC COPPER RE-
FINING, appearing in the Transactions of the American Institute
of Mining and Metallurgical Engineers, vol. 106, pages 427 to 440
(1934), M. A. Mosher describes the equipment and procedures
employed. The following paragraphs are adapted from his paper.
The raw sludge that collects during the electrolytic corrosion of
the copper anodes is washed, roasted, leached in sulphuric acid,
washed again, and furnace-refined with fluxes to remove as much
as possible of the base metals including arsenic, antimony, selenium,
tellurium, and others. (These processes are described in detail in
Mr. Mosher's paper, but are only summarized here because they are
not of immediate interest to the worker in precious metals.)
The metal that remains after these treatments it is called dore, or
dore silver averages 980 parts silver, 15 parts gold, and the rest plati
num group metals and others. This silver-gold alloy is then ladled out
into water-cooled double-face, reversible molds to produce anodes for

the subsequent electrolytic refining. Two types of anode are cast, one
19 1/4 by 211/2 by 1/2 inch for the Thum electrolytic process, and the other
6 1/4 by 9 by 1/2 in., with a small lug at one end, for the Moebius process.
Electrolytic Parting and Refining of Silver and Gold
The standard processes employed for the electrolysis of silver-gold

anodes do not lend themselves to the elimination of any appreciable
amount of base-metal impurities, since the latter may be rendered soluble
as the anode corrodes and then be plated or cemented out at the cathode
with the silver. The anode metal is brought to the highest possible
purity therefore in the dore furnace, 995 + parts of silver plus gold, and
the electrolytic treatment becomes simply a parting process to separate
the silver from the gold and any other precious metals the anode may
contain. Two types of electrolytic equipment are commonly employed,
the Thum or horizontal process, and the Moebius or vertical process,
each having its advantages and disadvantages, which will be discussed
later.
In the Thum process the anode is placed within a wooden frame lined
with muslin and covered with a close-fitting canvas sack. This frame
in turn is placed in a heavier frame which, resting on the sides of the
electrolytic cell, is partly immersed in the electrolyte and equipped with
slats across the bottom to support the weight of the anode. This anode
frame assembly covers about one-half the solution area of the cell, leav-
ing the other half open. The cells themselves, measuring 48 by 24
by 7 in., are asphalt-lined concrete tanks. The bottoms of these tanks,
which must serve as cathode surfaces, are lined with slabs of graphite
and carbon 1/2 in. thick. The current enters the cell through a silver
contact piece which rests on the submerged anode in the tray, and is
conducted from the cell through another silver contact which rests on
the graphite slabs underneath the solution at the open end of the cell.
The silver corroded from the anode passes through the muslin and
canvas diaphragms, deposits on the carbon-slab bottom underneath the
tray, and is scraped forward periodically by hand to the open end of
the tank to make way for continued deposition. The accumulated
deposit is removed from the open end of the tank to a filter-bottom
wash car once each shift. As the anode corrodes away, another is
placed on top of it so that each anode in turn is completely electrolyzed,
leaving no scrap to be remelted.
The anode current density is 50 amp. per sq. ft. with a maximum
current per cell of 150 amp. at 3 to 3 1/2volts, the cells being arranged
in series in sections of 21 tanks. The electrolyte is a neutral silver-
copper nitrate solution analyzing 60 grams per liter silver and 40 grams
per liter copper.
In the Moebius or vertical electrolytic parting process the same solu-
tion and current density are employed, but the anodes and cathodes are
1'low ahcct of ck'Ctrotylic of precious jnetaJs at the Ranun Copper Works.
suspended vertically in a cell measuring 24 by 29 by 22 in. The anodes,

9 by 61/4by 1/2 in. with a lug for hanging, are suspended in rows of
three in canvas bags, the anode bar across the tank carrying the sack
and its frame as well as the hangers supporting the anodes themselves.
Cathode sheets of chromium iron 12 by 20 by 1/16 in. are suspended

from the busbars at the sides of the tank. The cathode surfaces arc con-
tinually scraped free of the crystal silver deposit by wooden blades
attached to an oscillating rack which rests on the framework over the
tanks and is mechanically operated. The crystal silver scraped from the
cathode sheets falls through the solution into a wooden tray, which is
suspended from the electrode assembly frame over the tanks. As the
anodes corrode away new ones are hung in their places and the scrap
lugs, which otherwise would have to be remelted, are placed in the
Thum anode trays for complete corrosion.
Each tank contains five anode groups and six cathode sheets, con
nected in parallel. The cells are connected in series in sections of six
with a maximum current density of 450 amp. at 2. 1/2 to 3 volts.
After a full producing day the entire operating equipment of a sec
tion is mechanically raised as a unit to a position 1 ft. above the cells.
The trays, which have hinged bottoms, are then emptied of the accumu
lated crystal silver deposit into filter-bottom wash cars.
The advantages of the Thum process are: (1) Mechanical and cur
rent conductor simplicity, which make for low cost of installation and
maintenance; (2) complete corrosion of the anodes, which may be of
any size, shape, or weight within the capacity of the anode tray; (3) no
necessity for shutdown while the crystal silver production is being
removed.
The disadvantages are: (1) Corrosion from but one face of the anode,
resulting in low production per unit of floor space and volume of elec
trolyte; (2) dependence on and cost of hand labor for scraping the
deposit from the cathode; (3) lack of regular stirring or circulation of
electrolyte in the cell.
For the Moebius process the advantages are: (1) Low labor cost due
to mechanical scraping of the deposit from the cathode, with the added
benefits of regular stirring of the electrolyte, a more uniform deposit,
and a current efficiency of 98 percent, which is 5 to 10 percent higher
than that obtained in the Thum process; (2) close spacing of electrodes,
together with corrosion of both faces of the anodes, resulting in high
production per unit of floor space and volume of electrolyte; (3) no
necessity for muslin diaphragms, because the anode does not rest on the
accumulated gold mud.
Its disadvantages are: (1) Mechanical and current conductor com
plexity, resulting in a high installation and maintenance cost; (2) com
plete shutdown necessary for the removal of the crystal silver production.
In both the Thum and Moebius processes the gold accumulates
around the anodes in the sacks as a black slime or mud contaminated
with particles of silver and basic copper salts. This mud is removed
from the Thum trays after every three days of operation by taking out
about 45 min. being required for the casting of the usual 25,000-oz. lot
in 1000-oz. bars. The fineness of the silver is adjusted to 999, the
standard for the fine silver market, by adding to the melt, just before
pouring, a few ounces of pure copper.
Electrolytic Refining of Gold (Wohlwill Process)

The gold sand is melted in No. 25 graphite crucibles in a cylindrical
oil-fired crucible furnace, the Quigley-type burner consuming about
8 gal. of light fuel oil per hour. The metal is cast into vertical closed
molds to produce gold anodes averaging 980 fineness and 140 troy
ounces in weight, the dimensions being 9 by 5 by % in. The electrolysis
of these anodes is accomplished by the Wohlwill process, in which from
six to eight of them are suspended in pairs from silver anode bars in
the electrolyte in a small stoneware electrolytic cell measuring 20 by 12
by 12 in. Cathode starting sheets of fine gold rolled to a thickness of
0.003 in. and having an immersed area 12 by 2 in., are suspended in
rows of four from silver cathode rods spaced between the anode bars.
A current of 450 amp. is put through the cell, giving a voltage of 1 1/2
volts and an anode current density of 125 amp. per square foot with
six anodes. The electrolyte is a gold chloride solution containing from
80 to 100 grams per liter of gold and 10 percent free hydrochloric acid.
Heat to maintain the minimum operating temperature of 140° F. is
obtained by means of a steam coil placed around the stoneware cell and
enclosed by a lead-lined wooden box, sand being packed between the
lead and the stoneware cell to give even distribution of the heat from
the coil. Circulation of the electrolyte during operation is maintained
by means of a Pohle air-lift. The anodes are corroded to 5 percent
scrap in from 16 to 18 hr., the pure gold plating out on the cathode
sheets in a rough, somewhat nodularized, crystalline deposit which
adheres firmly.
The cathodes are removed from the cell, thoroughly washed, dried
and melted down in a clean crucible from which the gold is cast into an
open mold to produce a bar weighing from 700 to 800 oz. and assaying
999.7 + parts per thousand. The anode scrap is washed, dried and
remelted with the next batch of gold sand.
Silver, the chief impurity in the gold anodes, forms silver chloride
during the electrolysis and falls to the bottom of the cell as a slime
together with varying amounts of gold. This slime is removed periodi-
cally, leached with aqua regia for the removal of gold and the produc-
tion thereby of gold chloride solution to be used in replenishing the
electrolyte, the gold content of which is gradually depleted during
electrolysis. The silver chloride remaining after this leach is returned
to the dore furnace.
Strong wash waters are used to take up electrolyte evaporation, while

dilute waters are sent to the cementing tank. Auxiliary equipment
used in this process consists of two electric drying ovens and power-
driven 6-in. rolls for the rolling down of a small bar of fine gold into
cathode starting sheets when needed.
Fine silver bars are first weighed on a springless scale and finally
weighed with careful checking over a bullion balance of 3000-oz. capacity
and 0.005-oz. sensitivity. Fine gold is weighed in the same careful
manner.
Platinum and Palladium Recovery
Any platinum and palladium present in the original raw slime is
finally concentrated in the gold anodes and during their electrolysis goes
into solution in the electrolyte, the platinum and palladium content of
which can be permitted to go as high as 60 grams per liter. The elec-
trolyte is then removed, oxidized with nitric acid, and the combined
salts of platinum and palladium precipitated by the addition of ammo-
nium chloride. Separation and eventual refining of the platinum and
palladium are carried out by the usual chemical methods, both metals
being produced in the sponge form and of 99.5 + percent purity.
Summary of Major Equipment and Plant Capacities

The slime plant, with 10 slime-leaching tanks, can handle 120 dry
tons of raw slime per month and turn out from 60 to 90 dry tons of
treated slime per month, depending on the percentage of silver it
contains.
The furnace plant, with two of the three furnaces for the refining of
slime to dore in continuous operation, can refine 150 dry tons of treated
slime per month, there being ample capacity to handle, in addition to
the output of the Raritan slime plant, any treated slime or crude dore
bullion received from other plants of the Anaconda Copper Mining
Co. The capacity production of two furnaces working on slime varies,
according to the silver content of the slime, from 1,250,000 to 2,000,000
oz. of dore anodes per month.
The capacity of the parting plant, with the three 6-cell Moebius
sections and six 21-cell Thum sections in operation, is 2,250,000 oz. of
crystal silver per month. The melting of this amount and the casting
into fine silver bars can be taken care of with one of the two retort
melting furnaces in continuous operation.
Facilities for refining gold mud are ample to handle an amount
containing 35,000 oz. of gold per month. The Wohlwill electrolytic
plant with one of its two cells in operation will produce 25,000 oz. per
month, so that its capacity is greater than that required to refine the gold
resulting from full operation of the parting plant on dore running as
high as 15 parts per thousand gold. Platinum and palladium refining
facilities are adequate for the production of 200 oz. per month of the
two metals. The selenium plant is capable of a production of 3000 lb.
of powdered selenium per month and the tellurium plant can turn out
1500 lb. of metallic tellurium per month.
THE UNITED STATES ASSAY OFFICE
The Assay Office in downtown New York has always attracted

public interest. For many years it was situated on Wall Street, next
to the historic Sub-Treasury Building, and across the street from
the Stock Exchange and the office of J. P. Morgan & Co. The spec-
tacle of a first class refinery, with its furnaces, flues, acid fumes,
and so on, operated in conjunction with the storage of millions of
dollars in bullion, surrounded by skyscrapers, was bound to capture
the imagination. In 1933 the new and much larger Assay Office and
refinery was opened at Old Slip and the East River, but the Wall
Street building, still scarred with the marks of the unsolved Wall
Street explosion, is retained for office and storage purposes.
With few exceptions, the procedures used are familiar ones.
Electrolytic methods are used for refining both the gold and the
silver. The materials that are received by the Assay Office are of
many kinds, some being already of high quality. The regulations
under which the Assay Offices purchase gold and silver are subject
to change, and the reader is advised to obtain copies of the current
rules from time to time. (See Chapter C of the Appendix.) Each
deposit is melted down as received, a "dip" sample of the molten
metal is granulated in water, and the resulting shot metal is used for
an assay. Material running less than 900 fine in gold is handled in
the silver refinery; if more than 900 fine it is a gold proposition.
Metal for the silver refinery is cast into anodes and refined much
as in the Raritan plant, a modified Moebius or vertical cell arrange-
ment being used. In making up the anodes, care must be taken
to get the right proportions of silver, gold and other metals. This
is done by sorting and combining the various lots, and by adding
extra silver when necessary. Gold may run from 200 to 400 parts,
silver constitutes the remainder. Small amounts of the platinum

metals are frequently present.
The electrolyte contains about 3 percent silver nitrate and 2 per-
cent free nitric acid, and is kept in gentle circulation by an air
pump. Muslin bags surround the anodes and catch the powder of
gold and platinum that remains as the silver is dissolved away.
The cathode silver obtained in this process, when washed and
melted, runs about 999.5. The electrolyte, as it becomes foul, is
syphoned off and the dissolved silver is cemented out on copper
plates. This cement silver is melted in with new incoming material,
as necessary, in order to build it up to the desired silver content of
500 to 600 parts, to make new anodes of proper analysis.
The gold that is left in the anode bags is next washed and melted
down with other high quality gold to form the anodes for the Wohl-
will process, which is carried on in much the same manner as in the
Raritan plant. The electrolyte contains 30 to 60 grams of gold
chloride per liter, with about 10 percent free hydrochloric acid.
The silver content of the gold anodes must not exceed 5 percent,
or the silver chloride coating that forms will soon interfere with the
electrolysis. This coating, which formerly was scraped off by hand,
is now loosened by the scheme, mentioned in Chapter XXI, in which
an alternating current is superposed on the direct current. The
gold that deposits out on the cathode, when washed and melted
down, normally runs 999.8 fine, and frequently 999.9.
Whatever platinum metals were present are now accumulated in

the electrolyte used in the Wohlwill cells. Ammonium chloride is
added, and platinum precipitates as the familiar orange powder. The
next procedure is one that has not previously been described in this
volume: an electric current is now passed through the solution to
plate out the remaining gold (in somewhat impure form) on a
small cathode, and at the same time to oxidize the dissolved pal-
ladium, changing it over into such form as to precipitate the red
palladium-ammonium-chloride salt that we encountered in Chap-
ters XIII and XIV. The electric current here accomplishes the same
effect that we produced with sodium chlorate.
Since the Assay Office is not authorized by Congress to buy

platinum metals, but merely accepts them when offered, the recovery
of these metals represents almost clear profit.
The floor sweeps and other low grade wastes are burned, crushed,
sieved, sampled, and sold, as described in Chapter XVIII of this
book.
The rinse solutions from the various operations contain much
copper and some precious metal, both in solution and as finely
divided particles. These liquids are collected in settling tanks and
all the heavy metals are cemented out with scrap iron. Most of
this cement metal is sold to a copper refiner after assay.
Both oil and gas furnaces are used at the Assay Office, and a
Cottrell precipitator captures the fumes and dust.
In the Report of the Director of the Mint for 1912, Treasury Docu-
ment No. 2671, published in 1913, there is a full description of the
procedures and equipment in use at that time. In the magazine
The Brass World for September, 1913, an illustrated article based on
the Director's Report will be found. Elry J. Wagor, in charge of
the refinery at the Denver Mint, in his chapter on gold in Vol. 2 of
Liddell's NON-FERROUS METALLURGY, gives added details
and brings the story up to 1926. During the years since then there
have been improvements in detail, and much equipment has been
added, but there have been no radical changes in the general outline
of procedure.
THE PLANT AT ACTON, ENGLAND
In Chapter XXI we remarked that gold and platinum metals are

obtained as by-products in the refining of nickel. The nickel and
copper deposits of the Sudbury, Ontario, region contain considerable
platinum group metals. These ores are refined by electrolysis, and
also by the Mond nickel carbonyl process, in both of which methods
the precious metals accumulate as crude concentrates.
The following paragraphs are adapted or quoted from a paper
entitled REFINING METALS OF THE PLATINUM GROUP,
a description of the process at the Acton (England) refinery of the
Mond Nickel Company, Ltd. This paper, by C. Johnson and R. H.
Atkinson, was read before the British Institution of Chemical En-
gineers on May 19, 1937, and an abstract of it appeared in the
Industrial Chemist (London), pages 223 to 233, issue of June, 1937.
The Acton refinery receives residues from the Clydach nickel refinery,
which operates the nickel carbonyl process, besides concentrates from the
electrolytic nickel refinery and a smaller amount of concentrates from
the electrolytic copper refinery which treats the company's blister copper.
The residues from the Clydach refinery require further concentra-
tion before the platinum can be extracted economically with aqua regia.
For this purpose they are smelted with litharge, fluxes, and charcoal
Chemical stoneware vessels are available in large sizes.
in small tilting furnaces, with basic linings, to collect the precious

metals and at the same time to slag off silica and base metals.
The principal reaction during smelting is the reduction of lead oxide
to metallic lead which acts as a collector of the precious metals. An-
other important reaction is between lead sulphate, which forms approxi-
mately 50 percent of the Clydach residue, and soda ash according to
the following equation:—
PbSO4 + N a 2 C O 3 = PbCO 3 + Na 2 S O 4 .
The lead carbonate immediately decomposes into lead oxide and
carbon dioxide. The sodium sulphate separates in the moulds as a top
slag, which is removed and leached.
Subsequent cupellation of the ingots in similar furnaces removes the
excess of lead as litharge, which is used again in further smelting
charges, and yields a precious metal alloy rich in silver, about four
times richer than the Clydach residue.
If lead alloys are cupelled to a finish, that is, until litharge ceases to
form, there is a probability of producing a cupelled metal which is not
readily attacked in the course of the subsequent refining operations,
hence the reason for partial cupellation.
This alloy is treated with boiling concentrated sulphuric acid, which
removes most of the silver and about one-third of the palladium as
sulphates. The residue contains the platinum, gold, and the rest of the
palladium in a form particularly suitable for extraction with aqua regia,
which is the next operation. From the solution of chlorides thus
obtained, the gold is precipitated as brown gold by means of ferrous
sulphate, and then the platinum as ammonium chloroplatinate by the
addition of ammonium chloride, and lastly, the palladium is precipitated
as palladosammine chloride. The only one of these operations which
calls for special comment is the precipitation of palladosammine chlo
ride. The palladium present as chloride is converted into soluble tetram
mino-palladous chloride by the addition of excess ammonia solution
according to the equation:—
PdCl2 + 4NH3 = [Pd(NH 3 ) 4 ] Cl2.
When excess of hydrochloric acid is added, the sparingly soluble yel

low compound known as palladosammine chloride is precipitated ac
cording to the equation:—
[Pd(NH 3 ) 4 ] C l 2 + 2HCI = [Pd(NH 3 ) 2 C l 2 ] + 2NH4CL
The impure platinum salt is converted into metal by ignition, dis

solved in aqua regia and reprecipitated as pure ammonium chloro
platinate, which, on ignition, yields pure platinum sponge. The im
pure palladosammine chloride is purified by dissolving in ammonia
and reprecipitating with hydrochloric acid; the pure salt is converted into
sponge by ignition.
The silver and the gold are purified electrolytically, by the Moebius
and Wohlwill processes, respectively, special attention being given to
the recovery of small amounts of platinum metals. All liquors resulting
from the processes are treated with zinc, or iron, and acid to recover
traces of the precious metals. In practice, it is found to be more eco
nomical to smelt the final ihsolubles and reduction residues than to
attempt to recover rhodium, ruthenium and iridium from them by wet
processes.
The lead ingots obtained by smelting are cupelled to remove excess
lead and the resulting precious metal alloy is parted with nitric acid.
The solution is treated in order to recover precious metals, especially
palladium, platinum, and silver which dissolve during the operation.
Most of the rhodium, ruthenium and iridium present initially in the

lead alloy are concentrated in the insoluble form from which they can be
extracted and refined by chemical processes.
Rhodium is extracted by fusing the concentrate with sodium bisul-
phate. Subsequently the rhodium is refined by a modification of the
process described by Wichers, Gilchrist and Swanger.
The insoluble from the bisulphate fusion is given appropriate treat-
ments to remove platinum, gold and lead sulphate as a result of which
a concentrate of ruthenium and indium is obtained.
This concentrate is fused with caustic potash and potassium nitrate
in spun iron bowls at a dull red heat to convert the ruthenium into
soluble potassium ruthenate. The cakes are then dissolved in water;
after settling, the clear solution is decanted into glass flasks and treated
with chlorine. This treatment converts the ruthenate into volatile
ruthenium tetroxide which distils over on raising the temperature. The
tetroxide is absorbed in dilute hydrochloric acid containing methyl
alcohol. Evaporation of the contents of the absorption vessels gives an
oxychloride, RuOCl2, which is reduced to metal by igniting in hydrogen.
There appears to be very little osmium in the precious metal concen-
trates received from the nickel refineries. It amounts to less than 1
percent of the ruthenium content, and on account of the similarity in
properties of these two metals it will be found as an impurity in the
ruthenium unless it is eliminated, which can be easily done by heating
the ruthenium oxy-chloride to dull redness in a current of air. The
resulting osmium tetroxide is absorbed in an alcoholic caustic soda solu-
tion and the osmium recovered via osmyl tetrammine chloride,
[OsO2(NH3)4]Cl2, which can be ignited to metal in an atmosphere
of hydrogen. The caustic fusion, besides converting the ruthenium
into potassium ruthenate, which is soluble in water, also converts the
iridium into an oxidized form which, although insoluble in water, is
said to be soluble in hydrochloric acid. Actually aqua regia is used for
dissolving the indium.
The chloride solution thus obtained is evaporated, and a crude ammo-
nium chloriridate obtained by adding ammonium chloride and oxidiz-
ing with small amounts of nitric acid. The crude salt is separated
from the accompanying ammonium chloroplatinate by fractional crystal-
lization; the use of a mild reducing agent facilitates the solution of the
iridium salt. Recrystallization is continued until the desired purity is
achieved; the salt is then decomposed by heating in a gas-fired muffle
furnace. The partially oxidized metal is reduced by heating in hydro-
gen and is then further purified.
Refining of the richer concentrates from the nickel and copper re-
fineries which contain more than 50 percent platinum metals and very
little silver, commences with the treatment by aqua regia.
In precious metals refining, the order of operations depends on the
ratio in which the precious metals are present. The order described
here is followed because it has been found to be the most suitable for
these by-product concentrates, although it is realized that other sequences
of operations might be more suitable for other types of material.
The average purities of the metals recovered from these concentrates
are: platinum, 99.93 percent; palladium, 99.94 percent; iridium, 99.7
percent; rhodium, 99.7 percent; ruthenium, 99.7 percent; gold, 99.97
percent; and silver, 99.97 percent.
The high intrinsic value of the materials treated makes it imperative
to reduce working losses to a minimum and to obtain the maximum
yield as quickly and directly as possible. In practice, not more than 0.25
percent of the platinum metals contained in the concentrates is lost,
while 1.25 percent is temporarily retained in furnace slags returned to
the nickel refinery for retreatment. The remaining 98.5 percent is
produced directly as refined metal. The losses occurring in the processes
of extracting the nickel and copper are also very small, and it is esti-
mated that 90 percent of the precious metals, probably excepting
osmium, contained in the ore is ultimately recovered.
Chemical Plant for Refining Platinum

