Full text of "Alloys (non-ferrous)"

See other formats

Jit®

ALLOYS (Non-ferrous).

THE

SCIENTIFIC

PUBLISHING CO

ALLOYS

(NON-FERROUS)

— BY —

A. HUMBOLDT SEXTON,

F.I.C., F.C.S., &c.

Professor of Metallurgy in the Glasgow and West of Scotland Technical College.

PRICE 7s. 6d. net.

MANCHESTER :
THE SCIENTIFIC PUBLISHING COMPANY.

»«.*>*

Professor E. MARION HOWE, LL.D.,

Columbia University, New York, in recognition of the help
and stimulus derived from his various works on Metallurgy.

?/0!v

PREFACE.

THE literature of Alloys is not very large, and such great
advances have been made during the last few years that
little apology is needed for the preparation of another
book on the subject.

The object which the author had in view in the prepara-
tion of this book was to provide for students, and more
particularly for makers and users of alloys, a brief account
of some of the advances which have recently been made
in the study of Alloys. His object has not been in any
sense to provide a complete account of what has been
done, but rather to indicate the direction in which
research is going on, and to give to the maker and user of
alloys such information as may be of practical utility,
and which will enable him to clearly understand the
current literature of the subject, and to see in what
direction he must look for further developments that
may be of use to him.

A good deal of the work which has been published
can only be regarded as a preliminary explanation, and
in many cases the results await confirmation or modifica-
tion by fuller research on a larger scale than has yet been
attempted. The work of Dr. Carpenter and Mr. Edwards
for the Alloys Research Committee on the Alloys of
Aluminium and Copper is an example of the work which
needs to be undertaken for the various other groups of
alloys.

The scope of the work did not allow of the considera-
tion of many important points which are not yet fully
worked out, such, for instance, as the meaning of some
of the thermal changes of a minor character noticed during
the cooling of alloys, or of some recent methods of

337660

VI. PREFACE.

research such as the use of ultra violet light, from which
important results may be obtained in the future. The
alloys of iron are so many and so important that a
separate volume would be needed to consider them in
anything like useful detail. It has, therefore, been
thought better to omit them altogether in this work.

At present the results of the researches that have
been made are scattered through the journals of many
Scientific and Technical Societies and the Technical
Journals, and as there is no Society specially devoted to
this work, the field which has to be travelled over is
unusually large, and many valuable papers have been
published in the journals of societies which are difficult
to obtain. It is quite likely, therefore, that the author
may have overlooked valuable papers, and in some cases
he has had to depend on abstracts as the papers them-
selves were not available. Much of this information is,
therefore, quite inaccessible to the general reader, and
the author hopes that the bringing of a good deal of it
together into one book may not be altogether useless.

To previous writers on the subject the author is, of
course, deeply indebted, and he has not scrupled to
quote from them when necessary. To them and to the
Societies (especially the Society of Mechanical Engineers),
authors of papers, and publishers who have kindly allowed
him to use their results and illustrations, he gives his
best thanks. Acknowledgments have been made in the
text, and if by oversight any have been omitted, he
here expresses his regret.

His best thanks are also due to his assistants, Messrs.
J. S. G. Primrose, A.G.T.C., and J. A. C. Edmiston, for
the valuable help they have given in various ways.

CONTENTS.

CHAPTER I.

INTRODUCTION

CHAPTER II.

THE PROPERTIES OF ALLOYS AS RELATED TO THOSE OF

THEIR CONSTITUENTS ...... 7

CHAPTER III.

THE PHENOMENA OF SOLIDIFICATION 20

CHAPTER IV.

WHAT THE MICROSCOPE CAN TEACH 43

CHAPTER V.

CHANGES IN THE STRUCTURE OF ALLOYS IN THE SOLID

CONDITION 64

CHAPTER VI.

METALS USED IN THE PREPARATION OF ALLOYS . . . . 75

CHAPTER VII.

THE BRASSES (THE COPPER-ZINC SERIES) . . •'...'.. 87

CHAPTER VIII.
THE BRONZES (THE COPPER-TIN SERIES) 120

CHAPTER IX.

MACHINERY BRASSES AND BRONZES, BEARING BRONZES.

AND OTHER COPPER ALLOYS ....... .. .. 147

CHAPTER X.

WHITE ALLOYS ; ALLOYS IN \VHICH TIN is THE PRINCIPAL

CONSTITUENT 172

Vlll. CONTENTS.

CHAPTER XI.

WHITE ANTI-FRICTION ALLOYS , ... .. ./ ,. .. 184

CHAPTER XII.
LIGHT ALLOYS ; FUSIBLE ALLOYS ... . . . . . . . . 197

CHAPTER XIII.

NICKEL ALLOYS . ,. . . . . • . . . . . . 209

CHAPTER XIV.

ALLOYS OF THE PRECIOUS METALS .. .. .. .. 217

CHAPTER XV.

PREPARATION OF ALLOYS . .... 244

ALLOYS.

CHAPTER I.
INTRODUCTORY.

WHEN two or more metals are melted together they,
as a rule, remain intimately mixed, showing little
tendency to separate according to their densities, or
as it may be otherwise expressed, they remain in
solution one in the other, whilst they are in the liquid
condition. When the mass solidifies this state of uniform
distribution or mixture may continue, or it may be broken
up. In the former case the solidified mass will contain
the constituent metals in a condition of more or less
uniform diffusion, and such a mass is called an alloy.
In the latter case the metals will separate according to
their specific gravities, the heavier metal going to the
bottom and the lighter rising to the top, such separation
being as a rule the more complete the slower the solidifica-
tion. The separated metals in this case are rarely if ever
pure, but each retains a small quantity of the other,
and strictly speaking both are therefore alloys. In
practice, however, the term alloy is restricted to those
cases in which neither of the metals is present in very
small proportion ; the other cases being simply considered
as metals containing an impurity.

An Alloy is neither a Mechanical Mixture nor a Chemical Com-
pound.— An alloy is, then, an intimate mixture of two or
more metals, and the term mixed metals has sometimes
been used in place of alloys. This is, however, very mis-
leading, as the alloys are much more than mere mixtures,
and mixtures of metals may exist which are not alloys.
If lead and copper be melted together, and the mixture
be slowly cooled, the metals will separate. If, however,
the mixture be quickly cooled separation cannot take place

2 • INTRODUCTORY.

and the metals will remain mechanically mixed the one
with the other, they not having had time to separate
into distinct layers, but the mass will consist of intermixed
particles of the two metals, and if it be heated up to the
melting point of lead this metal may be, to a large extent,
melted out. This, then, would be a case of a mixture
of metals, but not of an alloy. In an alloy, the mixture
must be of such a character that the constituent metals
lose their individuality, and become blended into a new
substance which has properties, to some extent at ieast,
unlike those of its constituents.

As a rule, substances which are not elements are
divided into the two classes, chemical compounds and
mechanical mixtures, but the metallic alloy cannot be
made to fit exactly into either group.

In a mere mixture the particles, however small and
however intimately they may be mixed, always retain
their individuality, and the properties of the mixture are
always a mean of those of its constituents. If the con-
stituents be black and white the mixture will be grey,
if red and white, a paler shade of red, and so on through
all the other properties. This, as is well known, is not
the case with alloys. Brass containing, say, 50 per cent,
of copper and 50 per cent, of zinc, is yellow, and this
yellow colour is certainly not a mean between the red of
the copper and the bluish white of the zinc ; nor is the
specific gravity, or indeed any other property of the
brass, a mean between those of its constituents. The
only point in which alloys always resemble their con-
stituents is that they are distinctly metallic.

Alloys, therefore, are not mechanical mixtures.

A chemical compound contains the elements in fixed
proportions, these being always simple multiples of the
atomic weights, and some of the physical properties
follow from the molecular weight of the compound. This
is not the case with alloys. As a rule, the metals are
not present in any simple atomic proportion, and further
the proportions can be varied often within wide limits
without producing any great change in the properties
of the alloy.

The metals do not show any strong chemical affinity
one for another, but there is no doubt that in some cases

INTRODUCTORY. 6

definite chemical compounds of the metals do exist, but
in no case do they form alloys of any industrial inportance.

Solutions. — There is still another form in which sub-
stances can exist which, while not a mere mechanical
mixture, is something less than chemical combination.
If salt or any other soluble substance be stirred up with
water it disappears, or dissolves in the water, and the
result is a solution of the salt. This solution has some
of the properties of the salt ; it has, for instance, a
salt taste, yet its properties cannot be said to be a mean
between those of water and salt. The salt dissolves
without increasing the volume of the solution, so that
the solution is denser than the mean between salt and
water. The addition of the salt also lowers the freezing
point of the water, so that the freezing point of the
solution, instead of being a mean between that of water
and salt, is lower than that of either, and by the
addition of proper proportions of salt it may be reduced
to about -22 -5° C.

The essential character of a solution is that the con-
stituents are so intimately blended that they cannot be
separated or detected by mechanical means, whilst at
the same time they have not entered into true chemical
combination.

Solid Solutions. — As a rule, when a solution is frozen,
the constituents separate one from another to a larger
or smaller extent, but this is not always the case. We
can imagine a solution to become solid without any
other change, and the result would be a solid solution in
which the constituents would still be so intimately mixed
that no mechanical separation would be possible, and in
which the properties would not be a mean of those of
its constituents, but in which these constituents would
not be present in the definite proportions required for a
chemical compound.

Alloys. — True alloys are never mere mechanical
mixtures of metals, and though in some cases the metals
do combine, yielding definite chemical compounds
which often retain their metallic properties, none of
these are of any use in the arts. Alloys are very fre-
quently solid solutions of one metal in another, or of

Jfe . INTRODUCTORY.

a chemical compound of the metals in the metal which
is in excess. Many consist of mixtures of such solu-
tions with definite substances that have crystallised
out during cooling, so that the actual composition and
structure may vary very widely ; and each alloy, or rather
group of alloys, must be studied separately, as it is
impossible to lay down any except the most general rules.

Importance of a Knowledge of the Structure of Alloys. — It may be
thought that the structure of alloys, whilst interesting
enough as a scientific study, will be of very little im-
portance to the practical maker and user of alloys.
This is, however, far from being the case. The properties
of alloys depend to such a large extent on their structure,
that without a knowledge of the latter the former cannot
be understood. Many of the failures in the making of
alloys are due to changes in structure brought about by
small changes in the method of treatment or in other
ways, and failures cannot be prevented till the causes
which produce them are known ; and therefore such
knowledge is of great practical importance to all who
have to deal with alloys

Methods of Investigation. — Chemical analysis is, of course,
of primary importance, as giving the proportions in which
the various constituents are present ; but its uses are
limited to that, and that alone is not enough, for it is
often important to know not only what elements are
present, but how these are united ; that is, to know not
only the ultimate but the proximate composition. At
present the proximate analysis of metals and alloys has
made little progress, and only in a few cases has it been
found possible to separate any of the proximate con-
stituents by analysis. For chemical analysis it is neces-
sary to attack the whole alloy, and the reagents used
for this purpose have such a vigorous action that they
break up all the constituents of the alloy into their
elements, which then enter into new forms of combination.

The Microscope. — Within the past few years the appli-
cation of the microscope to the examination of metals has
developed very rapidly Accurate methods of work are
now known, and a vast amount of data has been gathered.
The microscope can supplement chemical analysis. It

INTRODUCTORY. 5

does not enable us to determine what chemical elements
are present, but it enables the structure of the body to
be made out, and particularly the constituents that may
have separated as the alloy has solidified, and therefore
to determine some at least of the facts which chemical
analysis leaves undetermined.

Phenomena of Solidification. — If alloys, when melted, can
be considered as being solutions of one metal in another,
then a study of the phenomena which take place during
the solidification and cooling of solutions should throw
great light on the changes which take place during the
solidification of alloys. This has actually been found to be
the case. Many solutions which solidify at a moderate
temperature are much more easily studied experimentally
than the alloys which melt at a high temperature, and it is
comparatively easy, with due care, to reason from the
one to the other.

Physical Properties. — The physical properties of alloys are
found to vary with changes in their composition and
methods of treatment, though the changes are often com-
plex, and in many cases do not seem to bear any simple
relation to the composition. Nevertheless, they throw
much light on the points at which changes take place.

Four Methods of Research.— These four methods of research
—viz., chemical analysis, microscopic examination,
study of the phenomena of fusion and solidification,
and study of the physical properties of the alloys —
have now been applied to most of the important series
of alloys. They supplement one another, and, taken
together, enable very many valuable inductions to be
made, so that what may be called a theory of alloys can
now be formulated.

Scope of this Book. — In this book it is intended first
to describe briefly the methods by which the structure
and character of alloys have been determined, and then
to give a brief account of the more important groups of
alloys and the methods of preparing them. The object
will be to give an account of the investigations that have
been made and the facts that have been discovered
during the last few years, in so far as they are likely
to be of any practical importance to the maker and

6 INTRODUCTORY.

user of alloys. In some cases it may be unavoidable that
the subjects discussed should be to some extent theoretical,
and perhaps appear somewhat far away from practical
utility ; but the needs of the practical worker are always
kept in view, as it is for him that the book is mainly
intended. At the same time, it will be a more or less
complete account of the present condition of our know-
ledge of alloys.

CHAPTER II.

THE PROPERTIES OF ALLOYS AS RELATED TO THOSE
OF THEIR CONSTITUENTS.

ALLOYS contain two or more metals in some form of mixture
or combination, and at the outset it will simplify matters to
consider such as contain two metals only. The properties
of the alloy will, of course, differ from those of its con-
stituents, but it remains to inquire how the two sets of
properties are related. It is evident that the properties
of an alloy must to some extent depend on those of its
constituents, for this is true even in the case of chemical
compounds ; but alloys, as remarked above, are not
definite chemical compounds, and therefore we should
expect a much closer resemblance between the properties
of the alloy and those of its components.

It must be remembered that a series of alloys can
usually be obtained in which the percentage of one metal
can be increased from 0 to 100, and that of the other
diminished from 100 to 0 ; the two extreme members of
such a series being the pure metals. And it is of interest to
study how the properties of such a series will vary.

If the alloy be a mere mixture, then the variation
should be always in the same direction, and should be
proportional to the amount of the second metal added ;
if it be a chemical compound, or set of chemical compounds,
sudden breaks or changes in properties might be expected
when compositions are reached corresponding to the
definite compounds ; if it be a solid solution, variations
might be expected continuously in the same direction,
but not necessarily proportional to the amount of the
added metal, as changes in solubility may take place as
the solvent metal becomes more nearly saturated, and if,
as is possible, two, or perhaps all these conditions may
co-exist, we should expect more or less irregular variations.
In this chapter some of the principal properties will be con-

8 THE PROPERTIES OF ALLOYS.

sidered, the alloys being selected so as to best illustrate
the property under consideration, and later the alloys
of commercial importance will be separately dealt with.

Colour. — There is no property which is subject to more
striking variations by admixture of metals than colour.
Unfortunately, we have no quantitative method of
valuing colour, so that comparisons can only be general
and qualitative.

Most of the metals are white, though the shade of
whiteness varies very much, from the silver- white of silver
or aluminium to the very bluish-white of lead. There
are only two metals in common use with well-marked
colours — gold, which is yellow, and copper, which is
usually said to be red, but which may best be defined as
being copper-coloured.

All alloys of the white metals one with another are
white, but the shade of white bears little or no relation
to the shade of the metals themselves. When a white
metal is alloyed with a coloured metal, great changes
of colour are often produced. These changes may be
looked at from two points. Starting with the white
metal, the gradual change of colour which is produced by
the addition of increasing quantities of the coloured
metal may be noted ; or starting from the coloured metal,
the changes in colour produced by the addition of in-
creasing quantities of the white metal may be considered.
In general, the second method is the most satisfactory,
because, strange though it may seem, the alteration of
colour produced by the addition of a comparatively
small quantity of a white metal to a mass of the coloured
metal is usually far greater than that produced by the
addition of a small quantity of the coloured metal to a
mass of the white one, and in either case the resulting
colour seems to bear little relation to that of the con-
stituent metals.

Almost any series of such alloys might be taken as
an example. In the case of the aluminium-copper
series for instance, whilst 5 per cent, of copper makes
very little difference to the colour of aluminium, 5 per
cent, of aluminium changes the red colour of copper to
a fine yellow.

THE PROPERTIES OF ALLOYS. 9

The most striking colour series, however, is probably
that of the copper-zinc alloys. Five per cent, of zinc added
to copper destroys the red colour and gives a yellow
alloy ; as more zinc is added the yellow colour becomes
more intense, though by no means uniformly so with each
addition of zinc, till the composition 50 per cent, copper
and 50 per cent, zinc is reached, this being the composition
of common yellow brass. As the proportion of zinc is
increased to about 60 per cent, a silver-white alloy,
white brass, is obtained, and with still increasing zinc
this silver-white gives place to the bluish-white colour of
zinc itself.

There are other cases where the colour change is
still more striking, such as the alloy of about 51 per
cent, of copper with 49 per cent, of antimony, known as
Regulus of Venus, which has a fine violet colour, and the
alloy of gold with 10 per cent, of aluminium, discovered
by Sir W. Roberts-Austen, which has a ruby-red colour.

Specific Gravity. — When metals are melted together and
solidified, the volume of the mixture is rarely the same
as that of the metals separately. There is either contraction
or expansion, and the specific gravity of the alloy is

L'O
£18

I-

10
.'.'.' 10

•^

""•

0 20 40 60 80 100) Percent
0 80 60 40 20 0 / volume.

Gold

Lead ... 100

FIG. 1.— SPECIFIC GRAVITY OF LEAD-GOLD ALLOYS.

therefore either greater or less than the mean of that of
its constituents in the same proportion. In most cases
the specific gravity of an alloy is greater than the mean
of that of its constituents, as is also usually the case with
solutions.

As a rule, the specific gravity falls as the quantity of
the lighter metal increases, but not in direct proportion

THE PROPERTIES OF ALLOYS.

to the quantity of the lighter metal present ; in such cases
a curve representing the densities of a series of alloys
will be a continuous curve rising above a straight line
connecting the specific gravity of the two constituent
metals, as is shown in the curve Fig. 1, which repre-
sents the densities of gold-lead alloys as determined by
Matthiesen. * On the other hand, antimony and tin expand
on alloying so that the specific gravity of the alloy is less
than the mean of its constituents, and therefore the curve
lies below the straight line, as shown in Fig. 2. In other

7-3

]
•

7'2

•

I7'1

o 7'0

6. 0

fi-7

>_ —

— »-

•

^>~~

)

inn

Tin

Antimony... 100 80 60 40 20

FIG. 2.— SPECIFIC GRAVITY OP ANTIMONY-TIN ALLOYS.

n [- cent, by
" I volume.

cases the increase or decrease is not regular, but there are
breaks at certain points. In the case of the copper-tin
series, the specific gravity of the alloy decreases as the
quantity of tin increases, till there is about 28 per cent, of
tin present ; then it rises till, when the percentage of
tin is about 38, it reaches 8- 9, and therefore the alloy
is denser than copper, then as the tin is further increased,
the specific gravity falls continuously. Richef states that
the curve of densities has two maxima corresponding to
Cu4 Sn (68-12 °/o Cu) and Cu, Sn (61-58 % Cu).

It is quite obvious, therefore, that the specific gravity
of an alloy cannot be calculated from that of its con-
stituents, unless we know exactly what changes take
place when the metals are alloyed, and that experiment
alone can decide.

* Phil. Tran. Royal Soc., 18GO, p. 183. t Comp. Rendu, 55, 162.

THE PROPERTIES OF ALLOYS. 11

Except when alloys are prepared specially for experi-
mental work it is difficult, if not impossible, to obtain a
series of varying composition, under exactly similar
conditions, so that an accurate comparison can be made,
and as the treatment to which the alloy has been subjected,
such as work put on it, annealing, rate of cooling, &c., all
may influence its specific gravity, concordant results are
very often not obtained by workers working under
different conditions.

The densities of the various alloys of commercial
importance will be considered later.

Tenacity. — The tensile strength of an alloy may be
greater or less than that of its constituents. Alloys of
copper and zinc, or copper and tin, in certain proportions,
are much stronger than either of the metals ; but, in
other proportions, they are much weaker. Here, again, no
definite rule as to the influence of one metal on another
can be laid down.

Ductility. — As a rule, increase of tensile strength is
accompanied by decreased ductility, but such is not
always the case. The addition of a foreign metal to a
soft ductile metal will as a rule harden it, and decrease its
ductility, but this is certainly not the case in every instance,
some alloys being not only stronger, but also more ductile
than either of their constituents. In some cases the
hardening influence of even a small quantity of foreign
element is well marked, as in the case of antimony on
lead, bismuth on gold, and carbon on iron.

The influence of foreign metals in diminishing the
tenacity and ductility of other metals seems to depend
on their atomic volume.* The larger the atomic
volume of the added metal the greater as a rule
is the reduction of tenacity and ductility. The late
Sir W. Roberts-Austen made a large number of experi-
ments on the influence of foreign metals on gold, and
found that the elements of high atomic volume reduced
the tenacity and also the ductility of gold, whilst the

* The atomic volume is the proportional volume occupied by this molecule,
and is Atomic weight j h f w th atomic volume i8 J^-L =l()-2.

Specific gravity.

That of bismuth is -^Sr = 21 '2. That of aluminium is ~= 10'6.

9o2 A o

12 THE PROPERTIES OF ALLOYS.

metals of low atomic volume either have no influence or
increase the tenacity. Pure gold has a tenacity of about
7 tons, and elongates about 30 per cent, before breaking,
with a test piece of the length used for the experiments.
When the gold was alloyed with bismuth, only • 21 per
cent, of that metal being added, the tensile strength was
only • 5 ton, and the elongation was imperceptible. When
alloyed with • 186 per cent, of aluminium, which stands at
the other end of the series, the tensile strength was
8-87 tons and the elongation 25-5 per cent. Silver,
which has the same molecular volume as gold, has little
effect one way or the other. It is not certain whether a
similar rule holds for all the other metals.

Fusibility. — The melting point of an alloy is in-
variably lower than a mean of those of its constituents,
and in some cases is even lower than that of the most
fusible. There is, however, great uncertainty as to the
exact meaning of the melting point of an alloy, as the term
is usually used, and as will be seen later, many alloys
have no distinct melting point, but a longer or shorter
melting range.

Expansion by Heat. — The expansion of alloys by heat
is very irregular, and does not seem to bear any definite
relation to that of their constituents.

Electric Properties. — The electric properties of alloys have
been in many cases carefully studied, and are of very great
interest. The electric properties can be accurately measured
and are thus capable of giving, not only qualitative but
quantitative results, and at the same time the relation
between electric properties and composition in substances
other than alloys has been so fully investigated that
light may be thrown by analogy on the structure of
alloys. There are three electric properties that are of
importance in this connection.

(1) Electric conductivity ;

(2) Action of an electric current on fused alloys

(electrolysis) ;

(3) Potential difference set up by the contact of

alloys with metals.

Electric Conductivity. — The influence of the addition of one
metal to another on the conductivity is well marked,

THE PROPERTIES OF ALLOYS.

and as a rule the conducting power is considerably
reduced. Alloys in general may be divided into
two groups, according to the way in which the
conducting power of the alloy is related to that
of the constituent metals. In the first group the alloy
behaves exactly as if it were a mere mixture of the two
metals, the conducting power rising or falling as the
percentage by volume of the second metal is increased,
according as the starting point is the metal of less or
greater conductivity. The curve for conductivity
when plotted, is therefore a straight line. This may
perhaps be called the normal curve, though as a matter of
fact a comparatively small group of alloys gives it. As
an example the tin-zinc series may be taken, and the
curve for these is plotted in Fig. 3.

Tin ...
Zinc...

60
28
26
24

20
18
16

14
12

. (
. 1(

•

1 20 40 60 80 100 (Per cent.

10 80 60 40 20 0 j" ,.„£„,.

FIG. 3.— CONDUCTIVITY OF ZINC-TIN ALLOYS.

The metals belonging to this group are lead, zinc,
tin, and cadmium, and the rule applies to alloys one with
another.

It is obvious that the addition of a small quantity of
one of these metals to another can have a very small
effect on the conducting power, because the resulting

THE PROPERTIES OF ALLOYS.

conductivity will be proportional to the volumes of the
constituents present.

With alloys of the other metals with the members of
this group or with one another, the result is very different.
The conducting power of the alloy is always less than
what it would be if the metals all behaved like the metals
of the zinc-lead group. That is, when the curve is
plotted, it will always lie entirely below the straight line
100,

Lead 0 20 40 60 80

Silver ... 100 80 60 40 20

FIG. 4.— CONDUCTIVITY OP SILVER-LEAD ALLOYS.

100) Per
n - cent, by
U j volume.

joining the conductivities of the two metals. The curve
itself may in general take two forms. Alloys of the
metals of the lead-zinc group with other metals show a
somewhat curious behaviour. If a small quantity of
a metal of the lead-zinc group, say lead, be added to
a metal not belonging to the group, say silver, it
causes a very rapid diminution of conducting power,
whilst the addition of a small quantity of silver to
lead has a small effect on its conducting power. When
the curve for such an alloy is plotted, it takes, there-
fore, somewhat of an L form, the vertical portion
falling sharply, it may reach a minimum, in which case
the continuation of the curve will rise, or it may con-
tinue to fall at a much reduced rate. The curve is

THE PROPERTIES OF ALLOYS.

not always continuous, but may show a sharp break or
change of direction, and these breaks do not seem to
correspond to the formation of definite chemical com-
pounds.

The curve for the silver-lead series is shown in Fig. 4.

With metals not of the lead-zinc series alloyed with one
another the curve is somewhat different, for here the

Oft

•

rt/\

f>J fift

1 50
o OU

*t)

o 40

.^^^

Gold

0) Per

Silver ... 1

(

r cent, by
) volume.

Silver= 100.
FIG. 5.— CONDUCTIVITY OF SILVER-GOLD ALLOYS.

addition of a small quantity of either metal to the other
causes a rapid diminution of conductivity, and the curve
therefore assumes a roughly U shape, the length of the
two vertical arms being unequal and in the proportion of
the conducting power of the constituent metals. The
curve for the gold-silver series is shown in Fig. 5.

These curves can only be taken as examples, for in
no two alloys are the curves the same. It may be that
the curves in cases 2 and 3 are not true curves, but
rather series of straight lines, the alloy behaving between
each turning-point like an alloy of the lead-zinc series.

The bearing of these facts on the constitution of alloys
will be discussed later.

16 THE PROPERTIES OF ALLOYS.

Electrolysis. — When an electric current is passed
through a compound either melted or in solution in a
suitable solvent, the compound is broken up, and the
two "ions," which may be either elements or groups of
elements, are separated at the electrodes. All attempts
to electrolyse fused alloys have failed, they seeming to
conduct in all cases like elements, i.e., not to undergo
decomposition. If any conclusion can be drawn from
a merely negative result, it is that apparently alloys are
not chemical compounds.

Potential Difference. — When two metals are connected
by a metallic conductor and put into an electrolyte, a
certain potential difference is set up between them and
a current is set in motion. The behaviour of alloys in
contact with their component metals may thus be of
importance in throwing some light on the constitution of
the alloys.

A. P. Laurie | has made a series of experiments on this
subject. He found that in a cell — a cuprous iodide cell
was used — using plates of copper, and of copper having
pieces of zinc soldered to them, the electromotive force
produced was the same as that produced with plates
of copper and zinc, even when the quantity of zinc ex-
posed was not more than 2^00 °^ ^e area of the plate
exposed.

He then tried alloys of copper and zinc, and he
remarks : —

" There are three possible ways in which zinc-copper
alloys may be constituted : —

" First : They may be merely mixtures of zinc and
copper. In that case they would give the electromotive
force of zinc in the voltaic cell.

" Second : They may be of the nature of the solution
of sulphuric acid in water " (i.e., solid solutions) ; " in
that case a series of such alloys beginning with 100 per
cent, of copper and ending with 100 per cent, of zinc
would probably show a gradual rise of electromotive force
in the cell, from the value for copper to that for zinc.

" Third : One or more of the series may be a definite
atomic compound, the rest being 'solutions of this

t Journal Chem. Soc., vol. liii., 1888, p. 105.

THE PROPERTIES OF ALLOYS.

compound or compounds in an excess of zinc or copper ;
in that case the electromotive force would probably
rise by a jump when a series of alloys were tested a slight
excess over that necessary for the compound, causing a
great alteration of electromotive force in the cell.
Further, this jump would probably occur where the
percentage of zinc and copper corresponded with some
simple molecular formula."

Fig. 6 shows the curve as actually obtained by
Laurie. The upper portion is very irregular, this being

•60

i
i

-X

•CO

•40

•30

•20

• 10

... (
... 1(

)

3
)

4'
(11

41)

80
20

100
0

Copper
/inc...

FIG. 6. — ELECTROMOTIVE FORCES OF COPPEK AND COPPER-ZINC ALLOYS
IN CUPROUS IODIDE.

due to irregularities in the alloy (these irregularities
are not shown in the figure), but the general character
of the curve is well marked. The electromotive force

THE PROPERTIES OF ALLOYS.

remains constant at nearly that due to pure copper and
zinc till about 32 per cent, of copper is present in the
alloy, then it suddenly drops to a very small amount,
and finally becomes nil, or, rather, in this case it becomes
negative and the current is reversed.

There are many difficulties in obtaining perfectly
concordant results with such experiments, but the
genera] result is quite clear ; and Laurie infers from this
and other experiments the existence of a definite compound
Cu Zn2 (32- 71 per cent. Cu). In the case of copper-tin

•bO
B

•50

•40

g -30

^ A

%
H -20

•10

Copper ...
Tin 1

•^

0 20 40 60 80 100) Percent

CO 80 60 40 20 0 j" „$£*.

FIG. 7. — ELECTROMOTIVE FORCE OF COPPER AND COPPER TIN ALLOYS.
AA in Stannous chloride.
BB in Stannous sulphate.

alloys (Fig. 7) exactly similar results were obtained, and he
inferred the existence of a definite compound Cu3 Sn
(Cu 61- 58 per cent.) and Cu4 Sn (68- 1*2 per cent Cu).

Putting all these considerations together, it will be
quite obvious that the constitution of alloys is not by

THE PROPERTIES OF ALLOYS. 19

any means simple, and that the physical properties of
the alloys do not enable us to form any conclusion as to
what that constitution really is.

CHAPTER III.

THE PHENOMENA OF SOLIDIFICATION.

WHATEVER may be the exact nature of an alloy in the
solid condition, in the liquid condition it is comparatively
simple, and may be regarded as being a mixture of the
two Liquid metals, or, rather, as a solution of the one in
the other. It therefore follows the law of solution, the
two constituents tend to diffuse until the composition is
uniform, and this diffusion can, of course, be greatly
accelerated by agitation, by stirring, or otherwise. The
few cases which are not solutions, but mere mixtures,
and in which, therefore, the constituents will tend to
separate according to their specific gravity, may be
neglected as being of little importance.

The essentials of a solution, as far as they are of
importance here, are :—

(1) The constituents are uniformly distributed, so

that the composition at all points is the same,
and the constituents do not tend to separate
even when very different in specific gravity ;

( 2) The constituents are in such a fine state of division

that they cannot be detected by physical tests,
and therefore the solution is essentially one
substance.

When the solution solidifies this condition of things
may or may not continue. The solution may solidify as a
whole, in which case the resulting solid still retains some,
at least, of the properties of the solution, and is called a
solid solution, but more usually the conditions of
equilibrium are disturbed, and a re-arrangement of the
constituents takes place during solidification.

The phenomena of solidification have been studied
in detail during the last few years, and much light has
been thrown on the structure of complex bodies which
have solidified from fusion, or from solution, for the

PHENOMENA OF SOLIDIFICATION.

conditions which hold in the two cases are much the same,
the principal difference being the actual temperature
at which the solidification takes place. It is quite obvious
that the fact that the solvent and the substance in
solution are metals, and that a high temperature is
required to keep the solvent in the liquid condition,
cannot in any way alter the laws according to which the
solidification will take place ; so that the solidification of
igneous rocks, alloys, and solutions of salts will follow
exactly the same principles.