The chemical engineering problems in connection with the refining of
platinum are of a specialized nature, with the result that experience
in other fields is not always directly applicable. Moreover, the tradition
of secrecy in the platinum industry up to the present has prevented
refiners from pooling their experience except in a very general way. The
state of affairs is in very marked contrast with what is known about
the plant and processes for the refining of silver and gold.
These remarks are not intended to apply to smelting operations,
which have received adequate treatment in text-books on metallurgy,
but rather to the wet processes for refining crude platinum and platinum
concentrates.
The very nobility of the platinum metals necessitates the use of
powerful acids and other reagents for refining them, a circumstance
ruling out many of the standard materials for the construction of chemi-
cal plant; for instance, even stainless steel can rarely be used. The scale
of operations is small, even the refining of platinum is frequently
described as large-scale laboratory work, and the refining of other
metals of the group is on a smaller scale still. Mechanical operations
are reduced to a minimum, and where machines have to be used the
vantage that there is no risk of bumping when heavy insoluble matter

settles to the bottom of the vessel.
On the other hand, the conditions are quite different for dissolving
mineral platinum, which is best done in gas-heated Vitreosil bottles of
70 liters capacity, using concentrated aqua regia at or near the boiling
point. Evaporation of nitric acid solutions is done in similar Vitreosil
bottles equipped with condensers.
Nitric acid treatment is carried out in gas-heated 20-liter basins made
of Vitreosil, which are also used for sodium bisulphate fusions. Sul-
phuric acid treatment is done in pans made of fine-grained grey cast
Vitreosil (fused quartz) retorts, for evaporating strong acids. Part of the asbestos
jacket in the foreground has been removed to show the retort.
Courtesy of The Thermal Syndicate
iron, each ot 25 liters capacity, and heated by gas. Lead-lined wood and
iron vessels are both used for liquors containing free sulphuric acid.
Wooden tanks up to 300 gal. capacity, lined with soft rubber, are
satisfactory for some of the operations, for instance, reactions where
sufficient free hydrochloric acid is present to make lead-lined vessels
unsuitable. Acid-resisting hose is used to convey acid liquors and to
make connections for removing fumes from reaction vessels.
Steel vessels lined with hard rubber are preferred to chemical stone-
ware vessels for operations involving the use of hydrochloric acid in
which the heat of reaction causes a rapid rise of temperature. Porcelain
plant is used for small-scale purification work.
Auxiliary PIant and Services

An oil-fired boiler supplies low-pressure steam for process purposes
and for general heating of the building.
The vacuum system consists of two 10 to 17-H.P. Nash Hytor
centrifugal vacuum pumps maintaining a vacuum equivalent to 10 in.
of mercury; a higher vacuum is considered inadvisable on account of
the fact that stoneware filters are used. The vacuum mains consist of
hard rubber pipes. Liquor catch pots and a vacuum scrubber protect
the vacuum pumps from corrosion.
Two Broom and Wade air compressors, with automatic control to
economise the power consumption, supply air for the acid eggs.
Drying of various intermediates and products, most of which are
slightly acid, is done in gas-heated drying ovens, some of which are
thermostatically controlled. Enamelled iron, Pyrex glass and ceramic
dishes are used.
Residues, which sinter during calcining, or cake during drying, may
have to be converted into powder for subsequent operations. For this
purpose there are four 8-in. and one 14-in. disc pulverisers. Care is
taken to collect and recover all dust produced during pulverising by
means of a fan and bag system.
Removal and Scrubbing of Acid Fumes

The fumes given off during reactions include sulphur dioxide,
sulphuric acid, chlorine, hydrochloric acid, oxides of nitrogen and
nitrosyl chloride. Although stoneware fans can handle mixed acid
gases such as these, it is more satisfactory to use an ejector in which
the gases do not pass through the fan. In the latter system a fan de-
livers a jet of air at high velocity into a specially designed nozzle, which
has the effect of creating sufficient suction to draw the fumes away from
the reaction vessels. The acid fumes are neutralised in chemical stone-
ware scrubber towers by a sodium carbonate solution which is kept in
circulation by centrifugal pumps.
The plant for the thermal decomposition of ammonium chloroplatinate
and palladosammine chloride has been selected for special mention be-
cause it possesses some features of chemical engineering interest.
Ammonium chloroplatinate can be converted into metallic platinum
by reducing the moist salt with zinc dust, but the resulting platinum
is contaminated by any impurities in the zinc and requires further treat-
ment, including ignition, before it is fit for sale. Consequently it is
preferable to decompose the salt by heat when the following reaction
occurs:
3(NH 4 ) 2 P t C l 6 = 3Pt + 1 6 H CI + 2NH4CI + 2N2.
Although it is an easy matter to ignite a few grammes of either of these

salts under laboratory conditions, the decomposition of 50 kg. a day
presents an engineering problem. The best solution of the problem
was not obtained at the first attempt, but a comparatively simple ar-
rangement of plant has now been in use for several years, and fulfils
the essential requirements of:—
(i) high purity of product;
(ii) trouble-free operation;
(iii) low cost of operation and maintenance;
(iv) efficient collection, removal, and neutralisation of the fume.
The dry salt is placed in Vitreosil trays which are slowly heated in
muffle furnaces fired by gas. Particular care must be taken whilst heat-
ing the material from 310-370° C, after which the temperature is
raised to 900° C. to complete the removal of volatile matter.
In accordance with the recommendation of Treadwell, the salt is
wrapped in filter paper to reduce losses during decomposition. Tread-
well also says there is a risk of the formation of volatile chlorcarbonyl
compounds due to the action of carbon monoxide on the decomposing
salt, but losses due to this cause can be avoided by careful operation.
The fumes evolved from the decomposing salt are drawn away
through a Cronite tube, an alloy containing nickel, chromium and iron,
into a sublimate box, where most of the sublimate settles, and thence
into a flue leading to a central scrubbing plant where the hydrochloric
acid is neutralised.
As the fumes are particularly unpleasant, it is important that the
ammonium chloride sublimate should not be allowed to settle out pre-
maturely and block the fume pipe. When decomposing palladosammine
chloride it is advisable to place a small gas burner under the Cronite
pipe as the fumes contain much more ammonium chloride than the
fumes from the platinum salt:
3 Pd(NH 3 ) 2 Cl 2 = 3 Pd + 4 NH4CI + N2 + 2 HC1

and there is more likely to be trouble from blockage of pipes and flues.
• * # #
SOME OTHER LARGE PLANTS
For the sake of the serious student interested in large-scale refining,

we shall summarize the procedures in three more large plants. No
new methods will be introduced, but it will be observed that the
choice and order of procedures will be determined by the ratio of
metals that are present.
The Copper Cliff Plant. A paper entitled RECOVERY OF

GOLD FROM BALBACH-THUM SLIMES AT COPPER
CLIFF, ONTARIO, by Frederic Benard, appearing in Metals Tech-
nology of February, 1938, gives an excellent description of this re-
finery. The problem resembles in some ways that of the Raritan
Copper Works. Crude copper is refined electrolytically; the sludges
from the copper cells are then refined by the Thum (or Balbach-
Thum) silver method, much like that used at Raritan. The sludges
from these silver cells, however, contain much more platinum and
palladium than those of Raritan, and the procedure changes here.
Because of this high percentage of platinum metals, acid methods
are now employed. The slimes are dissolved in aqua regia, and
the gold precipitated by ferrous salts, in a manner quite like that
given in early chapters of this book. The dissolved platinum metals
are concentrated (as in our Stock Pot) by a metallic precipitant,
in this case scrap iron, and shipped to the plant at Acton, England,
for further treatment. The gold, which came down as the familiar
brown powder, runs about 995 fine, and is put through a re-purifica-
tion in Wohlwill cells. The paper describes the equipment and the
method of disposing of fumes, and gives a flow-sheet.
The Plant at Montreal East. This plant refines copper that con-
tains some gold and silver, but practically no platinum metals. In
many ways the procedure is the same as that used at Raritan,
except that there are no gold cells only copper cells and silver cells.
The sludge obtained from the silver cells contains high quality gold
gold which after a treatment with boiling concentrated sulphuric
acid is ready to be melted down for market.
A paper entitled RECOVERY OF PRECIOUS METALS FROM
ELECTROLYTIC COPPER REFINING at the Canadian Cop-
per Refiners Plant, Montreal East, Quebec, by C. W. Clark and
A. A. Heimrod, describes the plant and its equipment in detail. It
is published in the Transactions of the Electrochemical Society,
pages 437 to 450, 1932.
The Recovery at Old Hickory Powder Plant. This particular
plant was unique, and probably will remain so. It was built hur-
riedly in 1920 for one purpose to recover the finely-divided plati-
num from the magnesia contact masses used in making sulphuric
acid, and left in the hands of the War Department after the Armi-
stice. It is mentioned here because the story proves how hastily-

contrived equipment can, in an emergency, accomplish much. The
procedure is now outmoded, the funds were limited, the time short,
and the personnel insufficient. But 343,241 pounds of dirty contact
mass were worked over and 8,812 troy ounces of platinum were re-
covered, all within about four months. An article entitled RECOV-
ERY OF PLATINUM FROM USED CONTACT MASS AT
THE OLD HICKORY POWDER PLANT, by A. L. Kibler, in the
Journal of the American Chemical Society, vol. 14, pages 636-641,
July, 1922, describes the equipment and procedure.
The Royal Mint at Ottawa, Ontario. In a handsomely illus-
trated paper entitled REFINING $2,000,000 OF GOLD PER
WEEK, A. L. Entwistle, writing in Canadian Metals and Metallur-
gical Industries (Toronto), vol. 1, January and February, 1938, de-
scribes the treatment of crude gold bullion at the Royal Canadian
Mint Refinery. When in 1914 it was found necessary to enlarge the
plant, the Miller chlorine process was installed, and is still the main
procedure in the new building erected in 1936. Only a few small elec-
trolytic cells are occasionally used on bullion suspected of containing
platinum metals. This bullion is chlorinated to about 97 percent
gold, then treated by the Wohlwill process and the platinum metals
recovered from the electrolyte. Entwistle's paper describes the
routine of the reception of bullion, its testing and purchase, the
equipment and procedure of the Miller process, and concludes with
statistics on the plant's efficiency.
APPENDIX
CHAPTER A. MELTING THE PRECIOUS METALS
Part I. Platinum and its group. Fuel and equipment. Melting small
lots of platinum scrap. Working and annealing. Melting platinum
filings. Defective platinum. Platinum sponge and black. Large melts.
Making up platinum alloys. The electric furnace. The oxygen torch.
Spectacles. Crucibles and refractories: lime; zirconia; sand; contamina-
tion from the crucible; other refractory materials. The platinum group
and occluded gases. Melting and working palladium. Melting the
other platinum metals.
Part II. Melting gold and silver. Sources of information. Melting the
fine gold powder. Melting lemel. Melting clean scrap. Deoxidizers.
Fluxes. Casting precious metals. Making up new alloys. New light
on old metals.
* # # #
PART I. PLATINUM AND ITS GROUP
The story of man's early efforts to melt platinum, culminating in
the invention of the oxy-hydrogen blowpipe, forms one of the most
dramatic chapters in the history of this metal. While much has
been written on this subject, as well as upon the chemical and
physical properties of the metal, the literature gives few details
of the actual present-day technique of melting it, specially in small
quantities. Platinum is being melted every day, not only in the re-
fineries where it is sold, but also in the host of small jewelry shops
and dental laboratories where it is worked. Since this book is
written primarily for the small organization, this chapter will devote
considerable space to this subject of small melts.1
FUEL AND EQUIPMENT
The melting point of platinum is so high 3191° F . , or 1 7 5 5

that the furnace commonly used for melting gold is not adequate.
A blowpipe burning city gas, in combination with oxygen from a
i This section is adapted from THE MELTING OF PLATINUM, by C. M. Hoke,
The Metal Industry, pp. 433-435, November, and pp. 468-470, December, 1936.
293
cylinder, is the device most generally used for melting platinum.

When city gas of proper composition or pressure is lacking, hydro-
gen from a cylinder may take its place. Acetylene with oxygen
provides heat enough for the purpose, indeed more than enough;
but because of its tendency to form the brittle platinum carbide, as
well as because it may melt the crucible, the oxy-acetylene flame is
not often used in this work. The electric induction furnace is used
in some cases, as will be described shortly.
This need for a high temperature has brought about some misunder-
standing. Thus a textbook will say: "Platinum can be melted only with
difficulty," meaning that the melting point is high. This does not mean
that the melter's task is laborious, however. On the contrary, if he has
proper equipment the task is accomplished quickly and without the
slightest hardship.
Many kinds of crucibles have been used, as we shall see later.

The most popular are those made of lime, or of a composition re-
sembling the so-called sand crucibles used in melting gold. No flux
is required.
The sketch shows a convenient arrangement of equipment. A
Melting a small quantity

of platinum.
Courtesy of
Hoke Incorporated.
one-inch gas supply pipe, with a 3/8 inch service cock, will furnish
enough gas for the average jewelry shop. The torch shown is suit-
able for melts up to 100 ounces, and smaller models are available
for the smaller shop.
Since platinum volatilizes appreciably under the oxygen flame
(possibly through simple evaporation, possibly through the forma-
MELTING THE PRECIOUS METALS 295
tion of oxides that are gaseous at high temperatures, then decom-

pose on cooling) it is wise to use a torch that is big enough to ac-
complish the melting quickly and without undue "soaking."
MELTING SMALL LOTS OF PLATINUM SCRAP
In a jewelry shop that makes platinum jewelry, clean scrap may

be melted five or six times a day. We shall first describe the melting
of a small quantity, 100 pennyweight or less. A melt of this size
is usually made in a sand or clay crucible wide at the base and low
on one side, as shown in the sketch. Usually this crucible is set
inside another larger crucible, or perhaps in a cavity carved out of
a silica insulating brick, to conserve heat and protect the table
beneath.
The sheet-iron screen serves merely to prevent the scattering of
small bits of precious metal. A canopy of the type used in anneal-
ing is not suited to this purpose as it throws the heat back into the
operator's face. Asbestos board protects the table.
The rest of the equipment is shown in the sketch tongs, hammer,
anvil, dark spectacles, the torch with its hose, and the oxygen cylinder
with its regulator and gauges. Since the melting of clean platinum
produces no unpleasant fumes no flue is needed, and the entire outfit
is portable.
The beginner is advised to learn how to melt scrap or clippings
before attempting to melt filings.
The clippings are stacked up against the high side of the crucible.
First the worker acquaints himself with the torch, turning on the gas
and lighting it, then slowly admitting oxygen until a clear blue flame,
with a noticeable roar, is obtained. If the crucible is new, he an-
neals it lightly by passing a flame over it to drive out moisture. The
flame is then turned directly upon the metal, and if the flame is
right, the edges of the metal will melt and run within a few seconds.
A slightly oxidizing flame is used; with too little oxygen the flame
is yellow and sooty, not hot enough to melt platinum but capable of
embrittling it through the formation of a carbide. If too much
oxygen is used, the flame is apt to blow out, is too cool to melt plati-
num, and is also capable of embrittling it through causing adsorp-
tion of oxygen.
A few minutes' practice is usually enough to enable even the be-
ginner to get a satisfactory flame. The metal soon flows down and
runs into a globule. The torch should be held at an angle not

straight up-and-down playing upon the bottom of the crucible, and
slanted so that the products of combustion are not reflected directly
back upon the torch nozzle. Nor should the torch be held too close
to the molten metal, otherwise the nozzle may melt.
Angle at which torch should be held.
Within a very few minutes 100 pennyweight of platinum will be

molten; three or four minutes is about average time with the torch
shown. If the metal does not melt promptly, it is because the flame
is not right; perhaps the gas volume is inadequate; or the oxygen
volume is too high or too low.
The next task is to mix the metal thoroughly by moving the flame
around over it so as to revolve the molten metal; in that way any
unmelted lumps will be liquefied. The metal should run around in
the crucible like a little pool of mercury.
The worker next removes the flame and shifts his dark spectacles
tip onto his forehead. Then, with a wet tongs, he pushes the button
(which by this time has solidified) out of the crucible, and places it
bottom upward on a piece of clean refractory. The button will be
rough, from its contact with the crucible, and will need smoothing.
The worker promptly replaces his spectacles and turns the flame on
this rough surface for a few seconds, long enough for it to run
smooth, while any beads of sand from the crucible will slip to the
edge. This process, usually called "sweating" or "flushing," should
always be carried out on a highly refractory surface such as a lump
of lime, or the bottom of an inverted sand crucible.
WORKING AND ANNEALING
While the button is still white hot, it should be hammered hard on

the anvil and shaped into a square block or bar. This hammering
does more than shape the metal, however; it proves its quality, for if
impure it will split under the hammer. Also, it makes the metal
dense and hard, so that it keeps its polish when made up into an
article of commerce. During this treatment the surface of the
platinum may take up a thin film of iron, which should be removed

by a hydrochloric acid dip.
The hammering sometimes leaves grooves in the metal which are
too deep to be obliterated by the rolls; these may be easily removed
by sweating. Sometimes the hot metal is placed under a drop-
hammer in a die; this gives a dense and firm texture with little effort
on the part of the workman.
Hammering a block of hot platinum.
Finally the block is cooled, then rolled, drawn or spun, during

which treatment it hardens and will require occasional annealing.
This is usually accomplished by passing a hot oxy-gas flame over it,
or it may be heated in a gas muffle that is open to the air. Anneal-
ing with a smoky flame, or in a reducing atmosphere, is not advised
as it might form carbides. Iridio-platinum requires a somewhat
higher annealing temperature than soft platinum,
(For further details on the working and annealing of this group, the
reader is referred to the books and papers mentioned in Chapter E, especi-
ally the following: PROPERTIES OF THE PLATINUM METALS,
by Wise and Eash; FABRICATION OF THE PLATINUM METALS,
by C. S. Sivil; and METALS OF THE PLATINUM GROUP, by
Atkinson and Raper. Sivil's paper also gives a particularly interesting
account of the early methods of melting, and Wise and Eash recommend
ontimum annealincr remneratures of the various alloys)
MELTING PLATINUM FILINGS
It takes longer to melt 100 pennyweight of filings than 100 penny

weight of clippings. For one thing, a gentler flame must be used or
the small filings will be blown about. In addition, the many facets
reflect the heat and make the process slower. Aside from these de
tails, it is no more difficult to melt clean filings than clippings.
Dirty filings are another story. This brings up one more difference
between platinum and gold: gold is readily freed from many im
purities by a simple melting, but it is not easy to refine platinum in
the same way. Under the flame, the platinum tends to combine
with the impurities to form brittle compounds, such as the carbide.
It is much easier to remove the impurities before melting the filings,
and if this cleaning is properly done, the melting is easy.
As in the case of scrap, the filings are piled up against the high
side of a sand crucible. Most workers prefer to handle 50 penny
weight or less at a time. When they have more than that on hand,
they first make several small buttons, then melt them together into
one large button.
A gentle flame is turned on at first, with enough oxygen to give
the proper temperature, but without blowing or noise. This is
directed at the bottom of the stack of filings, and slowly brought
toward them. Soon the outer layer of filings will melt and run
together. With that the flame is brought closer, and its pressure
increased gradually, while the worker takes care to keep the bottom
of the stack hot.
From here on the method is practically the same as for clippings.
The metal runs down into the bottom of the crucible, the flame is
moved around over it to make sure that all parts are molten. Finally
the button is lifted out and sweated as before, and hammered while
hot.
This is another instance where platinum differs from gold; gold is
not hammered when hot, but platinum is better the hotter it is when
hammered. When a crucible of molten gold cracks, the result is often
disastrous; the gold seems to find the smallest crack and go through it
immediately. But molten platinum acts quite differently; its surface
tension is so high that the metal rounds up like a drop of mercury,
and will remain on top of even a sizable aperture. This high surface
tension is a characteristic of clean metal; when impurities are present
the drop will flatten as it cools.
A melt of 100 pennyweight of clean filings should not take more

than ten minutes. A skilled worker will accomplish it in much
less time.
DEFECTIVE PLATINUM
Slovenly refining is the most frequent cause of defective metal.