Solidification of Pure Substances from Fusion. — When a pure
substance, whether it be an element or a com-
pound, solidifies from fusion, the thermal phenomena

Time.
FIG. 8. — FREEZING CURVE OP WATER (DIAGRAMMATIC).

are very simple. The temperature falls steadily till
solidification begins, then it remains constant till
all the substance has solidified, then it once more
begins to fall, and falls steadily. There is thus
always a fixed and definite melting point, and either
when the solid is melting or the liquid is freezing
the temperature remains constant at that point,
till the change is completed. If, therefore, a freezing

22 PHENOMENA OF SOLIDIFICATION.

curve be drawn, the ordinates being the temperatures,
and the abscissae times of cooling, it will take the form
represented in Fig. 8, which represents diagrammatically
the freezing curve of water.

Solidification, of course, always begins at the points
where cooling is most rapid, that is, almost invariably
at the outside. Crystals begin to form, and these grow
inwards into the still liquid mass, very likely
crossing and interlacing so as to form a net-work,
and ultimately the liquid within the net-work solidifies
and binds the whole together ; but there is no
difference in composition between the substance which
freezes first — that is, the first formed crystals — and that
which freezes last, the inter-crystalline material. Whether
the resulting solid mass will show a distinctly crystalline
structure or not will depend largely on the size of the crystals
which form and on their character. In the case of a
metal, if the crystals are large and have a good cleavage,
the fracture will almost always appear crystalline, but
if not it will probably be granular.

When crystallisation begins, as it very frequently
does, at many centres throughout the mass, the crystals
as they grow will press against one another, the true
crystal form will be obliterated, and a granular structure
will result, the grains being allotriomorphic or distorted
crystals.

Distinct crystals of visible size will only be produced
when, owing to shrinkage or any other reason, the still
liquid material is drained away before solidification is
complete, leaving the crystal projecting into a cavity.
Occasionally, large crystals of metals are formed in this
way.

Solidification of Solutions. — A solution may be considered
as a homogeneous liquid mixture of two substances with
different solidifying points, it matters not whether one
or both be solid at ordinary temperatures. The way
in which a solution will soljdify will vary very much with
the way in which the constituents behave to one another
in the solid condition.

As an example, the behaviour of a solution of common
salt (sodium chloride) in water will be considered, as

PHENOMENA OF SOLIDIFICATION. 23

this will give a key to the more complex phenomena
which take place in the solidification of alloys.

Solidification of a Solution of Salt. — Salt will dissolve in
water in varying proportions up to about 25 per cent.,
the exact amount required for saturation depending on
the temperature. The phenomena which occur during
solidification depend on the amount of salt present.

First : The case of a very dilute solution, one con-
taining not more than, say, about 1 per cent, of salt.
Suppose a thermometer capable of indicating small
changes of temperature to be immersed in it, and the
temperature to be slowly reduced. The temperature
falls, and when it reaches 0° the water does not freeze.
At -- • 6° O., however, solidification begins, the curve
representing the thermal change being exactly similar
to that representing the freezing of water, except that
the halt in the fall of temperature takes place at a
slightly lower temperature than in the case of pure water.
The solid mass is no longer pure ice, but contains salt,
and is, in fact, a solid solution of salt in ice.

This is one of the fundamental phenomena of the
solidification of solutions. The solution always solidifies
at a lower temperature than the freezing point of the
solvent. The lowering of the freezing point depends on the
quantity of the foreign substance present, and on its mole-
cular weight, the lowering being the same for the molecule
of any soluble substance, so that the lowering of the freezing
point gives a means of determining the molecular weight of
solid bodies in solution. The molecular depression of the
freezing point in the case of the solution of salt in water is
given as 35- 1° C. That is, if the molecular weight (58-5)
in grammes of salt could be dissolved in 100
grammes of water the freezing point would be reduced
to — 35 • 1°. This is impossible, as water will not dissolve
such a large amount of salt, but the reduction by the
solution of any given weight of salt will be in the same
proportion, so that 1 per cent, of salt will lower the
freezing point to — '6° C.

Now consider a solution containing, say, 10 per cent,
of salt. On cooling, the temperature will fall to about
— 6° before solidification commences. Then there will be a

PHENOMENA OF SOLIDIFICATION.

halt of cooling as before, and when the mass has
apparently solidified, the temperature will continue
to fall, but when a temperature of - 22- 5° is reached,
there is another halt in the cooling. In fact, the
solution has no longer solidified as a whole. At the
higher temperature the ice containing a small quantity
of salt solidified, or rather, commenced to solidify, the
solidification then going on continuously, the mother
liquor still left liquid being gradually enriched in salt,
tilLTat -22-5° it solidified. So that it is hardly correct to

-30°

Time.
FIG. 9.— COOLING CURVE OP 10% SALT SOLUTION (DIAGRAMMATIC).

say that a solution containing 10 per cent, of salt has a
definite freezing point unless it be the point at which
solidification is completed. It has, in fact, two freezing
points, the one that at which solidification begin-, the
other that at which it is completed. The former varies
with the percentage of salt, the latter is fixed.

Fig. 10 shows diagrammatically the phenomena of
the solidification of a solution containing 10 per cent,
of salt.

If now a still stronger solution be examined, an exactly
similar result will be obtained, the first freezing point

PHENOMENA OF SOLIDIFICATION.

S -

will be a good deal lower, but the second will be as before
at — 22 • 5°. If a solution containing 23 • 6 per cent,
of salt be taken, the temperature will fall continuously
to — 22 • 5°, when it will remain constant till the whole has
solidified, exactly as is the case with pure water.

If a solution slightly stronger in salt be used, salt will
crystallise out till the temperature falls to — 22 • 5°, when
the mother liquor containing 23 • 6 per cent, of salt will
solidify as a whole.

The solidifying temperatures, or freezing points, of a
series of salt solutions are shown in Fig. 10, the freezing
point curve. The ordinates represent temperatures,

c
40°

-40°

•20

Percentage of Common Salt.
FIG. 10.— FREEZING-POINT CURVE OP SALT WATER.

the abscissas the percentages of salt in the solution. The
line A B shows the temperature at which freezing begins,
and B C the temperature at which the precipita-
tion of salt begins, whilst D E shows the temperature
at which the mother liquor finally solidifies, which, as
will be seen, is constant, and begins to be observed as
soon as the total quantity of salt is in excess of that
which the solid ice can hold in solution.

The actual phenomena in the case of salt and water
must be carefully considered, as an understanding of
them will help to make the changes which take place
during the solidification of alloys clear.

26 PHENOMENA OF SOLIDIFICATION.

Consider again, what happens when a solution of salt
containing, say, 10 per cent, of salt freezes. As soon as
the freezing point, due to the percentage of salt present, is
reached, the water begins to freeze, and as it does so it
ejects some of the salt from solution, thus making the
remaining solution stronger in salt, and therefore, having a
lower freezing point. As the temperature continues to fall
the freezing point of this solution is reached, and more salt
is ejected, and this continues till the residual solution or
mother liquor contains 23-6 per cent, of salt, when it
solidifies as a whole. It will thus be seen that the upper
freezing point is that which is determined by the per-
centage of salt actually present, and is the temperature
at which the solidification begins, but this temperature
does not remain constant during solidification, as in the
case of pure water, because the freezing point is gradually
falling. As the water solidifies heat is evolved, so that the
fall of temperature is retarded, but the temperature line
does not remain horizontal ; it rather slopes downwards in
the form of a curve convex upwards, as is shown in Fig. 9.
The mother liquor, which freezes at — 22 • 5°, is of peculiar
character. It has a definite percentage composition, but
it is not a chemical compound, for the constituents are
not present in simple atomic proportions. It is found to
consist merely of an intimate mixture( of salt and ice,
the two being arranged in more or less parallel plates
of microscopic size. Such a solidified mother liquor is
called in the case of aqueous solutions a cryohydrate, or
more generally in the case of all solutions a eu tec tic.

Eutcctic. — A eutectic is the portion of any solution or
alloy which is the last to solidify as the solution is slowly
cooled. It has a definite composition and freezing point,
depending only on the substances which it contains, and
this is the same from whichever end of the series it is
approached. For example, in the case of salt and water,
whether it is reached by the cooling of a dilute solution or
a very strong solution. The constituents of a eutectic are
very rarely, and then only accidentally, present in pro-
portions approaching a simple atqmic ratio, and the
eutectic is always a mixture of the two substances in
distinct portions which have separated in contact.

PHENOMENA OF SOLIDIFICATION. 27

Eutectics in the case of metallic alloys will be fully
discussed later, but it is very important at the outset
to have a clear idea of the meaning of the term.

Solidification of Metallic Alloys. — The idea of a solid solution
has already been alluded to, and it is one that must be
kept in mind in considering the phenomena now to be
described. Molten alloys may be considered as being
solutions of the one metal in the other, but whether this
condition will be retained during solidification will
depend on circumstances.

Three possible conditions may be considered : —

(1) The metals are quite insoluble one in the other

in the solid condition.

(2) The metals are soluble to some extent the one in

the other in the solid condition.

(3) The metals are soluble one in the other in the

solid condition in all proportions.

Conditions 1 and 3 are probably not to be met with,
but they are limiting conditions which may be approached
if not actually reached, and which for simplicity may be
considered in illustration of the subject.

(1) Metals which are insoluble one in another in the solid
condition. — It is assumed that they are soluble in one another
in all proportions whilst in the liquid condition, but that
they are insoluble one in the other in the solid condition ;
and that, therefore, whilst the melted metals will form
a homogeneous solution, they will separate from one
another completely on solidifying.

Let the metals be called A and B, and consider first
the case of an alloy consisting mainly of B, but containing
a small percentage of A, B being the metal of higher
solidifying point. What will happen as the mass solidifies
will be something as follows : As the temperature falls,
solidification will begin at a temperature somewhat below
the actual freezing point of B, because A being in solution
in B will lower its freezing point. As B solidifies A will
be ejected, and thus the mother liquor will become richer
and richer in A, and the freezing point will continue to fall,
the cooling curve, owing to the evolution of heat during
solidification, being convex upwards. As the temperature

28 PHENOMENA OF SOLIDIFICATION.

continues to fall A will continue to be ejected, and there-
fore the mother liquor to become still richer in A, and
therefore its freezing point to fall, until a point may be
reached when the freezing point of the mother liquor
is the same as that of A, then A and the remainder
of B will solidify at once, separation, of course,
taking place at the moment of solidification,
and a solid eutectic, consisting of a mixture
of the two metals, will be formed. Or, if the freezing
point of the mother liquor, even when containing an
unlimited quantity of A, should be above the freezing point
of A, then the material to solidify last will not be a true
eutectic but pure A.

In either case, on the solidification of an alloy of A and
B in any proportions whatever there will be two solidifying
points, the upper one depending on the quantity of A
in solution ; the lower one being the freezing point of A,
and the material however solidified will be merely a
mixture of the two metals A and B. Such cases are
not by any means common, but the condition is nearly
approached in the case of a mixture of copper and bis-
muth, the latter metal always separating, and thus
rendering copper containing it very brittle.

An alloy which contains a eutectic is said to be
" eutectiferous," and in the case of metals related as
A and B are supposed to be, it will be eutectiferous in
whatever proportions the metals may be present.

(2) Metals which arc soluble one in another. — Now, the other
extreme case, that in which the metals are soluble one
in the other in all proportions both in the solid and
liquid conditions, may be considered, in this case the
conditions will obviously be very different from those in
the first case. In considering the changes which take
place, a liquid solution of a small quantity of a metal C
in a larger quantity of a metal D of higher melting point
may be taken as an example.

As the solution cools, solidification will begin at a tem-
perature below the freezing point of D, the exact tem-
perature being determined by the amount of C present.
As solidification goes on, the solidified mass will not be
pure D, but will be D containing a certain amount of C

PHENOMENA OF SOLIDIFICATION.

in solution, but it will be poorer in C than the original
liquid mixture, some C being ejected and the mother
liquor becoming richer and richer in C, and each layer or
portion of D as it solidifies being richer in C than that
which solidified before it, so that if a sample could be taken
it would be found that the percentage of C in the alloy
gradually increases from the first to the last portion solidi-
fied, so that the alloy will not be homogeneous. If a eutectic
be defined as solidified mother liquor, or as the portion of
the alloy which solidifies last, then, of course, there must
be a eutectic in every alloy, but in this case it will not
have the true eutectic structure, i.e., it will not be a mixture

QGp. Temperature Centigrade.

M". 5 S =

• • " O O O O C

^^^^J

0 20 40 60 80 100) Percent.
00 80 60 40 20 0 |" by weight.

FIG. 11.— FREEZING POINT CURVE OF SILVER-GOLD ALLOYS.

of the two metals, but will be simply a more concen-
trated solution of the one metal in the other, and such an
alloy is best called a non-eutectiferous or solid-solution
alloy.

Such an alloy can have no fixed freezing point, but a
more or less extended freezing range. There will be a
definite point at which freezing begins, and a definite
point at which it ends, and the first of these will always
be more sharply marked than the latter, indeed, the latter
may hardly be noticeable.

The best marked series of alloys of this character is
that containing gold and silver, the freezing points of which
are shown in Fig. 11, and these freezing points, of course
lie between those of pure gold and of pure silver.

PHENOMENA OF SOLIDIFICATION.

Metals which arc More or Less Soluble one in the Other. — The
two cases already considered are of a somewhat special
character, and are of rare occurrence. The case now to
be considered is one which is much more general.

An alloy consists of two metals, A and B, each of which
is to a certain extent soluble in the other in the solid
condition. Let it be assumed that A in the solid condition

B o

A 100

~AO
fcO

Ewteehe Poinf

80
20

»Sf}SKr

FIG. 12. — DIAGRAM OF FREEZING POINTS OF ALLOY OF A AND B WHEN A AND B

ARE EACH TO SOME EXTENT SOLUBLE IN THE OTHER.

can hold in solution 10 per cent, of B, and that B in the
solid condition can hold 10 per cent, of A ; here obviously
will be a combination of the two conditions previously
considered.

If an alloy of A and B contains less than 10 per
cent, of B it will behave exactly like the alloy of the
metals which were soluble in all proportions. The solidi-
fication will begin at a temperature dependent on the
quantity of B present, and will continue with a continu-
ously falling temperature till the whole is solid. There
will be a comparatively short freezing range, and no
eutectic in the ordinary sense will be formed, since the
portion of lowest freezing point will be a solid solu-
tion. If the alloy contains more than 10 per cent, of
B the phenomena will be exactly the same up to the
10 per cent, point, then the solidification will continue,
the mother liquor becoming more arid more concentrated
till at last a point is reached where the mother liquor
will solidify, separating at once into its two constituents

PHENOMENA OF SOLIDIFICATION.

A saturated with B and B saturated with A, a true
eutectic in either case. If the start be made at the
other end of the series the phenomena will be exactly
similar, and the eutectic solidifying point will be reached
at the same temperature and with the same propor-
tions as when the start is made from the A end. The
freezing point curve will be of the form shown in Fig.
12A. At either end alloys rich in one metal and poor in

Temperature, Centigrade.

Ot Ot Ot Oi

662°

"c5

lie* *

302° e.
1
122°

j;V

**#*?•

EutecKc

Tin 0 20 40 60 80

Lead.. 100 80 60 40 20 0 / cent,

FIG. 13. — FREEZING POINT CURVE OF LEAD-TIN ALLOYS.

another will solidify as solid solutions ; in the intermediate
range there will be a definite eutectic formed which
will, of course, always solidify at the same temperature.
There will be a range of solidification in which the alloy?
of A and B will be non-eutectiferous, and an intermediate
range in which they will be eutectiferous.

This is a very common condition. It is the condition
of ice and salt, and it is well shown in the freezing point
curve of alloys of tin and lead shown in Fig. 13.

ALLOYS OF METALS IN WHICH A DEFINITE CHEMICAL
COMPOUND is FORMED.

It has been assumed in the cases already considered that
the metals do not form any chemical compound, but in
many cases they do so combine, and this complicates still
further the conditions of solidification by the introduction
of new conditions. It is quite obvious that the chemical
compound formed may be completely soluble, partially
soluble, or quite insoluble in either or both of the metals
in the solid condition.

PHENOMENA OF SOLIDIFICATION.

It will be sufficient to consider one case. Suppose two
elements A and B to form a compound Ax By, and that
this compound is to some extent soluble in both the
metals.

Now, starting with an alloy containing a small
quantity of B and a large quantity of A, the whole of B
will unite with some of the A to form the compound Ax By,
and if there be not too much of this it will remain in solution
in the solid condition. Freezing will begin at a tempera-
ture dependent on the quantity of Ax By present, and
will continue at a falling temperature until the whole is
solid, the mother liquor gradually concentrating. If the
quantity of B be larger, it will still all unite to form Ax
By, which will dissolve. Freezing will begin as before at a
temperature dependent on the quantity of Ax By in solu-
tion, and will go on, Ax By being ejected as the saturation
point is passed, the mother liquor becoming more con-
centrated till ultimately a eutectic made up of A and Ax
By will separate at the minimum temperature. These
conditions will hold good till the quantity of B is such as

\ /

AxBy

A 1

3 2
)0 8

0 4
0 6

EwKeefic
A+AxBy

0 6
0 A

0 &

° Ewteche2
B + AxBv

0 10

o o

1 .

>LPER
JCENT

FIG. 14. — DIAGRAM OF FREEZING POINTS OF ALLOYS IN THE CASE OF METALS A AND
B, WHICH FORM A DEFINITE COMPOUND Ax By, WHICH IS TO SOME EXTENT
SOLUBLE IN BOTH METALS.

to give the eutectic composition for A and Ax By, when
the mass will solidify as a whole. As the quantity of B
is increased, the conditions will be different, the quantity
of Ax By will be greater than the eutetctic proportion ; the
freezing will therefore begin at a temperature higher than
the eutectic point, but lower than the freezing point of

PHENOMENA OF SOLIDIFICATION. 33

the compound Ax By. When the quantity of B in excess
is so small that it can remain in solid solution in Ax By,
the eutectic will disappear, and when the proportions
are such as to exactly form Ax By this will solidify as a
whole at a definite point. Alloys containing more B
will be first solutions of B in the compound Ax By, till
the point is reached when a eutectic of Ax By and B is
formed ; then Ax By will be in solution in B, and so on,
so that ultimately the curve will take the form shown in
Fig. 14.

There are, of course, other possible conditions, but
the cases explained will serve as illustrations, and will
enable the changes which take place under other con-
ditions to be understood.

When there are three or more constituents present,
the conditions may become more complex.

CONDITIONS OF HOMOGENEITY.

It will be seen that, as a rule, a solidified mass of metal
will not be homogeneous, there being indeed only three
conditions when perfect homogeneity may be expected,
viz., when the substance is a pure metal, when it is a
definite chemical compound, and when it consists of two
constituents either or both of which may be elements or
compounds in the exact eutectic proportion.

Conditions of Solidification. — The heterogeneity of a
solidifying solution must, of course, lead to a correspond-
ing structure of the solidified alloy, but what will be the
practical result will depend on the way in which solidifi-
cation takes place.

For simplicity, it may be best to assume first of all a
cylinder or sphere of the liquid material cooling uniformly
from outside, so that the solid material is formed in
thin layers or shells, one within the other. In that case,
at any moment there will be a solid mass outside, and a
liquid mass within separated by a thin layer just in
process of solidification. Using geographical terms, Prof.
Howe suggests that the portion already solidified should
be called the solid continent, the liquid portion the sea,
and the zone just between the two the littoral or shore
region. In the case under consideration, the solid

34 PHENOMENA OF SOLIDIFICATION.

continent will gradually extend in wards, the sea gradually
shrinking and ultimately disappearing. It is quite
evident that the centre of the mass will be the last to
solidify.

If the alloy solidifies as a whole, and yields a per-
fectly homogeneous solid, the portion which solidifies
last will not differ in any respect from the portion which

Li floral Zone
or Seaboard

FIG. 15. — IDEAL SECTION OF A COOLING INGOT.

solidifies first, and it will remain liquid to the last simply
because being surrounded by a mass of hot metal it can
only cool more slowly.

On the other hand, if the alloy be not one that has a
definite freezing point, the mass in the middle which last
solidifies will be that of lowest melting point, either the solid
solution of lowest melting point or the eutectic as the
case may be. In the former case, there will be a gradual
transition from the composition of the outer layers to
that of the inner, with no line of demarcation between the
two ; whilst in the latter there will be a more or less
sharp line of demarcation between the first solidified
solution outside and the eutectic mixture within.

In either case, the segregation or separation into por-
tions of different composition will be more or less well
marked, and chemical analysis would show a progressive
change in composition inwards. For very many reasons,
solidification never takes place in quite such a simple way.

The line bounding the continent and the sea is never
a plane surface parallel to the cooling surface, for the
solidification takes place by the irre'gular growth inwards
of crystals of the solidifying materials, so that the con-
tinent becomes extended into peninsulas projecting

PHENOMENA OF SOLIDIFICATION.

inwards into the sea, and the sea thereby becomes broken
up into bays, and as the crystal growth does not take
place by any means regularly, the crystals grow not only
directly inwards, but cross from one side of the bays to
the other, so ultimately the mother liquor becomes
broken up into a series of pools, or it may be squeezed into

Liquid

Fro. 16. — IDEAL SECTION OF A SOLIDIFYING INGOT, SHOWING HOW LAKES OF LIQUID
MATERIAL MAY BE ENTANGLED IN THE SOLID MASS.

thin strings. In this way, whilst there is still the separa-
tion of the various constituents, there is not the same
distinct segregation that would be under the conditions
described above.

These can only be regarded as examples, for there are
many other ways in which solidification may take place.
It may, for instance, begin at many places in the mass at
once, the solid matter growing from these solidifying

Ourside
Edge

Confmenl-

Interior

FIG. 17. — DIAGRAM OF MODE OF SOLIDIFICATION.

centres until they meet, the last solidifying mother
liquor being squeezed into the spaces between the solidify-
ing masses, and forming a network which may, of
course, be reduced to mere strings.
Examples of these will be seen later.

PHENOMENA OF SOLIDIFICATION.

Segregation. — From what has been said it is evident
that as an alloy solidifies, there may be more or less of
segregation or actual separation of the parts. If the mode

FIG. 18. — CRYSTALLISED IRON.

of solidification at all approaches to H the first condition
described above this may be very marked. It may
happen that the solidifying mass may be more or less

PHENOMENA OF SOLIDIFICATION. 37

free in the mother liquor, that is, unattached to the
already solidified continent, and in that case if there beany
great difference in specific gravity the heavier mass will
tend to sink and the lighter to rise, so that there may be a
distinct segregation in the casting not inwards but
upwards. The phenomena of segregation are well
known to the makers and users of certain alloys, and
cause no little trouble.

Prevention of Segregation. — From what has been
said, it is clear that the lack of homogeneity of the alloy
cannot be prevented, because it depends on the natural
laws of cooling, but the injurious segregation may be
minimised if not prevented. As a rule, the more slowly
the mass is frozen the more perfect will be the separation,
whilst the more rapid the freezing the better will be the
mixture of the constituents. For example, zinc will only
retain in solid solution about 1-5 per cent, of lead, and
yet if a mass of zinc be cast in such a way that it cools
very rapidly, a much larger quantity will apparently be
retained, but this excess consists of lead, containing a
small amount of zinc in solution, scattered through the
mass in very minute shots or masses. On remelting
and cooling very slowly, a considerable portion of the
heavy lead alloy may separate out.

Liquation. — As most solidified alloys consist of two
portions, of different melting points, it is sometimes
possible to melt or liquate out the more liquid portion.
This is only possible under certain conditions, for reasons
to be described later.

A very good example of this is in the case of copper and
lead. These metals do not remain alloyed, but on
solidification copper containing a little lead separates
and rejects lead containing a little copper ; but if the
solidification be sufficiently rapid the two may remain
intermixed. If the mixture be heated to just above the
melting point of lead, the lead will melt and run out.
Advantage was taken of this in the old Freiberg
method of separating copper and silver, the silver dis-
solving in, and liquating out with, the lead.

In many cases, there is a tendency for the portion of an
alloy of lowest melting point to " sweat " out on heating.

38 PHENOMENA OF SOLIDIFICATION.

This is well seen in the case of highly-sulphurous pig iron,
where globules of an iron and iron sulphide eutectic are
often to be found on the surface of the pig, forced out by
the pressure of the cooling and contracting metal on the
still liquid material.

SOLUTION OF GASES IN METAL.

Liquids always dissolve gases, but the conditions are
somewhat different from those which hold in the case of
solutions of solids or of other liquids. The amount of gas
dissolved depends, of course, on the character of the gas
and its solubility relation to the liquid, and it also depends
on the temperature. There is always a temperature,
or perhaps a range of temperatures, at which there is a
maximum solubility, the solubility decreasing both at
higher and lower temperatures, the gas being always —
except in cases of chemical combination — expelled at the
boiling point of the liquid, and also to a large extent,
though frequently not completely, at the solidifying point.
The solution of a gas usually causes expansion, so that
the resulting solution has a lower specific gravity than
that of the solvent.

The coefficient of solubility of gases in molten metals
and alloys has not been determined, all experiments that
have been made being on the solution of gases in water,
saline solutions, and a few other liquids, so that the
actual solubility of the gases in metals is not known.

The gases which are likely to be dissolved in molten
metal are few in number. The metal may come in con-
tact with oxygen or nitrogen from the air, hydrogen from
the decomposition of water, carbon dioxide and carbon
monoxide from the combustion of the fuel, and in certain
cases also sulphur dioxide from the fuel. Oxygen is
rarely likely to be present as such, but will prob-
ably be in the condition of dissolved metallic oxides,
since most of the metals are easily oxidised at tempera-
tures above their melting points. The gases in solution
would be completely expelled if the metal were heated to
near its boiling point, but this is not a practicable
condition. If one constituent of the alloy be volatile at
the temperature it may carry off with its vapour a
considerable quantity of dissolved gas. Thus, in brass

PHENOMENA OF SOLIDIFICATION. 39

making the volatilised zinc probably carries away with it
much of the dissolved gas.

On freezing, most of the gas will probably be given off,
and under certain conditions may lead to the formation of
blowholes. The gas is separated much in the same way
that solids are separated, but being much lighter than
the liquid always tends to rise and thus escape. But
bubbles of gas disengaged in a liquid are often very easily
retained. They tend to adhere to a smooth surface, and
thus may become enclosed in solidifying metal, and also
as the solidification of the metal does not take the form of
smooth surfaces, but of irregular growths projecting
outwards into the liquid mass, gas bubbles may easily
become entangled and surrounded by metal. Just as a
fragment of solid matter in a solidifying solution tends to
determine the solidification of the metal round about it,
so a bubble of gas once formed tends to increase in size
by the accumulation of more gas, the bubbles thus
elongating and growing inwards.

The phenomena and results of gas evolution from
cast metals may vary much —

(1) If the mass of metal or alloy be very fluid, i.e.,
not pasty, and the solidification be very slow, a large
portion of tho gas may escape and thus do no harm to
the casting. To ensure this, the upper parts of the casting
must be kept liquid to the last, or if this be not possible,
a head of liquid metal must be provided into which the
escaping gas can rise, and the unsound portion thus
produced can be cut off.

(2) If the mass be more or less pasty or the
solidification be very rapid, or if the gas be not
separated till solidification has gone on to a consider-
able extent, the gas may be retained in the form of
bubbles, and thus produce blowholes. These will usually
be more or less lenticular in form, with their long axes
at right angles to the surface of solidification.

The formation of blowholes in steel has been more fully
studied than in any other metal, but probably the same
laws would hold good in all cases. The term blowholes
should only be applied to these small bubbles, not to
the larger masses of gas which accumulate in the upper
part of the casting by the collection of gas which has

PHENOMENA OF SOLIDIFICATION.

actually been given off, but is unable to escape, and
which may produce larger or smaller gas cavities near
the top of the casting.

(3) Under some conditions gas may actually be
retained in solid solution, in which case of course it
cannot cause blowholes.

FIG. 19. — BLOCK OF ICE, SHOWING ARRANGEMENT OF AIR BUBBLES. THE OPACITY
OF THE CONICAL PORTIONS is DUE TO A MASS OF AIR BUBBLES. THE IRREGU-
LAR WHITE PATCHES ARE WHERE THE SURFACE HAS BEEN DAMAGED BY THE
DOGS USED FOR LIFTING.

Prevention of Blowholes. — The means of preventing
unsoundness due to blowholes may be of three kinds : —

(1) The addition of substances which will destroy
the gas in solution.

(2) Treatment of the metal so as to facilitate the
escape of the gases.

(3) Treatment of the metal so as to retain the gas in
solution.

PHENOMENA OF SOLIDIFICATION* 41

The first method is rarely practicable, since if the
gases are nitrogen, hydrogen, and the oxides of carbon
they are not likely to be removed chemically by the
addition of any reagent. The addition of easily
oxidisable substances such as phosphorus or silicon
may destroy the oxides of carbon, if present, but
their action is probably mainly on solid dissolved oxides
which may impair the strength of the metal, rather than
on gases.

The second method is in general use. Agitation
during solidification often has a good effect, but is
in general impracticable. Slow cooling and keeping
the upper portion of the metal liquid to the last, so as to
allow free escape of gas, is usually all that can be done.

The third method may be carried out either by
chemical or mechanical means. The addition of certain
substances, usually metals, seems often to suddenly stop
the evolution of gas, as it is unlikely, at least in most cases,
that chemical changes have taken place by which the gas
has been converted into a solid or liquid, the "quieting"
can only be produced by a change which enables the gas
to be retained in solution up to and after solidification.
This is probably the action of silicon and aluminium on
steel castings. As the solubility of gases in various alloys
is not known, it is impossible to say which metals would
increase the solubility in any particular case.

Casting Under Pressure. — Sound castings may very
often be secured by casting under pressure, the pressure
being obtained either by the use of a hydraulic press, gas
pressure, or liquid pressure obtained by a head of metal.

The way in which pressure acts is somewhat uncertain.
The action may be of three kinds, and probably all
three may take place together.

Since pressure facilitates the solution of gases, the gas
may actually be retained in solution, in which case
casting under pressure would fall under the third division
of methods above mentioned, and probably this action
always takes place to some extent. On the other hand
the pressure may actually squeeze the liberated gas out of
the casting. This probably always happens to some
extent, and in many cases the escape of gas can be dis-

42 PHENOMENA OF SOLIDIFICATION.

tinctly noticed. In other cases, the pressure may simply
compress the gas, and thus make the cavities which it
forms so small as to be of little moment. As the volume
of gas is inversely proportional to the pressure to which
it is subjected, it will be reduced to one-half the volume
which it would occupy under atmospheric pressure
by a pressure of 15lbs. per square inch, and a
pressure of 10 tons per square inch, which is often
exceeded when steel is cast under hydraulic pressure,
would diminish the volume of any separated bubbles to
about TT^OO of their normal volume, and under these
conditions they might be so small as to be of little impor-
tance.

CHAPTER IV.
WHAT THE MICROSCOPE CAN TEACH.

I. — METHODS.

THE structure of metals as shown by a fracture has always
been a factor in the judgment of the quality of the
metal, but until recently only the appearance to the naked
eye, or at anyrate as seen by a hand magnifier, could be
taken into account.

FIG. 20. — BAIRD & TATLOCK'S POLISHING MACHINE.

The fracture is of some importance, but for various
reasons it is always an uncertain guide, the appearance
depending as muclTon the way in which the fracture is
obtained as on the structure of the metal itself, so that

44 WHAT THE MICROSCOPE CAN TEACH.

it is always of limited value, except in certain special
cases.

In 1864 Dr. Sorby suggested, and actually used the
microscope for the examination of the structure of metals.
He was, however, much before his time, and but little
attention was paid to his work, and the wonderful results
he obtained were almost completely overlooked. Slowly,
however, the value of his work came to be understood,
and other workers entered upon the field, and now the
use of the microscope has become quite general for the
examination of metals, and soon a metallographic
laboratory will be regarded as being as essential as a
chemical laboratory in a well-equipped works. There
can be no doubt as to the real value of the results of
microscopic investigation, but as is always the case with
new methods, some workers over-rate its powrer and
value, and expect far too much from it. Like all
other methods of research, it has its own field, and there
only is it of value.

For the microscopic examination of metals three
things must be taken into account : The microscope
which is to be used ; the preparation of the sample for
examination ; and the methods of examination to be
adopted. These will be briefly considered in the reverse
order to that in which they have been named.

Principle of the Methods Used — In all ordinary microscope
work the object to be examined is transparent and is
viewed by light transmitted through it, and reflected
up by a mirror placed below the stage. Even in the
examination of rocks and minerals the samples are ground
so thin as to be transparent, and are examined in this
way. With metals this is impossible, as they cannot be
ground into such thin films as to be transparent, for
however thin the metal, it is always quite opaque, and
therefore the sample can only be examined by light
reflected from the surface.

If the fractured surface of a piece of metal be examined
with a microscope, nothing of the real structure can be
made out, as appearance of the fracture depends on so
many conditions that it throws but little light on the real
structure, and, further, the surface is sure to be so

WHAT THE MICROSCOPE CAN TEACH. 45

irregular that it is quite impossible to get more than a few
points of it into focus, and the surface will often be covered
with accidental markings which are of no importance.

To allow of focussing, the sample must have a perfectly
plane surface, and this is obtained by polishing. A
surface may seem perfectly smooth and bright to the eye,
but when examined under the microscope even with a low
power it is seen to be covered with scratches which look
like deep grooves and effectually hide the real structure.

Preparation of the Sample — The first^thing to be done
in the preparation of a sample is to obtain a surface
perfectly smooth and free from scratches. This is done
by means of a series of polishings, a finer polishing material
being used at each stage than for the one before, so that
at each polishing all the scratches are removed, and if the
surface is not left perfectly smooth, the scratches left are
much finer. This step-by-step polishing is essential because
the abrading power of the polishing materials used for
the last stages is so small that it would take a very long
time and a large amount of labour, even if it were possible
at all, to remove the amount of substance necessary to
reach the bottom of the deep coarse scratches.

The polishing may be done by hand by carefully
rubbing the sample on the polishing material mounted
on suitable blocks, and excellent results can be obtained
by this method, though it is laborious and somewhat
tedious. Where very careful work is required it is still
probably the best method which can be used. One of
the conditions laid down by the Director of the National
Physical Laboratory for an investigation into the struc-
ture of steel which is being undertaken by many indepen-
dent observers is that the polishing shall be done entirely
by hand.

Most workers, however, prefer the use of a machine of
some kind, and there are several on the market. The
machine consists of a disc of wood or metal suitably
mounted so that it can be rotated at a very high speed
by hand or foot, or by a small motor. On this disc the
polishing material is mounted and the sample, held
either in the hand or in a small holder, is kept in contact
with it till the polishing is complete.

WHAT THE MICROSCOPE CAN TEACH.

FIG. 21.— STEAD'S POLISHING MACHINE.
(Made by Messrs. Carling & Co., Middlesbrough.

WHAT THE MICROSCOPE CAN TEACH. 47

Emery cloth and similar materials may either be
glued to the disc or held in place by a metal ring, whilst
the powdered polishing, materials are spread on a piece of
cloth, velveteen, or selvyt, stretched tightly on the disc, or
for very fine polishing, a double layer of cloth may be
used, the powder being put between the two layers,
when enough works through to polish the specimen. As
the sample is apt to get very hot during polishing it is
kept wet, best by allowing water to drip on. to the disc
during the operation.

There is no difficulty in polishing hard substances,
but soft metals are exceedingly troublesome, as the
surface tends to flow rather than to be polished away.
The original structure of the metal may thus be com-
pletely destroyed, and scratches and other marks may
not be removed, but only covered by the surrounding
metal being forced over them. Very brittle substances
are apt to break away in minute fragments which, getting
on the polishing disc, may produce deep scratches.

No two workers use exactly the same set of polishing
materials or go through exactly the same routine. All
that is necessary is that each polishing material should
remove the scratches from the one before and that
the last one should leave no scratches.

The following order is a convenient one for alloys :
Very fine file, Nos. 0, 00, and 000 emery paper, the finest
rouge, and lastly diamantine, a polishing agent con-
sisting of pure alumina, which is specially prepared for
this purpose.

When the polishing is complete, the surface will
appear smooth and bright, but will show no structure,
except in cases where some of the constituents are very
soft, when these may be rubbed away and a structure
shown. This is well seen in the case of pig iron con-
taining large flakes of graphite. Use is made of this
occasionally under the name of the " polish attack."

Etching. — In order to " develop " or rather to reveal
the structure, the surface must be etched, that is, it is
treated with some reagent which will attack the surface.
If the metal be perfectly homogeneous no structure
will be revealed, but if it be heterogeneous the constituents

WHAT THE MICROSCOPE CAN TEACH.

will be attacked at different rates by the etching agent,
some portions being dissolved away much more quickly
than the rest, so that the portions least attacked are left
standing in slight relief, the relief of course being so
slight that both portions can be in focus under the
microscope at the same time. Sometimes the different
constituents are differently coloured by a reagent, and
thus can be distinguished the one from the other.

FIG. 23. — METHOD OF OBTAINING VEETICAL ILLUMINATION.

S, specimen for examination ; R, reflector inclined at 45° ; L, condensing lens ;
IM, incandescent mantle.

The etching agent must necessarily be selected so as
to suit the peculiarities of the metal under examination.
A reagent which will develop a structure in one metal
may be quite useless for another, thus a reagent that
is suitable for iron and steel may be quite unsuitable for
brass or other alloys.

In some cases structure may be brought up by gentle
heating, the constituents being differently oxidised and
therefore rendered visible. This is called heat tinting.

The Microscope. — The microscope is, of course, the
most important item in a metallographer's outfit. Good
work may be done with an ordinary microscope, if it be
of fair quality, but one specially made for and therefore
adapted to metallographic work is much better. The
makers of microscopes are now competing with each

WHAT THE MICROSCOPE CAN TEACH. 49

other in the production of instruments suitable for
metallographic work. Those of Messrs. Beck, Watson,
Ross, and Swift, in this country, and of Messrs. Reichart,
Zeiss, and other firms on the Continent leave little to be
desired, and are made at various prices.

In all ordinary microscopes the object to be examined
is transparent, and the light is reflected up from below ;
for metallographic work this cannot be done, and the stage
is best made solid. The samples to be examined may be
of considerable size, so that a much larger motion of the
stage or tube is necessary than in ordinary biological
microscopes. Messrs. Swift make a stand so constructed
that the stage can be dispensed with if necessary, the
microscope standing on the article to be examined, so
that large articles can be examined without the removal
of specimens.

For a reason which will be seen directly, the coarse
adjustments should be attached to the stage, and not to
the tube, and the stage should be provided with trans-
verse motions in two directions, and should be capable
of being accurately levelled. The light must obviously
be thrown upon the surface to be examined, and reflected
upwards into the objective. With low powers, fin. and
upwards, there is no difficulty — the light can be thrown
upon the sample by means of a bull's-eye condenser.
This is usually called natural or oblique illumination.
With high powers this method of illumination is im-
practicable, as the sample is so near the objective that
the light cannot reach it, and some other method of
illumination must be devised, and vertical illumination
by means of a reflector within the tube of the micro-
scope is therefore used. This method of illumination
can be used for low powers as well as for high, but it is
often undesirable for the former.

The simplest form of vertical illumination is that of
Messrs. Beck. At the lower end of the microscope tube,
just above the objective or at some other convenient place,
is fixed a short tube which contains a small mirror of
very thin unsilvered glass so arranged that it can be
rotated into any required position, and opposite this a
circular hole is made in the tube. A horizontal beam
of light is sent into the tube through the opening, and if

50 WHAT THE MICROSCOPE CAN TEACH.

Fid 22.— ROSENHAIN MICROSCOPE. MA1>E BY IVjESSRS. BECK & Co.

WHAT THE MICROSCOPE CAN TEACH. 51

the mirror bo placed at an angle of 45° the light will be
partly reflected downwards and partly transmitted, the
latter part being lost. The light which passes downwards
passes through the objective, illuminates the object, and
is reflected back again ; the upward beam striking the
mirror is partly transmitted and partly reflected, the
portion transmitted passes upwards and reaches the
eyepiece.

It is quite obvious that by this arrangement there is
a considerable loss of light by the various reflections and
transmissions, but enough reaches the eyepiece for the
purpose.

In place of this simple mirror, a prism may be used,
and Messrs. Beck have recently introduced a new form of
mirror, the one half of which is silvered to act as a reflector,
whilst the other half is clear to transmit the image.

Appearance of the Object — The appearance of the
object varies very much with the character of the illumina-
tion, a surface which appears bright with an oblique
illumination often appearing dull by vertical illumination,
and vice versa, so that when a specimen is described as
being bright or dull, the character of the illumination
should always be specified. Suppose a perfectly smooth
surface to be examined by oblique illumination, it

FIG. 24.— BRIGHT SURFACE, OBLIQUE ILLUMINATION.

will appear dull, or almost black, whilst on the
other hand by vertical illumination it will appear
brilliantly bright. The explanation of this is quite
simple. A beam of light falling obliquely on a bright
surface is reflected according to the law of reflection,
and none of the light enters the object glass which is
vertically above it, and thus the surface appears dull.

WHAT THE MICROSCOPE CAN TEACH.

On the other hand, when the light is sent down vertically
on to the bright surface, nearly the whole of it is reflected
back and thus the surface appears bright.
[ ~] If the surface be dull the oblique ray will not be
regularly reflected, but will be scattered, so that a fair

XL/

FIG. 25. — BRIGHT SURFACE,
VERTICAL ILLUMINATION.

FIG. 26. — DULL SURFACE,
OBLIQUE ILLUMINATION.

portion will enter the object glass and the object will
appear bright ; on the other hand, if the light is sent
down vertically upon the surface, but a small propor-
tion will be directly returned, so that the surface will
appear dull.

FIG. 27.— DULL SURFACE,
VERTICAL ILLUMINATION.

As already remarked, on the etched sample portions
stand up in relief, the surrounding portions having been
dissolved away, and with oblique illumination a distinct
shadow will be cast, which will make the distinction

WHAT THE MICROSCOPE CAN TEACH.

between the two constituents much more pronounced
than by vertical illumination, where there can \ be no
shadow. Similarly, a small hole or depression will be
much more strongly marked by oblique than by vertical
illumination, since in the former case it will be in shadow.

FIG. 28.— SHADOW CAST IN
OBLIQUE ILLUMINATION.

It very frequently happens therefore that a change
from oblique to vertical illumination quite alters the
appearance of the specimen — the surfaces which before
were bright become dull, and those which were dull become
bright, and holes or scratches become much less strongly
marked.

In order to preserve the results of an examination, a
photograph should always be taken if possible. The
magnification should be given thus, x 30, meaning that
the photograph is 30 times linear larger than the sample,
and the illumination should always be marked, o for
oblique illumination, and v vertical illumination.

Great judgment and experience is required for the
correct interpretation of the meaning of the structure
seen.

RESULTS.

THE microscope can give very valuable information a&
to the structure of alloys, but like every other instrument
it has its limitations, and to expect of it more than it can
do is to court disappointment. Its field of usefulness i&
strictly limited to the detection of differences in physical
structure, and more than this it cannot show. It can in

54 WHAT THE MICROSCOPE CAN TEACH.

no way replace chemical analysis, though it is a very
valuable adjunct to it ; for it can give no hint of the
presence of combined or dissolved impurities unless these
give rise to differences of structure or colour. Its usefulness
depends on the fact that the differences in structure on
which the properties of an alloy depend are on such a
small scale that the unaided eye is not able to detect
them.

The methods of microscopic examination have been
described, but it may be added that the difficulties in the
application of microscopic research lie not so much in the
observations themselves as in the interpretation of what
is seen.

It has been already mentioned that there are three
cases in which a metallic substance, or, indeed, any other
substance, will solidify at a fixed temperature : (1) when it
is an element, (2) when it is a definite chemical compound,
(3) when it is an alloy of eutectic composition.

An alloy which solidifies at a definite temperature
has a sort of identity or individuality which cannot be
claimed for one which solidifies in several parts at different
temperatures. It may be well, therefore, to see what
information the microscope can give as to the structure
of such substances.

(1) Pure Metals. — When a pure metal solidifies from
fusion it is obvious that the solid substance must be
chemically homogeneous ; it must, that is, have the same
composition in all parts, and any lack of homogeneity
can only be due to the formation of holes, by contraction
during solidification, or perhaps from the giving off of
dissolved gas, if the term pure metal be not held to
exclude metals holding gas in solution. As the metal
solidifies, it will always crystallise, and will therefore
yield a crystalline mass, and the form and size of the
crystals will depend on the metal solidifying and on
the rate of solidification.

The crystallisation may take place in various ways.
Very frequently when cooling takes place at the outer
surfaces it is by the growing inwards of crystals into the
still liquid interior, the liquid mass subsequently
solidifying between them, and thus forming a solid

WHAT THE MICROSCOPE CAN TEACH.

mass. In the case of metals which yield large crystals a
feathery crystalline structure is often seen on the surface.
The crystals being in slight relief from the contraction of

FIG. 29.— Sr K PACK OF AN INGOT OF ANTIMONY (NATURAL SIZE)

FIG. 30.— CRYSTALLINE SURFACE OF TIN (NATURAL SIZE).

56 WHAT THE MICROSCOPE CAN TEACH.

the last solidified portions, can often be seen with the
naked eye, and in other cases the structure can be brought
up by etching. Antimony and tin are very good examples
of this. When a section of such a metal is made, the
structure is often difficult to make out, as there is little
to differentiate the first-formed crystal from the subse-
quently solidified material. With soft metals, it is very
difficult to secure a surface smooth enough for etching,
except by casting on a surface of some very smooth
substance, such as mica.

FIG. 31.— A PURE METAL, NEARLY PURE IRON. FERRITE V 50x.

In the normal case of the solidification of a metal
crystallisation begins at a large number of centres, the
crystals growing outward in all directions into the mother
liquor, the space between them therefore gradually
becomes less and less, and at last they press one upon
another, so that the sharpness of the edges and the
regular crystalline form is completely lost, an irregular
polygonal structure only remaining. When such a metal is
polished and etched, these polygonal grains, which are
distorted crystals, can often be made out. The boundary
line is an optical phenomena, and does not indicate a real

WHAT THE MICROSCOPE CAN TEACH. 57

line of separation between the crystals. Such structures
are spoken of as " allotriomorphic " crystals, because
they do not show the true crystal form ; and sometimes
they are called crystal grains. It may happen in some

FIG. 32. -A CRYSTALLINE METAL. IRON WITH 4 PER CENT. SILICON. (STEAD.)

cases that the crystals retain their natural crystal forms,
and it sometimes happens that distinct lines indicating
cleavage planes can be distinguished.

The appearance of the fracture to the naked eye
usually depends on how the planes of fracture are related
to these crystals. If the fracture takes place along lines
between the crystals a granular structure will usually be
produced, whilst if owing to the. presence of cleavage
planes the fracture takes place across the crystals
brilliant cleavage faces are often visible. When the
metal is very malleable, so that it draws out before
fracture, a fibrous fracture may be obtained.

It is obvious that the information which the microscope
can give as to the structure of pure metals is not of much
value.

Definite Chemical Compounds. — None of the definite com-
pounds of one metal with another are of much prac-
tical importance. A chemical compound is just as much a
unit as an element, and on solidification it behaves

58 WHAT THE MICROSCOPE CAN TEACH.

exactly in the same way, showing usually a definite
crystalline structure, exactly similar to that shown by a
pure metal ; indeed, the microscope would give no indi-
cation whether a substance under examination was a
pure metal or a chemical compound.

Solid Solution. — A solution of one metal in another of
such composition that the metals remain in solution
in the solid condition may not show under the microscope
any sign of variation in composition, the solid will be
distinctly a unity, and no structure except that due
to crystallisation will be detectable. It does not
follow, of course, that the presence of the foreign metal
may not alter the structure of the whole, but simply
that the separate constituents will not be distinguishable.

Eutcctics. — When the two constituents are present in
the eutectic proportion, the mass solidifies at a definite
temperature, but the conditions are very different from
those already considered, for whilst, in the liquid condition,
the mass is a solution at the moment of solidification
the two constituents separate completely, so as to remain
only as a mechanical mixture ; and as the two constituents
of any mixture will certainly be differently acted on by
somd etching reagent, a structure can easily be brought up.

As the two portions of the eutectic solidify at the same
time, it might be expected that the separated portions
would be so small as to be indistinguishable, even if not
of molecular dimensions. This, however, is not the case ;
molecular attraction comes into play and the separated
molecules aggregate into masses of sensible size, — at
least, in most cases, though there are cases in which
the constituents remain mixed in such minute portions
as not to be distinguishable, forming what has been called
a eutectic emulsion. The actual amount of differentia-
tion in the constituents will depend, among other things,
on the rate of cooling.

The arrangement of the separated portions will also
vary very much. What may be called the normal
eutectic structure consists of a series of more or less
parallel plates of minute size, the parallelism, however,
only extending over small areas, the whole surface being
frequently broken up into series of eutectic areas in which

WHAT THE MICROSCOPE CAN TEACH.

the orientation of the plates varies considerably, it
often appearing as if incipient solidification has taken

FIG. 33.— A TYPICAL EUTKCTIC STRUCTURE, PEARLITK. (!RON AND Fe3C.)

place to a sufficient extent to form a series of crystals,
and then that independently within each of those areas
the eutectic has been formed.

FIG. 34. -THE EUTECTIC OF SILVER AXD LEAD. V-90, BY SAVILLE SHAW.

60 WHAT THE MICROSCOPE CAN TEACH.

Mr. Stead has classified eutectics into three groups* :—
(1) The curviplanar, in which the constituents con-
sist of curved plates in juxtaposition . Examples
of this structure are to be found in the alloys
of silver and copper, and in slowly-cooled carbon
steels.

P^ 1

fefe^S

K ls s^****?* * +**/ **

i/~ ?y*£ «&**L
t*-t sT ?rv 5KL* V* -;

FIG. 3.*.— AN ALLOY OP Two CONSTITUENTS. WHITE SWEDISH IRON x 20.
'\ he dark parts are pearlite, the white parts the solidified eutectic.

(2) The honeycombed or cellular, a very common

variety. Gold and lead, bismuth and tin, and
many other alloys yield eutectics of this type.

(3) The rectiplanar, in which the two constituents

separate in flat plates. Silver and lead eutectic
is an example.
To these may be added :—

(4) A spherulitic structure, usually produced when

alloys are very rapidly solidified, the growth
taking place from centres and forming a mass
resembling the spherulitic structure in certain
minerals.

(5) An emulsion structure, in which the constituent

particles are so small that they can only just be
detected, or perhaps may not be detectable at
all, and are apparently not arranged in any
definite form.

* Proceedings Cleveland Institution of Engineers, 1900-1, p. 36.

WHAT THE MICROSCOPE CAN TEACH.

No doubt other forms of structure also exist, but
these include all those commonly met with.

FIG. 36.— CRYSTALS OF GRAPHITE SEPARATED IN A GROUND MASS OF FERRITE.

Micro-structure of Alloys. — Most alloys are more complex
in structure, and two or more of the structures
described may be present.

Flu. 37.— INGOT CONTAINING 10-17 PKR CENT. PHOSPHORUS AND 88'9 PER CENT.
IRON, SHOWING SECTIONS OF RHOMBIC OB OBLIQUE IDIOMOKPHIC CRYSTALS
OF FE3 P EMBEDDED IN A GROUND MASS OF EUTECTIC. V x 60 (STEAD).

WHAT THE MICROSCOPE CAN TEACH.

In an ordinary alloy, as cooling goes on, one constituent
— usually a solid solution — separates, and then the mother
liquor solidifies, forming the eutectic, or there may be
more than one stage of solidification before the eutectic
point is reached. The visible structure of the alloy will

A mm

FIG. 38.— INGOT CONTAINING 10'2 PER CENT. PHOSPHORUS AND 89-8 PER
CENT. IRON (STEAD).

It is the eutectic of phosphorus and iron. It has only one critical point, at
about 980° C. Etched with nitric acid. V x 350.

vary very much, according to the relative quantities of
the constituents.

If the quantity of the eutectic be small, the crystals
first formed may go on growing and, of course, ejecting
the mother liquor, till when the eutectic point is reached
it may be reduced to mere threads separating the crystals,
or to isolated patches distributed through the mass.
On the other hand, if the composition approaches the
eutectic point the substance first solidified may form only
a network, in" the meshes of which the eutectic will
solidify, or it may be more or less definite crystals which
will be embedded in the subsequently solidified eutectic.
If the conditions be favourable the crystals may assume
their true form.

WHAT THE MICROSCOPE CAN TEACH. 63

The structure of alloys may, however, vary so much
according to the way in which the constituents crystallise,
the proportions in which they are present, and the con-
ditions of solidification, that at this stage only one or
two typical examples can be given. Others will be con-
sidered in connection with the various groups of alloys.

CHAPTER V.

CHANGES IN THE STRUCTURE or ALLOYS IN THE SOLID
CONDITION.

WHEN an alloy has solidified it by no means follows that
it has reached a perfectly definite and stable condition,
for changes in structure and in proximate composition
may still take place. A solid is not by any means the
fixed rigid thing that is sometimes imagined, for the
molecules retain some freedom and therefore can to a
certain extent, though often very slowly, undergo re-
arrangement. The higher the temperature the greater
is the molecular mobility and therefore as a rule the
greater the ease with which changes can take place, but it
must by no means be assumed that such changes do not
take place to an important extent at atmospheric tem-
perature, and the structure may be greatly modified by
various causes.

Internal Changes During Cooling. — When a mass of
an alloy has solidified it will be in a distinctly crystalline
condition, the crystals in the inner part of the casting
being probably much larger than those near the outside,
owing to the slower solidification ; but even after solidifi-
cation has taken place changes may continue, and the rate
of cooling after solidification may considerably modify
the. structure.

The molecules are still in a condition of comparative
freedom, and therefore can redistribute themselves,
and just as in a solidifying solution the crystals tend to
grow round a nucleus, so in the solid the crystals tend to
grow round another crystal, the larger crystals growing
and absorbing and thus obliterating the smaller ones.
Very slow cooling, as distinguished from slow solidification,
therefore tends to produce a largely crystalline structure,
with the comparative weakness usually following from it.
As a rule, therefore, as far as structure is concerned, the

CHANGES IN THE STRUCTURE OF ALLOYS.

more rapidly an alloy is cooled after it has solidified the
better.

In a casting which is irregular in section so that the
different portions cool at different rates, the crystalline
structure and therefore the strength may vary very much
in different parts. In many cases where a tensile or
other test is specified, but where the casting itself can-
not be tested nor a portion cut from it to test, a fin of
some kind is cast on it which can be cut off and shaped
into a test piece for testing. As this fin will usually
cool much more quickly than the bulk of the casting,

FIG. 39.— INGOT BRASS ROLLED DEAD HARD. (MAGNIFICATION 58 DIAMETERS.)

it will often show a much finer grain and be considerably
stronger than a test piece cut from the casting itself.
This is frequently seen in the case of propeller blades
and similar castings. In all such cases the conditions
under which the test piece is to be cast should be carefully
specified, and the strength required must only be that
of the test piece, as a guarantee of the quality of the
metal, and not as a criterion of the strength of the casting
itself.

No doubt in the case of an alloy a temperature is
soon reached at which the molecular mobility becomes

66 CHANGES IN THE STRUCTURE OF ALLOYS.

too small to 'produce any serious and rapid change, and
sucix a point, which in this connection might be called a
critical point, has considerable practical importance, though
in most cases it is not known with any great degree of

FIG. 40.— BRASS AFTER HEATING TO 500° C. (MAGNIFICATION 58 DIAMETERS.)

FIG. 41.— BRASS AFTER HEATING TO 600° C. (MAGNIFICATION 58 DIAMETERS.)

CHANGES IN THE STRUCTURE OF ALLOYS.

accuracy. It must not be assumed, however, that this
is the actual limiting point of crystal growth, because the

FIG. 42.— BRASS AFTEK HEATING TO 750° C. (MAGNIFICATION 58 DIAMETERS.)

FIG. 43.— BBASS AFTER HEATING TO 800° C. (MAGNIFICATION 58 DIAMETERS.)

68 CHANGES IN THE STRUCTURE OF ALLOYS.

growth may go on, though probably very slowly, at con-
siderably lower temperatures, especially under the in-
fluence of vibration or other mechanical stimulus.

It is obvious that this crystal growth at high tempera-
tures may take place just as readily in the case of a metal
heated from a low temperature to above the critical
point as with one cooled from fusion, so that continuous
heating at a high temperature is very apt to induce
coarse crystallisation and subsequent brittleness in an
alloy. This is well shown in the series of illustrations
Pigs. 39 to 45, which show the effect of heating to a high
temperature on the structure of brass. The illustrations

FIG. 44.— BRASS AFTER HEATING TO 900° C. (MAGNIFICATION 58 DIAMETERS.

are reproduced from " Technics," by the kind permission of
Messrs. Newnes & Co., Ltd. When brittleness is pro-
duced by overheating, the metal is usually said to be burnt.
This is, however, not correct ; it should be called over-
heating, and the term burning should be restricted to those
cases in which there is decided oxidation or other
chemical change.

Annealing — By annealing is understood the heating
of a metal or alloy to a high temperature, so as to allow
of a molecular rearrangement or re -crystallisation, and

CHANGES IN THE STRUCTURE OF ALLOYS. 69

thus the removal of stress which may have been induced
by work. The change produced is almost entirely one of
crystal growth. The crystal structure of the metal has
been broken down more or less completely by the work
which has been put upon it, and a hardness thereby
produced. When the metal is heated above the critical
point the molecular forces are able to come into play
and by restoring a normal crystal structure to restore
the properties of the metal. It is quite obvious
that the change must not be allowed to go too
far, or the crystal structure may become too coarse
and thus again injure the properties of the metal.

FIG. 45.— BRASS AFTER, HEATING TO 1,000° C. (MAGNIFICATION 58 DIAMETERS.)

Annealing is usually looked upon as a very simple
operation, and so, in fact, it is ; but there is no
operation in the whole range of metallurgy which requires
greater care, so as to conform strictly to the conditions
of success, and there is probably no operation in which
failures are more frequent.

In all cases there are the two conditions to be con-
sidered— the temperature and the time of heating ;
whilst, as is pointed out below, the nature of the
atmosphere in which the heating takes place may have a
profound effect.

70 CHANGES IN THE STRUCTURE OF ALLOYS.

If the temperature be too low, the molecules will not
have sufficient freedom to allow of re-arrangement,
whilst if it be too high the change may be too rapid and
the crystals may become unduly large and be so separated
as to greatly impair the strength of the metal. On the
other hand, even if the temperature be correct, if the
metal be exposed to it for too long a time, the crystal
growth may go on beyond the required point, and
brittleness may be produced.

It is very important that the annealing range of tem-
perature for various alloys should be carefully determined,
and this can only be done by those who have control of
works in which experiments can be made on a large
scale and extending over a considerable time.

It is quite obvious that the perfect annealing of a
large casting must be a matter of extreme difficulty
since the heat can only slowly reach the interior, and
therefore perfectly uniform heating becomes impossible.

In some cases the changes produced by heat
treatment may be much more complex. This has
been clearly made out in the case of steel, and
there is reason to believe .that similar phenomena take
place in certain alloys. It is well known that in the
case of steel, finishing rolling at a high temperature or
heating to a high temperature may produce a very
coarse structure with corresponding loss of strength, but
that if the coarse-grained steel be then heated up to a
temperature of about 900° C., but which varies according
to the percentage of carbon in the steel, the whole structure
is completely changed, the large grains breaking down
and giving a fine-grained structure.

In the case of iron and steel also the internal changes
are much more complex, owing to changes in the form of
combination of the carbon present, and perhaps to changes
in the allotropic condition of the iron itself, so that the
metal can be hardened and tempered. It is impossible,
however, to reason from the changes which take place
in steel to those which may take place in other alloys,
because the conditions are in many respects so different,
but the occurrence of these phenomena, in the case of steel
at least, warn us to be carefully on the watch for
similar phenomena in the case of other alloys.

CHANGES IN THE STBUCTURE OF ALLOYS. 71

Diffusion. — It is well known that gases, however
different their specific gravity, rapidly diffuse one into
the other so as to produce a homogeneous mixture, and
that with liquids diffusion takes place quite as surely,
though nothing like so rapidly as in the case of gases. If
pure water be placed above a saturated solution of sugar
or copper sulphate, the dissolved substance will gradually
diffuse through the liquid till the solution becomes of
uniform composition. There is thus as far as diffusion is
concerned a continuity between the liquid and the gaseous
state, the difference being one of degree and not of kind.
It is now known that solids behave in the same way, and
that one solid will diffuse into another so as to tend towards
uniformity of distribution. With most solids at ordinary
temperatures the molecular mobility is so small that the
diffusion is inappreciable, but if the temperature be
raised to the point at which the molecules have any con-
siderable amount of freedom, the diffusion may become
recognisable or even well marked. The only case in
which diffusion of solids at ordinary temperatures has
been determined is that of gold into lead, in connection
with which experiments were made by the late Sir W.
Roberts -Austen. The diffusion was well marked, though
of course, it was slow.

Solid diffusion may, and in some cases does, produce
changes in metals which are kept at a high temperature
for some time. J^

In the case of the growth of crystals in an alloy con-
sidered above, it was assumed that the alloy was homo-
geneous, i.e., that it consisted of definite crystals, of
one substance only, a condition only met with in a few
of the alloys of commercial importance. How will the
influence of high temperature or long heating be modified,
if instead of an alloy consisting of one constituent, one be
taken consisting of, say, two constituents, either a eutectic,
or a eutectic together with an excess constituent ?

The cause of change, if any, will of course be mole-
cular mobility as in the case already considered, but in
this case the molecular mobility may act in two directions.

The tendency to segregate, which is the same thing as
the tendency to grow into larger crystals, will tend to
cause a more complete separation of the constituents.

72 CHANGES IN THE STRUCTURE OF ALLOYS.

The molecules of the excess substance will tend to grow
together, thus ejecting the eutectic into well-marked
areas, and the constituents of this eutectic will tend
to aggregate into well-marked plates. When an alloy
is slowly cooled this is the structure which is usually
seen in the case of a eutectic alloy, whilst if the cooling
(not only the solidification) be very rapid, the constituents
of the eutectic may remain in the semi-emulsified or
unsegregated form in which they are hardly distinguish-
able. Slow cooling in such a case will tend to produce a
heterogeneous and coarse structure, in which not only
are the crystals of each constituent large, but the segre-
gated masses are also large, a condition necessarily
tending to weakness and brittleness.

If, on the other hand, diffusion is the predominating
influence, these conditions will tend to be reversed.
Instead of the constituents tending to separate or segre-
gate, they will tend to diffuse one into the other so as to
produce a more or less homogeneous mass. It is quite
evident that the influence of slow cooling on the properties
of a heterogeneous alloy will depend on which of the two
tendencies is the more powerful, and until this be known
no idea can be formed as to what the influence will be,
and no doubt the difference in the properties of alloys
produced by similar treatment is to some extent due to
this difference in behaviour.

If the phenomena during the two stages of solidifica-
tion and subsequent cooling be considered, it will be found
that in each case there is the double influence, i.e., an
influence in two directions, and the actual result will be
the algebraic sum of the two actions : —

(1) Slow solidification tends to increase segregation,

and thus to produce a less homogeneous alloy
by allowing more complete separation of the
constituents, whilst rapid solidification has
the reverse effect.

(2) The effect of slow cooling after solidification will

vary with the nature of the alloy. If the
constituents are mutually insoluble, slow cooling
may produce crystal growth and increased
separation of the constituents, but if the con-

CHANGES IN THE STRUCTURE OF ALLOYS. 73

stituents are soluble one in the other their
diffusion will come into play, the separated
constituents may re-dissolve, and the alloy thus
become more homogeneous. As a rule rapid
solidification will be best, and the cooling
after solidification should be slow or rapid,
according to the character of the alloy.

It will be seen, therefore, that the structure of an
alloy may be very considerably modified by the rate of
solidification and of subsequent cooling, and that once
the behaviour of the alloy is known, its properties may be
controlled by the rate of solidification and cooling. As a
rule, the rate of cooling can be much more easily modified
than the rate of solidification.

Much further work is necessary on the behaviour of
alloys during heating and cooling before much practical
use can be made of these facts, but many workers are
engaged on the subject, and no doubt much will be done
in the near future.

Effect of Work. — When an alloy is subjected to work
by hammering, rolling, or otherwise, the structure may be
much modified, and the results will vary according as the
work is done hot or cold ; by hot work being understood
work done above the temperature at which there is con-
siderable molecular mobility, and by cold work that done at
temperatures at which molecular mobility has ceased.