Incomplete refining may leave gold, silver chloride, copper, organic
dirt and possibly other things in the filings. Gold shows itself
by a bronze or purple stain on the crucible; silver chloride leaves a
gray or black stain and gives a characteristic smell; copper colors
the flame green during the melting operation. Organic dirt shows
itself in a smoke that rises from the metal during and after melting,
and the surface of the button may be dark or rough. Sometimes the
button will flatten as it cools, instead of solidifying in a smooth
globule.
A button that smokes or has a bad surface should be hammered
with care, as it may be extremely brittle. Some workers try to
"save" such a button by remelting it and letting the flame play on
it for several seconds in the hope that the dirt will burn out. Such
a procedure is rarely successful, and there is always a detectable loss
of platinum. It is much easier and more economical to clean the
filings thoroughly before melting them.
A second cause for defective platinum is improper melting.
Sometimes platinum that has been properly refined will yield a but-
ton which when hammered will prove to be brittle, or pitted, or
"double." All of these faults are caused by improper melting. They
are easily corrected in one remelting, and should not occur after
the operator has acquired a little experience.
Thus the beginner may find that his platinum bars will split under
the rolls in a way that suggests alligator's jaws, or his platinum sheet
will crack at the edges. When this happens, and the color of the
crucible indicates that the platinum contained no metallic impurity,
the chances are that an improper flame was used. Perhaps there
was not enough oxygen in the flame; this may give a warty button
of a dark color. Or perhaps there was too much oxygen, which was
absorbed while the platinum was molten and did not have an op-
portunity to escape as it cooled; this gives a rough white button,
which under close examination will be found to contain minute
bubbles or pores.
Pits in the surface of an otherwise good button may appear when
the metal is not properly sweated, being caused by beads of sand

from the crucible.
"Double" platinum is caused thus: part of the metal melts and
runs down into the bottom of the crucible, where it cools; later
more metal melts and runs down over it, and the two layers do not
flow together. Subsequently, when the button is rolled, a blister
will show where the two layers failed to join. If a worker suspects
doubles he will anneal the metal after rolling, as this makes the
doubles more conspicuous. The way to avoid this is simply to keep
the bottom of the crucible hot all the time, and to be sure that the
metal is truly molten before it is allowed to cool.
The remedy for all these imperfections brittleness, warts, pits,
and doubles is simply a remelting.
PLATINUM SPONGE AND BLACK
Platinum that has been through the acid refining is usually re-
covered as sponge, sometimes as black. The refiner does well, when
burning the orange powder, to heat it until the sponge has contracted
into a sort of clinker, so that it will not blow about under the flame.
Some workers compress it firmly in a mold. Clean sponge is a
pleasure to melt, and is handled the same as clean coarse filings.
The black, however, being much more finely divided, must be
handled with caution, a small and very gentle flame being applied
at first, until the powder on the surface has run together.
LARGE MELTS
When large melts 40 ounces or more are made, lime crucibles

are preferred. These will be described later in detail. Lime holds
the heat excellently, nor does its surface soften to permit the metal
to sink into it. Sometimes two torches are used at once. Usually
the metal is fed in a little at a time from ladles previously filled and
set close at hand. With large melts it is impossible to stir the
metal with the force of the flame; some other method of mixing
must be devised. Sometimes a carbon rod is used; sometimes the
crucible is set into a support with a long handle and semi-spherical
base; as the metal melts the worker tips and tilts this support on its
rounded base, effecting proper mixing.
When small amounts of metal are melted, the button cools so
quickly that it is practically impossible to pour it into a mold.
Larger melts retain enough heat to permit pouring if desired, and

usually they are cast into blocks, the forms being made of hard
carbon sometimes lined with thin platinum foil.
Pouring molten platinum.
MAKING UP PLATINUM ALLOYS
The lndio-platinum alloys are usually made by mixing platinum

sponge and iridium sponge in proper proportion, and then melting.
The combination has a higher melting point than that of pure
platinum, but the technique presents nothing new.
Within the last few years many other platinum alloys have been
studied, of which the most interesting are those used in dentistry.
These are sometimes quite complex, containing gold as the major
constituent, with silver, copper, platinum, palladium perhaps, and
sometimes other metals, in varying proportion. When the propor-
tion of platinum is low, these are prepared in the gold-melting
furnace, the low-melting ingredients being melted together first, and
platinum, usually in small pieces, being added toward the last. Or,
gold and platinum may be melted together with the oxy-gas flame,
the same as platinum alone, and this alloy is then melted in a gold-
melting furnace with the other ingredients.
However, as the requirements of the dental profession have

become more exacting, greater uniformity has been demanded in
the manufacture of alloys. The electric furnace of the high-
frequency induction type has proven very satisfactory for this work.
THE ELECTRIC FURNACE
Ever since the Ajax Northrup induction furnace was first de-
veloped, in about 1916, its application to the precious metal field
was inevitable. Its cleanliness, convenience, freedom from fumes,
Ajax Northrup High Frequency Induction Furnace. The converter is in the lef
background, the furnace proper in the foreground.
and the startlingly high temperatures it produced, have led man)

workers to experiment with it. Its first application in this field
was in the melting of silver, where it has been notably successful
and it is now being used also in melting the ordinary gold alloys, as
well as the special dental alloys. It has been extremely useful in

basic research on the platinum alloys.
The outfit consists of two main parts, the converter which changes
currents of commercial frequency into the high-frequency current,
and the furnace proper. The latter is essentially a metal coil through
which the high-frequency current oscillates. When a crucible con-
taining metal is placed inside this coil, the electric eddy currents
induced in the metal quickly raise its temperature to the melting
point or beyond. Theoretically there is no limit to which this
temperature can be raised; the practical limits are determined by the
refractory that contains the melt, and the wishes of the operator.
A very small
Ajax Northrup fur
nace. A type used
in research work on
The furnace proper, or coil, may be small enough to be lifted

bodily for pouring. Another convenience of the induction furnace
is that it automatically stirs the melt the eddy currents within it
cause active turbulence in all parts of the metal. This is especially
desirable in making up complex dental alloys with their many
constituents. It is calculated that one plant alone has already melted
up more than $10,000,000 worth of gold and platinum-gold alloys
in a furnace of this type.
A final advantage of the induction furnace is that the atmosphere
surrounding the melt can be controlled perfectly. Air can be ad-
mitted or excluded; hydrogen, nitrogen, or other gas can be substi-
tuted; or the whole operation can be carried on in a vacuum. These
possibilities have been exploited by the research metallurgist as
well as by the conscientious manufacturer.
Up to the present, however, the electric furnace has not sup-

planted the oxygen torch for the routine melting of the simple
platinum and iridio-platinum alloys, either in large or small lots.
Ajax-Northrup furnace for precious metals, suitable for melts of a few pennyweight
or as much as 1ooo ounces of gold.
This is due mainly to the great difference in costs, the converter

especially representing a sizable investment, while the oxygen
outfit is within the means of even the humblest jewelry shop.
THE OXYGEN TORCH
Dr. Robert Hare of Philadelphia, the inventor of the oxy-hydrogen

blowpipe, was the first to melt platinum in the modern manner;
that was in about 1847. In 1859 the French chemists Deville and
Debray designed an improved furnace for the purpose, consisting of
a blowpipe and lime crucible. Innumerable improvements in de-
tails have been made since then, but the same principles are being
used to-day.
The simplest oxy-hydrogen or oxy-gas blowpipe consists of little
more than two pipes and perhaps two stop-cocks. The early ones,
which used both gases under fairly high pressure up to 220 lb.
per square inch were apt to back-fire and burn inside the tubing,
with resulting damage to the blowpipe and possible danger to the
Melting platinum with the oxygen-gas torch.
operator. With the advent of platinum as a fashionable jewelry

metal, in the first and second decades of this century, many im-
provements of torch design appeared, attention being given to safety
and efficiency. The rapid growth of the compressed oxygen in-

dustry was a big factor in the vogue of platinum. Indeed, without
cheap and available oxygen, platinum would still be a laboratory
curiosity.
By 1912 or so, it was a common sight in the larger jewelry shops
to see the two large cylinders, one with low-pressure (200 lb.)
oxygen, the other sometimes containing hydrogen, but much more
often simple city gas compressed to the same 200 pound pressure.
But early in that decade, Sam W. Hoke of New York City
conceived the idea of a torch that would use city gas at ordinary
house pressure less than a pound in conjunction with the in-
expensive high-pressure oxygen. This was a much more convenient
and economical arrangement. A patent was granted, and many of
the early Hoke models are still in use.
The Hoke patent involved a mixing device close to the nozzle of
the torch, by which the gas in the outer tube was slightly rotated,
through rifling of the barrel, and the oxygen in the inner tube
.-#: Mouth of a modern oxygen-gas torch
was broken up into many small streams, with the result that the
two were mixed in the proper proportion, in spite of the difference
in pressures behind them. The small cut shows the mouth of the
torch; oxygen comes through the many small holes in the central
cap; gas through the four large outer passageways.
Other successful models utilized the old injector or Venturi
principle to attain the same end proper mixing of oxygen and fuel
gas. By means of minor adjustments the torches were adapted to
natural gas or to hydrogen, as well as to city gas.
One big problem was the material for the nozzle. The heat of the
flame is sufficient to melt most metals promptly, especially if the torch
be held in such a way as to let the flame play back upon it. Other
metals, such as some high temperature steels, do not melt promptly,
but form copious scale which soon drops off into the molten platinum.
The last, as any platinum worker knows, may be a real calamity. The
ideal material for a nozzle is of course platinum itself, and some large
plants are indeed using platinum nozzles. For the ordinary shop, how-
ever, other materials had to be found, of which the best so far are
certain nickel alloys; these do not melt if the flame is held at the proper
angle, and such scale as does form is closely adherent.
SPECTACLES
The light emitted by molten platinum is such that lenses

almost as dark as those used by the oxy-acetylene welder are re-
quired. Most workers find cobalt blue the easiest on the eyes. The
temples should be designed for quick doffing and donning; com-
plicated straps or buckles are not convenient.
CRUCIBLES AND REFRACTORIES
Lime
Lime was the first choice for a crucible material; fresh high-
burned lime. It was troublesome it quickly absorbs moisture and
becomes useless; since it is a natural product coming in irregular
lumps, the purchaser often had to pick over many pieces before
finding one of the right dimensions, free from cracks. A cup-like
cavity was carved in a square block, and the block bound in sheet-
iron and provided with a handle for tilting. A cover consisted of
a second block, carved in an arch, with two holes, one for the torch
nozzle, the other for the escape of the products of combustion. A
crucible cannot be used often, nor can it be stored, and it seemed
that the worker spent a great deal of his time carving lime. But
even with these disadvantages, lime is still in general use, especially
for large melts or where special purity is demanded. It holds heat
amazingly well. If bits of lime stick to the metal, they can be readily
soaked off in hydrochloric acid. A big factor in its popularity is its
cleansing effect on the melt, as it absorbs minor impurities after
the manner of a cupel.
To-day, in plants that habitually make large melts, instead of
carving the natural blocks, they granulate dry lime and press it into
blocks with a hydraulic press, using pressures up to 150 tons to
the square inch. These blocks are then bound in sheet-iron, in the
same manner as before. For a melt of 400 ounces 8000 pennyweight
a lime block 9 inches across and 5 inches high, outside dimensions,
properly carved, is adequate. Just before use the crucible should be
heated slowly to drive out moisture.
Zirconia
Zirconium oxide and zirconium silicate, highly refractory ma-
terials, have been shaped into crucibles for this work, especially by
some of the larger refineries and Mints. They resist heat remark-
ably well, and a crucible can be used scores of times, day after day.
They have not the cupelling effect of lime, but in most other respects
are excellent crucibles. Probably the one factor that stands in the
way of their general acceptance is their cost.
Sand
When platinum first came into use in jewelry, most, melts were
made in the familiar Hessian type sand or clay gold-melting cru-
cible, with one side broken down. Even to-day crucibles of this
class that get broken in transit are used for this purpose. Shortly,
however, J. Goebel of New York provided platinum-melting
models of a similar composition, with the high side and a flat
bottom. The flat bottom is important, especially when filings are
to be melted, because it permits the flame to reach and melt all the
small particles. One big fault of the broken-down gold-melting
crucible is its tapered bottom, in which particles will sink and escape
the flame.
Contamination from the Crucible
Some workers have reported that the sand or clay crucible is apt
to contaminate the melt with silicon or iron. However, for ordinary
shop and routine work, this contamination, if it occurs, is too slight
to affect the metal's usefulness. It should be recalled that an oxidiz-
ing flame should always be used; a reducing flame one of a yellow
color can reduce sand to silicon or lime to metallic calcium, both
of which elements are taken up by the molten platinum to its detri-
ment. No doubt some crucibles have been blamed for con-
taminations which in fact were caused by an improper flame.
Other Refractory Materials

Graphite or carbon is not a suitable crucible material, because of its
tendency to combine with platinum when hot. For the same reason,
if a carbon stirring rod is used, it should not be left in contact with
the molten metal for any length of time, and when graphite ingot molds
are employed, they should receive little if any pre-heating.
Crucibles made of fused quartz have been suggested, also those

made of aluminum oxide (Alundum). These substances, however,
melt at so nearly the melting point of platinum that they are unsuitable.
It is true that the sand-clay crucibles also soften under the flame, but
so slowly that the melting job is usually finished before this softening
has done any harm.
Beryllia and thoria, on which some experimental work has been done,
are promising refractory materials, but at present their cost discourages
general adoption.
THE PLATINUM METALS AND OCCLUDED GASES
The fact that the platinum group metals can absorb gases, espe-
cially when hot, has been mentioned repeatedly. It has caused
annoyance to the artisan and confusion to the student. The litera-
ture contains countless references to it. The exact nature of the
process has been the subject of much discussion. One student will
speak of it as "occlusion"; another as "absorption"; another will use
the term "adsorption" to suggest that it resembles absorption but is
not quite the same.
All the metals of the group show the tendency to some extent.
Thus when platinum is heated strongly, no visible oxide is formed,
but volatilization occurs. This may be a simple boiling away, but
perhaps it may involve the formation of oxides that are volatile while
very hot, then decompose on cooling into metal and oxygen. As we
have just pointed out, if too much oxygen is present in the flame
while platinum is being melted, some of it can be entrapped as the
metal solidifies, to occasion minute bubbles or blisters.
The reader will recall that osmium and ruthenium, when
heated in air, form volatile oxides. These oxides can be captured
and reduced back to metallic form without much difficulty; indeed
if osmium oxide touches the mucous membrane of the eye, it breaks
down and deposits a film of metallic osmium in the tissues. Molten
palladium absorbs very large volumes of oxygen, and sometimes
visible oxides form and disappear and form again. As we shall see
shortly, this peculiarity must be borne in mind when palladium is
being melted. To the serious student, this is one of the most
fascinating aspects of this fascinating group.2
2 For a comprehensive review, see DAS VERHALTEN DER EDELMETALLE
UND IHRER KUPFERHALTIGEN LEGIERUNGEN GEGENUBER SAUERSTOFF
IN FLUSSIGEM ZUSTANDE UND BEIM GLUHEN IN FESTEM ZUSTANDE.
By E. Raub and M. Engel. Mitt. Forsch. Inst. Probieramts Edelmctalle, pp. 1-12,
June, 1939.
MELTING AND WORKING PALLADIUM
Except for a few details, the procedures used in the melting and
fabrication of platinum are employed with palladium and its
ordinary alloys. Most workers prefer hydrogen (with oxygen)
instead of city gas, but both combinations are being used. Lime is
the approved crucible material, some workers finding that they
do not get good results with the sand crucible. For very small lots
a hollow carved in a charcoal soldenng-block is adequate No
flux is required, either with this or any of the other platinum
metals.
Because of palladium's ability to absorb much more oxygen while
molten, a slightly different melting procedure is demanded. Begin
work by turning a full hot flame on the scrap palladium, which
will melt promptly, the melting point being lower than that of
platinum. When all the metal is molten, instead of removing the
flame abruptly, as is done when melting platinum, the worker will
slowly reduce the oxygen supply, and at the same time slowly
withdraw the flame, so that the button may solidify while being
bathed in a reducing flame.
As the metal cools, the occluded gases are expelled, and some-
times the button swells suddenly and solidifies with a rough sur-
face. If this happens, turn a small gentle flame on the top surface
until it just runs smooth, then reduce the oxygen supply and
withdraw the flame slowly, as before. This gives the absorbed
oxygen an opportunity to escape. Do not melt the entire button—
only its top surface. This sweating may have to be repeated.
Palladium and its ordinary alloys are rolled and annealed in
much the same manner as platinum, being given their first reduc-
tion while hot, though they can be worked cold. If gas has been
occluded, the edges of the bar will crack, and sheet will open up
like alligator's jaws, and the metal must be remelted.
Films of colored oxides form when palladium is heated moder-
ately, and disappear as the metal is heated further, only to reappear
when it cools. One way to get rid of these films is to quench
the led hot metal in water, thus cooling it before the film has time
to form. Another way is to pickle the palladium while warm in
dilute formic acid.
MELTING THE OTHER PLATINUM METALS
In any history of the platinum metals, the student will find mention
of a mode of fabrication, used before the invention of the oxygen torch,
in which platinum sponge was strongly compressed, then heated, first
gently, then as strongly as possible in a wind furnace, and forged while
hot, then rolled down or drawn into wire. This method is called
Wollaston's process, and while not as convenient as the method of
fusion with the torch, it offers certain advantages, such as freedom from
contamination from the crucible or from the gases in the flame; and it
is still occasionally used to-day, especially in melting and working the
rarer metals of the group. (The same principle, sometimes called
"powder metallurgy," is also used in the manufacture of tungsten wire,
among other things.)
The oxygen flame and the electric induction furnace are also used,
sometimes all three methods in combination. When we recall that
these metals all have high melting points, all tend to absorb gases when
hot, and that at least two of them have highly volatile oxides, it is easy
to see why these metals present such fascinating vistas of speculation to
the research student.
Rhodium. The classic research on the melting and working of
rhodium was done by William H. Swanger of the National Bureau of
Standards, and reported by him in 1929. (See Chapter E of the
Appendix for details regarding this and other publications.) The melt-
ing point is higher than that of platinum, and the tendency to absorb
gases is even more annoying than with palladium, but it is perfectly
possible to melt rhodium with the oxy-hydrogen flame, and, by manipu-
lating the oxygen supply as the metal cools, to produce a fairly good
button. However, for better results, Swanger suggests the following
procedure:
First, clean rhodium sponge is pressed into pellets in a steel form,
then heated and cooled in hydrogen. These pellets, now fairly firm, are
next melted in vacuum in the high-frequency induction furnace. The
ingots thus obtained have a smooth silver-white surface. These ingots
are then remelted on lime with the oxy-hydrogen flame, and forged or
swaged while very hot. In this way ductile wire of small size can be
obtained.
Iridium. It is possible to melt iridium with the oxy-hydrogen
torch, but the melting point is so near to the temperature of the flame
itself that the process calls for great patience. Iridium melts more
readily in the hotter oxy-acetylene flame, and if the acetylene-oxygen
ratio is well adjusted, workable metal can be obtained. That is, metal
that is workable at white heat. Even the best iridium so far produced
cracks and crumbles when worked cold. Iridium has also been melted
in the high-frequency induction furnace. Thoria crucibles were used,

as with rhodium, and satisfactory results were obtained.
Ruthenium and Osmium. According to Atkinson and Raper,
there seems to be no definite reason to exclude the possibility of melting
and working these metals, in spite of their high melting point and
volatile oxides, by modern powder metallurgy technique, using high
temperatures and excluding air. Deville and Debray melted very
small quantities of ruthenium with the oxy-hydrogen torch, but were
unable to fuse osmium. Because of their rarity and restricted industrial
application they have not received as much study as their sister metals,
and the literature gives few details.
PARTII. MELTING GOLD AND SILVER