The result of work done at high temperatures is as a
rule to increase the strength of the metal. The structure
is more or less broken down by the work put upon the
metal, but the molecules are free enough to re-arrange
themselves, so that they are not left permanently in a
condition of stress ; the result is therefore usually
a finely crystalline structure.

When the work is done cold, the result is somewhat
different. As the pressure is put upon the metal it
reaches momentarily the flow point, i.e.. the metal behaves
as if it were plastic or fluid, and the crystals break up
into a more or less fluid mass. The passage from this
stage to the solid is so rapid, the whole change being
almost instantaneous, that the metal has no time to re-form
definite crystals, and it is therefore left in such a con-

74 CHANGES IN THE STRUCTURE OF ALLOYS.

dition that no crystalline structure can be made out.
As the change of condition is so sudden the molecules, or
particles perhaps would be more correct, have no time to
adjust themselves to the condition normal to the cooled
state and the metal is left hard and brittle, the particles
being in a condition of stress which is relieved by an-
nealing, when, as already explained, the metal is heated
to a temperature at which there is a certain amount of
molecular freedom.

Burning. — Certain metals, when heated to a high
temperature or when heated for a considerable time in
a particular atmosphere, are liable to a change which is
called burning. The crystals become large, more or less
distinctly separated, and the metal becomes very brittle,
often indeed quite friable, and the plasticity and
strength cannot be restored by annealing.

The cause of burning is not always the same, but it is
probably always a chemical change. The most common
cause of burning is oxidation. Plastic metals are usually
more or less permeable to gases, and if air finds its way
in, usually following the lines of separation of the crystal
grains, films of oxide may be formed which break up the
continuity of the metal, and thus make it brittle.

Copper, however, is burnt by being heated in a reducing
atmosphere, probably by the removal of the last trace of
oxygen.

CHAPTER VI.

THE METALS USED IN THE PREPARATION OF ALLOYS.

THE number of alloys in use is very large, but they may
be conveniently classed into a few groups.

(1) Brasses. Alloys of copper and zinc with or without

the addition of small quantities of other metals.

(2) Bronzes. Alloys of copper and tin with or without

the addition of small quantities of other metals.

(3) Machinery brasses or bronzes. Alloys of copper

with tin and zinc, and sometimes with other
metals.

(4) Aluminium alloys.

(5) White bearing metal alloys.

(6) Soft alloys, such as pewters, type metal, &c.

(7) Nickel alloys. German silvers.

(8) Alloys of the precious metals.

(9) Amalgams, or mercury alloys.

(10) Alloys of iron and steel. This group will not be
considered in this book.

The constituent metals used in the manufacture of
alloys, excluding those used only in small quantity, are
copper, zinc, tin, lead, antimony, nickel, aluminium, gold,
silver, platinum, mercury. Whilst it is not necessary to
discuss the metallurgy of these metals, it may be advisable
to describe briefly the forms in which they are obtainable
in commerce, and therefore in which they can be used in
the preparation of alloys.

Copper. — Copper is distinguished by its characteristic
red colour, and if present in large quantity it imparts a
colour to alloys containing it, though the colour produced
does not seem to be in any way related — at least in most
cases — to the colour of the copper itself. It melts at about
1,090° C., and is slightly volatile, sufficiently to impart a
green colour to a flame in which it is placed, but not
sufficiently for there to be any loss when it is melted.
It is malleable and ductile, and so can be obtained in
thin sheets or in fine wire. Its specific gravity is about
8-9, but the figures obtained vary with the condition of

76 METALS USED IN THE PREPARATION OF ALLOYS.

the metal examined, castings as a rule having a lower
specific gravity than metal which has been wrought.
The tensile strength of copper is not high, and varies
considerably, according to the condition of the metal,
being higher in the case of metal rendered hard by working
than when the metal has been annealed, and much less
in the case of castings than in the case of the wrought
metal. Thurston states that when copper is to be used
for structural purposes the strength specified should be
not less than 25,000 Ibs. (11-16 tons) per square inch for
castings, 35,0001bs. (15-62 tons) for bars, and 60,000lbs.
(26-8 tons) for wire. Dr. Watson* quotes some
samples of electrotype copper unworked as having a
tensile strength of 16 tons, with an elongation of 20 per
cent, on 4in., whilst when annealed the same copper gave
13-6 tons with an elongation of 42 per cent. Other sam-
ples of copper, pure copper deoxidised by phosphorus and
rolled down to Jin. thick without annealing gave the

following figures :— Tensile Elongation.

PerCent. Strength. in2in.

Pure copper 14-38 . . 62-5

Copper containing arsenic -050 = .. 14-29 .. 60-0

188= .. 14-39 .. 61-0

„ antimony -025- ..14-50 ..56-5

200= .. 14-777 .. 60-0

lead 200= .. 14-36 .. 58-0

Copper does not cast well, as in the molten condition
it absorbs a considerable quantity of gas which is given
out as the metal cools, and thus produces unsoundness.
The addition of a small quantity of phosphorus, to a large
extent, overcomes this defect.

Copper is a good conductor of heat and electricity,
and the electric conductivity is considerably reduced by
the presence of small quantities of impurities, quantities
of certain metals which would escape detection by the
ordinary chemical analysis having a marked effect on
the electric conductivity.

Copper combines very readily with most metals and
non-metals, and therefore may contain considerable
quantities of impurities. How far these impurities will
interfere with the use of the metal by the maker of alloys

* Proceedings last. Mech. Engineers, 1893, p. 169.

METALS USED IN THE PREPARATION OF ALLOYS. 77

will depend on the nature of the alloy being made and the
purposes for which it is to be used. It by no means
follows that an impurity which has little influence on the
properties of the copper itself will therefore be equally
uninjurious in an alloy, or that an impurity which has a
marked influence on the properties of the copper will have
an equally important influence on the properties of an
alloy. Arsenic, for instance, is always considered to be a
most objectionable impurity in copper, and yet for some
purposes • 5 per cent, is not only not injurious, but seems
to improve the quality of the metal. It would probably
not be objectionable in copper to be used for preparing
yellow brass which is to be cast, but it would render
the metal quite unfit for the manufacture of a brass
which had to be drawn cold.

Copper oxidises very readily on exposure to the air at
high temperatures, black flakes of copper oxide or copper
scale (a mixture of the two oxides CuO and Cu20) being
formed. If the metal be in a liquid condition, probably
only Cu20 is formed, and this is rapidly dissolved, making
the copper "dry," in which condition it is extremely
brittle, and breaks with a brick-red granular fracture
instead of the fibrous fracture of tough-pitch copper.
The amount of oxygen present in copper as dissolved
oxide varies very much. It is extremely difficult to
estimate accurately, and many of the published figures
are unreliable. In a series of analyses of copper fire-box
plates, published in the Proceedings of the Institution
of Mechanical Engineers for 1873, the amount of
oxygen in combination is given as varying in the 1 1 sam-
ples analysed from -019 to -248 per cent. It is doubtful
whether the presence of a small quantity of oxygen is
any serious objection in the case of copper to be used for
making alloys, since the other metal, zinc or tin, is so much
more easily oxidisable that it would probably decompose
the copper oxide forming zinc or tin oxide, which would
not dissolve but which would pass into the slag.

The removal of the oxide of copper which is formed
during the process of refining, and the presence of which in
excess is necessary for the complete removal of sulphur
and other easily oxidisable impurities, is brought about
by the process of " poling," which consists as is well

78 METALS USED IN THE PREPARATION OF ALLOYS.

known of immersing a pole of wood in the molten copper.
The wood undergoes decomposition, and the reducing
gases given off reduce the copper oxide and carry away
the oxygen. If the poling be carried too far the metal
passes beyond the tough stage, and becomes over-poled.
It is then brittle, but the fracture is quite different from
that of dry copper. The cause of over-poling has not yet
been completely made out, but it seems that in presence
of small quantities of certain impurities, especially arsenic
and antimony, the presence also of a small quantity of
oxygen in combination is essential to keep the metal
in its " tough " form, and if this be removed it becomes
brittle. Over-poling is closely related to the burning
produced when copper is heated in a strongly reducing
atmosphere. When copper is heated in an oxidising
atmosphere, it is rendered brittle by the formation of
oxide of copper along the planes separating the con-
stituent crystals.

Commercial copper may contain arsenic, antimony,
lead, bismuth, iron, nickel, cobalt, oxygen, sulphur, and
in rare cases perhaps other metals.

Many alloy makers in order to secure the best results
use the purest, and therefore the most costly, copper for
the manufacture of their alloys. Where the brass is to be
worked cold, as, for instance, where it is to be used for the
manufacture of boiler or other tubes, it is essential to use
a fairly pure metal, but even in this case it is doubtful
whether metal of extreme purity is of much, if any
advantage. For alloys which are to be cast it does not
seem that metals of great purity have any advantage over
the ordinary commercial forms, but at present the influence
of small quantities of impurities on the quality of alloys
has not been thoroughly worked out.

The purest copper obtainable is that known as
electrotype copper, which is obtained by electro-
deposition. This is almost chemically pure, and is now
used on a large scale for the manufacture of brass con-
denser and other similar tubes.

•^ Lake Superior copper made from the native copper
of the Michigan copper district is also extremely pure.

The purest form of copper usually used is that known
as B.S. (Best Selected) so-called because it was at one time

METALS USED IN THE PREPARATION OF ALLOYS. 79

made in Swansea from metal containing a considerable
quantity of arsenic and other impurities by a process
known as the Best Selecting process. Now it is almost
always made from materials so pure that that selecting
is not necessary. It should not contain more than
•05 per cent, of arsenic and a trace of antimony.

A test used by the Admiralty can be made use of to
ascertain whether a copper is of the B.S. quality or not.
Three pounds of the copper is melted in a crucible in the
ordinary way and 2lbs. of zinc is added, so as to make an
alloy of approximately the composition of 60 per cent,
copper and 40 per cent. zinc. The metal is then cast into an
iron mould about 4in. squareand lin. deep, and is allowed to
cool slowly ; the ingot is then nicked across the top with a
swage and is broken either under the steam hammer or
by means of a sledge. If the metal is B.S. quality the
fracture will be dull, granular, and of a buff colour, and
there will be few if any bright, brassy streaks crossing it.
If it contains a considerable quantity of arsenic or antimony
the whole fracture will be columnar, and will have a bright
yellow colour and metallic lustre.

The following is a scale of qualities as indicated by
this test : —

1. Very good. The fracture is of a uniform, dull
buff* tint.

2. Good. The fracture is mostly as in 1, but shows
a few bright, brassy looking streaks.

3. Tolerably good. The number of bright streaks
is greater.

4. Not good. The bright streaks are numerous, but
cover not more than about one-third of the area.

5. Bad. The bright streaks predominate.

6. Very bad. The fracture is entirely, or almost
entirely, bright and brassy.

Nos. 1, 2, and 3 would be passed as B.S.
The ingot must be allowed to cool naturally ; sudden
cooling considerably modifies the fracture.

Tough copper is the ordinary commercial copper. It
may be very impure and may contain -5 per cent, or
even more of arsenic. The name, of course, carries
with it no guarantee as to quality.

80 METALS USED IN THE PREPARATION OF ALLOYS.

It is perhaps hardly necessary to mention "Chili bar,"
though cases have occurred in which founders, seeing
this quoted at lower prices than other varieties of copper,
have purchased it for alloy making, with not very satis-
factory results. It is an impure unrefined copper im-
ported from Chili and may contain up to 2 per cent, or
even more of sulphur.

Copper can be obtained in many forms, and it is need-
less to say that any quality can be prepared in any form.
The usual form is that of ingots weighing about 141bs.r
cast with one or two nicks at the bottom, so as to
facilitate breaking. Ingots are, however, now cast of
any form and size that the user may require. Slabs
are of larger size and are usually used for rolling.
Roller ends are often used for alloy making ; they
are the ends cut off a calico-printer's rollers, and then
broken into pieces under a hammer. They are usually
high in arsenic. Bean shot and similar varieties are
obtained by pouring the molten copper into water.

No judgment can be formed as to the quality of a cop-
per by the appearance of the ingot, or by the fracture, except
that by the latter it can be seen if the metal is tough, dry,
or over-poled. The surface and colour of the ingot depend
largely on the way it is cast and cooled. Ingots cast in
copper moulds are much smoother than those cast in
iron moulds, and the red colour so often seen on the
surface of the ingot is due to quenching in water imme-
diately it has solidified. The sooner the ingot is turned into
the water the better colour it will be. The red colour
is best shown in rosette copper, which is solidified by
throwing water on the surface of the liquid metal, or
in Japanese copper, which is cast under water.

The following analyses will indicate the general
composition of commercial coppers.

1. Electrotype.

2. Lake Superior (Eggleston.)

3. B.S.

4. B.S.

5. Tough copper.

6. Copper fire-box plate.

METALS USED IN THE PREPARATION OF ALLOYS. 81

Arsenic ....
Antimony . .
Lead

•01
Trace
Trace

Nil
Nil
•016

•03

Trace

• 025

Trace
•024

.32

Trace
.07

•373
• 035

•408

Bismuth. . . .
Iron

Trace
•Nil

•05

Trace

•Oil
•006

.01
• 01

• 036
•007

Nickel ....

Nil

•041

• 06

•304

Silver

• 026

•03

•035

Oxygen ....
Phosphorus
Sulphur

Nil
Nil
Nil

•15

.143

.12

•018
•006

It should be noticed that refined copper never con-
tains more than a minute trace of sulphur or lead, but
that as lead is often added during rolling, rolled copper
may contain up to about • 5 per cent. lead.

Zinc. — Zinc is a bluish-white metal having a specific
gravity of about 7-1. It melts at about 415° C., and
boils at about 930° C., so that it can be readily distilled,
and there is always a sensible loss when it is used in the
manufacture of alloys. The metal in .fine shavings or
vapour burns readily with an intense bluish-ivhite flame,
forming dense clouds of white zinc oxide (philosophers'
wool). It is malleable and ductile through a limited
range of temperature only, and is largely used for rolling
into sheets for roofing and other purposes. It oxidises
only slightly on exposure to the air, with the formation
of a basic carbonate.

Zinc comes into the market in the form of rolled
sheets, and also in cast cakes of about lin. thick, in which
form it is known as spelter. The cakes are very brittle,
and break with a more or less crystalline fracture. If the
metal be nearly pure the crystal faces are large, bright, and
smooth ; if there be a small quantity of iron present dull
spots appear on the crystal faces, and if the quantity ot
iron rises to a few per cent., as in dross spelter, the fracture
becomes granular. The amount of iron present can be
fairly judged from the appearance of the fracture. Its
tensile strength is low, but it is never used for structural
purposes where it is subjected to great stress.

82 METALS USED IN THE PREPARATION OF ALLOYS.

Zinc casts well and contracts but little on solidifying,
and is largely used for the manufacture of statuettes and
other ornamental castings which are usually coated with
bronze or brass by electro-deposition.

Zinc is never pure. The principal impurities are
iron, lead, tin, copper, arsenic, and cadmium. . Iron is
always present in spelter. It does not distil over with the
zinc, for freshly distilled zinc hardly contains a trace,
but it is dissolved from the iron vessels in which the metal
is melted, and rods with which it is stirred. When zinc
is used for galvanising a hard zinc which contains several
per cent, of iron accumulates in the vats. Good com-
mercial spelter should not contain more than -05 per
cent, of iron, and this is about the maximum allowable
for alloy making.

Lead is invariably present in spelter in larger
or smaller quantity as it distils over with the zinc during
the process of manufacture. Lead is only slightly soluble
in zinc and in the solid condition zinc cannot retain more
than about 1 • 5 per cent. It sometimes happens that
from rapid cooling a sample may contain more, but in
that case some of the lead will almost always be found to
be distributed in minute shots or fragments through the
mass, and if the metal be melted and slowly cooled it will
separate. For making brass or other alloys, a spelter
containing more than 1 • 5 per cent, of lead should be
rejected. When spelter is redistilled, even if the redistilla-
tion be repeated, about • 2 per cent, of lead passes over, so
that it is almost impossible to obtain a spelter containing
less lead than this.

Cadmium is rarely present except in minute quantity,
and is so like zinc in all its properties that it does not
seem to be objectionable, at anyrate in any quantity
likely to be present in commercial zinc.

Copper is rarely present in any but the minutest
quantities, and is quite unobjectionable for the prepara-
tion of alloys.

Tin is often present in minute quantities, but rarely
in sufficient quantity to be objectionable.

Arsenic is rarely present except in quantities too small to
be of any importance for practical purposes, whilst anti-
mony and sulphur a,re sometimes present in minute traces.

METALS USED IN THE PREPARATION OF ALLOYS. 83

Tin. — This is a silver-white metal having a specific
gravity of about 7-3. It is soft and very malleable.
It melts at about 232° C., and boils at a white heat. It
does not tarnish rapidly on exposure to the air, but at a
red heat it is readily oxidised, forming the oxide SnO,.

Commercial tin is never pure, though the quantity of
impurity present is always small. The following analysis
of Queensland tin by Thurston will give an idea of the
composition of a good commercial tin : —

Lead M ..0-165

Iron 0-035

Manganese.. .. .. .. .. 0-006

Arsenic . . . . . . . . . . Trace.

Copper . . . . . . . . . . None.

Zinc

Antimony . . . . . . . . ,,

Bismuth . . . . . . . . . . ,,

Nickel . . . . . . . . . . ,,

Tungsten . . . . . . . . . . „

Molybdenum . . . . . . . . ,,

Banca tin is said to be the purest obtainable. Mr.
Parry states that Peruvian and Bolivian tin are the
most impure, and contain lead and antimony, and that
certain brands of Australian tin contain bismuth.

The impurities most likely to be present are lead,
iron, and copper. The quantity of any of the elements
should not exceed about -1 per cent., and the total
quantity of tin should be from 99 • 7 to 99 • 95.

Tin comes into the market cast in various forms, to suit
the convenience of users. Block tin is in small bars,
blocks, or cakes of various sizes. Stick tin is cast into
small sticks. These sticks emit a peculiar crackling
sound or " cry " when bent, and this sound is sometimes
taken as being a test of the purity of the metal, because a
small quantity of lead destroys it. Grain tin is made by
casting a large block, heating it till it becomes brittle,
and then breaking it up. Granulated tin is made by
melting the metal and pouring it into water,

Tin foil is tin which has be.en rolled out into very thin
sheets, often not more than J^in. in thickness.

84 METALS USED IN THE PREPARATION OF ALLOYS.

Lead is an extremely soft metal of a bluish
colour. Its specific gravity is 11-4, and therefore it is the
heaviest of the metals used in making alloys. It has very
little tenacity, is very malleable, but its low tenacity
makes it difficult to draw it into fine wire. It melts at
327° C., and is sensibly volatile at high temperatures. It
oxidises very slowly in air at ordinary temperatures, but
rapidly at a red heat, forming the oxide Pb 0 and finally, if
the temperature be not too high, red lead Pb3 O4.

Commercial lead is always very pure, the impurities
present being in very small quantities and of no practical
importance.

The following analysis by Thorpe will give an idea of
the usual degree of purity : —

Silver .. .. 0-00200 Antimony .. 0-00173
Copper .. .. 0-00228 Iron .. .. 0-00035

Cadmium.. ... Trace. Zinc 000014

Bismuth.. .. 0-00040 Sulphur .. .. 000076

Antimony. — Antimony is a bluish-white crystalline
metal, which melts at about 450° C. and is volatile at a
white heat. It has a specific gravity of about 6 • 7 and is
extremely brittle. When pure antimony solidifies slowly
the surface assumes a fern-like crystalline appearance,
and a structure called the antimony star is pro-
duced. To produce this appearance the solidifying sur-
face must be kept covered with a layer of slag. Impure
antimony does not give the star, but as it depends on the
conditions of cooling pure antimony does not always
show it. Commercial antimony usually contains sulphur,
arsenic, lead, copper, and iron. The tollowing analysis
will indicate its composition :—

Per Cent.

Arsenic . . . . . . . . v . . -06

Tin .. ...-.-. .. ^ ;. .. —

Lead .;. .. • : • • ^. : .. -46

Copper . . . . .v . . . . -07

Iron .. ..-- '. . . . .. -16

Zinc .. .. v, .. .* -08

Sulphur .. .. ^.. .. ... -20

Antimony (difference) .. .. .. 98-97

100-00
The sample showed a well-crystallised surface.

METALS USED IN THE PREPARATION OF ALLOYS. 85

Bismuth. — Bismuth is a pinkish- white metal having
a specific gravity of 9- 82. It melts at 266° C., and ex-
pands by over 2 per cent, on solidifying. It volatilises at
high temperatures, its boiling point being between
1,000° C. and 1,500° C.

Metallic bismuth may contain silver, lead, copper,
arsenic, iron, nickel, cobalt, and sulphur, and sometimes
the rarer metals. The following analyses from the
" Mineral Industry " for 1893, p. 72, will give an idea of the
composition of commercial bismuth : —

Bismuth . .
Antimony
Arsenic . .
Copper . .
Silver . .
Sulphur . .

Saxon. Peruvian. Australian.

99-77 ..

93-372 ..

94-103

Nil. . .

4-570 ..

2-621

Nil. . .

Nil ..

0-290

0-08 ..

2-058 ..

1.944

0-05 ..

—

0-01

•430

99-91 100-00 99-388

Bismuth is very little used, except for the preparation
of very fusible alloys.

Aluminium. — Aluminium is metal of a brilliant silver-
white colour. The most striking property of the metal
is its extreme lightness, its specific gravity being only 2 • 7.
It is soft, malleable, and ductile ; when cast it has a
tensile strength of about 6 or 7 tons per square inch,
which by working may be increased to 16 tons or there-
abouts. Its melting point is between 600° and 700° C., but
owing to its very high specific heat, it melts, solidifies,
and cools very slowly.

It does not oxidise readily in air, even at a red heat,
unless finely divided, when it will burn with a brilliant,
highly actinic flame ; but it decomposes metallic oxides,
alumina being formed.

86 METALS USED IN THE PREPARATION OF ALLOYS

Commercial aluminium, as now prepared by the
electrolytic methods, is very pure, containing about
99-50 of the metal.

The properties of gold, silver, platinum, and mercury
will be described as far as is necessary in connection with
the alloys for which they are used.

CHAPTER VII.
THE BRASSES.

BRASS is an alloy of copper and zinc, and strictly
speaking should contain no other added metal, the
impurities present being only those present in the
metals used in making the alloy or accidentally introduced
in the process of manufacture. Brasses to which
other metals are added in large or small quantity will be
considered separately.

Impurities in Brass. — It is obvious from what has been
said of the metals used in the manufacture of brass that
the impurities present should be only in small quantity ;
indeed, if metals of good quality are used, a brass should
not contain more than about —

Arsenic . . . . . . . . . . «03

Lead .... .. -50

and other impurities in still smaller quantity. As lead
and arsenic cannot be accidentally introduced during
manufacture, any larger proportion of these elements
must be regarded as being due either to the use of impure
materials or intentional addition. It must, of course, be
remembered that it is only for certain purposes that a
brass of high purity is necessary.

The quantity of iron present will usually exceed
that due to the iron in the metals used, because iron tools
are used for stirring, and some of the iron is always
dissolved. About 15 per cent, is a fair amount to be
present in a brass, but a slightly larger quantity is not
usually objectionable.

Sulphur, which is one of the most objectionable im-
purities in braps, is never present in appreciable quantity
in copper or zinc, and therefore is never present in brass
made from fresh metal unless the brass has been melted
under such conditions that sulphur can be absorbed
either from the fuel or the products of combustion.
Should coke fall into the crucible during melting, or

88 THE BRASSES.

should the metal be melted in a reverberatory furnace
with a fuel containing much sulphur, sulphur will almost
certainly be taken up by the metal.

In actual practice, however, new metals are very rarely
used alone, a portion of the charge being usually made up
of scrap. If the scrap be new clean scrap, this will, of course,
have no injurious effect, except to make the preparation
of brass of definite composition more difficult, but if as
is often the case the scrap is dirty, sulphur and other
impurities may be introduced into the metal. Old boiler
and condenser tubes returned as scrap are frequently
used in the manufacture of brass, and these may contain
deposit which consists partially, at anyrate in the case of
boiler tubes, of sulphur compounds. When such scrap
is melted, the sulphur passes into the metal, and may
seriously interfere with its quality. In some ingots of
brass made by melting old boiler tubes the following
percentages of sulphur were found.

Per Cent.

No. 1 .018

No. 2 .020

No. 3 .020

No. 4 .107

As a rule the quantity will be less, as only a portion of the

charge will be scrap.

Range of Composition of the Brasses. — The brasses used
commercially range from about 95 per cent, copper and
5 per cent, zinc to 40 per cent, copper and 60 per cent,
zinc, the most important, however, being those containing
from 70 to 50 per cent, of copper. The brasses not only
vary very much in composition, but necessarily vary very
much in all the properties on which their usefulness
depends. A large number of experiments have been made
at different times for the purpose of tracing the connec-
tion between composition and these properties, but they
have not until recently been attended with great success,
for one reason among others that care has not always been
taken that the alloys compared have been in the same
physical condition or of the same degree of chemical purity,
both of which conditions may have a marked influence on
the properties. Few of the properties form a continuous
series varying directly with the composition, but each

THE BRASSES. 89

one rises or falls as the case may be to a maximum or
minimum, and then decreases or rises again, and the only
way of satisfactorily showing the variations is by the use
of curves, and needless to say before these can be
determined accurate data must be at hand.

Many tables have been published giving details of the
results of the work of many observers ; a very extensive
table, for instance, which was drawn up by an American
committee in 1881 is contained in most of the books on
alloys. Such collections are of little scientific value
because being the work of many observers, and
the observations having been made under varying
conditions, the results are not strictly comparable.
If an attempt be made to plot a series of curves with the
data given in the table mentioned above, it will be found
that no curve can be drawn which will even approxi-
mately include all the observations. Such a table may,
of course, be of general value as indicating the sort of
variations that may be expected with alloys of very
similar composition, but for little else.

The American committee above mentioned collected a
large amount of useful information, and under the direction
of Prof. Thurston many experiments of great interest
and value were made. Later, the English Institution of
Mechanical Engineers appointed an Alloys Research
Committee, and under the direction of the late Sir W.
Roberts Austen further research was made into the
relationship existing between the composition and
properties of alloys. The work of this committee is of
special value, because it had at its command methods of
research that were not available to the earlier workers.
Much of our accurate knowledge of the alloys has been
derived from the work of these two committees, but
within the last few years the interest in alloys has greatly
increased, and many competent workers have entered
the field and published the results of their work.

Remembering that a brass is an alloy of copper and
zinc in any proportion, and therefore the brasses form a
series commencing with copper and ending with zinc,
the simplest method of study will probably be to take
the more important properties of the alloys and see
how these vary with changes in composition.

90 THE BRASSES.

Colour of Brasses. — The colour of brass is one of its
striking properties, and brass yellow is a fairly descriptive
term. Only certain members of the brass series, how-
ever, have a yellow colour, others being quite white.

The following is the colour series given by the American
Committee,* the observations having been made by Prof.
A. R. Leeds.

The percentages of copper and zinc are those obtained
by the actual analysis of the sample.

No. Copper. Zinc. Colour.

0. 100 0 Red.

1. 96-07 3-79 Brilliant yellow-red.

2. 90 56 9-42 More nearly approaching

yellow.

3. 89 -80 10 - 06 Light yellow.

4. 81-91 17-99 Brass yellow.

5. 76-65 23-08 Full yellow.

6. 71-20 28-54 Dark yellow.

7. 66-27 33-50 Gold yellow.

8. 60-90 38-65 Orange yellow (tarnished).

9. 55.15 44.44 Surface tarnished, of dull

reddish-yellow colour.

10. 49-66 50-14 Deep yellow.

11. 47-56 52-28 Reddish white.

12. 41-30 58-12 Nearly silver white, changed

to yellow by oxidation.

13. 36-62 62-78 More silvery.

14. 32-94 66-23 Bluish white.

15. 25 • 77 73 - 45 Dull bluish white.

17. 20-81 77-63

18. 14-19 85-10

19. 10-30 88-88

20. 4-33 94-59 Bright bluish white.

21. 100 Bluish white.

The colours are those of a fractured ingot. From
Nos. 10 to 14 the lustre is described as splendent or

* Thurston's " Materials of Engineering," vol. 3, p. 373, second edition.

THE BRASSES. 91

brilliant. It is, of course, impossible to give any numerical
colour standard, and therefore the descriptions must
necessarily be somewhat vague, but the general result is
clear. The addition of a small quantity of zinc speedily
destroys the red colour of the copper, producing first a
reddish yellow and then a yellow alloy. The yellow
colour varies much in shade, the exact shade seeming
to have little relation to the percentage of copper,
till this is reduced to about 40 per cent., when the yellow
colour gives place to a white and white brass is produced.

In the United States Committee's report, the
appearance of the fracture is also given, but this is of little
importance, and it must be borne in mind that in some
cases both the appearance and colour of the fracture may
be very much modified by the presence of impurities
even in small quantity.

Specific Gravity. — The specific gravity of copper being
much greater than that of zinc, as might be expected the
specific gravity of the alloy falls as the percentage of
zinc is increased. The alloy always has a density greater
than that which it would have if it were merely a
mechanical mixture of the two metals.

The following table gives the results of one series of
experiments made by the U.S. CommitteeJ which will be
quite sufficient for general purposes :—

No.

( 'opper.

Zinc.

Sp. Or.

1 ..

96-07

3-79 ..

8-825

90-56

9-42 ..

8-773

3 ..

89-80

.. 10-06 ..

8-656

4 ..

81-91

.. 17-99 ..

8-598

5 ..

76-65

.. 23-08 ..

8-528

6 ..

71-20

.. 28-54 ..

8-444

66-27

.. 33-50 ..

8.371

8 ..

60-94

.. 38-65 ..

8-405

9 ..

55-15

.. 44-44 ..

8-283

10 ...

49-66

.. 50-14 ..

8-291

11 ..

47-56

. . 52-28 . .

8-189

12 ....-

41-30

.. 58-12 ..

8-061

J Thurston's " Materials of Engineering," vol. 3, p. 377, second edition.

THE BRASSES.

No.

13
14
15
16
17
18
19
20
21

Coppe-.

36-62
32-94
25-77
25-92
20-81
14-19
10-30
4-35

Zinc.

62-78
66-23
73-45
73-06
77-63
85-10
88-88
94-59
100-0

Sp. G.

974
811
675
687
418
163
253
108

7-143

Fig. 46 gives these specific gravities plotted so as to
give a specific -gravity curve.

-S8'5

0 g
o
jd
3

O> 7-K

d, / 0

— -

-—

r — •

— •

Cu. % 0 10 20

80 90 100

Copper-zinc Alloys — Specific Gravity.
FIG. 46.

(It will be seen that the curve is not regular. Many of the variations may
be due to differences in the condition of the samples.)

Tenacity. — The tenacity or tensile strength is, of
course, one of the most important properties of alloys
to be used for structural purposes, and many series
of determinations have been made. Here again the
American committee's results are of great value. The
actual tensile strength of a brass will vary with the treat-
ment to which it has been subjected, but if all the samples
are treated in the same way, a series of figures will be
obtained which will at least roughly indicate the
relationship existing between the tenacity and the
composition.

THE BRASSES.

Copper.

Zinc.

Tensile
Strength,
Ibs. per
square inch of
Original
Section.

Elastic
Limit
per cent,
of Breaking
Load.

Total
Elongation
.per cent.

100

27,800 51-8

6-47

97-83

1-81

27,240

82-93

16-98

32,600

26-1 j 26-7

81-91

17-99

32,670

30-6 31-4

76-65

23-08

30,520

24-6 35-8

71-20

28-54

30,510

29-5 29-2

66-27

33-50

37,800

25-1 37-7

60-94

38-65

41,065

40-1

20-67

55-15

44-44

44,280

44-0

15-31

49-66

50-14 30,990 54-5

4-97

47-56

52-28

24,150 100

•79

41-30

58-12

3,727 100

—

36-62

62-78

2,656

100

—

32-94

66-23 1,774

100

—

25-77

73-45 9,680

100

•35

25-92

73-06

7,931

100

—

17
18

20-81
14-19

77-63
85-10

9,000
8,500

100
100

-16
-31

10-30

88-88

14,450

100

•39

4-35

94-59

18,665 ! 100

•49

—

100

5,400

•69

These figures are plotted in Fig. 47, and for com-
parison the curves obtained by the Alloys Research Com-
mittee and by M. Charpy are given.