Men who carry out the acid refining of gold are usually called
upon to perform a number of related tasks, such as melting clean
scrap into buttons for re-use, making up new karat golds, making
castings, and the like. They may also have to choose the furnace
and arrange its placement, purchase and care for crucibles, fluxes,
ingot molds, and other equipment. Some of these matters have
already been touched upon from time to time in this book.
SOURCES OF INFORMATION
To cover all the aspects of the gold melter's art would require a
volume much larger than this one. Fortunately the literature is
ample, so we shall call attention to some of the more valuable
books and discuss here only a few points, including some that
have been the subject of recent questioning or controversy. In
Chapter E of this Appendix the reader will find descriptions (and
in many cases prices) of many books and magazine articles, certain
of which should be in the hands of every melter. We mention the
following as outstanding: E. A. Smith's book Working in Precious
Metals, (1934); Edward A. Capillon's paper Melting and Casting
Some Gold Alloys, (1930); Leach and Chatfield's paper Manufac-
ture of Sterling Silver and Some of Its Physical Properties, (1928);
Daniel L. Ogden's paper Melting Fine Silver in High Frequency
Induction Furnace, (1932); George S. Overton's book How to Make
Jewelry, (1927); T. K. Rose's book The Metallurgy of Gold, (1937);
Eugene W. Skinner's book The Science of Dental Materials, (1937) ;
Thomas B. Wigley's book The Art of the Goldsmith and Jeweller,
(1911); Tracy C. Jarrett's paper Effect of Composition on Color

and Melting Point of 10-K, 12-K, and 14-K Gold Alloys, (1939);
the same author's Effect of Composition upon Physical and Chemi-
cal Properties of 10-K Gold Alloys, (1939); and E. M. Wise's paper
High-Strength Gold Alloys for ]ewelry, and Age-Hardening Phe-
nomena in Gold Alloys, (1929). This by no means exhausts the
list of helpful publications.
MELTING THE FINE GOLD POWDER
The fine gold powder produced in acid refining is normally

melted in a sand crucible in a gold-melting furnace. The purpose
is to remove any residual impurities and to obtain a button that
will be as near 1000 fine as possible.
Refiners often speak of this as one of their pleasantest tasks.
The pure gold melts tranquilly to assume a sort of luminous trans-
parency, very beautiful even through the melters' goggles, and it
can be poured without spitting or turbulence into a shining yellow
ingot.
As we say, it cools without spitting. That is because it does not
absorb oxygen, even when molten. It might be well for us to
pause here and recall a few facts that may offer an understanding
of metal behavior. Most metals, such as copper and iron and
nickel, unite with oxygen to form stable oxides; they unite slowly at
ordinary temperatures, rapidly at high temperatures, and the oxides
do not decompose on cooling. Platinum and palladium, as we have
just seen, occlude oxygen when molten, only to give it up on cool-
ing. Silver resembles the platinum metals in this respect, and the
well-known spitting and crabbing of fine silver, when poured, is
due to this peculiarity. But fine gold differs from all of these in
that it does not absorb, or unite with, or take up oxygen, even
when molten.
This characteristic is the basis of "fire" refining. It means that

if oxygen is introduced into molten gold it will unite with any
base metals that may be present, to form stable base-metal oxides.
If these oxides can be floated or fluxed away, the gold will be
purified.
This was brought out in Chapters V and XV of this book, to
which the reader is referred. When nitre or saltpetre is heated
strongly it releases oxygen gas; that is why we add a pinch of such

a salt to the molten gold. We also add some borax, in which the
base-metal oxides dissolve and are floated or fluxed away.
Chlorine is another gas which when hot forms stable compounds
with base metals, but not with gold. Accordingly we can purify
gold by introducing chlorine. This is the basis of the Miller
process described in Chapters XXI and XXII; also of the practice
of adding a little mercuric chloride, or ammonium chloride, or
copper chloride, to the melt when purifying gold that is almost
fine.
If platinum group metals are present in small amount in such gold,
they can be removed by repeated treatments with nitre, passing off in
the fumes. This fact is undoubtedly related to their tendency to take
up oxygen when molten. Their removal by this means is not suggested
as a method of separation, as there is no easy way to capture the
volatilized metal. Chlorine and chlorides do not remove the platinum
metals.
MELTING LEMEL
In Chapters V, XII and XVIII of this book, we mentioned the

procedure for melting filings (lemel) into a button for subsequent
acid refining. The purpose here is merely to get all the metallic
particles into one lump, without regard to karat or working qualities.
It is customary, but not essential, to carry out a few preliminary
steps, such as the removal of iron with a magnet, and the burning
out of grease and paper. In most shops the melting is done in a
tall sand crucible in a gold-melting furnace, with calcined borax
as a flux.
(There are on the market some prepared fluxes that are described
as capable of "refining" lemel or dirty filings, meaning that they
can remove certain of the impurities and produce a button that
can be hammered and rolled. Such a button might be of approx-
imately the same karat as the original metal, but on the other
hand it might be much higher or much lower, depending upon
circumstances that might change with every melt. It is obvious
that such "refining," while often a useful procedure, is quite dif-
ferent chemically from acid refining, in which fine gold is obtained,
free from silver or other alloying element.)
As we said, this work is traditionally done in a gold-melting
furnace. But if the reader has a platinum-melting torch, he is urged
to consider the following letter, written by a veteran refiner of

unusual skill:
"I used to melt my gold filings in a gold-melting furnace, using a
big sand crucible and considerable flux. When you count the cost of
the crucible, which can be used only once, and running the blower
and gas for two or three hours, it is rather expensive. I find it is better
to use the Hoke oxy-gas torch. Here is the idea:
"First burn the filings and take out iron with the magnet. Sometimes
I use nitric acid, but not always. Use a large cut-down crucible, the
same as for platinum, or break a cracked gold-melting crucible down
low on one side. If using a new crucible, dip it into water first, as this
prevents the crucible from breaking easily as moisture will be evenly
distributed.3
"Rub boric acid inside the crucible to prevent sticking. Let the
crucible stand in a clean iron tray while you melt, in case it tips over
or any filings blow out. Put in only a little material at first and use a
gentle flame, not much oxygen, gradually adding more material and
using a harder flame as it runs into a button.
"After all the filings have been melted down, run the flame around
the sides of the crucible, starting at the top, to melt down any little metal
particles that may cling to the wall.
"Let the melt cool somewhat, then grasp the crucible with round
tongs and turn it upside down and the button will fall out. If not,
loosen it gently with a sharp tool, so as not to damage the crucible,
which generally can be used several times.
"This plan is much cheaper than the old way, and takes no more
than twenty minutes."
MELTING CLEAN SCRAP
When clean gold or silver scrap is melted down for re-use, every
effort is made to keep the fineness and the working characteristics
as nearly as possible the same as before. The process therefore is
much more exacting than the melting of filings or scrap as a pre-
liminary to acid refining, and it is not surprising that the process
should have received considerable attention in the literature. Capil-
lon's paper Melting and Casting Some Gold Alloys, and E. A.
Smith's book Working in Precious Metals, are particularly helpful
in this connection.
3 Note: Many experienced workers will disagree with this. In general every
effort is made to keep crucibles as dry as possible.
Usually the scrap, after careful preliminary cleaning, is melted

in a gas-air furnace or in an electric furnace, and boric acid is a
popular flux, with or without powdered charcoal. It is well known
that repeated remeltings may change the karat of gold alloys, by
driving off part of the alloying elements, especially zinc. The
temperatures of fusion and pouring are important, and the size and
shape of the ingot molds have their influence on the success of the
operation. All these details are discussed fully in the paper and
book mentioned above.
Fine silver, in common with the platinum metals, absorbs con-
siderable oxygen when molten, and expels it on cooling, with
conspicuous spitting and crabbing. The same tendency to lesser
degree is observable in Sterling silver and in gold alloys that contain
much silver. The logical way to avoid this gas absorption is to
use an electric furnace and melt in a vacuum or in an inert gas,
but so far this practice has not become general. At present the
most convenient way to avoid this gas inclusion, which is prone to
cause blisters and brittleness, is to add a deoxidizer to the melt just
before pouring.
DEOXIDIZERS
There are several substances which, when added in small amounts

to gold and silver alloys, will take up the oxygen and pass it into
the slag. They are sometimes called scavengers, or oxide-removers.
Among the more successful ones are phosphor-copper and phosphor-
magnesium-copper. Calcium boride is a third, suggested for gold
alloys by E. M. Wise and tested by Capillon, who describes it as
well as several others, in the paper referred to.
We sometimes hear stories of mysterious substances, which when
added to gold alloys, to Sterling silver, aluminum bronze or other
metals, have a magic effect and produce amazingly good results.
Sometimes a workman will conceal this mystery treatment and
demand extra compensation because of his secret knowledge. In
almost every instance the magic is simply a good deoxidizer,
properly used.
The amount needed is very small. Thus, one ounce of phosphor-
copper is sufficient for a 100 pound melt; or 1/8 of an ounce for a
200 ounce melt. It should be added ten minutes or less before
the charge is pulled from the fire. It should be introduced in the
form of small grains, and pushed under the surface and forced
to mix well with all parts of the melt. The phosphorus unites with
the oxygen and passes off as fumes or slag; the copper goes into the
melt. In exact work, this small additional weight of copper must
not be forgotten.
These deoxidizers should not be used with platinum, as the
phosphorus will unite with it, instead of going off into the slag or
escaping in the fumes.
FLUXES
The word flux comes from the root meaning to flow, but it has
now come to apply to a wide variety of substances that are added
to molten metals for one purpose or another.
Fluxes may be divided into four groups. Those (1) that introduce
oxygen are used in removing oxidizable impurities from fine gold.
Potassium nitrate (called saltpetre or nitre) and sodium nitrate
(called Chile saltpetre) are the best of this group. The reducing
agents (2) are those that remove oxygen by reacting chemically
with oxides that are already present. To this class belong the many
carbonaceous and organic materials charcoal, sugar, dry sodium
carbonate (called soda ash), argol, dry potassium carbonate (called
pearl ash), potassium cyanide, and so on. Next we have those
(3) that neither add oxygen nor take it away, but which form a
coating that excludes it. Borax and boric acid are the commonest
of these, and they further dissolve and flux off any base-metal
oxides that may be present. To this same group belong fluorspar
and powdered glass. Then (4) come the so-called tougheners—
ammonium chloride, mercuric chloride and copper chloride, used in
purifying gold that is almost fine.
These materials can have a powerful influence for either good or
evil, and have commanded considerable attention in the literature.
They have been mentioned often in this volume; in E. A. Smith's
book, for example, several pages are devoted to the occasions and
manner of their use; and the dealers in crucibles and supplies for
metal workers are generous with helpful practical advice.
CASTING PRECIOUS METALS
The cire perdu or lost wax process of casting is very old, as one
may learn by visiting any art museum. It consists of making a wax
pattern, setting it in some material such as plaster of paris (called
the investment), then heating the investment until the wax melts
and runs away, and finally pouring molten metal into the cavity
thus provided.
The modern dental laboratory has done much to transform this
old art into a science a science by means of which intricate designs
can be reproduced accurately in metals of high melting point. In-
vestment materials capable of withstanding high temperatures have
been developed, and many tricks of technique have been learned.
Platinum and high-karat gold jewelry (as well as dentures) are
being cast to-day, in forms that were once considered impossible,
by special processes described as economical and artistically satisfy-
ing. Patent applications cover different phases of this technique,
about which more can be learned from dealers in metals and manu-
facturers of melting equipment.
MAKING UP NEW ALLOYS
This is the most exacting and the most interesting of the gold-
melter's tasks. Within the last generation a host of new alloys
have been made white golds, pink golds, high-temperature dental
golds, and others. Perplexing problems of manufacture and
manipulation have arisen. The formulas of the different alloys,
the conditions under which they are melted, poured, annealed and
machined all these have received deep study, as can be seen by a
glance at the literature, to which the reader is referred. One of
the best sources of information is E. A. Smith's book Working
in Precious Metals, but later articles, appearing in the technical
magazines, bring the subject up to date.
NEW LIGHT ON OLD METALS
The precious metals have been known to mankind for at least

5,000 years, and the melting and working of them was an art even in
the days of King Solomon. But the recent studies on these metals
have brought to light a wealth of new facts facts that will grant
wider scope to the artist, increased precision to the scientific worker,
financial advantage to the wholesale producer, and increased re-
sponsibility to the man in the furnace room.
The craftsman of the past, hammering each article by hand, could
tell by the feel of the metal when it was time to anneal, and if
different articles responded differently, no harm was done. But
with modern machine production, there must be no individual varia-

tion. We must know exactly when to anneal, and to what tem-
perature, and all pieces will be treated the same. Color, karat,
hardness, springiness, resistance to tarnish, and so on all must be
regulated and understood. A much more intimate and scientific
knowledge of the nature of our metals and alloys is now demanded.
This intimate knowledge could not have been secured without
certain useful tools, notably the microscope, the pyrometer, and the
X-ray. The story of what they have revealed constitutes modern
metallurgy.
When old-timers pick up a volume of modern metallurgy they are
at first repelled by the technical terms and scientific diagrams.
However, if they will start at the beginning and follow the au-
thor's train of thought, they will have the happy experience of
recognizing old truths in a new dress. They will find explanations
of facts that they had themselves learned by accident or experience.
They will be able to say: "Why, that is something I knew all along
but could not quite explain!" And finally, of course, they will see
how to turn these new facts to their own advantage.
CHAPTER B
HAZARDS
The National Safety Council. Comparison of chemical and heat burns.

First aid. First aid treatment for chemical eye burns. Neutralizers.
Alkalies or caustics. Hydrochloric or sulphuric acid. Hydrofluoric acid.
Nitric acid. Nitrous fumes. When handling sulphuric acid. Shower
baths. Cyanide poisoning. Mercury. Hazards of the melting room;
hazardous arrangement; crucibles; oxygen cylinders. Stains on the skin.
Gas masks. Some articles to read.
* * • •
Several of the most potent chemical agents known to man—
strong acids, caustic soda, and the deadly cyanide are used in the
refining of precious metals. Fumes such as sulphur dioxide, am-
monia, chlorine, mercury vapor, and the brown gases produced by
nitric acid, every one of which is a man-killer if present in sufficient
concentration, must be handled and disposed of safely. We also
have the hazards of the melting-room the white hot crucible, the
molten metal, and the terrific temperatures of the oxygen flame.
However, it is a fact that one hears only rarely of a serious
accident in this work. Probably that is because precious metals
are rarely entrusted to irresponsible workers. Possibly, also, their
very preciousness makes for a sobriety of mind and builds up a
psychology of carefulness. But the beginner must understand
these hazards and their avoidance. Hence this chapter.
The National Safety Council, Inc., is a non-profit association
devoted to the study of safety measures in industry. Among other
activities it publishes pamphlets on the hazards connected with
different processes, and recommends measures for their avoidance.
The association maintains offices at 20 North Wacker Drive, Chi-
cago, 111., and all employers of labor should be familiar with its
efforts.
We are indebted to the Council for its permission to quote the
following paragraphs, taken from its copyrighted pamphlet
Chemical Burns, Their Nature and Treatment.
320
HAZARDS 321
"Chemical burns are often thought to be similar to heat burns, but

this is true only in that both may cause destruction of body tissue. The
object causing a heat burn may cool, or the person may withdraw from
the object, and destruction of the tissues ceases. When contact is made
with a chemical substance, the contact continues as long as the agent
is present in sufficient concentration to destroy the tissue. Absorption
must also be considered in treating chemical injuries, as this may cause
illness or even death.
"The severity of chemical burns depends upon: (1) the concentration

of the chemical, and (2) the duration of the contact. Generally, the
first factor cannot be controlled, but the second may, and proper first
aid rendered immediately is of outstanding importance.
"The successful handling of chemical burns calls for the complete
removal of the irritating agent, or the administration of such treatment
as will render the chemical inert.
"The importance of immediate first aid for chemical burns is in-

dicated by the following experiments to test the action of several acids
at different strengths:
"Four tests were made using a 70 percent nitric acid, a 76 percent
sulphuric acid, a 37 percent hydrochloric acid, and a 99 percent acetic
acid on three persons. In each case the burning could be noticed in
five to fifteen seconds. Using the same acids in 25 percent concentra-
tions, no burning was noted.
"To compare the effects of neutralization (using a 5 percent solution
of sodium carbonate) with those of water, extensive tests indicated that
the use of water, either in a large tank or from a running tap, was
considerably better in treating acid burns than neutralization.
"If a patient's clothing becomes soaked with corrosive fluid, burning

will continue until the clothing is removed. Clothing should therefore
be removed immediately, but efforts should be made to keep the in-
jured person warm. The acid or alkali that remains on the skin should
be removed by flushing copiously but gently with large quantities of
water—warm if possible. Shock will be lessened if the water is warm,
but speed is of greater importance, and if possible a stream of running
water should be directed onto the burned area so that all the chemical
may be washed away. The burned area may then be treated as any
heat burn and covered with any suitable protection, preferably a water-
soluble jelly.
FIRST AID
"Persons giving first aid often make the mistake of first applying
an antidote. This treatment is not nearly so important as proper flush-
ing with water. The immediate use of a neutralizing agent on an
unwashed burn may increase the injury.
"According to some authorities, one of the most effective agents to
combat burns is tannic acid. (This applies mostly to third degree or
very serious burns. Sterile vaseline or anhydrous lanolin are often
recommended for burns of lesser degree.) Many doctors are now
using gentian violet in the form of a jelly for first, second, or third
degree burns.
FIRST AID) TREATMENT FOR CHEMICAL EYE BURNS
"The seriousness of an eye injury is determined by (I) the strength

and character of the chemical, (2) the length of time before treatment,
and (3) the type of the first treatment. The importance of immediate
first aid should be stressed. The mechanical removal of the chemical
is very important. This can best be accomplished by careful washing.
Water may be poured from a cup onto the bridge of the nose and
allowed to run over the eyeball, which is kept exposed by holding the
lid open, to flush out the inner surfaces of the lids. Do not pour water
directly onto the eyeball. If a hose is used, be sure to use a slow stream
so that the chemical may not be driven farther back under the eyelids.
NEUTRALIZERS
"There has been considerable controversy relative to the neutraliza-

tion of acids and alkalies with suitable neutralizers. Recent research
indicates that the best results may be obtained by the use of large
Remember: For chemical burns, wash immedi-

ately with plenty of plain water.
quantities of plain, clean water. It is impossible to use too much water.

The use of a soothing ointment or jelly (as recommended by the
medical director or physician) in the after treatment has proven
HAZARDS 323
beneficial. While the washing is going on, have the physician on his
way to the scene, that he may give any treatment that he may deem
necessary.
"The following paragraphs give brief descriptions of some important
corrosive agents and suggestions for further first aid treatment remem-
bering always that flushing the burned area with water is the im-
portant first treatment, and that a doctor should be brought upon the
scene as soon as possible.
ALKALIES OR CAUSTICS
"Here too, experiments have indicated that thorough washing with

water, as a first aid remedy, is more effective than neutralization. No
time should be lost before washing. Bathe freely with water, then
apply a 2% solution of acetic acid, weak vinegar, or lemon juice.
(Use care when making up caustic mixtures; for example the solution
on page 17 with which you removed shellac from filings. Add the lye
or potash slowly, and stir to avoid local overheating. Use similar care
when ammoniating an acid solution, or acidifying an alkaline one—
work slowly, stir well; and if the mixture spatters on skin or clothing,
wash it off immediately with much plain water. Ed.)
HYDROCHLORIC OR SULPHURIC ACID
"Bathe freely with water and apply a saturated solution of bicar-

bonate of soda.
HYDROFLUORIC ACID
"Bathe freely with much water. Apply powdered sodium bicarbonate

or a 5% solution of ammonium carbonate. Boric acid is also recom-
mended. Hydrofluoric acid is a violent corrosive, and skin burns heal
slowly.
NITRIC ACID
"Bathe freely with water, then apply a saturated solution of bicar-

bonate of soda. One doctor advises scrubbing until as much as pos-
sible of the yellow pigment is removed, using soap, ammonia, and
water.
NITROUS FUMES
"The fumes referred to are the red (or brown) fumes present when
nitric acid is being manufactured or used. Beware of them. Don't
think, even if you are big and healthy, that you can take chances with
them. Don't think for one minute that you can get used to them. If
you do get caught in them and have a suffocating cough, see a doctor.
WHEN HANDLING SULPHURIC ACID
"Never pour water into strong sulphuric acid. To dilute the acid,
pour it into the water, slowly, and with constant stirring. Keep the
acid away from combustible materials (such as wood or paper); it
may start a fire. Keep acid containers away from heat, and out of the
sun, to prevent building up pressure in them. Empty containers should
be flushed out with water to prevent burns to those who handle them.
Keep plenty of fresh water near at hand with which to wash acid off
the body in an emergency."
SHOWER BATHS
Laboratories and plants that handle acids in quantities are

equipped with shower baths under which a worker can quickly step
in case of emergency; a good type is one with a valve that operates
when the worker steps on a platform. In most plants it is thought
to be enough if a tub or barrel of water is available.
CYANIDE POISONING
The R. & H. Chemicals Department of the du Pont Company

publishes a trade paper called Modern Metal Finishing, in the June
1936 issue of which a new antidote for cyanide poisoning is given.
We quote:
ANTIDOTE
"Always have on hand a liter (quart) of a 1 percent solution of

Sodium Thiosulfate and a carton of Amyl Nitrite Pearls. Directions for
treatment:
1. Have patient inhale content of an amyl nitrite pearl.
2. Cause vomiting by giving soap water or mustard water.
3. Give 500 cc (about a pint) sodium thiosulfate solution by mouth.
"Repeat the above procedure every fifteen minutes. If necessary give

artificial respiration until physician arrives.
"The rapidity of death from cyanide poisoning has been justly
emphasized, for most patients die within 30 to 60 minutes. On the
other hand, many linger for several hours; one lived over three hours.
As long as the victim's heart still beats, the clinician should consider
the case hopeful and treat it without delay.
HAZARDS 325
"Be sure that cyanide never comes in contact with open wounds or
skin abrasions. Always wash the hands well with running water after
handling cyanide. Food should never be stored, handled, or consumed
during cyanide operations. Keep cyanide away from all acids. Keep
the container tightly closed when not in use; cyanide is weakened by
exposure to air."
MERCURY
Mercury poisoning can develop when mercury or amalgams are

handled or heated, and when corrosive sublimate (mercuric chlo-
ride) is used as a flux in the melting of gold. The fumes appear
at a low temperature and are extremely poisonous. They are heavy,
so heavy that they are not carried off by a weak draft or a small
exhaust fan; instead they settle to the floor, and seep through the
crevices of the furnace or flue, and accumulate in the building The
exhaust system must be designed with this fact in mind.
For five cents the Superintendent of Documents, United States
Government Printing Office, Washington, D C, will send a 4-page
pamphlet entitled Mercury Poisoning, Its Cause and Prevention,
issued by the United States Department of Labor. This is a concise
statement explaining how mercury poisoning occurs, the signs and
symptoms and what to do about it, and preventive measures. Any
person who works with mercury, its salts or its amalgams, should
obtain copies and post them where they will be read frequently.
To the precious metal refiner, the greatest hazard from mercury
is probably that involved in opening the retort; see Chapter XVI.
When the retort is opened it may still contain some mercury vapor.
Unless the operator thoroughly understands this situation, he may
receive enough mercury to cause poisoning.
Bare hands should not be used when handling mercury bearing
materials. Men should not smoke when there is a possibility of
mercury fumes in the air, and food should not be eaten at a retort
plant.
HA7ARDS OF THE MELTING ROOM
Hazardous Arrangement. The National Safety Council esti-

mates that hazardous arrangement is the principal mechanical cause
of all serious injuries. Unsafe clothing and failure to use proper
goggles can be included in this grouping Unsafe planning or
work, and unsafe methods and processes, are most frequently men-
tioned. For example, it is bad management to place the ingot
mold so far from the furnace that the operator will have to walk
many steps carrying a crucible full of molten gold. It is bad practice
to leave obstructions on the floor of any factory, most especially
ones in which hazardous materials such as strong acids or molten
metals have to be carried about. It is bad practice to omit the guard
crucible or other device to catch molten material in case a crucible
breaks or molten flux boils over inside a furnace. It is bad practice
to place gas cocks, electric switches, or other controls where they
cannot be easily reached in an emergency. It is bad practice to
leave off asbestos gloves and protective shin-guards and foot-gear
when making large castings. Instances of bad practice can be
multiplied indefinitely, but a little intelligent planning on the
part of individual shop foremen, and vigilance on the part of the
workmen, can reduce them to a minimum.
Crucibles. Many of the hazards of the melting room are con-
nected with crucible failure. In Chapter A of this Appendix we
mentioned the Crucible Manufacturers' Association, and their very
informative pamphlet How to Cut Crucible Costs, which gives
many safety suggestions. As we said before, this pamphlet may be
had without cost from your crucible dealer, or from the Association
at 90 West street, New York, N. Y.
Crucible users should realize, says this pamphlet, that a crucible
is made of a material that undergoes certain changes under the
influence of heat; that it must do work that no metal is capable of
doing that is, when the metal is molten and the crucible itself
is at almost a white heat, it must still be strong enough to hold
the heavy liquid metal and carry it safely to the mold. And it must
do this over and over again. Therefore it is entirely within reason
to say that a crucible must be handled carefully at all times, and
particularly so when it is new. Suggestions on the handling of
crucibles are given, some of which have already been mentioned in
Chapter A. Proper storage with precautions against dampness are
important. The use of properly shaped tongs is important. The
reader is advised to have this pamphlet posted in a prominent place,
and to refer to it often.
Oxygen Cylinders. The oxygen gas used when melting plati-
num is compressed in steel cylinders. The Interstate Commerce
Commission regulations governing the conditions under which tanks
HAZARDS 327
can be shipped are such as to remove practically all hazards con-

nected with tank construction. In addition, most cities have Fire
Department regulations on the storage and handling of these tanks.
If these regulations, and the suggestions offered by the oxygen
dealers, are followed, the user need have no fears. It will be recalled
that oxygen tanks can be found in almost every hospital that treats
pneumonia, asthma, heart ailments, and some other disabilities;
hence it is clear that such tanks bear a reputation for safety.
Cylinders are provided with a safety disc that will release the
pressure in case the cylinder becomes unduly heated as for instance
if the building should take fire and the gas itself is not poisonous.
Keep tanks in a fairly cool place, not in the direct sun nor too
close to a radiator; use no oil on any part of them; follow the
advice of the dealer and you will have no hazard.
The regulators (pressure-reducing valves) used to regulate the
oxygen flow, if handled as directed by the seller will work well
for many years. Use no oil on any part of the regulator, torch,
tubing, cylinder, or any connection or piping that carries com-
pressed oxygen. This is important, because compressed oxygen can
combine with oil to form an explosive mixture.
Pressure reducing valves have been known to rupture under abu-
sive mismanagement. When this happens all the oxygen escapes.
However, properly designed instruments are so made that there is
no hazard to the user.
STAINS ON THE SKIN
Shortly after handling a solution that contains gold, the worker

is almost sure to find purple stains on his fingers. Even a minute
amount of dissolved gold will make a conspicuous and character-
istic stain. Platinum and silver solutions cause unsightly black
stains. Nitric acid produces a yellow stain. These stains themselves
offer no special hazard to the worker, but ill-advised efforts to
remove them can be dangerous.
Careful rinsing and scrubbing of the hands while the work is
in progress will minimize these stains, all of which, once formed,
are difficult to remove and usually remain until the outer layer of
skin has been replaced.
Some foolhardy workers remove the purple gold stains by rub-
bing the spots with a lump of wet cyanide. This is unwise for
two reasons: first there is the obvious danger of absorbing the

poisonous cyanide through the skin; second, cyanide is sufficiently
caustic to cause the skin to chap and peel, and it leaves the hands
so roughened as to be rather worse off than before, both in ap-
pearance and in susceptibility to poisons and infections.
Different people respond differently to cyanide poisoning, but
all authorities agree that it can be absorbed through small abrasions
or tiny sores in the cuticle. Hence the worker is advised to use
every care to avoid these metal stains, rather than to use drastic
methods of removal.
GAS MASKS
The small shop, especially if it is equipped with a proper hood,

will have little need for a gas mask. A wet handkerchief draped
over the face like a highwayman's mask will serve most purposes.
The large plant, however, will probably want one. The chemical
supply houses provide several designs for use with the various
chemicals encountered in this work. In general it is well to depend
upon fans and blowers, rather than to permit fumes to collect in
such concentration as to suggest the need for a mask.
SOME ARTICLES TO READ
Publications of the National Safety Council, Inc., 20 North Wacker

Drive, Chicago, Ill., especially CHEMICAL BURNS, THEIR
NATURE AND TREATMENT.
CYANIDE POISONING. Modern Metal Finishing, issue of June,
1936; published by the R. & H. Chemicals Dept., E. I. du Pont
de Nemours & Co., Inc., Wilmington, Del.
MERCURY POISONING; ITS CAUSE AND PREVENTION. U. S.
Department of Labor. Price $0.05 On sale by the Superintendent
of Documents, Government Printing Office, Washington, D. C.
MERCURY POISONING. Pages 36-40 of the Information Circular
No. 6966, U. S. Bureau of Mines, "Occurrence and Treatment of
Mercury Ore at Small Mines," by M. W. von Bernewitz. This is
out of print, but copies can be seen at many public libraries.
HEALTH HAZARDS IN CHEMICAL INDUSTRIES. By Henry
F. Smyth, Jr., of the Industrial Toxicological Department, Mellon
Institute of Industrial Research, Pittsburgh, Pa. Published by
THE CHEMIST (bulletin of the American Institute of Chemists,
233 Broadway, New York), issue of April, 1939.
HAZARDS 329
HOW TO CUT CRUCIBLE COSTS. Obtainable gratis from the

Crucible Manufacturers Association, 90 West St., New York City.
CARE OF OXYGEN, ACETYLENE, AND OTHER COMBUS-
TIBLE GAS CYLINDERS. Distributed by the Air Reduction
Sales Co., which has offices in many cities. General Office, 60 East
42nd St., New York City.
CHAPTER C
LAWS AND REGULATIONS
Federal Licenses for handling gold. Regulations concerning Govern-

ment purchase of precious metals. National stamping law covering
gold, silver and their alloys. Laws for stamping platinum. The Com-
mercial Standards. Local ordinances. Standards in foreign countries.
# # # #
FEDERAL LICENSES FOR HANDLING GOLD
Shortly after the United States went off the gold standard in
1933, the Treasury Department instituted regulations for the licens-
ing of persons who acquire, transport, melt or treat, export, import
or earmark gold, or hold it in custody for foreign or domestic
accounts. This obviously is of interest to the jeweler and dental
technician handling gold. No fee is charged for these licenses,
and the legitimate dealer or manufacturer has no difficulty in
obtaining one.
At present writing, the person holding only small amounts of
gold at a time requires no license; however, the regulations are
subject to change without notice, and the reader is advised to obtain
copies of the current regulations from time to time. These may be
had from any Federal Reserve Bank, or from any Mint or Assay
Office.
No federal license is required for the handling of platinum or
silver.
In addition to the Federal license, some States and a few
municipalities require local licenses of buyers of old gold, of pawn-
brokers, dealers in second hand jewelry, and the like.
REGULATIONS CONCERNING GOVERNMENT PURCHASE OF PRECIOUS METALS
The United States Mints and Assay Offices will purchase gold,
and usually silver as well. They are not authorized to buy platinum
metals, and if these metals happen to be included with shipments of
330
LAWS AND REGULATIONS 331
gold or silver, no payment will be made for them, and the refining
charge may be increased by their presence.
The forms in which deposits will be accepted, the prices paid,
etc., are all subject to change. At present deposits will be received
in amounts of not less than one Troy ounce of gold, and the material
must run at least 200 fine by assay. A copy of current regulations
can be obtained for the asking from any Mint or Assay Office
Because the Government does not seek a profit on these opeiations,
the price it pays is generally higher than that obtainable from
commercial refiners However, there is a wait of several days
before the payment is made, and the conditions mentioned above
are such as to exclude some kinds of materials
NATIONAL STAMPING LAW COVERING GOLD, SILVER AND THEIR ALLOYS
The fact that gold and silver are used as money has meant that
their stamping or marking has been regulated by law for many
centuries. Such laws are usually thought of as a protection to
the purchaser of jewelry or silverware, but at the same time they are
a protection to the jewelry manufacturers as a group, since they
tend to sustain public confidence in the industry.
Handy and Harman, precious metal dealers of Bridgeport,

Connecticut, have published a book called Handy Book for
Jewelers in which some of these laws are quoted in full.
The United States National Stamping Law, forbidding the im-

portation, exportation, or interstate commerce of falsely or spuriously
stamped articles of merchandise made of gold or silver, or their
alloys, was enacted June 13, 1906 The full text of the act may be
found in the Handy Book and in almost any law library or big
public library. Ask for Rev. Stat. U. S., vol. 34, pt. 1, p. 260, 59th
Cong., 1st Sess., Public Law 226. A summary quoted from the
Handy Book covers its main provisions, as follows:
Gold. If an article is made of gold and is stamped gold, it must
also bear a quality mark such as "10 karat" (10-K), "14 karat" (14-K).
If an article of gold is given a quality mark, the fineness by assay
must not be lower than:—
Watch Cases and Flatware 003 less than stamped quality.
Other articles, not including solder . .0208 ( 1/2 karat) less than the
stamped quality.
However, the assay of a complete article, including solder, must not
be more than .0417 (1 karat) under the stamped fineness per karat.
For example, the gold in a 14-karat watch case, free from solder must
be at least .5803 by assay. The entire case, including solder, must assay
at least .547 (13 karat). A gold ring, not soldered, stamped "14-K"
must assay at least .5625 (131/2karat). The gold in a brooch stamped
"10-K" must assay at least .3958 (91/2karat) and the entire brooch,
solder and all, must assay at least .3750 (9 karat).
Silver. The silver in any article stamped "Sterling Silver" should

assay .925, and the silver in an article marked "Coin Silver" should
assay .900. The silver in an article, not including solder, must not be
less than this by more than .004. For example, an article marked
"Sterling Silver," free from solder, must assay at least .921.
Soldered parts must not reduce the assay of the entire article, includ-
ing solder, by more than .010 under the standard assays of .925 and .900,
respectively, for sterling silver and coin silver. For example, an article
marked sterling silver when melted, including solder, must assay at
least .915.
Gold and Silver Used with Inferior Metals. An article made

of an inferior base metal combined with gold may be marked "Rolled
Gold Plate," "Gold Plate," "Gold Electroplate" or "Gold Filled" as the
case may be. If any mark is used to indicate the fineness of the gold,
it must be accompanied by one of these terms.
For example:"10 K. Rolled Gold Plate," "14 K. Gold Filled."
The words "Sterling" or "Coin" alone or in combination with other
metals cannot be used to describe a plated article.
For example:—"Sterling Plate" is prohibited.
Penalties. Violators may be fined up to $500 or be imprisoned
up to three months, or both, at the discretion of the court.
LAWS FOR STAMPING PLATINUM
For some years after the introduction of platinum as a jewelry

metal there was confusion regarding its marking, and much mis-
branding, adulteration and fraud took place. Three of the States
in which considerable platinum jewelry was manufactured New
Jersey, New York, and Illinois passed laws regulating the stamp-
ing of platinum and its alloys. And finally, on June 20, 1938, the
National Bureau of Standards made effective a series of regulations,
based on these State laws, that covers the entire nation. The full
text of the New York State law will be found in the Handy Book
Copies of the new National Standard, known as Commercial
Standard 66-38, may be obtained from the Superintendent of Docu-
ments, Washington, D. C, for 5c. Briefly its main provisions are
as follows:
Articles may be stamped "platinum" or "plat," provided all parts of
the article purported to be of platinum shall constitute at least 985/1000
parts platinum. If platinum assaying 985/1000 parts pure has been
combined with gold the article must be stamped with the karat mark
indicating the fineness of the gold in conjunction with the word or
abbreviation of platinum, as "14K & Plat." When platinum is alloyed
with iridium, palladium, ruthenium or osmium, these articles must be
marked in fractions designating the content of these metals. Merchandise
bearing quality marks must also be stamped with a registered trade
mark.
THE COMMERCIAL STANDARDS
The National Bureau of Standards, in co-operation with members

of the precious metal industries, has formulated several other
"Commercial Standards" that may be of interest to the reader.
These Standards have their origin in the Bureau, rather than in
the halls of Congress or the State Legislatures, and are a crystalliza-
tion of established trade practices rather than any sort of govern-
mental regimentation. They are subject to amendment when the
interested industry feels that changes are advisable, and they are
enforced with the co-operation of the Federal Trade Commission.
Copies of the texts may be obtained from the Superintendent of
Documents, Washington, D. C, price 5c. Other standards may be
proposed and possibly adopted in the future; the following are now
effective:
Marking Articles made of Silver in Combination with Gold—

Commercial Standard 51-35.
Marking Articles Made of Karat Gold Commercial Standard
67-38.
Marking of Gold Filled and Rolled Gold Plate Articles other
than Watchcases Commercial Standard 47 34, with amendments
of February 25, 1939. Bulletin TS 1942, of July, 1933, defines the
terms "Gold Filled" and "Rolled Gold Plate "
LOCAL ORDINANCES
In efforts to improve trade practices and to discourage traffic

in stolen goods, a number of cities and states have passed laws of
interest to the precious metal worker. Some of these regulate the
sale of second hand jewelry, especially second-hand watches. Others
apply to itinerant buyers of old gold Others require licenses of
watch repair men or regulate the handling of jewelry "on memoran-
dum " Information on such local ordinances can usually be ob-
tained from the local Chamber of Commerce, or from the National
Better Business Bureau in the Chrysler Building, New York, N Y
STANDARDS IN FOREIGN COUNTRIES
E A Smith gives the following facts in his book Wording in

Precious Metals, page 395 We quote
Platinum. For platinum the standard 950/1000 has now found
general acceptance in most countries for the purpose of a quality standard
for jewelry (The International Jewelers' Congress at Rome, May 5 7th,
1933, confirmed 950/1000 as the International Platinum Standard See
report in Goldsmiths Journal, June 1933, page 257)
Gold. For coin the gold standard in most countries is 900, excep
tions being Great Britain, which has adopted 916 6 This higher
standard is also in use in Brazil, Chile, India, Newfoundland, Peru,
Portugal and Turkey, whilst Egypt is 875
For gold plate and jewellery the most universal standards are 750
and 585 (or 583)
Lower standards of 375 are used by Great Britain, Esthonia and
Tunisia, and 333 by Germany, Italy, Bulgaria and Portugal
Silver. The standard of 900 is very general for silver coin In
Great Britain the standard is 925 (temporarily lowered to 500)
In some countries lower standards are in use, especially for subsidiary
coins.
For silver wares various standards are in use. The 925 standard is
the one current in the British Empire and the United States, the so
called Sterling silver. The silver exported by England, too, is generally
of this standard.
France manufactures 950 in larger objects for the home market, and
frequently 800 in smaller articles.
The Scandinavian countries, the Netherlands, Czecho-Slovakia
Austria, Portugal, Egypt and Bulgaria use standards between 850
and 830.
Germany manufactures 800 for the home market as well as for
export; and in many countries, owing to German influence, the 800
standard is current. Manufacturers in Germany, however, are discussing
a change from 800 to 835 as the lower quality alloy tarnishes somewhat
readily. The 835 on the contrary is regarded as satisfactory.
The Canadian Precious Metals Marking Act is quoted in the

Handy Book The magazine The Manufacturing Jeweler for Janu
ary 4, 1940, page 7, gives a helpful summary of the principal pro
visions. Complete copies of the Act may be obtained from the
King's Printer, Ottawa, Canada; price $0.10
CHAPTER D
A LIST OF DEALERS
Buyers of precious metals. Dealers in chemical apparatus and supplies.

Dealers in supplies for Jewelers. Dealers in special
equipment and machinery.
# # # #
This list is by no means complete. There are many reputable
dealers in addition to these few. Several of the firms here listed
have branches in other cities. The telephone Red Books, and
the trade catalogues mentioned in Chapter XX, will provide
many more names.
Buyers of Precious Metals

With few exceptions, the firms listed below also sell gold, silver
and platinum in one form or another, for industrial or scientific
use. The reader is advised, before shipping precious metals, first
to confirm these addresses, and second, if dealing with persons not
on this list, to ask his bank for a report on their credit rating.
CALIFORNIA
Pacific Platinum Works, 814 So. Spring St., Los Angeles.

Wildberg Brothers Smelting & Refining Co., 635 So. Hill St.,
Los Angeles.
Shreve and Co., Post and Grant Sts., San Francisco.
S. B. Gracier & Sons, 212 Stockton St., San Francisco.
Western Gold and Platinum Works, 589 Bryant St., San Francisco.
Wildberg Brothers Smelting & Refining Co., 742 Market St., San
Francisco.
CONNECTICUT
Handy & Harman, Bridgeport.

J. M. Ney Co., Hartford.
336
A LIST OF DEALERS 337
ILLINOIS
Thomas ]. Dec & Co., 55 East Washington St., Chicago.

Goldsmith Bros. Smelting & Refining Co., 58 East Washington St.,
Chicago.
Kilgallon & Co., 31 North State St., Chicago.
LOUISIANA
H. Schmidt & Son, 208 Chartres St., New Orleans.
MASSACHUSETTS
Attleboro Refining Co., Union St., Attleboro.

John Hood Co., 178 Tremont St., Boston.
Eastern S. & R. Co., 107 West Brookline St., Boston.
Glines & Rhodes, North Attleboro.
Jewelers' & Silversmiths' Co-operative Refining Co., North Attle-
boro.
OHIO
Cincinnati Gold and Silver Refining Co., 523 Elm St., Cincinnati.
Peter Henry & Son, 439 Race St., Cincinnati.
G. W. Seifried Co., 127 Opera Place, Cincinnati.
MICHIGAN
C. L. Pratt, Jr., 15 Gratiot St., Detroit.
MINNESOTA
General Refiners, Inc., 27 No. Fourth St., Minneapolis.

Spyco Smelting & Refining Co., 55 So. Third St., Minneapolis.
NEW JERSEY
Irvington Smelting and Refining Works, Nye Ave., Irvington.

American Platinum Works, 225 N. J. Railroad Ave, Newark.
Balbach Smelting & Refining Co., 580 Market St., Newark.
Baker & Co., 54 Austin St., Newark.
Clinton Refining Co., 91 East Kinney St., Newark.
Enright Refining Co., 42 Marshall St., Newark.
L. Lelong & Bro., 449 Washington St., Newark.
H. A. Wilson Co., 97 Chestnut St., Newark.
NEW YORK
Hoover & Strong, Inc., 119 W. Tupper St., Buffalo.

Williams Gold Refining Co., 2978 Main St., Buffalo.
Sigmund Cohn, 44 Gold St., New York.
Joseph B. Cooper & Son, 26 John St., New York.
Dentists' Supply Co., 220 West 42, New York.
Goldsmith Bros. Smelting & Refining Co., 74 West 46, New Yorl
Moe Hamerschlag, Inc., 174 Park Row, New York.
Handy & Harman, 82 Fulton St., New York.
J. F. Jelenko & Co., 136 West 52, New York.
Johnson Matthey & Co., 15 West 47, New York.
Kastenhuber & Lehrfeld, 24 John St., New York.
Kushner & Pines, 21 West 46, New York.
Orange Bros., 40 John St., New York.
I. Stern & Co., Inc., 218 West 40, New York.
Charles W. Williams, 106 Fulton St., New York.
PENNSYLVANIA
J. Bishop & Co., Malvern.
T. B. Hagstoz & Son, 709 Sansom St., Philadelphia.
Masel Dental Laboratory, 1108 Spruce St., Philadelphia.
S. S. White Dental Co., 211 S. 12, Philadelphia.
Dentists' Supply Co., York.
MARYLAND
Gilbert Cummins & Co., 1 North Eutaw St., Baltimore.
RHODE ISLAND
Geo. M. Baker Co., 91 Page St., Providence.