From these results it is possible at least to roughly
summarise the influence of composition on the strength
and other properties of a brass.

As zinc is added to copper the tensile strength increases,
at first slowly, and then more rapidly till the maximum
is reached with something between 55 to 60 per cent, of
copper, the exact point being different in the three sets of
determinations. As the amount of zinc is further increased
the tensile strength falls off very rapidly, till when white
brass is reached, with about 40 per cent, copper, the
tenacity has become so small as to be negligible. The

THE BRASSES.

10 20 30 40 50 60 70 80 90 100
Cu. % 100 90 80 70 60 50 40 30 20 10 0

Copper-zinc Alloys — Tenacity.

A.R. — Alloys Research Committee (worked rods). Th. — Thurston (castings).

Ch. — Charpy (annealed brass).

FIG. 47.

Cu.
Zn.

^/^

;

Th.

0
)0

\
s

0
0

{

50
50

0
0

A
(

10
iO

5
5

6
4(

3
)

7
3

0
0

(

100
0

Copper-zinc Alloys — Extensibility.
Ch = Charpy. Th = Thurston.
FIG. 48.

THE BRASSES. 95

maximum tenacity for cast brass is about 44,0001bs., or
about 19 tons per square inch, whilst for worked rods it
may reach about 80,0001bs., or nearly 36 tons.

Extensibility. — When the extensibility is considered,
similar results are obtained. As zinc is added the
alloy becomes more and more ductile, the ductility
reaching its maximum when about 70 per cent, of
copper is present. As the copper is diminished the
ductility falls off very rapidly, and becomes negligible
when the proportion of copper falls to about 50
per cent. It follows that where a brass is required
for strength only, which, of course, may be accom-
panied by hardness and to some extent brittleness, it
should contain about 60 per cent, of copper, but that
where ductility and toughness are required, as is almost
always the case, the percentage of copper must be about
70 per cent. On the whole, the alloy containing 70 per
cent, of copper and 30 per cent, zinc is the strongest and
most generally useful of the whole series. When the
copper falls below 66 per cent, the alloy is difficult to
work cold, though alloys poorer in copper may be readily
worked hot.

Hardness. — The hardness of the alloy is greater than
that of the copper, and like the other properties this
reaches a maximum, then falls off. Hardness is a property
which it is not easy to define or measure.

Fusibility. — The melting point of the brasses gradually
falls as the quantity of zinc is increased. Reference
should be made to what has been said on the meaning of
the melting point of an alloy, because a study of the melting
and freezing phenomena will throw much light on the
structure of the alloys.

Fig. 49 is the freezing-point curve of the copper-zinc
series of alloys as determined by the Alloys Research
Committee. .-

It will be seen that copper melts at about 1,0 82° C.,
and that as zinc is added the solidifying or melting
point of the alloy falls, the mass, however, solidify-
ing as a whole until an alloy is reached which contains
about 70 per cent, of copper, and which melts at

THE BRASSES.

about 950° C. Beyond this the mass solidifies in two parts :
Firstly, the copper containing zinc (which has a lower
melting point as the percentage of zinc is increased), and,
secondly, a eutectic solidifying at about 890° C., or a
little below. As the percentage of zinc is increased
till there is about 55 per cent, of copper present, the mass
solidifies as a whole at about 890° C., the first eutectic
temperature. As the amount of zinc is still further
increased, the freezing point falls, but the alloy still

1100°
Cul082°C.
1980°F.

1000°

900°

| 800°
'•+»

C
o>
0 700°

h
2

1 600°

o
P-

H 500°
400°
300°

2010°
1832°

1652°

1472° |

0)
M

12920 g

a
1112° H

932»

Zn.419°C

786°F
732°

572°

>>^ ,

fecfic

"•— ~.

*-**.

— • — |

•

Tufa

tr-

zt>

te"

— K

uteti

ui.

Zn % 0 10 20 30 40 50 60 70 80

Cu % 100 90 80 70 60 50 40 30 20
Copper-zinc Alloys. — Freezing-point curve.
FIG. 49.

90
10

100
0

solidifies as a whole till a composition of about 45 per
cent, of copper is reached, when, whilst the mass of the
alloy solidifies at about 860° C., a second eutectic separates
which freezes at about 840° C.* As the quantity of zinc is
increased when it reaches about 72 per cent., whilst the
freezing commences at about 786° C., a third eutectic is
formed which freezes at about 680° C., and the alloy

* The line of this eutectic is short and it was not noticed by the A.R.C., so is not
shown in the diagram.

THE BRASSES. 97

again solidifies as a whole, but immediately a fifth
eutectic which has the freezing point of zinc begins to
separate and continues until 100 per cent, of zinc is
reached.

It will thus be seen that the freezing phenomena
of the copper-zinc series are somewhat complex. On the
whole the freezing point or melting point does fall steadily,
though not uniformly, as the percentage of zinc is increased.
For certain ranges, e.g., from 100 per cent, to about 70
per cent, copper and through several other ranges of
temperature, the alloy has only one freezing point and
at certain (eutectic) points it solidifies as a whole,
but for other ranges of composition it has always two
freezing points, a definite eutectic being separated.
All the useful alloys fall within the range of the com-
positions when the metal either solidifies as a whole or
when the first eutectic separates.

It becomes of importance to see if any relationship
can be traced between the physical properties of the alloy
and the freezing temperatures. It will be seen that as
zinc is added to copper, the whole retaining its homogeneity,
the tensile strength and the ductility as measured by the
extension gradually increase. When the eutectic begins
to separate the ductility begins to fall, but the tensile
strength continues to increase, and reaches its maximum
at about the point when its whole mass is composed of
the eutectic or a little beyond, after which the tensile
strength also falls off very rapidly, soon becoming so small
that the alloys are valueless for all practical purposes.

It will be seen, therefore, that the changes in physical
properties are dependent on changes in the constitution
of the alloy itself.

It is stated that several chemical compounds of copper
and zinc exist, three at least having been described, viz. :

Cu. Zn. 49 • 3 per cent, copper, 50 • 7 per cent. zinc.
Cu. Zn, 32-7 „ „ 67-3

Cu. Zn3 24-6 „ „ 75-6

Mr. Campbell, however, states that no definite com-
pounds occur, the six solid phases being all solid solu-
tions which he indicates by the Greek letters a 8 y S c £•
H

THE BRASSES.

Before the composition represented by the first of
these is reached, the alloy has lost to a large extent the
properties which render brass so valuable in the arts,
except the yellow colour, and it may therefore be safely
stated that none of the useful alloys of copper and zinc
are definite chemical compounds.

Microstructure. — The microstructure of the brasses throws
some light on the reasons for the change of character at
the various critical points.

As will be seen from the cooling curve (p. 96}
an alloy of copper and zinc containing more than 70 per
cent, of copper solidifies as a whole, and is therefore
probably a solid solution of zinc in excess 01 copr>er, and

FIG. 50.

Brasses, 70 per cent, copper, 30 per cent. zinc. Crystals large in the one, small in

the other. This is the normal structure of alloys containing 70 per cent, or more

of copper after annealing, V, 100 x.

as might be expected, such alloys show a more or less
uniform structure, the whole being made up of crystals
variously oriented.

' It must be remembered that a solid solution is not
necessarily perfectly homogeneous, or, rather it should be
said, is not normally homogeneous, for the solidification
is always selective, the mother liquor growing stronger
and stronger in the dissolved substance, and therefore if
the cooling be slow, the mother liquor may be enriched
up to the eutectic point, even though the mass of metal,
is not saturated. It therefore sometimes happens
that distinct segregation takes place, and eutectic can be

THE BRASSES.

detected even in alloys richer in copper than the proper
eutectic. Such mixtures are, however, distinctly un-
stable, so that on annealing the eutectic diffuses, dissolves,
and disappears. For this reason the structure is always
better determined on the metal as annealed rather than as

A. 40 X.

OBLIQUE ILLUMINATION.

B. FIG. 51. V, 100 x. C.

A. An alloy of 92 per cent, copper, 8 per cent. zinc.

B. An alloy of 67 per cent, copper, 33 per cent, zinc, as cast.

C. Alloy 59 per cent, copper, 41 per cent. zinc. This is the normal structure of

alloys from 67 to 45 per cent, copper.

100

THE BRASSES.

cast. The appearance of a brass containing about 70 per
cent, copper is fairly uniform, but with higher copper it
varies considerably with the circumstances, no doubt
because the portion of the solution last solidified is richer
in zinc than that first solidified, and therefore is diffe-
rent in colour.

A. FIG. 52. B.

A. Alloy, copper 23 per cent., zinc 77 per cent.

B. Alloy, copper 20'5 per cent., zinc 79'5 per cent.

V, 100 X.

A. FlG. 53. B.

A. Alloy, copper 10'26 per cent., zinc 89'74 per cen*.

B. Alloy, copper 6'65 per cent., zinc 93'40 per cent.

V, 100 X.

As soon as the copper falls to about 67 per cent., the
exact point varying somewhat according to the conditions
of cooling, the whole character of the alloy changes, and
it is seen to consist of irregular masses of a light yellow
material (vertical illumination) embedded in a darker
ground mass, and as the quantity of zinc is reduced the

THE BRASSES.' 101

ground mass increases in quantity. The yellow material
seems to be the solution of zinc in copper which con-
stitutes the whole mass of the higher alloys, whilst the
ground mass is a eutectic probably made up of this solu-
tion and a solution much richer in zinc. As the composi-
tion of about 50 per cent, copper and 50 per cent, zinc is
reached the structure again begins to change. The struc-
ture becomes more uniform, consisting only of crystals
surrounded by a thin layer of the ground material. As
the quantity of zinc is increased the quantity of ground
mass increases, and other changes take place in the
appearance ; but these alloys, with such a large per-
centage of zinc, are of no commercial importance.

The constitution of these alloys is still somewhat
uncertain, and the statements above may need revision.
According to Mr. E. S. Shepherd :— *

High copper alloys crystallise in more or less definite
crystals which are a solid solution of zinc in copper (the
a solution). As the copper falls to about 67 per cent.,
the limit of saturation of solution a is reached, and the
brasses show two distinct constituents ; as the percentage
of copper falls, the a crystals disappear and at 52-2 per
cent, of copper the structure again becomes uniform,
only one constituent, the solid solution /3, being present.
As the percentage of copper is still further reduced, a
second constituent again makes its appearance, the solid
solution y, the crystals of which are white, and the alloy
is homogeneous y crystals from 40 to 31 per cent, copper.
Below 36 per cent, a second constituent makes its
appearance, the <$ solid solution, and then a third (the
e solution), so that three constituents are detected, at
20 to 13 per cent, the alloy is again homogeneous, and is
the e solution. Between 13 per cent, and 2-5 two con-
stituents are visible and below this there is a solid
solution of copper in zinc, the Y\ solution.

The colour of the fracture is in some cases different
from that of the filed surfaces when two constituents are
present because the fracture is often determined along the
crystals of one of the constituents, whilst the filed surface
gives the average colour.

* Journal of Physical Chemistry, Vol. 8, No. 6.

102 THE BRASSES.

Mr. Shepherd gives the following table : —

Composition
Per Cent.

Crystals
Present.

Colour of Filed
Surface.

Colour of Fracture.

of Copper.

100-63

a + /8

Red, changing 1o

Yellow

pale yellow

63-54

/3 + a

Reddish yellow

Yellowish red

54-51

Reddish yellow

Yellowish red

51-43
42-10

ft + y
y + j8

Reddish yellow
Yellowish red

Yellowish red
Silvery, with pinkish

tinge

40-30

Silvery

Silvery, very bril-

liant

30-90

7 + e + 5

Silvery grey to

Silvery grey, becom-

bluish grey

ing duller

20-13

Bluish grey

13-2-5

e + 17

Bluishgrey, becom-

Zinc

ing lighter

2-5-0

Zinc

It must be remembered also that the composition of
the solid solutions varies with the temperature, as
differences in stability are determined by the temperature,
and where there are two or more constituents time is
required for their separation, so that the structure may
differ according as the alloy is slowly or quickly cooled, or
annealed after cooling.

A reference to Prof. Roberts Austen's diagrams will
show that the structure described is much what would
be expected, there being four periods when two con-
stituents separate out, and for the rest you would expect
only one constituent. Mr. Shepherd says that one eutec-
tic solidifying between 800° and 900° has been over-
looked.

Mr. Shepherd gives the map, Fig. 54, which gives his
idea as to how the constituents of a brass are arranged at
different temperatures. The line A B C D E G represents
the commencement of solidification, the dotted and full
lines A 6? b1 cx C dL et fl represent the completion of
solidification, the part between these lines represents the
condition during solidification, and the part below shows
the condition of the solid. That some of the lines in
this portion are curved indicates changes taking place in
the solid alloy during cooling.

THE BRASSES.

103

Mr. Shepherd states that there are no " definite
compounds of copper and zinc," but that the six phases
are all solid solutions.

1100°
1000°
900°
800°
700°
€00°
500°

400°
1

S
\rf

+ JiquftK

^»^

i ^

7^-t^

\
\

\\qu\ft\

*\5+li

'UIOV

f/9

p+y

\fa€

\6-f- li^

£l\

A G

65 ^

€+y7\i7

00 90 80 70 60 50 40 30 20 10 /3 0

Per cent, copper.
FIG. 54.— MAP OF THE CONSTITUTION OF THE BRASSES (Shepherd).

It is evident that if Mr. Shepherd's views are correct
there should be five sets of brasses showing only one
constituent, viz. : —

(1) More than 67 per cent, of copper.

(2) From 51-53 „

(3) „ 30-40 „

(4) „ 13-19 „

(5) „ 0-2-5 „

and this seems to be borne out by microscopic examina-
tion of the alloys.

It will be noticed that the composition of the various
solutions depends to some extent on the rate of cooling.

Figs. 51 to 53 are those given by M. Guillet in illustra-
tion, and are taken from the " Metallographist," vol. 9.

Classification of Brasses. — For practical purposes brasses
may be divided into groups according to composition and
use.

104 THE BRASSES.

(1) Brasses containing 70 per cent, and upwards of
copper.

This series includes the most useful of the brasses, as
it includes those alloys of greatest tenacity and exten-
sibility (see curves, p. 94). For all ordinary purposes the
alloy of 70 per cent, copper and 30 per cent, zinc seems
to be the most suitable, and nothing is to be gained by
increasing the percentage of the more costly copper ;
but 70 per cent, copper may be taken as being about the
lowest limit that should be touched on account of the
possible separation of a eutectic when it is passed. All the
brasses of this series will work well cold. Commercial
names have been given to many of these alloys, but it
would be well if all these names were discarded and the
composition of the alloys always stated by the percentage
of the metals present.

Among the more important alloys of this class may
be mentioned :—

Tombac (Oreide, French gold) —

Copper, 90 per cent. Zinc, 10 per cent.
Pinchbeck varies very much, about —

Copper, 88 per cent. Zinc, 12 per cent.
Bed brass —

Copper, 80 per cent. Zinc, 20 per cent.
Chiefly used for ornamental work. When pickled in acid

it has a reddish colour.
Brazing metal —

Copper, 90 to 80 per cent. Zinc, 10 to 20 per cent.
The 80/20 alloy is known as quarter-metal and is largely

used.
Dutch metals —

Copper, 80 to 85 per cent. Zinc, 20 to 15 per cent.
This alloy can be hammered into very thin leaves, and is

used for Dutch metal gilding.
Standard English brass —

Copper, 70 per cent. Zinc, 30 per cent.

This is the alloy almost always specified for loco-
motive boiler and condenser tubes, and other purposes
where a high-quality brass is required ; it can be drawn
into tubes cold or rolled into sheets, and it resists
corrosion as well as the brasses richer in copper and
better than those which are poorer.

THE BRASSES. 105

(2) Brasses containing from 66 to 70 per cent, copper.
These brasses are similar to the first class, but extensi-
bility has begun to fall off, though the tensile strength has
not diminished. As a result they do not work so well cold.
They are just approaching the borderland, when a definite
eutectic separates, and they do not resist corrosion so
well as the members of group 1. They are, however,
still ductile enough to be drawn cold into tubes or rolled
into sheets.

Much ordinary brass has the composition of 67 per
cent, copper and 33 per cent, zinc, and is therefore
nearly a 2/1 alloy, which would give 66'67 and 33'33.
The Admiralty brass used for boiler tubes for steamships
has this composition.

The alloys used for rolling into sheets or drawing cold
will necessarily belong to either series 1 or 2.

(3) Brasses containing from 55 to 66 per cent, of
copper.

In these brasses the extensibility has fallen off very
seriously, and they cannot be worked cold. They are
still, however, malleable at a red heat, and can, therefore,
be used when they are to be worked hot. Alloys of this
series are often called yellow metal or Muntz metal,
though the latter name should be restricted to the alloy
containing 60 per cent, copper and 40 per cent, zinc,
which was patented by Mr. G. F. Muntz in 1832, and
which is largely used as a sheathing for wooden vessels.
The metal is attacked fairly readily by sea water, and the
poisonous zinc salts formed prevent the adhesion of the
living organisms which so soon foul the bottoms of ships
at sea. The same alloy is used for the manufacture of
bolts and other ship's fittings. The alloy is rolled at a
red heat. Hard-brass solder has about this composition.

(4) Brasses containing from about 48 to 56 per cent,
of copper.

In these alloys the extensibility is so far reduced that
they cannot be worked either hot or cold ; but as they cast
well, they are used for making brass castings of all sorts
where great strength is not required. As the metal has
a fine yellow colour, it is known as yellow brass.
For ordinary castings a mixture of about 50 per
cent, copper and 50 per cent, zinc is frequently used ;
but as a rule, since variations in the percentage are of

106 THE BRASSES.

little moment, little trouble is taken to ensure a definite
composition.

When an alloy of about this composition is heated to
just visible redness, it becomes exceedingly brittle and
can be readily powdered in an iron mortar with a heavy
pestle. It then constitutes the yellow solder which is
largely used for brazing purposes. In brazing or solder-
ing it is, of course, essential that the brazing material
should have a lower melting point than the materials
being united.

(5) Brasses containing from 34 to 45 per cent, of
copper.

As the copper falls below 50 per cent., the extensi-
bility and the malleability almost disappear, the tensile
strength rapidly falls off, and the colour of the alloy,
which had previously been yellow, becomes whiter and
whiter till, when the amount of copper falls to 40 per
cent., the colour becomes silver-white, and the metal
becomes very brittle. Tins alloy can be powdered in a
mortar. It is known as white brass, and is used as a
solder for brazing brass under the name of white solder.
The composition of white solder varies from about 36
per cent, to 40 per cent, copper.

(6) Brasses below 34 per cent, of copper.

These alloys are of little commercial importance,
as they are too brittle and weak to be of any use in the
arts. Alloys of zinc with a small quantity of copper, up to
about 10 per cent., are used for casting statuettes and other
similar articles. The addition of the copper increases
the strength of the alloy and destroys the very largely
crystalline structure of pure zinc. Such statuettes are
always bronzed by electro-deposition or otherwise, and
are sold under the name of French bronze. They are of
course much cheaper than true bronze, not only because
the metals used in the composition of the alloy are much
cheaper, but because the castings can be made more
cheaply and at a lower temperature.

INFLUENCE OF FOREIGN CONSTITUENTS IN THE

BEASSES.

As already remarked, a brass made from pure commercial
materials should contain very small quantities of impurities.
It sometimes happens, however, that the materials are not

THE BRASSES.

107

pure. The copper may contain arsenic, antimony, and
other elements, and the zinc may contain iron and lead,
and in other cases small quantities of foreign metals may
be added to modify the character of the brass. It
therefore becomes important to study the influence of
various impurities on the alloys.

^30

_o
H

o\o I/

\&.

**.

25 30 35 40 45

Tensile strength in tons per square inch.

- A luminiura Bronze (cast).
Aluminium Brass (cast).

FIG. 55.— TENSILE STRENGTH OF ALUMINIUM BRASS AND ALUMINIUM BRONZK.

The data for many of the elements is very scanty, so
that in many cases a full account cannot be given. The
added elements will be considered alphabetically.

Aluminium. — Aluminium is never likely to be present as
an accidental impurity, but is sometimes added in
considerable quantity up to about 3 per cent., and such
an alloy is known as aluminium brass. The copper is

108

THE BRASSES.

usually from 60 to 71 per cent., so that the brass may be
considered as an ordinary brass in which the zinc is
partly replaced by aluminium. Such a brass has &
fine yellow colour, has a high tensile strength, and
elongation as is shown by the diagram Fig. 55. Figs.

60
50
40

—

/uuu
6000(
5000(
400u<
30001
2000(
1000(

.X''

^"* -^.^^

~7_

\
\

0 0-3 0-8 2-9 4-7 A1.
40 39-7 39 2 ^rJ 37 '1 35-3 Zn.
60 60 60 •• 60 60 On.
— Extension.
Ultimate strength.

FIG. 56.— EFFECT OF ALUMINIUM ON 60/40 BRASS.

56 and 57 show the influence of aluminium in small
quantities on brass containing 60 and 70 per cent, of
copper. It casts well, and even when containing only 60 per
cent, of copper can be forged readily at a dull red heat if
it contains about 3 per cent, of aluminium. The forging

§50
^40
I 30

la,
Sio

_^-

70000 g
60000 &

S_i

50000 ^
40000 J
30000 §
20000 |
10000 I

Xxx

\
\

^-2

—

\
\

0-4 0-9 3-1 5-2 Al.
29-6 29-1 26-9 24 -8 Zn.
70 70 70 70 Cu.
Elongation.
Ultimate strength.

FIG. 57.— EFFECT OF ALUMINIUM ON 70/30 BRASS.

THE BRASSES. 109

temperature is lower the less aluminium is present,
so that with J per cent, it can only be forged cold. The
alloy is easily made, and casts well, but the castings
should always be allowed to cool slowly. Suddenly-
cooled aluminium brass can be recognised by the deep
gold colour and glittering lustre of the fracture.

Antimony. — Whilst aluminium is never likely to be
present as an accidental constituent, but is always added,
with antimony the reverse is the case ; it is never added,
but is always, when present, an accidental constituent,

FIG. 58.— FRACTURE OF BRASS CONTAINING NO ANTIMONY.

and is derived from the copper. It is probably, with the
exception of bismuth, the most injurious constituent that
can be present in brass, the smallest quantity being
objectionable. Antimony hardens the alloy and destroys
its ductility, making it cold short, so that it will crack on
rolling. For cold drawing brass tubes, or other similar
purposes, '01 per cent, of antimony renders the metal
quite useless, and even '001, an amount which is
sometimes present even in electrotype copper, is said to
be objectionable. For rolling brass high in copper the
maximum allowable is about O'Ol, and for safety less than
O005 per cent, should be specified to be present in the
copper to be used.

110

THE BRASSES.

The presence of antimony produces a remarkable-
effect on the fracture of the alloy, which is best seen
with an alloy containing about 60 per cent, of copper, and
this, as already remarked, is sometimes used as a test for
the quality of the copper.

The illustrations, Figs. 58, 59, 60, of the fracture of
brasses containing antimony, from a paper by Mr. E. S.
Sperry, read before the American Institution of Mining
Engineers,* will illustrate the effect of antimony on the
structure of the' metal.

The table below gives some of Mr. Sperry's results,
the brass being approximately 60 and 40.

60/40
Brass.

60/40
•01 Sb.

60/40
•02 Sb.

60/40
•05 Sb.

Tensile strength

Ibs. per square inch. .

58,700

61,290

58,800

56,600

Elongation in Sin.

per cent.

33-2

34-5

Reduction of area

per cent.

43-9

43-7 •

With "01 per cent, the brass seemed to behave, on
rolling, much like ordinary brass, but with 0'2 per cent,
the metal rolled badly ; the sheet could not be bent over to
45 degrees without cracking, but could be forged hot.

With even a very small percentage of antimony the
presence of a dark-coloured constituent can be detected
under the microscope distributed through the mass,
and in larger quantities it forms a network through the
mass.

Arsenic.— Arsenic is alwrays present, at least in traces, in
copper, even electrotype, and it is usually regarded a&
being a very objectionable constituent in brass. The
maximum quantity allowable in brass which is to be used
for cold drawing is about '05 per cent., and this is the
amount which is usually specified for B.S. copper which
is to be used for such purposes. The effect of arsenic is
to harden the metal and render it brittle, but to a less
degree than antimony.

* Transactions. Vol. XXVIII., page 176.

THE BRASSES.

Ill

FJG. 59.— FRACTURE OF BRASS CONTAINING '05 PER CENT. OF ANTIMONY.

FIG. 60. — FRACTURE OF BRASS CONTATNING '1 PER CENT. OF ANTIMONY.

Mr. Spsrry "oand that a 60/40 brass, with 0-50 of
arsenic, would not roll at all, but ." cracked to pieces "
in " breaking down " ; but it cast well, and " resembled
phosphor-bronze in its limpid nature." With 0-25 of

112 THE BRASSES.

arsenic the fracture showed traces of crystallisation.
The bar was rolled from a thickness of lin. to -049in.
with five annealings, and cracked badly during the rolling.
With 0-10 per cent, of arsenic there were only slight
traces of crystallisation visible on the fracture, but the
bar cracked on "breaking down," and even with 0-05
per cent, the cracking was quite marked. With 0-02
per cent, of arsenic the qualities of the brass were better
than most pure copper.

Mr. Sperry sums up his results :—

(1) Arsenic when present in brass to the extent of
over 0-02 per cent, is injurious and causes it to crack
on rolling.

(2) Arsenic produces great fluidity in melted brass.

(3) Brass containing arsenic makes a cleaner casting
than when it is not present.

(4) When present in an amount not over 0-02 per
cent., arsenic imparts ductility to brass, probably by a
reduction of the oxide of copper formed during melting.

Bismuth. — This is probably the most objectionable con-
stituent in brass, but fortunately it is very rarely present
n any quantity. Mr. Sperry has investigated this sub-
ject, and the results of his experiments are published in
the " Transactions " of the American Institution of Mining
Engineers, Vol. XXVIII. He comes to the conclusion that
as far as inducing cold shortness bismuth is less injurious
than antimony, but that it produces very marked red
shortness. An alloy containing 0*05 per cent, would
forge on a thin edge, but if bent over cracked at the
bend, whilst an alloy with 0'25 per cent, would not forge
at any temperature. He also conies to the conclusion
that bismuth is the cause of fire cracks, and that even
when these are not actually visible they are present
latent, not perhaps as actual cracks, but more likely as
lines of inferior cohesion in an apparently homogeneous
mass. Boiling develops them, and to all appearance they
then partake of every character of true fire cracks. Mr.
Sperry gives O'Ol per cent, as the maximum that should
be allowed in brass for rolling, and even less than this
will be injurious to the copper.

THE BRASSES.

113

Brass containing bismuth breaks with a highly crystal-
line fracture, as is seen by the photograph of the brass
containing 0*09 per cent, as from Mr. Sperry's paper.
It will be seen that the crystalline structure is much less
marked than in the case of antimony.

Under the microscope the separation of a substance
containing bismuth can be seen.

Iron. — Iron is only likely to be present in very small
quantities as an accidental impurity. The zinc may
contain a small quantity, and the remainder can only

FIG. 61.— FRACTURE OF BRASS CONTAINING '5 PER CENT. OF BISMUTH.

be derived from the iron tools used to stir the alloy.
It is considered as being objectionable, so that zinc which
is to be used for the manufacture of high-quality brass
is usually specified not to contain more than 0'05 per cent,
of iron. It is sometimes added in considerable quantity, and
produces somewhat remarkable alloys, which have been
known for a long time as sterro-metal and Aich metal,
and in more recent form as delta metal. Prof.
Roberts Austen found that a brass having a tensile
strength of about 20'7 tons at 20° C. had this increased
to 25'6 tons by the addition of 1'5 per cent, of iron, and
though both the alloys lost strength very considerably as
the temperature was increased, the loss was less with
J

114 THE BRASSES.

the alloy containing the iron than with that which was
free from iron, so that the difference became greater as
the temperature rose.

The reason for this difference is explained by Sir W.
Roberts Austen as being due to the fact that iron
raises the solidifying point of the alloy, and that it also
prevents the formation of a eutectic which otherwise
forms at a comparatively low temperature. Sterro-metal
consists of copper, 60 per cent. ; zinc, 38 to 38' 5 ; and
iron, 1*5 to 2. It is therefore as 60/30 of brass, in which
a small part of the zinc is replaced by iron. Aich metal
is almost if not quite the same. Delta metal contains
varying quantities of iron according to the purpose for
which it is to be used, and other metals, such as iron
and manganese, may also be present, a little phosphorus
being added to deoxidise the copper. The alloy has a
brass-yellow colour, is strong, having a tensile strength
up to from 25 to 35 tons, with an elongation of from 11 to
39 per cent, on an 8in. test-piece. It is therefore very
tough, and can be rolled and worked quite satisfactorily.
It is also said to resist corrosion much better than
ordinary brass.

Lead. — Lead is present in brass in small quantities
derived from the spelter used in its manufacture, and is
sometimes intentionally added. The amount of lead
which brass will take up is not large, and it tends to
separate on cooling. With 5 per cent, or over the
tendency of the lead to separate is marked, and it may
squeeze out during working, and even with much smaller
quantities it can be detected as a separate constituent
under the microscope. The presence of lead reduces
both the tensile strength and the extension of the
metal, the effect being very much more marked with
cast brass than with that which had been drawn and
annealed, probably on account of the segregation of the
lead in the latter case. The presence of lead makes the
metal softer for working, and for that reason it is some-
times added to brass, and its effects can be well seen by
the difference in the turnings that are obtained, the
turnings from the brass itself coming away in long curled
pieces (Fig. 62), while those from the leaded brass come
away in short chips (Fig. 63). The amount of lead added
usually varies between 2 and 3 per cent. Mr. Sperry, in

THE BRASSES.

115

FIG. 02. — BRASS CHIPS FREE FROM LEAD.

FIG. 63.— BRASS CHIPS CONTAINING 2 PER CENT. OF LEAD.

a paper read before the American Institute of Mining
Engineers, in 1897, gives the analysis of a considerable
number of brasses for various purposes containing lead,

116

THE BRASSES.

60000
55000

| 50000

£
g 45000

S 40000

| 35000

o'
^ 30000

•2

§> 25000

I
* 20000

| 15000
10000
5000

40 |

30 ^
a"

20 '•§

10 |
0

**•

"*"**-

— ^

--,«

""•-

— —

*-»«.^

•

• —

—

-1

-c

~"~~ — -

— -~

01 234567 89 10

Lead, per cent.
FIG. 64.— ALLOYS OF COPPER, ZINC, AND LEAD, CONTAINING 60 PER CENT. COPPER.

a a Tensile strength of annealed sheet.
bb „ „ cast metal.

c c Elongation in lin. of cast metal.
d d Sin.

2 50
£-40

§ 30

'§20

N O *"d i-' to w rfi. en os <i

Ultimate strength,
pounds per square inch.

-^^ — ==

— •» ^.

0 05 1

ro 69 6

50 30 3

2 3
8 67
0 30

tension,
bimate strength.

-.- Ul

FIG. 65.— EFFECT OF LEAD ON 70/30 BRASS.

THE BRASSES.

117

and discusses the influence of lead on brasses very fully.
The diagram (Fig. 64) is from Mr. Sperry's paper.

Manganese. — Alloys are now largely made under the
name of manganese bronze which are really nothing
more than brasses, to which a small quantity of manganese
has been added. The composition of the commercial
manganese bronzes (or rather brasses) is very variable,

Extension, per cent.

i— > to 00 **. Or OS -.
0 0 0 0 0 0 C

•===5.

•*>«. ~

=~^

===== <=

"*^^

70000 ^
60000 g
50000 ^"|
40000 "&2

§ 03

30000 £ g,

50 a]

20000 | £

10000 .g P

0123
40 39 38 37
60 60 60 60

6 7 Mn.

35 33 Zn.

60 60 Cu.

Extension.
Ultimate strength.

FIG. 66.— INFLUENCE OF MANGANESE ON 60/40 BRASS.

Extension, percentage.

5 g g S § § 2

QNlg § 8 g § § § 2
sag" o 8 o § 8 o c

Ultimate strength, pounds
per square inch.