Conley & Straight, Inc., 236 Eddy St., Providence.
Pease & Curren, 780 Aliens Ave., Providence.
Horace Remington & Son Co., 91 Friendship St., Providence.
TEXAS
Southwest Smelting & Refining Co., 1809 Main St., Dallas.

Southwest Smelting & Refining Co., 108 E. 10th St., Fort Worth.
WASHINGTON
Speyer Smelting & Refining Co., Medical & Dental Bldg., Seattl
Dealers in Chemical Apparatus and Supplies

In addition to the following large organizations, there are many
small chemical supply houses throughout the country, most of
which carry in stock the chemicals and equipment used in refining,
or can obtain them on short notice. It is advisable wherever pos-
sible to purchase from a local dealer, since transportation costs are
often a large part of the cost of chemicals.
CALIFORNIA
Braun Corporation, 2250 East 15th St., Los Angeles.

Braun-Knecht-Heimann Co., 576 Mission St., San Francisco.
Justinian Caire Co., 268 Market Street, San Francisco.
COLORADO
Denver Fire Clay Co., Box 5510, Denver. (Branches in Salt Lake
City and El Paso.)
ILLINOIS
Central Scientific Co., 1700 Irving Park Blvd., Chicago.

E. H. Sargent & Co., 155 East Superior St., Chicago.
MARYLAND
Gerald K. Heller, Eagle and Smallwood Sts., Baltimore.

James B. Ferguson, 1642 Warner St., Baltimore.
MASSACHUSETTS
Central Scientific Co., 79 Amherst St., Cambridge.

Macalaster Bicknell Co., Washington and Moore Sts., Cambridge
MISSOURI
Heil Corporation, 210 S. Fourth St., St. Louis.
NEW JERSEY
New Jersey Laboratory Supply Co., 235 Plane St., Newark.

NEW YORK
Buffalo Apparatus Co., 180 Main St., Buffalo.

J. & H. Berge, 145 Hudson St., New York.
Eimer & Amend, 18th St. and Third Ave., New York.
Emil Greiner Co., 161 Sixth Ave., New York.
Palo-Myers Inc., 81 Reade St., New York.
Will Corporation, Rochester.
PENNSYLVANIA
Arthur H. Thomas Co., West Washington Sq., Philadelphia.

Fisher Scientific Co., 709 Forbes St., Pittsburgh.
VIRGINIA
Phipps & Bird, Inc., 915 E. Cary St., Richmond.
WEST VIRGINIA
B. Preiser Co., 109 Washington St., Charleston.
CANADA
Fisher Scientific Co., 898 St. James St., Montreal.

Central Scientific Co. of Canada, 119 York St., Toronto.
Dealers in Supplies for Jewelers

For a much more extended list of dealers, the reader is referred
to the Jewelers Buyers Directory, published biennially by the Jew-
elers' Circular-Keystone, 239 West 39, New York City; and to the
Jewelers Handbook, published annually by Walter B. Frost & Co.,
42 Weybosset St., Providence, R. I.
CALIFORNIA
Friedman-Gessler Co., 220 W. 5th St., Los Angeles.

E. W. Reynolds Co., 315 W. 5th St., Los Angeles.
Justinian Caire Co., 268 Market St., San Francisco.
COLORADO
Denver Fire Clay Co., Box 5510, Denver. (Branches in Salt Lake
City and Utah.)
FLORIDA
Clifford R. Maxwell, 117 N. E. 1st St., Miami.
GEORGIA
Ewing Bros., 5 Plaza Way, Atlanta.

Jewelers Supply Co., 84 Peachtree St., Atlanta.
ILLINOIS
C. & E. Marshall Co., 1445 West Jackson Boulevard, Chicago.

(Branches in Detroit, Duluth, Columbus, New Orleans, Hous-
ton, Dallas, Oklahoma City, St. Paul, Milwaukee, Los Angeles,
Kansas City, Minneapolis.)
Swartchild & Co., 29 East Madison St., Chicago. (Branch in New
York City, Boston, and Dallas.)
Henry Paulson & Co., 37 S. Wabash Ave. Chicago.
J. Henri Ripstra, 5 S. Wabash Ave., Chicago.
INDIANA
Bardach & Gran, 448 N. Capitol Ave., Indianapolis.

Indiana Jewelers' Supply Co., 15 East Washington St., Indianapolis.
KENTUCKY
Geo. Katzman Co., 3rd and Walnut Sts., Louisville.
LOUISIANA
New Orleans Jewelers Supply Co., 208 Chartres, New Orleans.
MARYLAND
John A. Tschantre, 25 N. Liberty St., Baltimore.
MASSACHUSETTS
Kirby & Co., 387 Washington St., Boston.

Wm. J. Orkin, 373 Washington St., Boston.
Travis, Farber Co., 373 Washington St., Boston.
MICHIGAN
Michigan Jewelers' Supply Co., 1022 Michigan Bldg., Detroit.
MISSOURI
Clark Tool & Material Co., 925 Grand Ave., Kansas City.
Edwards-Ludwig-Fuller Co., 1115 Walnut St., Kansas City.
Meyer Jewelry Co., nth and Grand Ave., Kansas City.
NEW JERSEY
William Dixon, Inc., (tools, supplies, equipment, machinery, saws,

buffs, materials,) 32 East Kinney St., Newark. (Branch in
New York City.)
NEW YORK
Anchor Tool and Supply Co., 12 John St., New York.

Arrow Supply & Tool Co., 109 Lafayette St., New York.
Buhl, Peer & Keefe, (watch and jewelry supplies), 41 Maiden
Lane, New York.
William Dixon, Inc., 36 West 47, New York.
Paul H. Gesswein & Co., 35 Maiden Lane, New York.
Hammel, Riglander & Co., (watchmakers' supplies,) 209 W. 14,
New York.
M. J. Lampert & Sons, (watch and jewelry supplies) 35 Maiden
Lane, New York.
OHIO
E. & J. Swigart Co., 32 West 6th St., Cincinnati.
PENNSYLVANIA
Wm. S. Waples, 727 Sansom St.* Philadelphia.

Martin Gluck & Son, 313 Clark Bldg., Pittsburgh.
RHODE ISLAND
Z. Berberian Co., 147 Chestnut St., Providence.

Geo. L. Claflin Co., 156 Dorrance St., Providence.
Johnson & Raftery, 104 Point St., Providence.
Jewelers Supply Co., 1144 Eddy St., Providence.
TENNESSEE
Young-Neal Co., Chamber of Commerce Bldg., Nashville.
TEXAS
Jack Gould & Co., Santa Fe Bldg., Dallas.
VIRGINIA
Herr & Kline, 231 Granby St., Norfolk.
WASHINGTON
Butterfield Bros., 1504 Third Ave., Seattle. (Branch in Portland,

Ore.)
CANADA
E. & A. Gunther Co., Ltd., Brantford, Ontario. (Branch in Van-

couver.)
Dealers in Special Equipment and Machinery

Alberene Stone Corp. of Virginia, (hoods, sinks,) 419 Fourth
Avenue, New York. (Quarries at Schuyler, Va.)
Alloys and Products, Inc., (deoxidizers,) Oak Point Ave., Bronx,
New York.
American Cuttlefish Bone and Crucible Co., 75 Cliff St., New York.
American Gas Furnace Co., Elizabeth, New Jersey.
American Oil & Supply Co., (acids, crucibles,) 238 Wilson Ave.,
Newark, New Jersey.
American Seitz Filter Corp., (filtering equipment,) 480 Lexington
Ave., New York.
H. J. Astle & Co., (blowers, polishing equipment,) 118 Orange St.,
Providence, Rhode Island.
Autovent Fan & Blower Co., 1805 No. Korstner Ave., Chicago, 111.
Sigmund Cohn, (casting equipment,) 44 Gold St., New York.
DeBothezat Ventilating Equipment Division, 100 Sixth Ave., New
York.
Dentists' Supply Co., 220 West 42, New York.
Jos. Dixon Crucible Co., 233 Broadway, New York.
E. A. Eddy Machinery Co., (jewelry factory machinery,) 211 Eddy
St., Providence, R. I.
1AA RFFTXFTXrn PRFrTOTTS M F T A T . W A S T K S
General Ceramics Co., (acid proof stoneware,) 225 Broadway, N e w

York.
J. Goebel & Co., (crucibles, tongs, fluxes, furnaces, casting equip-
ment,) 95 Bedford St., New York.
W. Green Electric Co., (plating equipment and salts,) 192 Broadway,
New York.
H. Henrich, Inc., (alloying materials,) 68 Fulton St., New York.
Hoke Incorporated, (oxy-gas torches and accessories,) 122 Fifth
Avenue, New York.
H. E. Jacoby, (filter-presses,) 205 East 42, New York.
J. F. Jelenko & Co., (casting equipment,) 136 West 52, New York.
Jewelers Technical Advice Co., (information and advice, books on
precious metals,) 123 William St., New York.
Maurice A. Knight, (chemical stoneware,) Kelly Ave., Akron, Ohio.
Laboratory Construction Co., (hoods, fume disposal equipment,)
113 Holmes St., Kansas City, Mo.
Laboratory Furniture Co. Inc., (hoods, tables,) 37-18 Northern
Blvd., Long Island City, N. Y.
Lapp Insulator Co., Chemical Porcelain Division, (ceramics,) 400
Gilbert St., LeRoy, N. Y.
Leiman Brothers, (dust collecting equipment, jewelers' benches,) 23
Walker St., New York.
McWilliams Mfg. Co., (rolling mills, driers,) 237 Eddy St., Provi-
dence, R. I.
The Merrill Co., (filter-presses,) 343 Sansome St., San Francisco,
Calif.
Oliver United Filters, Inc., (filter-presses,) 33 West 42, New York.
Pangborn Corp., (dust control equipment,) Hagerstown, Md.
Research Corp., (Cottrell precipitation systems,) 405 Lexington
Ave., New York.
E. H. Sheldon & Co., (hoods, laboratory furniture,) 720 Nims St.,
Muskegon, Mich.
T. Shriver & Co., (filter-presses,) 802 Hamilton St., Harrison, N. J.
D. R. Sperry & Co., (filter-presses,) Batavia, 111.
Thermal Syndicate, Ltd., (quartz vessels,) 12 East 46, New York.
U. S. Stoneware Co., (ceramics, tanks, jars,) Akron, Ohio.
Van Dorn Metal Furniture, (hoods, tables,) 41 East 42, New York.
Western Precipitation Corp., (Cottrell precipitation systems,) 1016
West 9th St., Los Angeles, Calif.
CHAPTER E
BOOKS TO READ
The books and articles that are listed below are mentioned either
because they are useful or because they are interesting to the worker
in precious metals.
Several of the dealers in precious metals have published cata-

logues containing many facts of interest to the student and of
practical use to the worker. Some of these trade publications are
mentioned by name in Chapter X.
The large encyclopedias, and all text and reference books in

chemistry give valuable information under the headings of "Gold,"
"Silver," "Platinum Metals," "Alloys," and so on.
The monumental BIBLIOGRAPHY OF THE METALS OF

THE PLATINUM GROUP—1748 to 1917, by Dr. James Lewis
Howe and H. C. Holtz, published by the United States Geological
Survey, is of particular value to students of the history of these
metals. See also the titles listed at the end of Chapter XVII.
Some of the magazines that cater to the precious metal industries

also sell books, and take pleasure in advising their subscribers.
Among such magazines are the following: The Jewelers Circular-
Keystone, 239 West 39th St., New York City; The Manufacturing
Jeweler, 42 Weybosset St., Providence, R. I.; Horology, 747 So. Hill
St., Los Angeles, California; The National Jeweler, 531 So. La Salle
St., Chicago, 111., and The Laboratory Technician, 220 West 42nd
St., New York City.
For books on gem stones, consult the Gemological Institute of

America, 541 South Alexandria Avenue, Los Angeles. Associated
345
with this institute are the American Gem Society, and the publica-
tion Gems and Gemology.
Several bulletins on the hazards connected with precious metal
refining are mentioned at the end of Chapter B of this Appendix.
For laws, standards and regulations relating to the precious
metals, see Chapter C of this Appendix.
• * # *
BOOKS
JEWELRY, GEM CUTTING AND METALCRAFT. By W. T.
Baxter and Henry C. Dake. 224 pp. $2.50. McGraw-Hill Book
Company, Inc., New York. 1938.
THE SCIENCE OF DENTAL MATERIALS. By Eugene W.
Skinner. (Contains texts of American Dental Association Specifica-
tions for dental amalgam alloys; casting investment; impression
compound; casting wax; inlay casting golds; mercury; wrought
gold wire alloys; cementing medium.) 411 pp. $4.50. W. B.
Saunders, Philadelphia. 1937.
THE METALLURGY OF GOLD. By Sir T. K. Rose and W. A. C.-
Newman. 561 pp. $15.00. Seventh edition. J. B. Lippincott
Co., Philadelphia. 1937.
TESTING PRECIOUS METALS. By C. M. Hoke. (See Chapter 9
of this book.) 59 pp. $1.00. Jewelers Technical Advice Co.,
New York. 1937.
PLATINUM AND ALLIED METALS. With bibliography. 137 pp.
Price 3s 4d. The Imperial Institute, Mineral Resources Depart-
ment, South Kensington, London, England. 1936.
ANALYTISCHE CHEMIE DER EDELMETALLE. By Dr. Alfred
Wogrinz. (Band XXXVI, die Chemische Analyse.) 141 pp.
Price, paper, 13 RM; bound, 14.80 RM. Ferdinand Enke Verlag,
Stuttgart, Germany. 1936.
HANDBOOK FOR PROSPECTORS. By M. W. von Bernewitz.
372 pp. $3.00. McGraw-Hill Book Co., Inc., New York. 1935.
THE METAL CLEANING HANDBOOK. By Robert W. Mitchell.
214 pp. $1.00. Published by the Magnus Chemical Company
(manufacturers of cleaning materials), Garwood, New Jersey. 1935.
HANDBOOK FOR THE AMATEUR LAPIDARY. By J. Harry
Howard. 140 pp. $2.00. Published by J. H. Howard, 504 Crescent
Ave., Greenville, South Carolina. 1935*
WORKING I N PRECIOUS METALS. By, ( E . ^ ^^
$6.75. N. A. G. Press, Ltd, 26-34 ° ' d *">
BOOKS TO READ
1934. (Obtainable also from Henry Paulson & Co., 37 So. Wabash
Ave., Chicago.)
PRINCIPLES OF ELECTROPLATING AND ELECTROFORM-
ING. By William Blum and George B. Hogaboom. 424 pp.
$4.50. McGraw-Hill Book Co., New York. 1930.
A LIST OF ALLOYS. By William Campbell. 65 pp. Paper bound,
$1.50. Cloth bound, $2.00. American Society for Testing Mate-
rials, Philadelphia. 1930.
HOW TO MAKE JEWELRY. By George S. Overton. 298 pp. $3.00.
Walter B. Frost & Co., Providence, R. I. 1927.
HANDBOOK OF NON-FERROUS METALLURGY. Edited by
Donald M. Liddell, written by a staff of experts. In two volumes.
(Chapters 28 and 29, in vol. 2, are of interest to workers in precious
metals.) 1500 pp. $12.00. McGraw-Hill Book Co., New York.
1926.
CHASING AND REPOUSSE. By Albert Wehde. 24 pp. $1.00. The
Tremonia Publishing Co., 621 Plymouth Court, Chicago. 1924.
THE PLATINUM METALS. By E. A. Smith. 120 pp. Sir Isaac
Pitman & Sons, London. 1924.
A TEXTBOOK OF FIRE ASSAYING. By Edward E. Bugbee.
254 pp. John Wiley & Sons, Inc., New York. 1922.
HANDBOOK OF METALLURGY. By Carl Schnabel, translated by
Henry Louis. Vol. 1, Copper, Lead, Silver, Gold. 1171 pp.
$13.50. 1921.
GEM-STONES AND THEIR DISTINCTIVE CHARACTERS. By
G. F. Herbert Smith. 314 pp. $3.00. Brentano's, New York.
1920.
RECOVERING PRECIOUS METALS FROM WASTE LIQUID
RESIDUES. By George E. Gee. 380 pp. $4.80. E. & F. N.
Spon, Limited, 57 Haymarket, S.W.I. London. 1920.
JEWELRY MAKING AND DESIGN. By Augustus F. Rose and
Antonio Cirino. 463 pp. $5.00. Metal Crafts Pub. Co., Provi-
dence, R. I. 1917.
SILVERWORK AND JEWELRY. By H. Wilson and Unno Bisei.
496 pp. $2.50. D. Appleton and Co., New York. 1912.
THE ART OF THE GOLDSMITH AND JEWELLER. By Thomas
B. Wigley. 264 pp. $3.25. Chas. Griffin & Co., Ltd., London.
1911.
THE ELECTRO-PLATING AND ELECTRO-REFINING OF
METALS. By Alexander Watt and Arnold Philip. 677 pp.
$4.50. D. Van Nostrand Co., New York. ion.
34« REFINING PRECIOUS METAL WASTES
ARTICLES
The reader will line] that some of these publications are out of
print, or difficult to obtain for other reasons. However, many libraries
—for example the Engineering Societies Library, at 29 West 39th
Street, New York City—are equipped to make photostatic copies, at a
small sum per page, of any publication.
PLATINUM AND ALLIED METALS. By H. W. Davis. 11 pp.

5 cents. Chapter from Minerals Yearbook, Bureau of Mines. 1939.
EFFECT OF COMPOSITION UPON PHYSICAL AND CHEMI-
CAL PROPERTIES OF 10-KARAT GOLD ALLOYS. By Tracy
C. Jarrett. 9 pages. Metals Technology, October, 1939.
EFFECT OF COMPOSITION ON COLOR AND MELTING
POINT OF 10-k, 12-k, and 14-k Gold Alloys. By Tracy C. Jarrett.
8 pp. Metals Technology, September, 1939.
NEW STONEWARE IN THE ELECTROCHEMICAL FIELD. By
Percy C. Kingsbury. Transactions Electrochemical Society, pp.
303-310,1939.
DAS VERHALTEN DER EDELMETALLE UND IHRER KUP-
FERHALTIGEN LEGIERUNGEN GEGENUBER SAUER-
STOFF in fliissigem Zustande und beim Gliihen in festem Zustande.
By E. Raub and M. Engel. Mitt. Forsch. Inst. Probieramts Edel-
metalle, pp. 1-12, June, 1939. (Comprehensive review.)
ELECTROLYTIC SILVER REFINING. By Edmund Downs. (A
general discussion.) The Metal Industry, London, pp. 543-546,
vol. 54, May 19, 1939.
RECOVERY OF GOLD AND SILVER FROM PLATING SOLU-
TIONS. By Frank K. Savage. The Metal Industry, New York,
p. 159, vol. 37, April, 1939.
REFINING $2,000,000 OF GOLD PER WEEK. By A. L. Entwistle.
(Description of the treatment of crude gold bullion at the Royal
Canadian Mint Refinery at Ottawa. See Chapter 22 of this book.)
Canadian Metals and Metallurgical Industries, Toronto, Vol. 1,
January, pp. 3-8; February, pp. 53-56, 1938.
THE RECOVERY OF SILVER, GOLD AND MERCURY FROM
PRECIOUS METAL AMALGAMS. By C. M. Hoke. The Metal
Industry, New York, pp. 22-23, January, 1938.
PROPERTIES OF THE PLATINUM METALS, II. Tensile strengths
of platinum, palladium, and several of their commercial alloys at
elevated temperatures, with a few .notes on high-temperature corro-
sion resistance of platinum. By E. M. Wise and J. T. Eash. Metals
Technology, pp. 282-294, February, 1938. Also in Trans. A.I.M.E.,
Vol. 128, Institute of Metals Division, pp. 282-294,l9i '
BOOKS TO READ 349
NEW PROCEDURE FOR THE ANALYSIS OF DENTAL GOLD

ALLOYS. By Raleigh Gilchrist. Research Paper RP 1103, Journal
of Research of the Bureau of Standards, Vol. 20, pp. 745771, June,
1938.
A LIST OF LABORATORIES THAT MAKE ASSAYS, ANALYSES
AND TESTS ON ORES, MINERALS AND OTHER SUB-
STANCES. By C. W. Davis and M. W. von Bernewitz. 25 pp.
Bureau of Mines Information Circular I.C. 6999. March, 1938.
RECOVERY OF GOLD FROM BALBACH-THUM SLIMES AT
COPPER CLIFF, ONTARIO. By Frederic Benard. 5 pp. Metals
Technology, February, 1938.
OCCURRENCE AND TREATMENT OF MERCURY ORE AT
SMALL MINES. By M. W. von Bernewitz. 40 pp. Bureau of
Mines Information Circular I. C. 6966, Oct., 1937.
PLATINUM IN THE WORLD'S WORK. By Lonnelle Davison.
The National Geographic Magazine, Washington, D. C, Vol. 62, pp.
345-360. Sept., 1937.
PLATINUM AND ALLIED METALS. SOME POINTS OF IN-
TEREST TO THE CHEMICAL ENGINEER. By C. Johnson
and R. H. Atkinson. Trans. Institute of Chemical Engineers,
London, Vol. 15, pp. 131-144, 1937. (See Chapter 22 of this
book.) An abstract of this article was printed in The Industrial
Chemist, (London), pp. 223-233, June, 1937. A shorter abstract
was printed in The Chemical Age, (London), pp. 459-462, May,
r937-
JUST HOW MUCH PRECIOUS METAL IS LOST IN POLISH-