\
\

-V

'\^'

0123 5

0 29 28 27 25 2,
0 70 70 70 70 7(

J 1(
J 21

) 7

nsion.

na.f.p ntrftnorfh

.... _... . TTlfir

FlG. 67-lNFLUENCE OF MANGANESE ON 70/30 BRASS.

and sometimes they contain no manganese, the man-
ganese having been all oxidised out, owing to its
ready oxidisability. It does not, of course, follow
that the manganese has been of no use because
it will have acted as a deoxidising agent. Again,
the manganese is usually added in the form of ferro
manganese, so that even if the manganese is completely

118 THE BRASSES.

oxidised out, the residual iron may have an influence on
the properties of the resulting metal. The presence of
manganese seems to harden the alloy, increase its strength,
and at the same time diminish very considerably the
extension before fracture. Figs. 66 'and 67 show the
influence of small quantities of manganese on brass.

Now that comparatively pure manganese can be
obtained, it can be used as a deoxidiser in brass and
bronze casting, and it has the advantage that an excess
is not injurious, as in the case of phosphorus, but rather
the reverse. The melting point of pure manganese is
so high that the manganese should always be added in
the form of a copper-manganese alloy.

Manganese bronze contracts very much on solidifica-
tion, so that large gates and rising heads are essential.

Nickel. —When present in very small quantity nickel
has no important influence on the properties of the alloy.
Alloys containing a considerable proportion of nickel will
be considered under the head of German silver.

Oxygen. — Copper dissolves oxygen very readily and
becomes very brittle or dry. When such copper is used
in the manufacture of brass it has been thought that it
would produce an alloy of inferior quality. This, how-
ever, does not seem to be the case. The oxide of copper
is apparently decomposed by the zinc. Whether brass
contains any oxygen in solution is uncertain, but it seems
unlikely.

Phosphorus. — Phosphorus is largely used for deoxidising
copper, and has occasionally been used for a similar
purpose in the manufacture of brass, whether with any
real advantage or not the writer is unable to say. He
has never come across a brass containing phosphorus.

Tin. — Tin is sometimes added to brass for various
purposes. In the manufacture of condenser tubes and
other articles 1 or 2 per cent, of tin is sometimes added.
Naval brass, largely used for condenser tubes, contains —

Copper 70

Zinc 29

Tin 1

100

THE BRASSES.

119

Whether the addition of this small quantity of tin is
any real advantage is somewhat uncertain, but the
author thinks not. The influence of tin is to harden the
alloy, increase its strength, and diminish the extension
before fracture. With 1 per cent, of tin the influence is
not very great, but as the percentage is increased the

a 60

!*>

^40

•I30

| 20

W 10

(
41
61

60000 j-'

50000 c3
2
40000 ^

30000 -g
20000 J,
10000 ^

^^ __

•

—

^v^

"^x

J -7 1-3 2-2 Sn.
) 39-3 38-7 37'8Zn.
) 60 60 60 Cu.

Elongation.

Ultimate strength.

FIG. 68.*

action becomes very marked. It is sometimes said that
the brass containing tin is less liable to corrosion than
brass ; this, however, is not the case. The tin seems to
dissolve in the alloy, and does not alter the micro-
structure. Fig. 68 shows the influence of small quantities
of tin on brass.

* Figs. 64 to 68 are from an article by L. S. Augtin in the " Mineral Industry,"
Vol. XIV., p. 151.

CHAPTER VIII.
THE COPPER-TIN SERIES.

NEXT in importance to the brasses may be placed the
bronzes, or copper- tin alloys. For many reasons this
series, in addition to its great practical importance, is of
very great theoretical interest, and it has therefore been
very fully investigated from a purely scientific point of
view, and whilst much has been done there are still some
obscurities to be cleared.

The influence of tin on the properties of copper is
very marked, and though alloys of the metals in all pro-
portions have been obtained it is only two series that
are of practical importance, those properly called bronzes,
containing 80 per cent, or more of copper, and the
speculum metals, containing about 50 per cent, of copper,
and these are used for quite different purposes.

Bronze has been known and used from times before
the commencement of the historic period. In prehistoric
times it was used for weapons and tools, and is generally
supposed to have been in use for such purposes before
the discovery of iron. In early-historic times it was
used for the manufacture of coins, medals, and similar
articles, the composition of the alloy used being almost
identical with that used to-day.

Range of Composition of the Bronzes. — The range of
composition is much less in the case of the bronzes than of
the brasses, as all those of any importance for engineering
purposes contain 80 per cent, of copper or over, the in-
fluence of tin in modifying the character of "the alloys being
much more marked than that of zinc. An immense
amount of information has been collected as to the proper-
ties of bronze and the influence of its composition on its
qualities.

The United States Board Committee on Alloys pub-
lished an elaborate report in 1878, giving the composition
and properties of a large number of bronzes used for
various purposes, and this table, in whole or in part, has

THE COPPER-TIN SERIES. 121

been reproduced in most books on alloys. The Alloys
Research Committee of the Institution of Mechanical
Engineers took up the subject, and published very valu-
able material in its third Report in 1895, and since then
the work has been carried on by many workers, and results
have been obtained which may be of great practical
importance.

Freezing - point Curve. — At the outset it may be ad-
visable to describe the phenomena which are met with as
an alloy of copper and tin is cooled, since the position of
an alloy on the freezing-point curve seems largely to
determine its properties.

As tin is added to copper (which melts at 1,090° C.)
the melting point slowly falls, as usual, and the alloy
solidifies as a whole, i.e., is a solid solution, till the per-
centage of tin reaches about 5. As soon as the percentage
of tin becomes larger a double freezing point occurs,
exactly as in the case of brass, one portion solidifying at
a temperature constantly falling as the percentage of tin
increases till with about 25 per cent, of tin it reaches about
790°, whilst the other portion, the mother liquid, solidifies
uniformly at about 790°, so far the alloy behaving-
exactly likean ordinary eutec tic alloy. When thepercentage
of tin reaches about 10 per cent., however, another phe-
nomenon makes its appearance. The solidification begins
at about 1,000° C., the second freezing takes place at
about 790° C. as before ; but there is still some material
left unsolidified which freezes at 500°, so that with the
alloys containing from 10 to 20 per cent, of tin there are
three distinct freezing points ; that is, the alloy solidifies
in three separate portions. Fig. 69 shows the freezing-
point curve as determined for the Alloys Research Com-
mittee. It will be seen that it is very complex, that with
alloys containing a little over 60 per cent, of copper there
are no less than four distinct freezing points, that with
40 per cent, of copper these have fallen to three which
persist till there is less than 10 per cent, of copper, and
that up to about 97 per cent, of tin there are still two
freezing points. Reference will be made to this freezing-
point curve later, and explanation given of some of
its peculiarities ; but enough has been said to indicate its

122

THE COPPER-TIN SERIES.

5£ *

TEMPER/TTURE FAHRENHEIT. . g . .

Vi Vt M Cf ^1 Cl ^* - . O* . r _ *

d S5 . i* '* S "5^

= S £ C^ ; " t-5-

§*

00
»n>ra

oc*

CEKTICPW)E.

general character. Other observers have worked on the
subject, and have on the whole confirmed the results
obtained by the Alloys Research Committee.

The properties of the alloys will now be studied in
the light of this freezing-point curve.

THE COPPER-TIN SERIES. 123

Colour. — Tin very rapidly destroys the colour of copper,
and imparts to the alloy a bronze-yellow colour. The
following table of the influence of an increasing percen-
tage of tin on the colour of the alloy is taken from the
report of the American Committees : —

Copper. Tin.

96-27 . . 3-73 reddish yellow.

90-00 .. 10-00 greyish yellow.

80-00 .. 20-00 yellowish red.

75-00 . . 25-00 reddish white.

70-00 . . 30-00 white.

60-00 .. 40-00 light grey.
All below this being shades of greyish white.

It will be seen therefore that generally all alloys below
75-00 per cent, of copper are white or grey in colour.

It will be seen from the cooling curve that in the
alloys containing over 80 per cent, of copper there will
be a considerable portion of the constituent which
solidifies at the highest temperature, and this is appa-
rently a solid solution of tin in copper, similar to the
solution of zinc in copper described under the brasses,
and which is yellow in colour. The eutectic, which
solidifies at about 760° and contains only about 70 per
cent, of copper, is white, so that it gradually dilutes
the colour of the alloy, and when it becomes the
predominating constituent the colour becomes white.
It will be seen that for alloys containing a smaller
percentage of copper the constituents will vary, and
with this variation in composition there is a small
change in the colour of the alloy.

When the alloy contains about 66- 6 per cent, of copper,
it has a fine silver-white colour, and it is then called
speculum metal, because it was at one time used for
making the mirrors or specula for reflecting tele-
scopes.

Specific Gravity. — The specific gravity of the bronzes is
greater than that of a mean of its constituents, the metals
therefore contracting on mixing.

124 THE COPPER-TIN SERIES.

The following table from Thurs ton* gives the specific
gravity of the alloys : —

S.G.

Copper.

100

Tin.

S. G. Actual.

8-79

Calculated.

8-74

DiffereiN

0-05

100

8-78

8-71

0-07

100

8-76

8-68

0-08

100

8-76

8-66

0-10

100

8-80

8-63

0-17

100

8-81

8-61

0-20

100

8-87

8-60

0-27

100

8-83

8-43

0-40

100

8-79

8-05

0-74

Tenacity. — Very many experiments have been made
to determine the tensile strength and other mechanical
properties of the bronzes.

The following figures are taken from the American
report before quoted :—

Tenacity, 11}S.
Copper. Tin. per Square Itach.

96-27 .. 3-73 .. 32,000

92-80 .. 7-20 .. 28,560

90-91 .. 9-09 .. 32,093

89-29 .. 10-71 .. 37,688

85-71 .. 14-29 .. 44,071

84-29 .. 15-71 .. 36,004

81-10 .. 18-90 .. 39,648

80-00 .. 20-00 .. 32,980

76-29 ... ' 23-71 .. 21,728

72-80 .. 27-20 .. 10,976

70-00 .. 30-00 .. 5,585

68-25 .. 31-75 .. 1,620

61-71 .. 38-29 .. 638

50-00 .. 50-00 .. 725

These are only examples taken from the table. An
examination of the whole table will show that the figures
are extremely discrepant, and that no definite inference
could be drawn from them except that of a general rise
of tenacity up to about 80 per cent, of copper, then a fall
which soon becomes very rapid, and the tenacity almost
disappears, to increase somewhat as the tin end of the

*" Materials of Engineering," part 3, p. 141

THE COPPER-TIN SERIES.

125

series is reached. Fig. 70 gives the curve for tensile strength
and elongation as given by the Alloys Research Committee
in its third report. A glance at that will show that tensile
strength rises, but not uniformly, till it reaches a maxi-
mum of about 36,000lbs., with rather more than 80 per
cent, of copper, and then it falls off very sharply.

£ H Lbs.

• o **.

-0 36.

Q
0 3t O

018 "1

a
o z+

- 0 20 r
016.

- 013,

008

• o :• i

0, Tilt
? PERCEHT.

3 £

P. 30,000.-
« n
q c/j 2t-,000

£J S 1 8. o o o
M O 13,000.

pc 6.000,

0
PEP CtNT. v
COPPER 1C

TENSILE

/'\...

/ \

\
\

<i.

'':j _- — -

-4--

> i

o g

STRE1

0 20 3Q *Q :

o. so. *m eo i
iiTvr

JQ ea

(0. *0. J

uLOKGATIO

70. so oa
ro 20 10

{

iijin, ,

)

FIG. 70.— TENSILE STRENGTH AND ELONGATION OF COPPER-TIN ALLOYS.

Ductility. — The ductility which marks the toughness
of the metal, and which is in many cases quite as important
as actual tensile strength, is seen to reach a maximum
when there is about 5 per cent, of copper, then to fall
off and almost disappear, and then to reappear when
about 80 per cent, of tin is present, and, as might be
expected from the known softness and ductility of tin,
to increase considerably as the tin approaches purity. As
ductility is the reverse of brittleness it will be judged
that those alloys which are devoid of ductility will be
very brittle, and this is the case.

The passage from the strong to the weak condition
is attended with a complete change in the character of
the fracture. The strong alloys break with a more or
less granular fracture, but the brittle ones are almost
glassy, and the fracture is often distinctly conchoidal.

126

THE COPPER-TIN SERIES.

Electric Conductivity. — Fig. 71 gives the curves for
electric and heat conductivity as determined by the
Alloys Research Committee. It will be seen that as the
percentage of tin increases the conductivity falls, becom-
ing very small, then with about 9 per cent, of tin there is
a sudden break and a continued fall at a slower rate till
about 32 per cent, of tin is present, then a slight rise, and
then a nearly-constant conductivity as the percentage
of copper falls, until pure tin is reached.

Heat Conductivity. — The heat conductivity, as will be
seen by reference to Fig. 71, varies very irregularly.

-L,

PERCENT
COPPER

0
100.

10
90

4-0
60

60
•H)

100. TIN
o PER CZJfl

.ELECTRICAL C OKDl/CT I Y ITY. ; HEKT C OMD UCT IV IT X

FIG. 71.— CONDUCTIVITY OF COPPER-TIN ALLOYS.

Microscopic Structure. — As might be expected, the
microscope throws considerable light on the structure of
the alloys of the copper-tin series ; but, owing to their
complexity and the great influence of the rate of cooling
on the structure, it is not possible to give a clear outline
of the results that have been obtained.

It may be simplest to take the classification and de-
scription of Dr. Campbell, who, in his paper read before the
Society of Mechanical Engineers, as an appendix to the
Reports of the Alloys Research Committee, has given a
full account of his researches into the microstructure of
the copper-tin alloys. He begins with tin, adding small
percentages of copper. He classifies the alloys into groups.

THE COPPER-TIN SERIES.

127

(1) 0 to 1 per cent, copper. The mass consists of tin
with a second constituent surrounding the grains of tin.

(2) 1 to 8 per cent, of copper. Crystals which are
hollow rhombic crystals, which form groups, and which
appear in section as three or six rayed stars. The more
slow the cooling the larger and less numerous are the
stars.

(3) 9 to 40 per cent, copper. A new constituent
crystallises out, and as the copper reaches 40 per cent.
" plate-like crystals are grouped in parallel bunches,"
and in the eutectic between them small bright, hollow
crystals are seen.

(4) 41 to 61-7 per cent, copper. Dr. Campbell says
" the difference between the alloy containing 40 per cent.

FIG. it. — BRONZE. COPPER, 2 PER CENT.; TI.N, 98 PER

511.

This shows its constituents, but not the star-like crystals which the author has been
unable to obtain with alloys of this composition.

and that containing 41 per cent, of copper is very marked.
The crystals in the latter are small and lath-shaped, and
arranged more or less in groups, and separated from one
another by eutectic." With each addition of copper
the groups of crystals become more and more
compact, and the amount of eutectic diminishes
till at 50 per cent, copper it disappears altogether.

128

THE COPPER-TIN SERIES.

The bright constitutent of the crystals grows
smaller and smaller ; at 50 per cent, it takes the
place of the eutectic and forms the ground mass in which
the constituent containing the higher percentage of
copper has solidified. "When 61-7 per cent, of copper is
reached the bright constituent disappears, and we have
a homogeneous mass, probably the definite compound
Sn Cu3."

(5) 61-7 to 68-2 per cent, copper. Each addition of
copper to Sn Cus brings out more and more of the bright

FIG. 73.— COPPER, 11 ; TIN, 89 V x 40.

constituent Sn Cu4. " The alloys set as a whole at
the first break, and tend to rearrange themselves sub-
sequently in the solid."

(6) 68-2 to 74-5 per cent, copper. "A second
eutectic makes its appearance, enveloping the grains of
Sn Cu4. As the copper increases the grains split up
into veins and dendrites which attain their full develop-
ment in the neighbourhood of 72 per cent. As the total
copper increases the eutectic increases also, the veins of
Sn Cu4 gradually disappear, and finally the dendrites go,
leaving the mass entirely made up of the eutectic, about
75 per cent, copper."

THE COPPER-TIN SERIES.

129

(7) 75 to 100 per cent, copper. " With 76 percent, of
copper present two new constituents make their appearance,
and the alloy assumes a yellow tint. It loses its brittleness.
In section are found yellow grains surrounded by a bright
white border set in the second eutec tic, in which small bright
white grains also occur." As the total copper is in-
creased, the yellow grains increase, forming dendrites
and skeleton crystals ; the white borders and grains merge
together, and the eutectic decreases till at about 90 per
cent, it disappears. The yellow grains become darker

FIG. 74.— COPPER, 21 ; TIN, 79 V x UO.

and darker, containing less tin in solid solution, till they
become copper colour. The light borders disappear,
leaving only copper dendrites at 95 per cent.

It will thus be seen that the structure of these alloys
is excessively complex, and that, as a rule, they are made
up of copper containing tin in solid solution, probably the
definite compounds Sn Cu4 or Sn Cu3 and Sn Co> with per-
haps other compounds containing less copper, and eutectics
made up a mixture of the two of these. The ordinary useful

130

THE COPPER-TIN SERIES.

FIG. 75.— BRONZE. COPPER, 38; TIN, 62V x 20.

FIG. 76.— BRONZE. COPPER, 65 ; TIN, 35 V x 140.

bronzes being composed mainly of copper containing tin in
solution, and the eutectic of the copper-tin solution and
the compound Sn Cu4. The metal becoming brittle and
ceasing to be of value as soon as the eutectic predominates.

THE COPPER-TIN SERIES.

131

The four known compounds of copper and tin which
may be present in alloys are :—

Per cent. Copper.

68-1

Sn Cu4
Sn Cu,

Per cent. Tin.

31-9

;. 61-6 .. 38-6

Sn Cu, .. .. ! 51-7 .. 48-3
Sn Cu" .. .. 34-8 .. 65-2

It will be noticed that the two first-named are the only
two which seem to exist as definite compounds in the
copper-tin series of alloys.

From the complex character of the alloys it is to be
expected that great changes would be produced by the rate
of solidification ; and this is so, the structure of the alloy

FIG. 77.— BBONZE. COPPER, 87; TIN, 13 V x 20.

suddenly solidified being in some cases quite different from
that which has been allowed to solidify slowly. The reason
for this is quite obvious. All the separations that take
place during the solidification of an alloy require time,
longer or shorter according to the nature of the change ;
and if the solidification be very rapid, the constituents
may not have time to separate, but may remain so
mixed as to be indistinguishable. This is especially the

132

THE COPPER-TIN SERIES.

case where the constituents usually occur in more or less
distinct crystals. On the other hand, very slow solidifi-
cation allows of more complete separation.

Messrs. Shepherd and Blough have recently investi-
gated the copper-tin series very completely in certain
directions, and as a result they have somewhat modified
the results previously obtained. They have replotted
the cooling curve, and mapped out the constituents of
these alloys. They find four solid solutions of copper and
tin, a /3 y 8 and e solutions, Cu3 Sn and the definite
compound pure tin.

The map is shown in Fig. 7?A. The line ABODE
represents the commencement of solidification, the line

Cu.

I. = ft

II. = a + liquid

III. = a + j8

IV. = a + 5

V. = j8 + liquid
VI. = B

VII. = p + 7
VIII. = 7
IX. - 7 + Cu., Sn

X. = 7 + 5
XL .= a + 5
XII. = 5

FIG. 77 A.

XQI. = 5 + Cu3 Sn
XIV. = 7 + liquid

XV. = Cu,Sn + liquid
XVI. = Cu', Sn + c
XVII. » e
XVIII. = e + liquid

A&! the completion of solidification, and the horizontal
lines the eutectic solidification. It will be seen at once
from the position of the lines that complex changes
take place during solidification, and during cooling
after solidification. The region I. is that in which solid
solution a, containing over 92 per cent, of copper, exists
alone. In the region II. a is solidifying and leaving a

THE COPPER-TIN SERIES. 133

liquid, and in the region III. there is a mixture of the two
solid solutions « and /3, but at 486° the /3 solution breaks
up, yielding another form <5, so that the region III. con-
sists of the two solid solutions a and S, the latter con-
taining from 68 to 75 per cent, of copper. The field V.
is a small one, containing 3 crystals in contact with still
liquid matters, which solidifying gives in region VI.
/3 solution. Region XIV. contains y solution in contact
with liquid matter which solidifying gives in region VIII.
pure y solution. Its shape is very peculiar. Region
VII. contains both 8 and y solutions, the crystals of the
former being yellow, those of the latter white. Region
X. consists wholly of white crystals, but these are seen
to consist of two constituents. In region XI. below
the lines d:i d4, the structure changes, and there is a mix-
ture of brilliant yellow crystals in a white matrix. " The
phase o, which is formed entirely through transformation
in the solid, was for a long time considered to be the
compound Cu4 Sn." The components of this phase
vary much with changing temperature. Region XIII.
consists of a mixture of 8 and Cu3 Sn, resulting from the
breaking down of the y crystals along the line dz d^
Region XV. is Cu3 Sn, in presence of liquid matter
passing over as it cools into XVI., Cu3 Sn + t, XVII. e.,
and XVIII. e., with still liquid matter, after which pure
tin or a solid solution, which is mostly tin, separates.

" In addition to the changes in the solid which have
been recorded for the copper-rich alloys, we found two
other heat changes in the alloys containing from 41 to
61-5 per cent, of copper. The first of these changes is at
218°, and occurs at the same temperature in all the above-
mentioned concentrations. The second change is like-
wise one at a constant temperature, and is found at
182°."*

Heat Treatment of Copper-Tin Alloys. — Not only are there
the changes due to sudden solidification, but a complex
body like bronze may undergo changes after solidifica-
tion, which may be altered very considerably by rapid or
slow cooling, and which are in some respects similar to those

* Shepherd and Blough. Journal of Physical Chemistry. Vol. 10 (1906),

p. 651.

134

THE COPPER-TIN SERIES.

which steel undergoes, and therefore it might be expected
that heat treatment would considerably modify its
properties. Messrs. Heycock and Neville have shown that
this is the case as far as the microscopic structure is
concerned The sudden cooling of the alloys under
certain conditions almost obliterating the microscopic
structure, much in the same way as does the quenching
of steel, and they found, moreover, that the structure
could be restored by heating the chilled specimen to
below its melting point and allowing it to cool slowly,
so that changes in the structure of bronzes may take
place not only during but also after solidification. Mr.
Campbell has also pointed out how greatly the micro-
structure of the bronzes varies according as the alloy is
allowed to cool slowly or quickly — structures quite visible
in the one case often being indistinguishable in the other.

M. Guillet, following up the work of Messrs. Heycock
and Neville, thought that perhaps the useful properties of
the alloys might be modified by heat treatment, and his
experiments show that with some of the alloys the changes
are well marked.

0 100 200 300 400 500 600 700 800 000 Quench-
Cu. 87 per cent. ; Sn. 13 per cent. ing Temp.
FIG. 78.

Fig. 78 shows diagrammatically the results obtained
with a brass containing 87 per cent, copper and 13 per
cent. tin.

THE COPPER-TIN SERIES.

135

With copper 84, tin 16, the results were : —

Quenching
Temperature.

Not quenched

300

400

500

550

600

650

700

750 ..

Tensile strength. Elongation Reduction of area
Kilos, per sq. mm. Per Cent. Per Cent.

24-4

19-4

40-1

42-6

36-3

34-4

29-6

4-7

5-6
10-1

1-4
0
0

1-4
5-9
36
1-4
2-9
5-0

100 200 1 300 400 600 600 700 800 900 Temp.
Cu. 91 per cent.; Sn. 9 per cent.

With coppe

Quenching
Temperature.

Not quenche
400

T 91, tin 9, the i

Tensile Stress.
Kilos, persq. mm.

sd 25-4

18-4

500

18-4

600

700

800

20-7

900 ..

3-9

Elongation.
Per Cent.

10-3
10-5
10-5

9-2
10-5

7-1

3-9

Reduction.
Per Cent.

16-5
14
11 5
23-5
23-5
30
2

The results are shown diagrammatically in Fig. 79.

136

THE COPPER-TIN SERIES.

With copper 90, tin 5, the results

Quenching Tensile Strensrth' Klone-r

were

Quenchin
Temperatu

Not quern
300
450
550
600
650
700
750
800

aj
O

^20

J3
tt

I 10

"03

Ten
re. Kil<

tfied.

sile St

»s. per

19-i
24M
24-1
23-
21-(

19-:

19-^
6-"

rength '
sq. mm.

Elongation
per Cent.

10-2
11-1

6-8

6-1

6-8
7-5

Eedu
per C

ction
ent.

0
7-5

7
8
7
5
0
2
3

'-4

1
<
|

|
i
<+
<

I
1

) ...

1 ...

)
3

. 2
. 2

. 2

. — - —

,--
_

i — •

0 100 200 300 400 500 600 700 800
Cu. 95 per cent. ; Sn. 5 per cent.
FIG. 80.

900 Temp.

These results are shown diagrammatically in Fig. 80.

M. Guillet draws the following conclusions from the
results of his experiments : —

(1) In the case of alloys containing over 92 per cent, of
copper the tenacity is slightly increased by quenching be-
tween 400° and 600°, and the elongation is similarly affected.

(2) In the case of alloys containing less than 92 per cent.
of copper the tenacity and the elongation increase decidedly
as soon as the quenching temperature exceeds 500°.

(3) Maximum strength is reached, whatever the composi-
tion of the alloy at a quenching temperature of about 600°.

THE COPPER-TIN SERIES. 137

(4) Maximum elongation is reached by quenching
from temperatures which vary with the composition of
the alloy. With 91 per cent, copper, maximum elongation
corresponds to a quenching temperature of 800°, while
with 79 per cent, the maximum elongation corresponds
to a quenching temperature of 600°.

(5) The difference between the tenacity of the cast
alloy and that of the metal quenched at the most desirable
temperature is the greater the less the percentage of
copper.

The heat treatment of the bronzes opens up a wide
field for research, and the influence of the temperature of
annealing needs investigation.

Segregation. — Bronze is very subject to segregation
during solidification, the segregation increasing as the
percentage of copper falls . Of the bronzes within the range
of those of commercial importance, those with 95 per
cent, of copper and over seem to be solid solutions of a
definite compound of tin and copper or of tin itself in
copper ; but it must be remembered that even in such
cases the freezing is selective, the first-solidified metal
being comparatively pure, and the mother liquor growing
richer and richer in the second metal, as it falls in tem-
perature, and if the mother liquor be lighter than the
solidifying copper solution, as it will be, since it contains a
larger proportion of the lighter metal, very distinct
segregation may take place. With less copper than
this there are one or two distinct eutectics. It is
uncertain how far during annealing diffusion might
restore uniformity of composition.

VARIETIES OF BRONZE.

Gun Metal. — This alloy is so-called because it was at one
time largely used for casting guns. It contains about 90
per cent, of copper and 10 per cent, of tin, and, as will
be seen by reference to the diagram Fig. 70, it is about
the strongest of the copper-tin alloys. Like all bronzes,
gun metal liquates considerably during slow solidifi-
cation, and guns were therefore always cast mouth
upwards, with a long, sinking head. The term gun
metal should be restricted to alloys containing about 10
per cent, of copper, but it is now often loosely used as

138

THE COPPER-TIN SERIES.

a synonym for bronze, and sometimes even for triple
alloys containing zinc as well as tin.

Bell Metal. — This alloy is largely used for bell founding,
and consists approximately of 80 per cent, of copper and
20 per cent, of tin. It therefore approaches the lower
level of the useful bronzes. It must not be supposed
that this is the only alloy used for bells, as they are often
cast in brass and in the triple alloys, but the name has
come to be associated with alloys of this composition.
Bell metal is hard and brittle, but is very resonant,
whence its use for bells.

The addition of small quantities of other metals has
sometimes been thought to improve the tone of bells,
and silver has sometimes been added for this purpose.
Foreign metals do not seem, however, to have any good
effect, and the pure copper-tin alloy is probably the best
that can be used. Brandt states that an alloy containing
78 per cent, of copper and 22 per cent, of tin gives the
best results. Bell metal seems to lose its resonance if
remelted several times, probably by the formation of
oxides, which dissolve in the alloy.

Brandt gives the following as being the composition
of alloys actually used in bell founding :—

Copper

Tin.

Ziuc.

Lead.

Silver.

Iron.

Anti-
mony.

Normal metal

Alarm bell at Rouen ...
,, at Darmstadt
,, at Reichen-
hall, 13th century

76-1
73-94

22-3
21-67

1-6

1-19

1-6
•17

...

House bells

Small bells

Sleigh bells

84-5

15 42

Clock bells (Swiss)

74-5

•5

It will be seen that the composition of alloys used for
bell founding may depart considerably from the standard
bell metal.

Coin or Medal Bronze. — Bronze has been largely used for
the manufacture of medals and coins, and as these are
made by striking by means of a die the metal must be
plastic in the cold. Copper is of course well suited for
such a purpose, and so-called bronze medals are frequently

THE COPPER-TIN SERIES.

139

copper, electro-bronzed to give them the proper colour.
For coins, however, which have to stand considerable
wear copper has been almost altogether replaced by
bronze, German silver, or some other alloy which is
harder than copper.

As the ductility of bronze reaches its maximum with
about 5 per cent, of tin, the best composition would be
about 95 per cent, of copper and 5 per cent, of tin. This
is the composition of British bronze coins, except that
about 1 per cent, of the tin is replaced by zinc. Alloys
with up to 8 per cent, of tin are used, sometimes
even with more, but in that case the alloy is usually
tempered by heating to redness and quenching in water.

When an impression has to be obtained by repeated
blows of the die, the metal must be annealed from time to
time, as it becomes very brittle under the influence of
work. A small quantity of lead or zinc softens the alloy
somewhat. Coins and small medals are stamped on
blanks cut out of rolled sheet, but larger medals are
stamped on cast discs.

A bronze specially suited for medals is given by
Brandt as containing : Copper, 97 per cent. ; tin, 2 per
cent. ; lead, 1 per cent. Greek and Koman coin bronzes
often contain about 97 parts of copper to 3 parts of tin.

Statuary Bronzes. — For casting statuary a metal is
required that will flow freely and cast well. A bronze
containing about 94 per cent, of copper and 6 per cent,
of tin answers the purpose best, and is largely used for
small castings. For large castings triple alloys contain-
ing zinc are more usually used.

The following analyses of bronzes will show the corn-
position of some alloys used for various purposes : —

Copper.

.Tin.

Zinc.

Lead.

Iron.

Macedonian Coin ...

87-95

11-44

...

Coin of Alexander

the Great

95-96

3-28

0-76

...

Coin of Alexander

Severus

89-0

10-2

...

080

...

Celtic Weapon

92-00

6-70

0-69

0-29 0-30

Egyptian Dagger ...

85-00

14-00

...

Small Statue found

at Oldenburg

92-58

6-33

...

0-99

Column, Place de

Vendome

89-16

10-24

0-49

0-10

...

French Coin

95-00

4-00

1-00

...

0-56

THE COPPER-TIN SERIES.

Only alloys are included here in which the quantity
of foreign metal is small. Obviously in most cases what
is present is due to accidental impurities in the metal
used.

Speculum Metal. — Before perfection was attained in the
manufacturing of glass and the silvering of glass reflectors
the specula for reflecting telescopes were always made of
metal. For this purpose an alloy was required of a white
colour, which should be hard enough to take a good
polish. Such a material was found in an alloy contain-
ing about 66 per cent, of copper and 34 per cent, of tin,
which was manufactured for the purpose and came
to be called speculum metal. The composition is very
near that of the definite compound, Sn Cu4, which would
contain 68' 1 per cent, of copper.

In practice, the composition of the alloy may vary a
little, up or down, without alteration in the properties,
and various makers have a composition to which
they adhere very closely. Either the proportions of
2 of copper to 1 of tin, or those given by the formula
Cu4 Sn, being usually used. Increase in the
quantity of copper tends to give the alloy a yellowish
colour, whilst increase in the percentage of tin tends to
give it a bluish tinge and at the same time to make it so
brittle that it will not polish.

Some makers add small quantities of foreign metals
such as arsenic, antimony, or nickel, but as a rule the
pure alloy of copper and tin is best.

The following analysis of specula, from Brandt, will
show the variations : —

Copper.

Tin.

Zinc.

Arsenic.

Nickel.

Standard alloys, Cu4 Sn. . . .
Otto's

68-1
68.5

36-9
31-5

...

Richardson's
Little's

65-3
65

30.0
30-8

0-7
2-3

2.0
1.9

...

Sallit's ...

64-6

31-3

4-1

Oxides in Bronze.— One great difficulty in the casting
and working of bronzes is the tendency which the
metal has to retain oxide, either in solution or
in admixture, and this very seriously interferes

THE COPPER-TIN SERIES.

141

with the useful properties of the alloy. To over-
come this difficulty, good results have been obtained
by the addition to the alloy of some powerfully reducing
substance which decomposes the oxide of tin, and at
the same time the oxide of which is insoluble in the
alloy and light enough to rise readily to the surface.

Phosphor Bronze. — This is bronze to which a small quan-
tity of phosphorus has been added. The phosphorus
may be added in the free condition, since phosphorus
combines readily both with tin and copper, but this

FIG. 81.— PHOSPHOR COPPER, 8'79 PER CENT. PHOSPHORUS.
MAGNIFIED 40 DIAMETERS. VERTICAL ILLUMINATION.

method of adding it is inconvenient, and the compo-
sition of the resulting alloy is uncertain, as a consider-
able quantity of phosphorus may be lost. The phos-
phorus is usually, therefore, combined either with copper
or tin to form phosphor copper or phosphor tin, the
compound being then used in the manufacture of the
alloy, a certain weight of it being used to replace some
of the metal.

Phosphor Copper. — This is an alloy of copper made by
melting together copper and phosphorus under suitable
conditions. It is a hard brittle substance with a white
metallic fracture, and may be obtained up to about 16 per

142 THE COPPER-TIN SERIES.

cent, of phosphorus. Under the microscope it is seen to
consist of two constituents. (Fig. 81.)

Phosphor Tin. — This is an alloy of tin and phosphorus
made by melting the two elements together under suit-
able conditions. It has a white colour and a metallic
lustre, and is extremely brittle, breaking with a largely
crystalline fracture. Under the microscope it is seen to
be composed of two constituents, a ground mass of
metallic tin through which are scattered plate-like
crystals of a phosphide of tin (Fig. 82). When the quantity

FIG. 82.— PHOSPHOR TIN, 9-78 PER CENT. PHOSPHORUS.
MAGNIFIED 40 DIAMETERS. VERTICAL ILLUMINATION.

of tin is very small the crystals are merely isolated
plates, but when the amount of phosphorus reaches
9 per cent, the mass is made up almost entirely of the
interlacing crystals of the phosphide. The proportion
of copper to tin in bronzes is usually about 8 : 2, and
an alloy of the metals in these proportions containing
about 6 per cent, of phosphorus is made commercially
under the name of hardener.

THE COPPER-TIN SERIES.

143

PHOSPHOR BRONZE.

The addition of phosphorus to bronze has a remark-
able effect upon its properties. The tensile strength is in-
creased, the limit of elasticity is enormously raised, and the
power of resisting repeated stresses is also largely increased.

Many figures of tests made have been published. The
following, issued by the Phosphor Bronze Company, will
be sufficient to indicate the character of the metal.

Two samples of rolled phosphor bronze, tested by Mr.
Harry Stringer, M.Inst.C.E., of Westminster, gave : —

Diam.
of

Test. .

Area
of
Test.

Reduction
. of
Area.

Extension
per cent.
011 2in.

Elastic Limit.

Breaking Stress.

Pounds
per
sq. inch.

Tons
per
sq. in.

Pounds
per
sq. in.

Tons
per
sq. in.

•757
•747

•4501

•438

58-1
22-6

24
9

Not per
90-360

ceptible
40-34

75,577
90,360

33-74
40-34

These were, of course, of different composition.

Another series of tests of cold-rolled phosphor bronze
by M. E. G. Izod gave :—

Area in
Square Inch.

Breaking Load.
Pounds per sq. inch.|

Breaking Load.
Tons per sq. inch.

Elongation per cent.
on 4in.

•6235
•625
•614

54,900
55,200
54,420

39-6
39-4
39-5

10-75
10

Mean.

39-5

10-58

The resistance to repeated pulls is shown by the
following figures from tests made in the Royal Berlin
Academy of Industry, by order of the Prussian Govern-
ment. The tests were of cast phosphor bronze, and
they were compared with similar tests on ordinary cast
gun metal.

(a) Trials by repeated pulls :—

Phosphor bronze.
Highest pulling

stress

Nuniber of pulls before
rupture.

Cast gun metal.
Highest pulling
stress

Number of pulls
before rupture.

per square inch.

Tons.

1. 10

408,350

Broke before

this stress

was reached.

2. 12i

147,850

4,200

3. 7-|-

3,100,000

6,300

144 THE COPPER-TIN SERIES.

(b) Trials by repeated one-sided bends

862,980

102,650

4,000,000)^ g

150,000

3,000,000 Y&*

837,760

2,000,000 j ^ ^

A bar of forged phosphor bronze has resisted without
rupture 2,500,000 twists at a strain of 12 tons, whilst a
bar of Krupp's cast steel under a 12-ton strain broke
after 879,700 twists.

It is obvious, therefore, that the properties of
phosphor bronze are so different from those of ordinary
bronze that it may almost be regarded as a new metal.
Indeed it may be said that phosphor bronze has all the
good qualities of the same bronze without the phosphorus,
very greatly extended.

As, however, phosphorus can be added to bronzes of
any composition, the term phosphor bronze is very vague,
and any comparison should be between a bronze and
another of the same composition but containing phos-
phorus. The data for such comparisons in detail is not
available.

Phosphor bronze of suitable composition can be
rolled and drawn cold, forged, and cast. Hence its
uses are very numerous. It seems to resist the corro-
sion of sea water better than most alloys, and therefore
is often used for propeller blades and other purposes
where it will be exposed to the influence of sea water.

The phosphorus in most cases seems to have little
direct action, but to exert its influence mainly as a
deoxidising agent, and the quantity of phosphorus left
in the alloy is often veiy small, so that the appearance
of the metal and its structure. as seen under the micro-
scope is not changed. Most phosphor bronzes contain
about '1 per cent, of residual phosphorus, present as
phosphide dissolved in the alloy, but in some cases where
hardness is required there may be as much as 1 per
cent., the phosphorus then apparently exerting a dis-
tinctly hardening influence. With 4 per cent, the alloy
becomes useless.

THE COPPER-TIN SERIES.

145

SILICON BRONZE. .

If the action of phosphorus in phosphor bronze
is as is generally believed merely one of deoxida-
tion, then similar results should be obtained by the use
of other easily oxidisable non-metals, or metals, and
this has been found to be, to a large extent, the case.

Silicon bronze, for instance, is a bronze to which a
small quantity of the element silicon has been added for
the purpose.

Silicon bronze has great tenacity, resists atmospheric
corrosion very strongly, and at the same time is a much
better conductor of electricity than is phosphor bronze,
the presence of a small quantity of phosphorus greatly
diminishing the electric conductivity of the alloy. For
that reason wires of silicon bronze are used for telephone
and other wires. The quantity of silicon left in the alloy
is very small.

The following tests of silicon bronze sheet are published
by the Phosphor Bronze Company :—

Dimensions.

Area.

Reduction
of Area
at

Extensions.

Elastic
Limits

Breaking Stress.

Per

On2in.

Fracture.

Frac-

Tons.

Sq. in.

Per cent.

12in.

ture.

Per sq. in.

lin. x 0-06in.

0-06

52-3

5-3

19-7

23-33

26 unannealed

lin. x O'OGin.

0-06

67-7

39-0

51-5

5-83

20 annealed

MANGANESE BRONZE.

Manganese, as is well known, is added to steel
in the process of manufacture for the purpose of
removing the oxygen which has been dissolved in
the metal, and it may be used for a similar purpose in
the manufacture of alloys ; but, also, as manganese and
copper alloy readily, the resulting alloys may contain a con-
siderable quantity of manganese. It is only recently that
manganese was obtainable in anything like purity, so
that the manganese alloys have usually been made by the
addition of ferro-manganese to the alloy, or some-
times a copper-ferro manganese was prepared, and this was
then used for making the alloy. Manganese oxidises so
readily when the alloy is melted that there is often con-
siderable loss, and it often happened, especially in man-
ganese bronzes where the manganese was only intended
L

146 THE COPPER-TIN SERIES.

for deoxi elation, and, therefore, where the quantity added
was small, that the whole of it was oxidised out, but
part of the iron from the ferro-manganese remained, so
that the alloy was rather an iron bronze than a man-
ganese bronze. Now that nearly pure manganese can be
obtained the presence of iron is not so likely.

Many of the so-called manganese bronzes contain a
large quantity of zinc, and are thus rather manganese
brasses than bronzes, and these have been described
under the brasses.

The manganese bronze which does not contain zinc
usually contains about 80 per cent, of copper, and a very
small quantity of manganese. It can be cast in the usual
way, has a high tensile strength, and resists the corrosion
of sea water strongly.

Ferro-manganese is an alloy of manganese and iron,
often containing 80 per cent, manganese, about 7 per cent,
carbon, and small quantities of other impurities.

Mr. Parsons states that a bar of manganese bronze
cast in sand in the ordinary way, and of one square inch
section, placed on supports l'2in. apart, requires upwards
of 4,2001bs. to break it, and before breaking will bend to a
right angle, and it will sustain from l,7001bs. to l,8001bs.
before taking a permanent set. Manganese bronze is
used for gearing wheels, many parts of machinery, and
largely for screw propellers.

FIG. 82A— MANGANESE BRONZE, -8 PER CENT. MN. V -70 x

CHAPTER IX.

MACHINERY BRASSES AND BRONZES, BEARING BRONZES,
AND OTHER COPPER ALLOYS.

As already remarked, brass should contain only copper
and zinc, and bronze should contain only copper and tin ;
but there are a large number of alloys which consist of
copper alloyed with both zinc and tin and sometimes with
other metals also, and which are largely used in engineer-
ing. For want of a better term these may be called
machinery brasses or machinery bronzes, the names
being used almost indiscriminately. Prof. Thurston has
suggested the name kalchoids for the series, but this
has not been generally adopted.

These alloys are used for various purposes, but in
general either for castings or for bearings.

CASTINGS.

The addition of zinc to bronze makes it cast better,
and for that reason a large proportion of the so-called
bronzes contain a little zinc. The British coin bronze,
which contains 1 per cent, of zinc, has been already men-
tioned. A very common bronze alloy is one containing
8 per cent, of tin and 2 per cent, of zinc, the object of the
addition of zinc being, at least partly, to improve the colour.
The quantity of zinc may be .very largely increased,
sometimes to nearly 10 per cent., the alloy still retaining
the name of bronze. On the other hand, a considerable
amount of tin may be added to brasses, an alloy of copper
58, zinc 25, tin 17, being said by Thurston to be excellent
for general castings and for casting statues. An alloy
of copper 90 per cent., tin 6 per cent., and zinc 4 per cent,
has a fine golden colour, and is used as an imitation gold.

Other metals are often added to these alloys for.
various purposes.

No complete investigation has been made as to the
strength -and qualities of these alloys as depending on their
chemical composition. Much has been published, but as a

148

MACHINERY BRASSES AND BRONZES.

rule the alloys are only indicated by their names, and no
* data are given as to the actual percentages of the metals
present ; they are usually of little value.

It will probably therefore be sufficient to give a table
of the composition of alloys which have been used for
various purposes.

Alloy.

Copper.

Tin.

Zinc.

Lead.

Use.

Oun metal

84
77-93
91-4
80-0
75-0
89-2
70-29
66-80
57-9
63-88
63-60
53-30
44-00
88-00

14
16-3
1-4
30
3-0
•5
9-28
2-0
5-3
5-55
2-60
1-30
3-30
10-00

2
6-4
5-5
17-0
20-0
10-2
29-39
32-80
36-8
30-55
25-00
43-0
49-90
2-00

1'7

2'0
•1
•17
'40

8-80
'30
1-20

Valves, screws,.&c.

Hard brass
Statuary bronze

Statue of Napoleon I.Paris
Column Vendome, Paris .
Brass wire

Bra^s leaf

White alloy

» »
Sheathing nails
Yellow solder, hard
White soft

For buttons
Castings for pumps

Lafond's alloys

BEARINGS.

Copper alloys are largely used for bearing metals,
i.e., for metals on which iron and steel shafts are to
revolve. In modern practice, the bearing metal is-
always made of a softer material than that of the
shaft which is to revolve on it. The conditions which are
required for a good bearing metal are to some extent
incompatible, so that all that can be done is to try and
get the most useful mean. The metal should be hard
because, as a rule, the harder the metal the lower will be
the coefficient of friction ; and at the same time it should
be soft enough to allow it readily to adapt itself to the
form of the running surface, so that the bearing may be
uniform over the whole surface ; whilst in order that the
metal may be durable both conditions must come into play.
It is found in general that alloys which give the best
results are those which are made up of hard portions-
embedded in a softer matrix.

The bearing metals in general use may be divided into-
two classes, those in which copper is the principal consti-
tuent, which maybe called bearing bronzes, and those which

MACHINERY BRASSES AND BRONZES.

149

consist principally of tin or lead, with other metals,
which may be called white bearing metals ; the former
only will be considered here, the latter are dealt with
in Chapter XI.

Owing^to its hardness and plasticity, bronze with 5 to
10 per cent, or even 15 per cent, of tin has been used,
but the higher the tin the harder is the alloy, and, therefore,
whilst it fulfils the former condition better, it at the same

FFG. 83.— BEARING BRONZE.

Copper 85, tin 5, lead 10, V x 90 diameters.

The lead is seen scattered through the mass.

time fails to fulfil the latter, so that hard bronzes cannot
be considered to be suitable for this purpose, and with such
alloys the wear is always unduly great.

Mr. G. H. Clamer has made a large series of experi-
ments on bearing metals of this class, and his results
were published in a paper read before the Franklin In-
stitute in 1903. He experimented with three bronzes,
(1) containing copper 85 per cent., tin 15 per cent. ; (2)
copper 90 per cent., tin 10 per cent. ; and (3) copper 95
per cent., tin 5 per cent. Of these, of course, No. 1 was
by far the hardest. The two first named have a duplex

150 MACHINERY BRASSES AND BRONZES.

structure consisting of copper, or a solution of tin in
copper, and a separated eutectic, the hard eutectic being
less in quantity in No. 1 than in No. 2. The results
obtained in one set of experiments in which both the
frictional resistance and the amounts of the metal worn
away were determinsd, were : —

Friction in Pounds. Wear in Grammes.

No. 1 ... 13 ... -2800

No. 2 13 ... 1768

No. 3 ... 14 '0776

The experiments were conducted in such a way that the
results could be directly compared. It will be seen that
the loss of the hardest alloy was much the greatest,
but that the friction was the least.

Bearing in mind the structure of an alloy that would
be likely to give the best result, it will be seen that the
addition of lead would probably be beneficial, for lead
only dissolves to a small extent in copper or copper-tin
alloys, and therefore it might be expected that lead or
a soft alloy of tin and lead would separate so that the
resulting alloy would consist of the copper containing some
tin in solution, and a copper-tin eutectic, and the softer
constituent would be mechanically intermixed through it.
It is, of course, essential that the alloy should not be
of such a character that extensive liquation would take
place, or it might be too irregular in composition.

When lead is added to bronze, a eutectic of lead and
tin separates on solidification, and in general the larger
the amount of this soft eutectic the better will the metal
be for bearing purposes. Mr. Clamer found that with
5 per cent, of tin the alloy could take up as much as
30 per cent, of lead without the lead separating during
solidification, but that as the tin was increased the amount
of lead that could be taken up diminished. With 7 per
cent, of tin not more than 20 per cent, of lead could be
taken up. The explanation which Mr. Clamer gives of
this phenomena is as follows : The copper-tin alloy is
made up of dendrites of copper, a chemically constituted
alloy of copper and tin and a eutectic, the eutectic being
made up of laminae of Sn Cu3 and laminae of copper, and
is found by analysis to contain 73 parts of copper and

MACHINERY BRASSES AND BRONZES.

151

27 parts of tin, and to have a solidifying point of approxi-
mately 930° Fah., whilst the copper solidifies above
1,800° Fah. As the tin is increased, the eutectic
is increased, and one can readily imagine that when a
large bulk of the alloy must cool down from the casting
temperature above the melting point of copper to
930° it must necessarily remain a long while in the liquid
state in the mould. The lead is but mechanically held
by the network of copper and tin, and having a solidifying

18- (

17-5

g
517-0

•c

16-5
16-0
15-5

15-0
(

•08
•07
•06
•05
•04
•03
•02

•01
JO

\l/l

ffl

) ' 5 10 15 20 25 :

Per cent.

Friction-
Wear- -

FIG. 84.— VARIATION OF FRICTION AND WEAR IN A BEARING ALLOY, WITH
THE PERCENTAGE OF LEAD.

point more than 300° below the eutectic, owing to its
higher specific gravity, has abundant opportunity to
liquate to the bottom of the casting, and this, in fact, is
exactly what happens ; but in the absence of eutectic or
the presence of only a small amount of it, solidification
takes place soon after the metal enters the mould, and a
copper-tin network is formed which envelops and upholds
the still-liquid lead.

152

MACHINERY BRASSES AND BRONZES.

The following are the results of the tests made with
bronzes containing lead : —

Composition.

Friction
in Ibs.

Wear in Grammes.

Copper

Tin.

Lead.

•0542

18*

• 0308

18J

•0327

184

•0277

•0204

• 0130

It will be seen, therefore, that as the lead is increased,
the friction is not materially increased, but the wear
becomes less and less.

The following table, from Mr. Clamer's paper, gives
the actual results obtained, and the composition of the
samples used, by analysis.

Copper per
Cent.

Tin per Cent.

Lead per
Cent.

Friction
in Ibs.

Temperature
above
Room.

Wear in
Grammes.

1—85-76

14-90

-2800

2—90-67

9-45

-1768

3—95-01

4-95

-0776

4—90-82

4-62

4-82

-0542

5—85-12

4-64

10-65

18-5

-0380

6-81-27

5-17

14-14

18-5

-0327

7—75 ?

20?

18-5

-0277

8—68-71
9-64-34

5-24
4-70

26-67
31-32

18
18

58
44

-0204
-0130

The addition of nickel was found by Mr. Clamer to
greatly improve the alloy by diminishing the segregation,
probably by causing more rapid solidification of the metal.
An alloy of copper and tin containing lead and about 1
per cent, of nickel is known commercially as plastic bronze,
and is made by the Ajax Metal Company, Philadelphia.

Many other bronzes are used for bearing metals.
Ordinary bronzes are sometimes used, but as a rule lead
is added.

The following figures, from Mr. Clamer's paper,
show the results with some other alloys tested exactly
in the same way as those described above.

MACHINERY BRASSES AND BRONZES.

153

Copper.

Tin.

Lead.

Zinc.

Friction
in Ibs.

Temperature
above
Room.

Wear
in
Grammes.

85-12
82-27
79-84
77-38
74-28

4-64
5-28
4-71
5-62
4-68

10-64
10-25
10-30
11-42
10-61

2-07

5-44
6-54
11-04

18-5
18-5

18-5
18-5
18-5

56
68
66
68
69

•0380
-0415
-0466
-0672
•0848

An analysis of Ajax plastic metal by Mr. C. N.
Forest gave : —

Copper

Tin

Lead

Nickel

99-97
Ajax standard metal contains : —

Copper .. 77-0 Lead .. 11-5 Tin .. 11-5
but alloys with much larger quantities of lead are made.
The following experiments illustrate the influence of a
large percentage of lead : —

Actual Wear

Friction

Temp, above

in Grains

in Ibs.

Room Temp.

1,000,000

revolutions.

Phosphor bronze

16)

10-5

(Composition not given)

Ajax standard metal

IS*

32£ 7-2

„ 21% lead ..; m

44 6-7

„ 30% „

3-0

,, 47% „

13J

1-65

Another series of alloys tested by the Pennsylvania
Kailway Company gave :—

Composition.

Copper.

Tin.

Lead.

Phos.

Arsenic.

Relative
Wear.

Phosphor bronze...

79-70

10-0

9-6

•80

1-00

Ordinary bronze ...

87-50

12-50

1-49

Arsenic bronze ...

79-70

10-0

9-50

—

•8

1-01

Bronze K

(77-00
(77-00

10-50
8-0

12-50
1250

—

•92

•86

154

MACHINERY BRASSES AND BRONZES.

The figures that have been given, and the experiments
that have been quoted, show the extreme importance of
care in determining the composition of bearing metals.

An alloy very largely used for bearing metals is
phosphor bronze, the bronze for this purpose containing
a considerable quantity — about 10 per cent. — of lead.

Arsenical bronzes are sometimes used. Mr. J. F.
Buchanan mentions one containing :—

Copper 80

Lead 10

Tin .. .... 10

and to this mixture 8 parts of arsenic is added. This makes
the metal very fusible, and helps it to carry the lead.

The following analyses of alloys used for bearing
metals have been collected from various sources : —

Copper.

Tin.

Zinc.

Lead.

Connecting rod bearings . .

Locomotive driving axle
bearings

82
74

16
9-5

2
9-5

Locomotive driving axle
bearings

85-25

12-75

2-0

Car and locomotive axle
bearings

2-0

Fenton's metal j 56

1-6

For heavy friction ( Laf ond) 83-00

1-50

•5

Locomotive bearings (Ger-
man) 81-17

15-20

14-60

0-90 iron

Kochlin's alloy for bear-
ings I 90-0

10-0

Anti-friction metal . . . . 1.6

98-23

Delta metal

92.39

2-37

5.10

•10 iron

Graney bronze
Damar bronze
Ajax bronze

78-50
76-61
81.24

9-20
10-60
10.28

••

15-06
12-52

7-27

0 "Phos-
phorus

Phosphor bronze .. ..79-17

10-22

••

9-61

°%hor°us~

MACHINERY BRASSES AND BRONZES. 155

OTHER COPPER ALLOYS.
ALUMINIUM BRONZE.

An alloy of copper and aluminium containing 10
per cent, or less of the latter metal is known as alu-
minium bronze, or aluminium gold, and is largely used
in the arts. As it contains no tin it is not a bronze,
but the name being in general use must be adhered to.
The aluminium bronzes generally used contain from 5 to
10 per cent, of aluminium, though those with less than 5
per cent, have some valuable properties, but with much
above 10 per cent, they cease to be of any value. The
higher limit might probably be put at about 11 per cent.

Colour. — The addition of a small percentage of
aluminium to copper destroys the red colour, and 1 per
cent, is sufficient to change this to yellow. The 5 per
cent, alloy has a fine yellow colour, very closely re-
sembling that of pure gold, whilst the 10 per cent, alloy
is a little darker and resembles ordinary 22-carat gold,
this alloy being, therefore, very largely used for the manu-
facture of jewellery. As the percentage of aluminium is
increased beyond 10 per cent, the colour becomes paler,
with 15 per cent, it is yellowish- white, and with a little
more it becomes white. The addition of a small quantity of
copper to aluminium has but little influence on its colour,

Specific Gravity. — Aluminium is a very light metal, and it
might be expected therefore that the alloys would be
lighter than copper. They are so to a small extent, but
contraction takes place when the metals mix, so that the
resulting alloy has a density greater than that of a mean
of its constituents, the contraction being greatest with an
alloy containing 7 -5 per cent, of aluminium, when it,
according to Kichards, amounts to about 5 per cent. The
following table of the specific gravity of aluminium alloys
may be of use. The figures marked * are from a table given
by Richards f from determinations made by Saarburger
and given by Messrs. Cowles & Bell Bros. ; the rest are
from the recently published report of the Alloys Research
Committee of the Institution of Mechanical Engineers. *

f Aluminium, page 553.

+ This exhaustive Report of Researches made under the direction of its
Committee by Professor Carpenter and Mr. C. A. Edwards, is referred
to in what follows by the letters A.R.C. It should be carefully studied
by all who are interested in the aluminium-copper alloys.

156

MACHINERY BRASSES AND BRONZES.

The metal used by the Committee was in the form of bars
rolled down to in. diam.

Aluminium.
Per Cent.

Specific Gravity
Determined.

Specific Gravity
Calculated.

Contraction.
Per Cent.

-1

8-92

1-06

8-73

—

2-10

8-62

*2 • 5

8-60

8-40

2-3

2-99

8-47

—

-3-00

8-69

8-33 4-1

-4-00

8-62

8-13 5-7

4-05

8-31

8*° (o'i

f8-37

\8-20

i ^

5-07

8-18

—

5-76

8-07

—

6-73

7-95

—

7-35

7-85

—

7-50

8-00

7-60

5-0

8-12

7-78

—

8-67

7-69

—

9-38

7-61

—

9-90

7-56

—

•10

[7-69
17-56

7-25

5-5
4-1

10-78 7-45

—

*11

7-23

7-10

1-8

11-78

7-35

—

13-62

7-25

—

The figures of the A.R.C. may be taken as being
accurate.

The specific gravity of chilled castings is sensibly the
same as that of the worked bars, and that of sand cast-
ings is a little lower.

Hardness. — Hardness is a property very difficult to
estimate. The A.R.C. used the method of Brinell, which
consists in pressing a ball of steel of known diameter by
a known weight into the alloy, and measuring the
size of the concavity produced. The hardness number

MACHINERY BRASSES AND BRONZES. 157

being obtained from the formula of Benedicks, viz.,

71= —- X A5/ where Z/ = the load in kilogrammes, S the

o v p

superficies of the cavity, (mm2) and p the radius of the ball.
Obviously if a ball of uniform size be used, a series of
numbers could be obtained which would give the relative
hardnesses without the use of the last factor. This
method of determining the hardness is the best that has yet
been proposed, but so far it has been little used except for
testing steel, and the ordinary determinations, or perhaps
it should rather be said statements as to the hardness
of alloys, are very uncertain and often misleading.

The hardness of the alloys was found to increase con-
tinuously but not quite regularly as the aluminium was
increased, the alloy with 15 per cent, of aluminium being
harder than the lower members of the series. The hard-
ness of a 15 per cent, alloy is just about that of a *45 per
cent, carbon steel quenched at 20° C. Work hardens the
alloys and they are softened by annealing ; the presence of
impurities, especially silicon, which is often present, is said
to harden the alloy very considerably.

Tensile Strength. — The high tensile strength is one of the
most marked properties of the aluminium bronzes. Many
determinations of the tensile strength have been made, but
those of the A.R.C. on account of their completeness and
the care with which they have been made should supersede
all others. In most results that have been published,
either the composition of the alloy is not stated or no full
account is given of the treatment to which it has been
subjected, and either of these omissions renders results of
very little value. The A.R.C. examined the alloys under
various conditions, (1) sand castings, (2) chilled castings,
and (3) rolled bars. The castings were examined under
various conditions of cooling, and the properties
measured were ultimate tensile strength (ultimate stress)
and ductility as measured by the extension on a 2in. test
piece.

It was found that the tensile strength increased with
the percentage of aluminium till a maximum was reached,
after which it fell considerably. The increase of strength.

158

MACHINERY BRASSES AND BRONZES.

was gradual till about 7 '5 per cent, of aluminium was
reached, after which it rose more quickly, and then when
the maximum was passed fell away again very rapidly.

The maximum tensile stress and the composition of
the alloy which gave it varied with the treatment. In
slowly-cooled castings, made in sand moulds, the strongest
alloy contained 1078 per cent, of aluminium, and gave a
tensile strength of 29'52 tons per square inch; when the
casting was cooled in water from 800° C. the strongest alloy
was that with 10 per cent, of aluminium, which had a
tensile strength of 50 tons per square inch ; with chill

80>

12*

FIG. 85.

Tensile strength, yield point, and elongation of aluminium bronzes.
Sand castings slowly cooled from 800° C. (A.R.C.)

castings the strongest alloy was that containing 10 per
cent, of aluminium, which under those conditions gave a
tensile strength of 36*93 tons, which, however, was
reduced to 27-72 tons when the alloy was slowly cooled.
With rolled bars also the strongest alloy was that contain-
ing about 10 per cent, of aluminium, with a maximum
tensile strength of about 38 tons. In some experiments
made at Zurich oa aluminium bronze made by the Heroult
process, and quoted by Richards, the 10 per cent, alloy
was found to have a strength of 88,3251bs. (39 tons), and
the 10J per cent, alloy 83,9151bs. (37 tons) and 91,0001bs.
(40 tons).

MACHINERY BRASSES AND BRONZES.

159

60*

FIG. 86.

Tensile strength, yield point, and elongation of aluminium bronzes.
Sand castings quenched from 800° C. (A.R.C.)

"""a

Yield

80>

12*

FIG. 87.

Tensile strength, yie'd point, and elongation of aluminium bronzes.
Chill castings slowly cooled from 800° C. (A R.C.)

160

MACHINERY BRASSES AND BRONZES.

The following table of tensile strengths from the
Report of the A.R.C. and other sources may be of
interest : —

Ultimate Strength (Tons per square inch).

Alumin.
Per Cent.

ifllj

Sand
Castings
Quenched
from 800°.

5.||lj

Chill
Castings
Quenched
from 800°.

Neuhausen.
Figures.

•10

11-34

10-67

11-07

11-30

14-50

1-06

12-9

—

12-9

12-60

15-88

—

2-10

14-0

13-9

13-90

17-46

—

2-99

14-8

14-6

14-9

19-97

—

4-05

17-0

16-5

16-4

17-7

23-80

5-07

18-8

18-6

18-0

19-1

26.41

—

*5-5

—

27-95-

5-76

19-4

18-1

19-5

20-5

28-60

6-73

18-82

19-9

—

18-9

28-85

—

7 * 00

(21-97

"(24-59

7-35

20-1

20-0

20-2

19-8

29-68

—

(24-38

7-50

—

"(25-98

8-00

—

(23.11

'(28-58

—

8-12

22-15

26-0

22-23

24-94

33-22

—

8-5

—

(29-44

"(30-48

31 -7&

8-67

23-5

30-7

27-1

29-8

36.67

—

9-00

—

(32-25
"(32-76

36-5*

9-38

23-41

38-16

27-46

35-59

38-00

9-50

—

(33-14
(35-56

39-8*

9-90

—

50-6

26-4

35 57

38-10

—

10-0

—

(35-11
(39-44

40-12

10-5

—

(37-67
J40.62

—

10-78

29-21

35-45

28-72

32-26

38-62

11-00

—

43 -1&

11-5

—

50-76

11-73

19-37

24-79

—

25-04

33-85

—

13-02

15-83

23-18

16-51

25-97

37-16

—

The results in the first five columns are shown diagram-
matically in Figs. 85 to 89, which are taken from the
Report of the A.R.C.

It will be seen that work considerably increases the
strength of the alloy, and that as regards tensile strength

MACHINERY BRASSES AND BRONZES.

161

the alloys may be divided into two groups — those con-
taining less than 7*5 per cent, of aluminium, which are
very little affected by heat treatment, and those containing
more than 7 -5 per cent., in which the heat treatment
greatly modifies the strength of the alloy, sudden cooling
raising and slow cooling lowering it.

The ductility also varies much in the same way, but
the variations are much less regular. The extension was
measured on 2in., the test pieces being turned down parallel
and of uniform size for 2 Jin. With slowly cooled sand
castings an extension of 81 per cent, was obtained with
the alloy containing 6 '73 per cent, of aluminium, when the

FIG. 88.

Tensile strength, yield point, and elongation of aluminium bronzes.
Chill castings quenched from 800° C. (A.R.C.)

casting was quenched from 800° C. and an extension of
70 per cent, with an alloy containing 7 -35 per cent, of
aluminium. With a chill casting slowly cooled an elonga-
tion of 89 per cent, was obtained with an alloy containing
4-05 per cent, of aluminium and the same casting
quenched from 800° C. gave an elongation of 81
per cent. Prof. Tetmayer's experiments, quoted by
Kichards, give a maximum elongation of 64-0 per cent,
for an alloy containing 5 -5 per cent, of aluminium.

The yield point of all the alloys is very low, and does
not vary much till the alloy contains about 7 '5 per cent.
M

162

MACHINERY BRASSES AND BRONZES.

of aluminium, after which it rises rapidly till in some
cases it coincides with the ultimate breaking stress.

A 10 per cent, aluminium bronze tested by Le
Chatelier at various temperatures was found to lose

100*

12*

FIG. 89.

Tensile strength, yield point, elongation, and reduction of area of aluminium

bronze.

Bars rolled to £|in., and slowly cooled from 800° C. (A.R.C.)

strength very slowly at first, then very rapidly. Richards
gives his results as being : —

Temperature.

15° C.

100
150
200
250
300
350
400
450

Tensile Strength.
Tons per sq. in.

3-4

Elongation,
per cent.

33-4
32-5
31-3
29-9
28-1
23-6
147
6-4

22
21
22
21
19
15
21
23

MACHINERY BRASSES AND BRONZES.