ING? By Ernest A. Smith. The Manufacturing Jeweler, Provi-
dence, pp. 30-31, October 15, 1936.
SILVER: ITS PROPERTIES AND INDUSTRIAL USES. By B. A.
Rogers, Irl C. Schoonover & Louis Jordon. Bureau of Standards,
C412. 72 pp. 10 cents. October 2, 1936.
METALS OF THE PLATINUM GROUP. ORES, RECOVERY
AND REFINING, FABRICATION AND USES, AND PROP-
ERTIES. By R. H. Atkinson and A. R. Raper. Journal of the
Institute of Metals, (London). Vol. 59, No. 2, pp. 179-210, 1936.
An abstract of this paper appears in The Metal Industry, (London),
pp. 114-116, January 15, 1937.
BULLION PARTING AT THE HOMESTAKE MINE. By
Nathaniel Herz. (See Chapter 22 of this book.) Mining and
Metallurgy, pp. 500-504, December, 1935.
A PROCEDURE FOR THE SEPARATION OF THE SIX PLATI-
NUM METALS FROM ONE ANOTHER AND FOR THEIR
GRAVIMETRIC DETERMINATION. By Raleigh Gilchrist and
Edward Wichers. Journal of the American Chemical Society,
Vol. 57, pp. 2565-2573, 1935.
STUDIES ON THE RECOVERY OF GOLD AND SILVER. By

Ed. S. Leaver, M. B. Royer, J. A. Woolf, R. E. Head, B. W.
Gandrud, R. E. Evans, and F. W. Thackwell. Bureau of Mines,
Report of Investigations R. I. 3275. 65 pp., June, 1935.
PROPERTIES OF THE PLATINUM METALS. I. Strength and
annealing characteristics of platinum, palladium, and several of
their commercial alloys. By E. M. Wise and J. T. Eash. Trans.
A.I.M.E., Vol. 117, Institute of Metals Division, pp. 313-328,
February, 1935.
JEWELERS' LOW GRADE WASTES: AN ASSET OR A LIA-
BILITY? By C. M. Hoke. (See Chapter 18 of this book.) The
Metal Industry, New York, November and December, 1934.
PLACER MINING IN THE WESTERN UNITED STATES. PART
II. HYDRAULICKING, TREATMENT OF PLACER CON-
CENTRATES, AND MARKETING OF GOLD. By E. D.
Gardner and C. H. Johnson. Bureau of Mines Information
Circular 6787. 89 pp., October, 1934.
A LIST OF PUBLICATIONS ON THE PLATINUM METALS.
9 pp. Bureau of Standards Letters Circular LC-426, October 1, 1934.
A METHOD FOR THE SEPARATION OF RUTHENIUM FROM
PLATINUM, PALLADIUM, RHODIUM AND IRIDIUM. By
Raleigh Gilchrist. Research Paper 654, Bureau of Standards
Journal of Research, Vol. 12, pp. 283-290, Mar., 1934.
A NEW SYSTEM OF ANALYTICAL CHEMISTRY FOR THE
PLATINUM METALS. By Raleigh Gilchrist and Edward
Wichers. Publicado en: Trabajos del IX Congreso Internacjonal
de Quimica Pura y Aplicado, Toma VI, Quimica Analitica.
Madrid, Spain, 1934.
RECOVERY OF PRECIOUS AND SECONDARY METALS FROM
ELECTROLYTIC COPPER REFINING. By M. A. Mosher.
(See Chapter 22 of this book.) Trans. A.I.M.E., Vol. 106,
pp. 427-440, 1934.
THE ROLE OF THE PLATINUM METALS IN DENTAL AL-
LOYS, III. The influence of platinum and palladium and heat
treatment upon the microstructure and constitution of basic alloys.
By E. M. Wise and J. T. Eash. Trans. A.I.M.E., Vol. 104, Inst.
of Metals Division, pp 276-307, 1933
ELECTROLYSIS OF SILVER-BEARING THIOSULFATE SOLU-
TIONS. By K. Hickman, W. Weyerts and O. E. Goehler. In-
dustrial and Engineering Chemistry, New York, Vol. 25, pp. 202-
212, February, 1933.
TREATING WASTE FROM SILVER MANUFACTURE. By
fa%&6^&8&fa$P$272-
BOOKS TO READ m
TREASURE TROVE IN LOWLY "SWEEPS." Scientific American,
p. 163, Sept., 1932.
THE RECOVERY OF PRECIOUS METAL FROM FACTORY
WASTE. By John S. Isdale. The Brass World, pp. 138-139,
l y , 1932.
MELTING FINE SILVER IN HIGH FREQUENCY INDUCTION
FURNACE. By Daniel L. Ogden. Tra
chemical Society, New York, pp. 221-223, I932-Reprin
Brass World, p. 144, July, 1932.
RECOVERY OF PRECIOUS METALS FROM ELECTROLYTIC
COPPER REFINING AT THE CANADIAN COPPER RE-
FINERS PLANT, MONTREAL EAST, QUEBEC. By C. W.
Clark and A. A. Heimrod. Trans. American Electrochemical
Society, pp. 437-450, 1932.
THE ROLE OF THE PLATINUM METALS IN DENTAL AL-
LOYS. I & II. By E. M. Wise, Walter S. Crowell & J. T. Eash.
Trans. A.I.M.E., Vol. 99, pp. 363-412, 1932.
ASSAYING PLATINUM METALS. By Martin Schwitter. Metal and
Mineral Markets, pp. 6-10, Jan. 21; pp. 6, 7, Feb. 18; pp. 6-11,
Mar. 17, 1932. A reprint of this article under the heading REFIN-
ING PLATINUM METALS, will be found in the Brass World,
Sept., Oct., Nov. and Dec, 1932.
A METHOD FOR THE SEPARATION AND GRAVIMETRIC DE-
TERMINATION OF OSMIUM. By Raleigh Gilchrist. Research
Paper 286, Bureau of Standards Journal of Research, Vol. 6, pp.
421-44S, March, 1931.
FABRICATION OF THE PLATINUM METALS. By C. S. Sivil.
Trans. A.I.M.E., Institute of Metals Division, pp. 246-261, 1931.
THE MAKING OF MIRRORS BY THE DEPOSITION OF METAL
ON GLASS. Bureau of Standards Circular 389, 17 pp., 5 ^ ,
January 6, 1931.
FIELD TESTS FOR THE COMMON METALS. By George R.
Fansett. 51 pp. University of Arizona, Tucson, Ariz., March 15,
1930.
MELTING AND CASTING SOME GOLD ALLOYS. By Edward
A. Capillon. 20 pp. Mining and Metallurgy, February, 1930.
HIGH-STRENGTH GOLD ALLOYS FOR JEWELRY, AND AGE-
HARDENING PHENOMENA IN GOLD ALLOYS. By E. M.
Wise. Trans. A.I.M.E., Vol. 83, pp. 384-408, 1929.
MELTING, MECHANICAL WORKING AND SOME PHYSICAL
PROPERTIES OF RHODIUM. By Wm. H. Swanger. Re-
search Paper 127, Bureau of Standards Journal of Research. Vol. 3,
pp. 1029-1040, December, 1929.
A GRAVIMETRIC METHOD FOR THE DETERMINATION OF

RUTHENIUM. By Raleigh Gilchrist. Research Paper 125, Bureau
of Standards Journal of Re-earch, Vol. 3, pp. 993-1004, December,
1929.
PHYSICAL PROPERTIES OF DENTAL MATERIALS. By R. L.
Coleman. Research Paper 32. Bureau of Standards Journal of
Research, pp. 867-938. December, 1928.
PURIFICATION OF THE SIX PLATINUM METALS. By Edward
Wichers, Raleigh Gilchrist and Wm. H. Swanger. (See Chapter 17
of this book.) 29 pp. Mining and Metallurgy, March, 1928.
REFINING AND MELTING SOME PLATINUM METALS. By
J. O. Whiteley and C. Dietz. (Osmium is not included. This
paper describes methods of refining dental and jewelry scrap and
laboratory ware, in 100 ounce lots. See Chapter 17 of this book.)
9 pp. Mining and Metallurgy, March, 1928.
USE OF THE NOBLE METALS FOR ELECTRICAL CONTACTS.
By E. F. Kingsbury. 19 pp. Mining and Metallurgy, March, 1928.
THE PLATINUM METALS AND THEIR ALLOYS. By Frederick
E. Carter. 24 pp. Mining and Metallurgy, February, 1928.
MANUFACTURE OF STERLING SILVER AND SOME OF ITS
PHYSICAL PROPERTIES. By Robert H. Leach and C. H.
Chatfield. 13 pp. Mining and Metallurgy, February, 1928.
SAMPLING AND EVALUATING SECONDARY NON-FERROUS
METALS. By T. A. Wright. 17 pp. Mining and Metallurgy,
February, 1928.
GOLD, SILVER, COPPER ALLOYS. By Frederic E. Carter. 17 pp.
Mining and Metallurgy, February, 1928.
MELTING STERLING SILVER IN HIGH FREQUENCY INDUC-
TION FURNACES. By Robert H. Leach. Trans. American Elec-
trochemical Society, pp. 223-227, 1928.
THE GLOW TEST FOR METALS OF THE PLATINUM GROUP.
By C. W. Davis. Journal of the Franklin Institute, pp. 679-700,
May, 1927.
IRVINGTON SMELTING AND REFINING WORKS. By Adolph
Bregman. (See Chapter 22 of this book.) The Metal Industry,
New York, pp. 183-187, May, 1927.
ALLOYS OF PRECIOUS METALS AND THE HIGH FRE-
QUENCY INDUCTION FURNACE. By Reginald V. Williams.
The Metal Industry, New York, pp. 141-143, April; pp. 188-190,
May, 1927.
SOME NEW ANALYTICAL REACTIONS OF THE PLATINUM
METALS. By S. C. Ogburn, Jr. Journal of the American Chemi-
cal Society, Vol. 48, pp. 2494-2507, Oct., 1926.
BOOKS TO READ 353
ANALYSIS OF DENTAL GOLD ALLOYS. By William H. Swanger.
Scientific Papers of the Bureau of Standards, No. 532. 10^. Vol.21,
pp. 209-239, August, 1926.
For a discussion of the dangers connected with the heating of osmium,
see The Keystone, Philadelphia, p. 92, January, 1924; Chemical and
Metallurgical Engineering, p. 743, November 10, 1924; Brass World,
Vol. 20, p. 242, 1924; Jewelers' Circular, May 28, 1924.
METHODS FOR THE RECOVERY OF PLATINUM, IRIDIUM,

PALLADIUM, GOLD AND SILVER FROM JEWELERS'
WASTE. By C. W. Davis. 14 pp. 5 ^ . Bureau of Mines
Technical Paper 342, January, 1924.
THE PREPARATION OF PLATINUM AND OF PLATINUM-
RHODIUM ALLOY FOR THERMOCOUPLES. By Robert P.
Neville. Trans. American Electrochemical Society, pp. 125-136,
1923.
RECOVERY OF PLATINUM FROM USED CONTACT MASS AT
THE OLD HICKORY POWDER PLANT. By A. L. Kibler.
(See Chapter 22 of this book.) Journal of Industrial and Engineer-
ing Chemistry, Vol. 14, pp. 636-641, July, 1922.
THE METALLURGY OF PLATINUM. By Louis duParc, trans-
lated by Allen C. Redding and Geo. M. Nelson, with Frank San-
born. A translation of Chapter XV of duParc's "Le Platine du
Monde." Issued in Mining in California, a publication of the
California State Mining Bureau, San Francisco, Vol. 18, pp. 158-172,
April, 1922.
THE DETECTION AND ESTIMATION OF PLATINUM IN
ORES. By C. W. Davis. 27 pp. Bureau of Mines Technical
Paper 270. January, 1921.
BIBLIOGRAPHY OF THE METALS OF THE PLATINUM GROUP
—1748-1917. By Jas. Lewis Howe and H. C. Holtz. 558 pp. (Out
of print.) United States Geological Survey, Bulletin 694, 1919.
ELECTRODEPOSITION OF GOLD AND SILVER FROM
CYANIDE SOLUTIONS. By S. B. Christy. 171 pp. 25#.
U. S. Bureau of Mines Bulletin 150, 1919.
TEXT BOOKS FOR THE JEWELER'S LIBRARY. By Clement W.
Coumbe. (A bibliography.) The Jewelers' Circular-Weekly, pp.
123-129. Feb. 7, 1917.
HANDLING OF PRECIOUS METAL SCRAP, FILINGS, SWEEPS,
ETC. By C. M. Hoke. The Keystone, Philadelphia, pp. 289-295,
September, 1917.
ELECTROLYSIS OF LOW GRADE GOLD BULLION. By Theo.
W. Bouchelle. Engineering & Mining Journal, January 25, 1913.
TREATMENT OF GOLD AND SILVER AT THE UNITED

STATES ASSAY OFFICE IN NEW YORK CITY. The Brass
World, pages 307-314, September, 1913.
ANNUAL REPORT OF THE DIRECTQR OF THE MINT FOR
1912. Treasury Department Document 2671, pages 44-52, 1913.
DE L'ANALYSIS DU PLATINE IRIDIfi EMPLOYE PAR LA SEC-
TION FRANCAISE DE LA COMMISSION INTERNATION-
ALE DU MfiTRE A LA CONFERENCES DES PROTOTYPES.
By E. H. Sainte-Claire DeVille and J. S. Stas. Proces-verbeaux,
Comite International des Poids et Mesures. Annexe No. II, 1878.
INDEX
ACIDS, to remove from skin, 32, 49, 321 Arrow Supply & Tool Co., 342
Acquaintance tests, aqua regia, 37, 38 Asbestos, 14, 15, 17
brass, 16 Assaying not covered in this book, 2
copper, 15 to learn, 233
gold, 16, 36, 65 Astle, H. J., & Co., 343
lead, 69 Atkinson, R. H., 198, 202, 206, 207,
nickel, 16 282, 297, 312,
palladium, 108 349
platinum, 15, 106 Attleboro Refining Co., 337
silver, 16, 37, 38 Autovent Fan & Blower Co., 343
tin, 69
zinc, 16
Agglutinants, 243 BAKER & Co., 105, 337
Ajax-Northrup furnace, 302, 304 Baker, George M., Co., 338
Alberene Stone Corp , 343 Balbach Smelting & Refining Co., 337
Alkalies, to remove from skin, 17, 32, Balbach-Thum process. See Thum
49. 323 Bardach & Gran, 341
Alloys and Products Incorporated, 343 Barium dioxide, 203
Alloys, dental, 131 Baxter, W. T., 346
Aluminum as precipitant, 83, 164 Beakers, glass, 34, 35
Amalgamation in sweeps recovery, 222 Benard, Frederic, 291, 349
Amalgams, to refine, 187 Berberian, Z., Co., 342
American Cuttlefish Bone & Crucible Berge, J. &H., 340
Co, 343 Bisei, Unno, 347
American Dental Association, 131 Bishop, J., & Co., 104, 338
American Gas Furnace Co, 211, 343 Blowpipe, oxygen, 293, 304, 306, 311
American Platinum Works, 105, 337 Bluestone, recovery of, 264
American Oil & Supply Co., 343 Blum, William, 347
American Seitz Filter Corp , 343 Bouchelle, Theodore W., 353
Amino-compounds, 128, 152 Brass, 6, 16
Ammonia, price of, 114 as added metal 52, 135, 220
compounds with copper or nickel, Braun Corporation, 339
99. 152. 177 Braun-Knecht-Heimann Co., 339
Ammonium chloride and the platinum Bregman, Adolph, 261, 352
metals, 106, 108, Brogan, Byard A., 5
119, 120, 176 Buchner funnel, 237, 238
for purifying gold, 174 Buffalo Apparatus Co., 340
not poisonous, 126 Bugbee, Edward E., 347
price of, 114 Buhl, Peer & Keefe, 342
Ammonium chloro-platinate. See Plat- Burner or heater, 12, 14, 15, 29
mum-ammontum- Burning the red palladium-ammonium-
chlonde chloride, 149
Anchor Tool & Supply Co , 342 platinum-ammonium-chloride, 107,
Antidotes for poisons, 322, 324 121, 127, 289
Aqua regia, composition of, 37, 43, 62, Burns, treatment for, 321, 323
116, 135, 146, Butterfield Bros., 343
171. 175 Buttons, to melt filings into, 39, 52, 133,
dissolving gold in, 43, 116, 135, 3M
171 silver-rich, 106, 140
tests with, 37, 38 By-product metals, 250
355
356 INDEX
CADMIUM, 7 Crucibles, handling of, 242, 326
Caire, Justinian, Co., 339, 340 lime, 307
Campbell, William, 347 old, to refine, 212, 215
Canadian Copper Refineries, 247 porcelain, 113
Capillon, Edward A., 312, 315, 316, 351 quartz, 309
Carborundum, to remove, 183 sand, 308
Carter, Frederick E., 208, 352 thoria, 309, 312
Casserole, 12 Cummins, Gilbert, & Co., 338
Casting precious metals, 317 Cupellation, 220, 224, 264
Caustic, price of, 15 Cyanide, antidote for, 324
to dissolve shellac, 17 danger of handling, 81, 118, 328
washing from skin, 17, 32, 49, 323 dissolving silver chloride with, 118
Cement silver, 78 price of, 114
Cementation of gold, 60, 249 solution, for stripping with the cur-
Central Scientific Co., 339, 340 rent, 185
Ceramics, new forms, 239 to recover gold and silver from,
Chatfield, C. H., 312, 352 81, 186
Chemical ware, to refine, 196 Cylinders, glass, 35, 36
Chemically pure acids vs. commercial
acids, 23, 62 DAKE, Henry Co., 346
Chlorine gas, 168 Davis, C. W., 349, 352, 353
in Miller process, 174, 251, 268 Davis, H. W., 348
in platinum-group separation, Davison, Lonnelle, 349
157 DeBothezat Ventilating Equipment Divi-
Chloroplatinic acid, 106 sion, 343
Christy, S. B., 83, 353 Debray, H., 312
Chromium plate, to remove, 183 Dee, and Co., Thomas J., 105, 337
Cincinnati Gold & Silver Refining Co., Dental alloys, composition of, 130
337 new procedure for, 206
Cirino, Antonio, 347 to refine, 129
Claflin, Geo. L., Co., 342 pins, 191
Clark, C. W., 247, 291, 351 Dentists' Supply Co., 141, 338, 343
Clark Tool & Material Co., 342 Denver Fire Clay Co., 339, 340
Clinton Refining Co., 337 Deoxidizers, 316
Cohn, Sigmund, 105, 338, 343 DeVille, E. H. Sainte-Claire, 202, 204,
Coleman, R. L., 352 209, 312, 354
Coning and quartering, 223 Dietz, C, 141, 156, 202, 208, 352
Conley & Straight, Inc., 338 Differences between gold and platinum,
Conn, C. G., 82 105, 106, 176,
Contact points, electrical, 194 298
Cooper, Joseph B., & Son, 338 Dimethyl glyoxime, as precipitant, 160
Copper, 6 tests with, 97
acquaintance tests, 15 Dishonesty, 231
as precipitant, 60, 164 Dixon, Joseph, Crucible Co., 343
compound with ammonia, 152, 177 Dixon, William, Inc., 342
to identify, 99 Dore, definition of, 273
Copper Cliffs Plant, 291 Downes, Edmund, 348
Copperas, nature of, 64, 84 Dropping bottle, 92, 93
price of, 36 DuParc, Louis, 203, 208, 353
recovering gold with, 46, 123
test involving, 46, 90
Cottrell precipitation process, 222, 246 EARTHENWARE, 35, 36
Coumbe, Clement W., 353 Eash, J. T., 297, 348, 350, 351
Crowell, Walter S., 351 Eastern Smelting & Refining Co., 337
Crucible Manufacturers Association, 242, Eastman Kodak Co., 85
326, 329 Economics of refining, 7, 210, 212, 232
Crucibles, Alundum, 309 Eddy, E. A., Machinery Co., 343
beryllia, 309 Edwards-Ludwig-Fuller Co., 342
for Miller process, 267 Eimer & Amend, 12, 340
graphite, 308 Eldorado Gold Mines, Ltd., 240
INDEX 357
Electrolytic methods, 174, 253 Gee, George E., 347
advantages of, 258 Gemological Institute of America, 345
for copper, 254 General Ceramics Co., 344
for gold, 256, 265, 278 General Refiners, Inc., 337
for silver, 255, 265, 274 Gesswein, Paul H., & Co., 342
for thiosulphate (hypo) solu- Gilchrist, Raleigh, 131, 137, 170, 204,
tions, 85, 86 206, 207, 208,
Emery, to remove, 183 349. 35°. 35i. 352
Enamel, to dissolve, 182 Glass, to dissolve, 182
Engel, M., 309, 348 Glines & Rhodes, 337
Enright Refining Co. 337 Glow test, 99
Entwistle, A. L., 292, 348 Gluck, Martin & Son, 342
Equipment used in refining, 12, 25, 34, Goebel, J., 308
92, 113, 158, 214, Goebel, J. & Co., 344
238, 239, 279, 286 Goehler, O. E., 86, 350
catalogues of, 12, 236 Gold, acquaintance experiments on, 16,
dealers in, 339, 343 36,65
where and how to and stannous chloride, 94
order, 12, 236 as added metal, 65, 135, 220
Erlcnmeyer flask, 136, 171 to cast, 317
Evans, R. E., 350 to granulate, 40, 135
Evaporating dish, 34 to identify, 90, 94
Ewing Bros., 341 laws concerning, 330
leaf, 193
purity of recovered, 63, 124
FANS and blowers, 26, 30, 241 to melt, 39, 48, 52, 124, 278, 312
Fansett, George R., 351 with the oxygen torch, 315
Ferguson, James B., 339 to recover from acid solution, 46,
Ferrous sulphate. See Copperas
Filter paper, 12, 13, 172, 238 55. 123, 173
how to fold, 18 from cyanide solution, 81,
old, to refine, 49, 129, 216 186
Filtering flask, 238 to repurify, 171, 313
Filter-press, 243 to wash, 47, 61, 124
Fire refining, 174, 313 Gold alloys, dental, composition of, 131
First aid, 31, 49, 322 green, 65, 75
Fisher Scientific Co., 340 18-k with platinum, 184
Flue dust, 216, 229 white, composition of, 66
Fluxes, 39, 134, 218, 317 Gold-filled scrap, to refine, 67
Gold-plated scrap, to refine, 67
prepared, 314 Goldsmith Bros. Smelting & Refining Co.,
Formic acid, 158
Friedman-Gessler Co., 340 337. 33«
Frying pan, 14, 17 Gould, Jack & Co., 343
Fumes, disposing of, 26, 241, 289 Gracier, S. B., & Sons, 336
in Miller process, 252 Graduate glass, 34, 35
in Wohlwill process, 259 Green, W., Electric Co. 344
trap for, 41, 42 Greiner, Emil, Co., 340
Funnel, glass, 12, 13 Gunther, E. & A., Co. Ltd., 343
Buchner, 237, 238 HAGSTOZ, T. B., & Son, 338
Funnel support, 13, 14 Hamerschlag, Moe., Inc., 338
Furnaces, cupel, 264 Hammel, Riglander & Co., 342
electric, 229, 302, 311 Handy & Harman, 105, 217, 218, 221,
for melting, 267, 277
for parting, 267 331, 336. 338
Hazards connected with refining, 31, 49,
"Further treatment" for palladium solu- 273, 320
tions, 150 Head, R. E., 350
Heil Corporation, 339
GANDRUD, B. W., 350 Heimrod, A. A., 247, 291, 351
Gardner, E. D., 190, 350 Heller, Gerald K., 339
Gauze, wire, 14, 15, 17 Henrich, H., Inc., 344
358 INDEX
Henry, Peter, & Son, 337 KASTENHUBER & Lehrfeld, 338
Herr & Kline, 343 Katzman, George, Co., 341
Herz, Nathaniel, 266, 349 Kibler, A. L., 292, 353
Hickman, K., 86, 350 Kilgallon & Co., 337
Hillebrand Prize, 204 Kingsbury, E. F., 194, 352
Hogaboom, George B., 347 Kingsbury, Percy, 239, 348
Hoke, C. M., 92, 210, 293, 348, 350, 353 Kirby & Co., 341
Hoke, Incorporated, 294, 344 Kjeldahl determinations, 242
Hoke, Sam W., 306 Knight, Maurice A., 344
Holtz, H. C, 345, 353 Kushner & Pines, 338
Homestake Mine, 266
Hood, John, Co., 337
LABORATORY Const. Co., 242, 344
Hoods for fumes, 26, 28, 241 Laboratory Furniture Co., Inc., 28, 344
Hoover & Strong, Inc., 338 Lampert, M. J., & Sons, 342
Howard, J. Harry, 3, 346 Lapp Insulator Co., 344
Howe, James Lewis, 345, 353 Laws and regulations concerning pre
Hydrochloric acid, price of, 36 cious metals, 33c
Hydrofluoric acid, 182 Leach, Robert H., 312, 352
Hydrogen sulphide, 128 Lead, 6
Hypo baths, to recover silver from, 85
acquaintance experiments, 69
as carrier, 219, 224, 262, 283
IDENTIFYING metals in solid form, 102
buttons, 219
in solution, 88 chloride, 69, 92
Imperial Institute, 346 dish, 183
Indiana Jewelers' Supply Co., 341 in soft solder, 67
Indium, 131, 206
Inquartation, 52, 54, 249 in the Stock Pot, 154
International Nickel Co., Inc., 105 nitrate, 70
Iridio-platinum, 103 sulphate, 71, 72, 283
Leaver, Ed S., 350
Iridium, its characteristics, 103, no, 136, Leiman Bros., 344
197. See also Lelong, L., & Bro., 337
Platinum Group Lemel, definition of, 39
its rarity, 197
to ignite, 146, 181 to melt, 39, 133, 314
to melt, 311 Licenses to handle precious metals, 330
to purify, 179, 285 Liddell, Donald M., 347
to recover from silver chloride resi- Litharge, 219, 224, 262
due, 137 Litmus paper, 100
Losses, 226, 229, 286
from Stock Pot, 147, 167 Low grade wastes, 210
Iridium-ammonium-chloride, 120, 176
Iron or steel, 6
as precipitant, 60, 83, 164, 291 MACALASTER Bicknell Co., 339
to identify, 100 Magnet, 12, 16, 51
to remove, 16, 21, 22 Marshall, C. & E., Co., 341
passive state of, 21 Masel Dental Laboratory, 338
Irvington Smelting & Refining Works, Masks, gas, 328
261.337 Maxwell, Clifford R., 341
Isdale, John S., 230, 351 McWilliams Mfg. Co., 344
Melting filings into a button, 39, 5:
JACOBY, H. E., 344 133. 3M
Jarrett, Tracy C, 313, 348 Melting gold, 39, 48, 52, 124, 278, 31:
Jelenko, J. F., & Co., 338, 344 315
Jewelers & Silversmiths' Co-operative Re- Melting palladium, 150, 310
fining Co., 337 Melting platinum, 293
Jewelers' Supply Co., 341, 342 Melting silver, 75, 79, 312, 316
Jewelers' Technical Advice Co., 344 Mercuric cyanide and palladium salts, 9
Johnson, C, 202, 206, 282, 349 Mercury, 6
Johnson Matthey & Co., 338 poisoning, 325
Johnson & Raftery, 342 to identify, 92, 95
Jordon, Louis, 349 to recover, 187
INDEX 3SP
Merrill Co , 344 "Ous" and "ic" conditions, 157, 169
Meta-stannic acid, 70, 72 Overton, George S., 312, 347
Meyer Jewelry Co , 342 Oxalic acid, 59, 173
Michigan Jewelers' Supply Co , 342 Oxidation and reduction, 157, 169
Miller, F B , 251 Oxygen, 23
Miller process, 174, 251, 258 cylinders, handling of, 326
at Homestake Mine, 266 flame and torch, 293, 304, 306, 311
at Royal Mint, Ottawa, 292 melting gold with, 315
Mining methods, 225
Mirror solutions, 83 PACIFIC Platinum Works, 336
Mismanagement a cause of losses, 232 Paints, 240
Mitchell, Robert W , 346 Palladium, 103. See also Platinum
Moebius process, 255, 274, 280, 284 Group
Mond Nickel Co, Ltd, 202, 207, 231, its solubility in acids, 109, 133, 178
282 its oxides, 109, 146, 178, 310
Montreal East, 291 to anneal, 310
Mortar and pestle, 113, 136 to dissolve, 109, 133, 146, 177, 178
Mosher, M. A., 273, 350 to identify, 94, 97, 99
to melt, 150, 310
NATIONAL Safety Council, 32, 320, to recover from solution, 148, 160,
325. 328 166, 279
Nelson, George M , 208, 353 to repurify, 177, 284
Neutralization of acid burns, 321 to work, 310
Neville, Robert P , 353 Palladium solutions, and ammonia, 150,
New Jersey Laboratory Supply Co , 339 177
New Orleans Jewelers Supply Co , 341 and ammonium chloride, 108
Newman, W A C , 346 and dimethyl glyoxime, 97, 160
Ney, J M , Co , 336 and mercuric cyanide, 99
Nickel, 6 and stannous chloride, 94
acquaintance experiments, 16 "further treatment of", 150
identifying, 98 Palladium-ammonium-chloride, 108, 149
in white golds, 67 to reduce, 149, 158, 160
Nitric acid, action on gold-silver and Palladosamine chloride, 152, 284, 289
platinum-silver al- Palo-Myers, Inc., 340
loys, 41, 52, 106, Pan, gold-washing, 51, 213, 216
133. M° Pangborn Corporation, 344
price of, 15 Paper storage, to refine, 49, 216
quantity required in refining, 42 Passive state, 21
to remove base metals with, 17, Paulson, Henry, & Co., 341
40, 53, 72, 115 Pease & Curren, 338
to remove excess of, 44, 55, 119, Pen points, 192
126, 147, 175 Philip, Arnold, 250, 347
Phipps & Bird, Inc., 340
OCCLUDED gases in metals, 309, 313, Photographers' wastes, 85
3.6 Pickling solutions, 80
Ogburn, S C , Jr , 352 Plating baths, to recover gold or silver
Ogden, Daniel L, 312, 351 from, 81
Old Hickory Powder Plant, 291 Platinum, 11, 103, 105, 197. See also
Oliver United Filters, Inc , 344 Platinum Group
Orange Bros , 338 acquaintance experiments, 15, 106
Orange powder, 107, 120, 176 action with nitric acid, 15, 106, 133,
to reduce, 121, 127, 158, 289 140
Orkin, William J , 341 action with sulphuric acid, 106,
Osmindium, 192 141
Osmium, 197, 200 See also Platinum alloys, to make up, 301
Group and ammonium chloride, 106, 120
danger from oxide fumes, 200, 209, and stannous chloride, 94
3°9 "black," 167, 300
to melt, 312 -clad nickel and copper, 196
to recover, 147, 285 defective, 299
360 INDEX
Platinum, melting point of, 293 Redding, Allen C, 208, 353
solders, 103 Refractories. See Crucibles
sponge, 108, 122, 300 Remington, Horace, & Son Co., 338
stamping laws, 333 Research Corporation, 344
to anneal, 296 Retort for mercury, 188
to cast, 317 Reversed current, to strip with, 184
to identify, 94 Reynolds, E. W., Co., 340
to melt, 119, 122, 220, 293, 298 Rhodium, 197. See also Platinum Group
to recover from aqua regia solution, to melt, 311
119, 138, 148 to recover, 147, 285
to recover from nitric acid solution, to strip plate, 198
141 Ripstra, J. Henri, 341
to repurify, 175, 180, 284 Rods, for stirring, 14
to work, 296 Rogers, B. A., 349
volatilization of, 294, 309 Rolled gold, to refine, 67
Platinum-ammonium-chloride, 106, 120 Rose, Augustus F., 347
to reduce, 121, 127, 158, Rose, SirT. K., 193, 229, 246, 249, 251,
289 312, 346
to wash, 121 Royal Mint at Ottawa, 292
Platinum Group, 103, 197 Royer, M. B., 350
alloys involving members of, Ruthenium, 197, 199, 200. See also
66, 103, 130, 191, Platinum Group
192, 194, 199, oxides, 309,
301 to melt, 312,
books on, 104, 206 to recover, 147, 285
colors of, no
glow test for, 99 SAMPLING, 222, 262
in the Wohlwill process, 257, Sanborn, Frank, 208, 353
259,279,281,284 Sand, to dissolve, 182
occluded gases in, 309, 310, 313 Sand bath, 34, 35
oxides of, 198, 200, 309 Sargent, E. H., & Co., 339
rarity of, 197 Savage, Frank K., 82, 348
separation of the, 200, 202 Schmidt, H., & Son, 337
volatilization of, 294 Schnabel, Carl, 347
Polishings, to refine, 217 Schoonover, Irl C, 349
Porcelain, to dissolve, 182 Schwitter, Martin, 195, 207, 351
Porcelain ware, care of, 17 Sea water, recovering gold from. 10
Potato as condenser, 188 Secondary metals, 251
Powder metallurgy, 311, 312 Seifried, G. W., Co., 337
Pratt, C. L., Jr., 337 Sheldon, E. H., & Co., 344
Precipitants, choice of, 55 Shellac, to remove, 17
metallic, 60 Shreve & Co., 336
Preiser, B., Co., 340 Shriver, T., 8c Co., 344
Professional methods, 248 Sieve, 14, 16
in the small shop, 260 Sieving sweeps, 213
Purity of recovered metal, 63, 124, 170, Silver, 74
286 acquaintance experiments, 16, 37, 38
and stannous chloride, 95
QUALITY marks, 332 as added metal in inquartation, 52,
Quartation, 52, 54, 249 75. 79. 84
Quartz vessels, 240, 288, 309 cement, 78
electrolytic refining of, 255, 265,
R. & H. CHEMICALS Department, 324 274
Radium, 240 in the Miller process, 252, 271
Raper, A. R., 198, 207, 297, 312, 349 price of, 74
Raritan Copper Works, 254, 257, 273 salts of, their solubility, 79
Raub, E., 309, 348 scrap, to refine, 75
Red powder, 108, 149 stamping laws and standards, 332
to reduce, 149, 158, 160 to identify, 91, 92
to repurify, 177 to melt, 75, 79, 312. 3i6
3fa
Silver, to recover from cyanide solution, Swartchild & Co., 341
81, 186 Sweeps, procedure for, 213, 262
from nitrate or sul- Sweeps reducer, 211
phate solution, 7y Swigart, E. & J., Co., 342
to repurify, 179
Silver chloride, to form, 37,-48, So
to reduce to silver, 77, 83, 272 TAILINGS, disposal of final, 152
to dissolve, 118 Taylor, N. O., 131
Silver Jar, definition of, 42, 48 Test for acid and alkalies, 100
to recover silver from, 76, 116 copper, 99
Silver-rich buttons, 106, 140 gold, 90, 94
Sivil, C. S., 208, 297, 351 iron, 100
Skinner, Eugene W., 130, 312, 346 lead, 92-
Smith, Ernest A., 208, 230, 312, 315, mercury, 92, 95
317. 3i8, 334. nickel, 98
346, 347, 349, palladium, 94
350 platinum, 94, 97
Smith, G. F. Herbert, 347 silver, 91, 95
Smyth, Henry F., Jr., 328 Test tubes, 13, 14
Sodium chlorate, 55, 108, 114, 148 to handle, 24
Sodium nitrite, 59, 172 Test tube brush, 13, 14
Sodium sulphide, 85 Test tube clamp, 13, 14
Soft solder, 67, 71 Test tube support, 13, 14
"Solution", definition of, 89 Testing Solution A, 93
Solutions, old, to recover, 217 Thackwell, F. W., 350
Southwest Smelting & Refining Co., 338 Thermal Syndicate, Ltd., 240, 288, 344
Sperry & Co., D. R., 245, 344 Thermo-couples, 196, 199
Speyer Smelting & Refining Co., 338 Thiosulphate (hypo) solutions, 85
Spoon-spatula, 14 Thomas, Arthur H., Co., 340
Spot-plate, 92, 93 Thompson, Lewis, 251
Spyco Smelting & Refining Co., 337 Thum process, 255, 274, 291
Stains on skin, 327 Tin, 6
Stamping laws, 331 acquaintance experiments, 69
Standard solutions, to make up, 93 in soft solder, 67
Standards, commercial, for precious metal in the Stock Pot, 154
articles, 333 oxide, 171, 172
in foreign countries, 334 paste, 70, 72, 75
Stannous chloride testing solution, 92 salt (stannous chloride), 93
Stas, J. S., 202, 204, 208, 354 Tongs, 113, 114, 295
Steam bath, 34, 35 Touchstone tests, 102
Steel. See Iron Travis, Farber Co., 341
Stern, I., & Co., Inc., 338 Triangle, pipeclay, 113
Stock Pot, definition of, 124, 143 Tschantre, John A., 341
to refine, 143, 153, 162, 168 Tungsten, 192
Stoneware, 35, 36, 239, 283, 287
Stripping solutions, to recover metal
from, 80, 186 UNITED States Assay Office, 221, 280
Stripping with the current, 185 U. S. Stoneware Co., 344
Sulphur dioxide gas, 56, 157, 172
Sulphuric acid, action on platinum but-
tons, 106, 141 VACUUM pump, 237, 238
added to aqua regia, 44, 72, Valence, 157, 169
117, 120, 141 Van Dorn Metal Furniture, 344
price of, 36 Ventilation, 30, 241
~> removing base metals with, 53, Venturi, 241, 306
72, 141 Virgin metals, 249
to mix with water, 54 Vitreosil, 240, 288
Sulphuretted hydrogen, 128 Volatilization of platinum metals, 294,
Swanger, Wm. H., 137, 170, 208, 311, 3"9
35i> 352. 353 Von Bcrncwitz, M. W., 328, 346, 349
362 INDEX
WAGOR, Elry J., 282 Will Corporation, 340
Waples, William S., 342 Williams, Charles W., 338
Wash-barrel residues, disposal of, 153, Williams Gold Refining Co., 338
215 Williams, Reginald V., 352
Wash-barrels, arrangements of, 244 Wilson, H., 347
vs. filter-presses, 243 Wilson, H. A., Co., 194, 337
Wash-bottle, 12, 13, 113 Wise, E. M., 297, 3r3, 316, 348, 350,
Watch glasses, 113, 114 35i
Water and plumbing in the refinery, 31 Wogrinz, Alfred, 346
distilled, 23 Wohlwill process, 256, 265, 278, 281,
Watt, Alexander, 250, 347 284, 291, 292
Wehde, Albert, 347 Wollaston's process, 311
Western Electric Co., 73 Woolf, J. A., 350
Western Gold & Platinum Works, 336 Wright, T. A., 223, 352
Western Precipitation Corp., 344 Wrought gold alloys, for dentistry, 131
Weyerts, W., 86, 350
White, S. S., Dental Co., 338
White gold, composition of, 66 YELLOW powder, 152, 284
attracted by magnet, 115 to reduce, 158,
Whiteley, J. O., 141, 156, 202, 208, 352 to reprecipitate, 177
Wichers, Edward, 170, 204, 206, 207, Young-Neal Co., 343
208, 349, 350,
352
Wigley, Thomas B., 312, 347 ZINC, 6, 144
Wildberg Bros. Smelting & Refining Co., as precipitant, 60, 81, 144, 160, 163
336 to reduce silver salts, 78,
83

Refining Precious Metal Wastes - C M Hoke

Uploaded by

Copyright:

Available Formats

Refining Precious Metal Wastes - C M Hoke

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Refining Precious Metal Wastes - C M Hoke

Uploaded by

Copyright:

Available Formats

REFINING METAL WASTE

PALISADE, NEW JERSEY. C. M. HOKE

This book is based on twenty-four years of experience in teaching

ing the orange powder and melting the sponge. (8) Re

THE PURPOSE OF THE BOOK

It is pertinent to digress here and recall the origin of chemistry. . . .

color-perceptions, a fairly steady hand, and will perform in order

Jeweler's wastes. Dental wastes. Photographer's wastes. Base metals

to be worth recovery in most cases. Then we have old plating

(h) Cadmium a metal closely related to zinc in many of its

OLD ELECTROPLATING AND STRIPPING SOLUTIONS

Solutions for pickling, stripping or plating gold and silver are

ECONOMICS OF REFINING THE VARIOUS KINDS OF WASTES

Let us compare the expense of selling a lot of high grade material to a

and samples these materials, will need to question the economics of

THE SIMPLEST CASE PLATINUM FILINGS THAT

Equipment and chemicals. "Acquaintance experiments." Beginning work;

These are obtainable at almost any chemical supply house; noth-

Casserole; stirring rod;

small samples. You may need additional casseroles later, especially

Magnet. Wash-bottle. Funnel.

One kind of filter paper. One kind of funnel support.

Test Test tube Spoon- Test tube

Sieve. One type of burner. Frying pan.

Test tube brush No. 19058 or equivalent, about $0.10.

Let us become acquainted with our materials through exposing

gas of extremely unpleasant nature. Thin wire will disappear within

BEGINNING WORK: SIEVE; MAGNET; CAUSTIC; HEAT

Let us start with a small quantity of filings or scraps. A few

If the filings contain much shellac or grease, place them in a

THE NITRIC ACID TREATMENT

HOW TO FOLD A FILTER PAPER

Use a paper of the right size. Fold it in half.

Add the acid slowly, or it might boil over. If there is much

To repeat the procedure, in one sentence, is the removal of

QUESTIONS AND ANSWERS ON CHAPTER III

Q. Should I use distilled water in this work?

ARRANGEMENT OF EQUIPMENT. DISPOSING OF

Disposing of fumes. Material for hoods. Lighting. Burners. Fans and

Every jewelry shop needs a method of carrying off acid fumes.

One inexpensive device for removing

almost never in operation overnight. This means that the acid

Another suggestion is that the jeweler visit some neighboring school or

and will take pleasure in explaining to a visitor how the fume-hoods

The chemists normally use glass-walled hoods, with flues of terra

A standard chemical fume

wood be coated with acid-proof paint. Back walls of Transite or

The arrangement of the gas burners is important. Run the gas

FANS AND BLOWERS

The width of the slots at top

plished by locating the fan low; sometimes a baffle plate is so ar-

Running water should be available if possible; otherwise provide

In large shops nitric and hydrochloric acids are purchased in car-

With one very minor exception (sodium chlorate) the chemicals

In general, when chemicals are spilled upon hands or clothing,

Ihe best immediate

water; after which it may be well to neutralize with a little vinegar,

In Chapter XX will be found a further discussion of equipment and

THE COMMONEST CASE—REFINING GOLD SCRAP

The first refining. Equipment and chemicals. "Acquaintance experi-