163

PSS The A.R.C. made many tests on other properties of
the alloys, some of which should be noted. For torsional
strength it was found that the alloys containing from O'lO
to 2 -10 per cent, of aluminium gave results "far higher than
those of either pure copper or any of its alloys of which
the authors have found mention in literature."* "Alloys
containing up to 7£ per cent, behave extremely well, but
beyond this percentage there is a rapid deterioration of
properties. "f

Tests were also made as to the influence of alternating
stress both in the National Physical Laboratory and by
Professor Arnold in the University of Sheffield. " The
alloys containing 0 to 10 per cent, of aluminium behaved
very satisfactorily in their tests, but the most valuable
range is from 5 to 10 per cent."

FIG. 90.
Aluminium bronze, 10 per cent, aluminium V X 75

Cooling Phenomena. — As the percentage of aluminium in
the alloy is increased the solidifying point of the alloy
slowly falls, till it contains 7 '8 per cent, of aluminium,
then it rises slightly and then falls till there is 67 per

* Report of the A.R.C. p. 133. t p. 134.

164 MACHINERY BRASSES AND BRONZES.

cent, of aluminium, after which it rises again. The

solidifying points of the alloys already given are :—

°C. °Fah.

0-00 Pure Copper 1,085 1,985

2-99 per cent. Aluminium 1,070 1,958

5-76 1,051 1,923

7-35 1,035 1,895

8-12 1,032 1,889

8-67 1,034 1,893

9-90 1,041 1,906

10-78 1,043 1,909

11-73 1,044 1,911

13-02 1,042 1,908

FIG. 91.

Aluminium bronze, 10 per cent, aluminium.
Heated to 900° C. and quenched V X 75.

The evolution of heat in each case being very con-
siderable. All the alloys containing less than 13 per cent,
of aluminium seem to solidify as a whole, there being only
one solidification point. With alloys containing larger
proportions of aluminium there are two or more tempera-
tures at which heat is evolved, either on account of the
solidification taking place in stages or of changes occurring
in the mass after solidification, but with alloys containing
from 7 to 13 per cent, a small evolution of heat was
noticed at a temperature below 500° C.

MACHINERY BRASSES AND BRONZES. 165

Microscopic Structure. — The micros truct are of the alloys
shows nothing of very great importance as far as the
alloys within the range of the aluminium bronzes is con-
cerned. The structure of alloys with less than 7*35 per
cent, of aluminium is distinctly and uniformly crystallised,
and only one constituent seems to be present, with alloys
containing more than this two constituents seem to be
present. With the alloy containing 8 -12 per cent, of
aluminium, a very small heat evolution occurs during
cooling at 563°-566° C. and this increases with rising
aluminium to a maximum at 12-13 per cent. This coin-
cides with the appearance and growth to a maximum of
a dark needle-like structural constituent in the alloy.

Heat Treatment. — Aluminium bronze is softened by chill-
ing, but with little if any change in structure. " To get
the bronze to its maximum elasticity and hardness it
must be cooled very slowly. Articles of bronze can be
heated red-hot in charcoal powder, and allowed to cool
embedded in it."* With the alloys containing over 7*5
per cent, of aluminium the influence of heat treatment
on the microstructure is very marked. With sand castings
the structure is, as might be expected, coarser than that
of the chill castings, and the structure is not much
modified by rolling. By short annealing the constituents
are broken up, so as to produce a banded structure, and
by prolonged annealing they are separated so as to
produce a coarse-grained structure. On quenching from
900° C. the structure becomes finely acicular.

Richards states that aluminium bronze can be worked
well at full redness, but that above this (bright red), or
below it (low red), it works with much less ease, and that
if it be rolled at this temperature it does not become
brittle by working; whilst if worked cold it rapidly
becomes hard and brittle, and needs frequent annealing.
For rolling, owing to the hardness of the metal, strong
rolls are required, and for drawing very hard dies. " In
regard to forging aluminium bronze, the statement that it
can be forged perfectly at all temperatures from a bright-
red to cold does not coincide with the experience of many
workers. At a cherry-red, the suitable temperature for

* Richards.

166 MACHINERY BRASSES AND BRONZES.

rolling, it hardly forges at all. A much lower temperature
must be used — a low redness — and at that it forges
perfectly. Metal hammered from this heat till it is cold
has its strength much increased."*

Aluminium bronze can be rolled into thin sheets drawn
into wire, spun, stamped, or pressed like ordinary brass.
It is said also to be very suitable for an anti-friction metal.
It is tough and malleable, and has "a peculiar unctuousness
or smoothness which seems to resist abrasion and to
render it one of the best anti-friction metals known "*

Corrosion. — -Aluminium bronze has always had a high
reputation for its power of resisting corrosion, and it does
seem to tarnish much less in air than ordinary bronzes, and
it seems to resist the action of fresh water, but sea water
attacks it slightly. The experiments of the Alloys Ke-
search Committee show that it is much less acted on
than Muntz metal or naval brass, providing the percentage
of aluminium be not too low.

The following table (A.R.C.) shows the change in
weight in pounds per square foot experienced by exposure
to sea water for one month : —

Aluminium per cent.

1-06 0-00281bs. loss.

2-99 0-0001 „ „

5-07 0-0000

7-35 0-0000

9-90 0 OOOllbs. gain.

Muntz Metal O'OOUlbs. loss.

Naval Brass 0-0018 „ „

Aluminium bronze does not oxidise readily when
heated to redness in air — it is stated that it has been
kept at a bright red heat for several months without
showing any oxidation.

Comparison with Steel. — The Alloys Research Committee
in the report give a most interesting comparison of the
alloy containing 9 -90 per cent, of aluminium with a
Swedish Bessemer steel of about 0-35 per cent, of carbon.

* Richards.

MACHINERY BRASSES AND BRONZES.

167

(a) Similarity of Mechanical Properties (Rolled
Materials Tension Test): —

Yield
Point.

Ultimate
Stress.

Elastic
Ratio.

Elongation
on 2 inches.

Steel

15-2

38-0

0-40

26-0

Aluminium bronze

14-8

38-1

0-39

28-8

Alternating Stress.

Maximum limit of resistance for 1,000,000 reversals
at 800 per minute with a ratio of tension to compression
of 1-4:—

Steel 29-5

Aluminium bronze ... ... 28 -3

Impact.

Foot-pounds absorbed on fracture (without deforma-
tion) :—

Steel ... 4-3

Aluminium bronze ... ... 4*5

(6) Similarity of physical properties, hardness, and
hardening capacity.

Hardness Numbers.

Annealed.

Quenched in
Water at 20° C.

Hardening
Capacity.

Steel

156

402

2-58

Aluminium bronze . . .

210

349

1-66

The similarity is probably closer than the above table
represents. The hardness number (210) is that of the
unannealed alloy ; a lower value would probably be
obtained for the annealed alloy. The hardening capacity
figures would thus be raised and brought nearer that of
the steel.

Structures.

(a) Rolled Bars. Structural Constituents.

Steel Ferrite, white (soft), pearlite, dark (harder)

Aluminium bronze Yellow crystals (soft), dark crystals (harder)

(6) Bars annealed for a short time.
The structure remains duplex in both cases, a banded
structure is developed in the harder of the two constituents.

168 MACHINERY BRASSES AND BRONZES.

The pearlitic structure eventually disappears. In both
cases the resulting structure is almost featureless.

Structural Constituents.

Steel Ferrite ( white) . . . Massive cementite white.

Aluminium bronze... Yellow crystals... Light brown crystals.

(d) Bars quenched from a high temperature in water.
The structure becomes markedly acicular in both cases.

Uses of Aluminium Bronze.— Aluminium bronze has been
and is used for a large number of purposes, some of
which have been mentioned above. It is largely used
for articles of jewellery on account of its colour and its
resistance to tarnish, and for many parts of machinery on
account of its great strength. M. Cowles has urged the
value of the metal for casting heavy guns, and it has
been used for propeller blades for ships, its great strength
and resistance to corrosion rendering it specially suitable
for this purpose, and there are innumerable other purposes
for which it could be used.

Aluminium Bronze with Other Elements. — The Cowles
Company have prepared silicon aluminium bronze with
2 to 6 per cent, of silicon and aluminium in equal
quantities. They claim to have made an alloy which
is strong, tough, and does not oxidise, and with 10 per
cent, of aluminium and with 2 or 3 per cent, of silicon an
alloy which is the strongest known. Phosphor-aluminium
and boro-aluminium bronzes have also been prepared.

COPPER-LEAD ALLOYS.

When copper and lead are melted together in equal
proportions and allowed to cool slowly they separate —
the lead going to the bottom and the copper rising
to the top. The separation is never complete, nor
is there a sharp line of demarcation, but the lead
seems to pass gradually into the copper. The copper
at the top is not free from lead, nor is the lead at the
bottom free from copper, however slowly the cooling may
have been brought about ; but there seems to be no true
alloy formed, for on examination under the microscope
the copper is found to contain the lead distributed
through it in the form of globules, and similarly the lead

MACHINERY BRASSES AND BRONZES.

169

G. 92.
Lead-copper alloy.
Copper 1 per cent.
The white mass is a separated crystal of copper V X 75.

FIG.

Copper-lead alloy.
Lead '25 per cent. V x 230.

contains separated copper. This is well shown in the
microsections, Figs. 92 and 93. Probably the separated
lead contains a small quantity of copper in solution, and

170 MACHINERY BRASSES AND BRONZES.

the copper a small quantity of lead. Alloys contain-
ing a large proportion of copper do not separate into
layers on cooling — probably, as in the case of bronzes
containing lead, because the copper, having a very high
melting point, solidifies so rapidly that it forms a net-
work which entangles and retains the lead.

COPPER, MANGANESE.

These metals alloy readily, forming alloys which may
be brittle or malleable according to the proportions of
the metals present. Manganese copper is made com-
mercially, and is used in the manufacture of manganese
bronze and other alloys.

COPPER AND IRON.

Copper and iron alloy in all proportions. Alloys
containing less than 2 -73 per cent, of iron are homogene-
ous, and under the microscope show only one constituent,
and the presence of the iron does not destroy the
colour of the copper. Alloys with from 273 to 9 7 '20 per
cent, of copper show two distinct constituents, the new
substance rich in iron appearing at first as six-rayed crystal-
lites, then as the quantity of iron is increased they gradually
occupy the greater portion of the mass and by interference
yield rounded crystal grains which are separated by the
solid solution of iron in copper. As the fracture follows the
lines of the copper-iron solution it is still copper coloured,
but a polished section may show little or no copper colour.

When the iron is above 9 7 '2 there again seems to be
only one constituent — a solution of copper in iron.

None of these alloys are of any commercial im-
portance.

COPPER AND ANTIMONY.

Copper and antimony alloy in all proportions, but
the alloys are very brittle and are of no practical im-
portance.

The most striking feature about the series is the
occurrence of an alloy of a fine purple colour (Regulus of
Venus) which contains 51 '5 per cent, of copper and has
approximately the formula Sb Ca>. The structure of the

MACHINERY BRASSES AND BRONZES. 171

alloys is very complex, as different constituents crystallise
out during solidification of alloys of varying composition.
Five classes of alloys have been determined, viz.,

(1) Antimony, 100- to 75*8. Copper, 24-2 to 0-

(2) „ 75-8 to 48-5. „ 51-5 to 24-2.

(3) „ 48-5 to 38-57. „ 61-43 to 51-5.

(4) „ 38-57 to 31-00. „ 69-00 to 61-43.

(5) „ 31-0 to 0-00. „ 100-0 to 69-00.

A full account of these alloys will be found in a paper
by Mr. J. E. Stead, F.R.S., in the " Journal of the
Society of Chemical Industry," December 31st, 1898.

Magnetic Alloys. — A series of very remarkable alloys
have recently been prepared by Dr. Huesler which,
though they contain no iron, are distinctly magnetic.

Two samples of these alloys contained* : —

A B

Copper 60-49 68

Manganese ... 22-42 18

Aluminium 11-65 10

Lead — ••,•••• ^

A also contained intermixed slag mostly consisting of
oxide of manganese and silica and carbon 1*52, silicon
0-37 per cent., and iron 0-21. The alloys are brittle and
cannot be forged. The magnetic power is weak com-
pared with that of iron.

The limits of composition between which the magnetic
properties can be detected are not yet known, but the
magnetic power seems to reach a maximum when alumi-
nium and manganese are present in atomic proportions.

* Fleming and Hadfield. "Page's Weekly," July 7, 1905, p. 29.

CHAPTER X.

WHITE ALLOYS.
ALLOYS IN WHICH TIN is THE PKINCIPAL CONSTITUENT.

TIN alloys readily with most metals, and some of the
alloys are of considerable practical value, and only such
will be considered here.

Tin and Lead. — Tin and lead alloy in all proportions,
forming a series of alloys to which the name Pewter may
be given, small quantities of other metals being some-
times added and special names being given to the alloys.

As tin is added to lead, the alloy becomes harder
than lead, and therefore the power to mark paper is
gradually diminished; but all lead alloys containing less
than 75 per cent, of tin mark paper, though less than
pure lead, and the darkness of the mark increases with
the percentage of lead. No determinations of the hardness
of this series of alloys seem to have been made. Tin
and lead alloys expand on alloying, so that the density
of the alloy is less than that calculated from the
densities of the constituents assuming them to be mere
mixtures. The following table gives the results of
Kupffer's experiments:*—

Lead.

Per Cent.
Pb. Sn.

Specific Gravity.
Calculated. Found.

Diff.

1 0

11-3803

SnPb

63-7

36-3

9-4366

9-4263

0-0103

SnPb2

77-82 22-18

10-0936

10-0782

0-054

SnPb3

84-04

15-96

10-4122

10-3868

0-0254

SnPb4

87-42

12-58

10-6002

10-5551

0-0431

Sn2Pb

47-73

53-27

8-7518

8-7454

0-0064

Sn3Pb

36-90

63-10

8-3938

8-2914

0-0069

Sn4Pb

30-49

69-51

8-1516

8-1730

0-0096

Sn5Pb

25-85

74-15

8-0372

8-0279

0-0093

Sn6Pb

17-04

82-96

7-9526

7-9210

0-0116

100

—

7-2911

* Watt's Dictionary, 1st Edition, Vol. III., p. 534.

WHITE ALLOYS.

173

Melting Points. — As tin is added to lead the solidifying point
falls from the freezing point of lead (326° C., 619° Fah.) till
at 180° C. a eutectic separates, which contains a little less
than 70 per cent, of tin ; as the quantity of tin is increased
the freezing point steadily rises to the melting point of
tin, 231° C. The lead solidifies as a whole, that is as a
solid solution of tin in lead, till there is about 4 per cent.

.
6I9'F

250

200

FkrCentO
Lead 100

o'Wn

PerCerrf*

earf- Tin

EuTect

d80°C

"^^

0 10 20 30 40 50 60 70 SO 90 /O
00 90 SO 70 60 50 40 30 20 W 6

FIG. 94.— FREEZING-POIKT CURVE OF LEAD-TIN ALLOYS.

of tin present, after which it ceases to solidify as a whole,
but commences to solidify at a temperature which con-
tinually falls as the percentage of tin is increased, there
being a second solidification point at 180°, the freezing
point of the eutectic. Starting from the other end the
phenomena are exactly similar, the tin solidifies as a
whole till it contains about 2 per cent, of lead, after which
the eutectic begins to separate. These phenomena are
indicated in the diagram Fig. 95.

Various tables of the freezing points of these alloys
have been published, the freezing point given being
always that at which freezing begins and corresponding
therefore to the upper branches of the freezing curve in
Fig. 95.

The following abstract of a table of melting points
determined by Messrs. Parkes & Martin, which is
given by Mr. Hiorns in his " Mixed Metals," may be
useful, though some, at least, of the temperatures do
not agree exactly with those given by more recent
work.

174

WHITE ALLOYS.

Tin.

Lead.

M.P. C*.

85-7

14-3

194-6

83-3

16-7

192

81-8

18-2

189

186

77-8

22-2

183

179

71-4

28-6

175-5

66-7

33-3

170} The

169 j eutectic

189

211

33-3

66-7

228

28-6

71-4

243

250

259

16-7

83-3

267

14-3

85-7

270-5

12-5

87-5

275

10-5

89-5

279-5

281

283

286-5

291-7

5-5

94-5

291-7

The microstructure of the alloy containing between
4 per cent, and 98 per cent, of tin shows the two con-
stituents distinctly.

Mr. E. S. Sperry, of New York, has made an exhaus-
tive research on the physical properties of the tin-lead
alloys. The results of his experiments are shown in the
diagram Fig. 96, and he sums up his results thus : — *

(1) Tin and lead combine in all proportions.

(2) The colour of the alloys ranges from that of pure
tin to that of lead.

(3) All the alloys can be rolled in the same manner
as that employed for rolling tin.

(4) The plastic alloyst are not so fluid as the non-
plastic compositions unless superheated.

(5) The yellow colour (due to surface oxidation of the

* J.S. C. I., 1899, p. 113.

f By plastic alloys is meant those which pass through a long plastic stage
during solidification.

WHITE ALLOYS.

175

tin) can be produced on alloys up to and including tin 44
per cent. After this point the lead characteristics begin
to predominate. In order to obtain the best results, the
metal must be poured at the proper temperature. It was
noticed that metal which had been poured "hot," and
consequently devoid of the yellow film, became coloured
by it on standing exposed to the air for some time.

(6) The " tin cry " can be produced from the alloys
(starting from pure tin) up to and including about 50 per

6000

Han

Tin 0 5 10 15 20 25 M 554O45 SO 55 6O 65 70 75 80 85 90 35 /CO Lead

lompressive Strength

Tens/fa fori°gth
/ny

E'/orJgOfion
in y$ Inches

FIG. 95.— COMPRESSIVE STRENGTH, TENSILE STRENGTH, AND ELONGATION
OF ALLOYS OP TIN AND LEAD.

cent. ; it is nearly absent, however, in the last few
combinations.

(7) The strongest alloy in tension is, tin 72'5 per cent.,
and lead 27 '5 per cent.

(8) The strongest alloy in compression is, tin 71 per
cent., lead 29 per cent.

(9) The most ductile alloy is, tin 40 per cent., and
lead 60 per cent.

(10) The alloy with most reduction of area is, tin 5
per cent., lead 95 per cent.

(11) The best alloy for ordinary use is, tin 50 per
cent., and lead 50 per cent., as the surface of the bar is
perfectly smooth and free from the matt surface found in
some other alloys.

(12) The alloys from tin 15 per cent, and lead 85 per
cent, to tin 30 per cent, and lead 70 per cent., inclusive,
are not so homogeneous. The outside of the test bar

176 WHITE ALLOYS.

fracture showed a fibrous nature, while the core consisted
of granular material.

(13) The alloys begin to assume a plastic nature at
ftn 34 per cent, and lead 66 per cent., and end at tin 15
per cent, and lead 85 per cent. The other alloys do not
pass through a true plastic state, but pass almost imme-
diately from the solid to the liquid condition. If an
attempt is made to use such alloys in place of the plastic
compositions, it will be found that the whole mass will
be filled with hard lumps, which prevent the successful
attainment of the end. The alloys between the limits
just mentioned are the so-called wiping solders, and in
commerce are known as 3 and 1 and 2 and 1.

(14) The alloys showing the most crystalline nature
are those containing from 10 to 20 per cent of lead,
inclusive.

(15) The shrinkage more nearly approaches that of
tin, and is considerably less than that of lead. The
average shrinkage of solder may be said to be O'OGin. to
the foot.

Pewter was at one time largely used for the prepara-
tion of drinking vessels and other articles of domestic
use, but they have almost completely disappeared, having
been replaced by more durable and cheaper materials,
pewter, owing to its softness, being very liable to be
dented and put out of shape. Ordinary pewter consists
of tin 80 per cent, and lead 20 per cent., a little antimony
or copper being often added to harden it, but the quantity
of tin may sometimes be as low as 50 per cent. Pewter
was decidedly objectionable as a material for drinking
vessels owing to the possible solution of the poisonous
lead; but it has been stated that, provided the percentage
of lead was not above 20 per cent., none was dissolved.
In the days when pewter was largely used, the Pewterers'
Company made attempts to keep up the quality of
pewter, but, owing to the great difference in the price
of lead and tin, this was always difficult to do.

The Solders. — The most important use of the tin-lead
alloys is probably the preparation of soft solders for
the use of plumbers and whitesmiths. In soldering metal
surfaces, it is necessary to use for a solder an alloy which
will unite with the two surfaces to be joined, and at the
same time it must melt at a lower temperature so as to
avoid risk of melting the metals. As one of the metals

WHITE ALLOYS. 177

which it is most frequently required to solder is lead, it is
evident that solder must have a lower melting point than
lead, and an alloy of lead and tin, if near the eutectic
composition, fulfils that requirement.

Best plumbers' solders are made by melting together
two parts of tin to one part of lead, and therefore would
contain 66'7 per cent, of tin, which is almost exactly the
eutectie proportion, but as variations up or down are of
little importance, the alloy is usually made by mixing
the metals in approximately the required proportion by
guess. The solder is usually cast in sticks for convenience,
and these sticks can at once be distinguished from sticks
of tin by the absence of the cry which is so characteristic
of the metal.

As tin is now a very costly metal, solders are often
made very much poorer in tin, equal portions of each
metal being often used. An alloy of 33'3 per cent, tin
and 66*7 per cent, of lead is also often used under the
name of plumbers' sealed solder, and is marked by the
Plumbers' Company. It is used for wiping joints, and
its value for this purpose depends on the fact that it
passes through a pasty stage, the lead, containing of
course a small quantity of tin in solution, solidifies as
cooling goes on, but the mass does not solidify completely
till the eutectic temperature is reached, so that as the
alloy is being worked between the temperature at which
freezing begins arid the eutectic temperature it consists of
a mass of the still-liquid eutectic entangled in a network
of the solid and solidifying lead.

TIN AND ANTIMONY ALLOYS.

Tin and antimony alloy readily in all proportions
with the production of alloys that are of little commercial
importance. The antimony does not affect the colour
of the alloy, but renders it much harder so that it can
be used for purposes for which tin, owing to its softness,
would be unsuitable.

As antimony is added to the tin, a definite eutectic,
which contains about 7*5 per cent, of antimony, seems
to be formed ; but according to Mr. Stead this eutectic
does not consist, as is the case of most eutectics, of more
or less parallel plates, and on treating it with dilute
hydrochloric acid a very fine black amorphous powder
is obtained which seems to be a definite compound of

178

WHITE ALLOYS.

FIG. 96.— TIN-ANTIMONY ALLOY (Antimony 7 %) V X 75 diams.

FIG. 97.— TIN-ANTIMONY ALLOY (Antimony 25 %) V X 75 diams.

WHITE ALLOYS.

179

antimony and tin, having the formula Sb Sn (Sb 50*21,
tin 49'79). As the tin is increased beyond 7'5 per cent.,
definite crystals separate and these seem to be cubes.
As the quantity of antimony is still further increased,
the crystals increase in quantity, gradually interfere
with one another so as to break up the true crystal form,
and when the composition 50/50 is reached they practi-
cally occupy all the space; but as the percentage of
antimony reaches 40 per cent, the crystals change in
form and probably in composition, the crystals being no

FIG. 98.— TIN-ANTIMONY ALLOY (Antimony 40 %) V x 75 diams.

longer cubes but thick plates. As the percentage of anti-
mony is further increased, the crystals change their forms
and appearance to those of antimony crystals.

All alloys of tin and antimony with less than 1' 5 per
cent, of antimony seem to have only one freezing point.

The specific gravity of the alloys is less than that of a
mean of the constituents, so that the metals expand on
union.

Antimony hardens tin, but up to 20 per cent, of anti-
mony the alloy is quite malleable, and can be hammered
or rolled cold, and can also be cast ; and all the useful

180

WHITE ALLOYS.

alloys of tin and antimony contain not more than this
proportion of antimony, except where greater hardness
and less ductility is required, when it may exceed this
limit.

Britannia Metal. — The best known alloy of antimony and
tin is Britannia metal, and this, though essentially an
alloy of the two metals, usually contains some other
hardening metal in small quantity. It has a white,
almost silver-white, colour, and is capable of taking a

FIG. 99.— TIN-ANTIMONY ALLOY (Antimony 50 %) V X 75 diams.

high polish. The object of the antimony is to harden
the metal and at the same time diminish its ductility as
little as possible. Other metals are sometimes added, iron
or zinc increase the hardness, but at the same time
increase the brittleness ; copper increases the hardness,
and does not interfere much with the ductility, but except
in very small quantities interferes with the colour, giving
a yellowish tinge. Lead does not harden the metal, but
makes it give cleaner castings, and also darkens the
colour. The best Britannia metal would probably contain

WHITE ALLOYS.

181

tin 90 per cent., antimony 10 per cent., and would be
free from other metals. Such an alloy could be cast,
stamped, or spun into the required form.

When articles were largely made of polished Britannia
metal the colour was of the utmost importance ; now that
such articles are usually electro-plated with silver the
colour is of much less importance, and therefore the
colour of the alloy need not be considered.

The following figures given by Brandt will indicate
the general composition of Britannia metal : —

Tin.

Antimony.

Copper.

Zinc.

Bismuth.

English

81-90

16-25

1-84

90-62

7-81

1-46

90-1

6-3

3-1

•05

85-4

9-66

0-81

3-06

Queen's JMetal

88-5

7-1

3-5

0-9

German

Birmingham (sheet)
(cast)
Karmaischs ..

90-60
90-71
85-00

7-80
9-20
5-0

1-50
0-09
3-60

T40

1-60

The cast surface of Britannia metal is dull and
crystalline, and before use it is always polished.

Other Tin Alloys. — Tin and zinc do not alloy readily,
and the alloys are of no commercial importance. Tin
and aluminium alloy, yielding white alloys. Owing to
the very high price of tin, tin is now used as little as
possible in the manufacture of alloys.

LEAD ALLOYS.

Lead alloys on the whole are of very little importance.
The alloys with tin have already been considered, and
with copper and zinc lead does not form useful alloys.
Lead will dissolve small quantities of other metals, which
impart to it a certain amount of hardness and at the
same time diminish its ductility, whilst in some cases
they increase the sharpness of castings which can be
obtained from the alloy.

The most important addition is probably that of anti-
mony, which very rapidly hardens the lead and makes it
brittle. Lead containing antimony when accidentally pro-
duced is known as hard lead. The eutectic seems to

182

WHITE ALLOYS.

FIG. 100.— LEAD-ANTIMONY ALLOY (Lead 90, Antimony 10) V x 75 diams.

FIG. 101.— LEAD-ANTIMONY ALLOY (Lead 60, Antimony 40) V X 75 diams.

WHITE ALLOYS.

183

contain about 13 per cent, of antimony. When there is
more lead than this the mass seems to consist of crystals
of lead embedded in a matrix of a very brittle eutectic,
whilst when the quantity of antimony is over 13 per cent,
it is the antimony that crystallises out. All the alloys
are very brittle.

Type Metal. — This is the only lead-antimony alloy that
need be considered. The alloy must cast readily, and be
capable of taking a very sharp impression ; it must be hard
enough to resist crushing by the pressure of the press, and
must be so soft that the edges do not cut the paper. Lead
and antimony are always the basis metals of this alloy,
but other metals may be present in small quantity.

The following table is that given by Brandt, but the
mixtures are calculated into percentages :—

1234 5678

75 83 91 77 70 60 55 55
25 17 9 15'4 18 20 25 30

Lead

Antimony

Copper...

Bismuth

Zinc

Tin

Nickel

— — 7-6 — — — —

— — — — 10 20 20 15

59-5
18-0
4-8
1-7

11-2

4-8

Lead and Arsenic. — The only interest in these alloys lies
in the fact that for shot making lead is always alloyed
with a small proportion of arsenic. The addition of

arsenic hardens the metal
and increases the fusi-
bility, and it is on account
of the latter property
that it is used in shot
making, as the increased
time taken in solidifying
gives the metal a better
opportunity of form-
ing a spherical drop in
its fall from the top of the
shot tower The quan-
tity of arsenic added is
small, always under 1 per
cent., probably usually

.V *

*». 102-LEiD

CHAPTER XI.
WHITE ANTI-FRICTION ALLOYS.

FRICTION causes such a large loss of energy in all machines
that it is important to reduce it to the lowest possible
amount, and as the friction between two metals, the
one sliding over the other, depends on the nature of the
metals it is important to select the metals so as to give as
little friction as possible. The one surface, the moving
part, the rotating axle or shaft, will necessarily be of
steel, but the other, the bearing surface on which it runs,
can be made of almost any metal. The bearing may
be of some hard, strong metal, such as phosphor bronze,
or it may consist of a shell of iron, bronze, or other
strong metal, with a bed of some softer metal cast in to
form the actual working surface. Lead seems to have
been the first metal suggested for the purpose, but this
was soon replaced by alloys of some kind, and now the
number of such alloys on the market is very large.
Gradually white metals came largely into use, it being
found that suitable alloys were very durable, and
diminished the friction very considerably, the friction
being with such alloys less than that of hard bronze,
and the wear being also less. As a rule, bearings are
wrell lubricated, so that the metal is only in contact with
the film of oil or other lubricant, and therefore it should
be independent of the metals in contact, and to a
certain extent, if the surfaces were perfectly smooth,
this would be so.

According to the ordinary law (Coulomb's law)
of friction, the amount of friction should be directly
proportional to the load, and should be indepen-
dent of other conditions, except the nature of the
surfaces in contact. This is found to be the case within
limits, but as the pressure becomes considerable the law
ceases to hold good, and the resistance increases much
more rapidly than the load, and the bearing therefore may

WHITE ANTI-FRICTION ALLOYS. 185

heat, or, to put it in another way, for moderate loads the
coefficient of friction remains constant, but it increases
considerably as the load is much increased. The point at
which this increase takes place is much higher with hard
metals than with softer ones ; and hence hard alloys such
as phosphor bronze have been largely used for bearing
metals.

With a hard metal, however, it is almost if not quite
impossible to secure a uniform bearing surf ace. The bearings
will be more or less rough, so that instead of the weight
being uniformly distributed over the whole surface, it is
borne on a larger or smaller number of points on which
therefore the pressure becomes very great, and heating
may take place. To avoid this, the bearing metal should
be sufficiently plastic to accommodate itself to the form
of the shaft, and thus to give the maximum of bearing
surface. The wear, of course, during running is also
more or less irregular, and with a hard unyielding metal
the irregularities might increase, but if the metal be
soft and plastic it will to a certain extent flow under
the pressure, and it will automatically correct such in-
equalities. On the other hand, the metal must not be too
plastic, or it will flow too readily, and at the same time
will tend to "cut." Two properties are therefore
required which seem incompatible, hardness and plasticity.
The best result is obtained, as already pointed out in the
consideration of the plastic bronzes, by combining a soft
metal with a hard one, that is, by the use of an alloy
which consists of a soft ground mass in which harder, but
not too hard, particles are embedded. Indeed, M. Charpy
states that this is characteristic of all bearing alloys.
He says, " The load is carried by the hard grains which
have a low coefficient of friction, and the cutting of which
can only take place with great difficulty. The plasticity
of the cement makes it possible for the bearing to adjust
itself closely around the shaft, thus avoiding local pressures
which are the principal cause of accidents."

In testing a bearing alloy, therefore, M. Charpy says :
" The plasticity may be ascertained by a compression
test ; it must be sufficient to enable the bearing to
adjust itself round the shaft, and must not exceed a
certain limit in order to prevent its undergoing any per-

186 WHITE ANTI-FRICTION ALLOYS.

manent distortion under the action of the load. The com-
pression test provides also a means of ascertaining whether
the alloy is brittle, which would be a serious defect."

The microscope will, of course, be necessary to make
out the structure, and M. Charpy says that the difference
in hardness will usually be so great that the different por-
tions can be detected after polishing without etching ; but,
of course, they will be made more distinct by etching. As
the boxes will usually be filled by casting, it is necessary
that the alloy should cast well. If the above con-
siderations are true, the value of a bearing alloy depends
mainly on its physical structure, and little directly on
its chemical composition.

The frictional resistance, however, does seem to
depend to some extent on chemical composition,
certain elements increasing and others diminishing it.
Prof. Goodman has pointed out that it sometimes
happens that alloys supposed to have the same com-
position gave frictional results which differed by 100
per cent., whilst on analysis they were found to have
their principal constituents in almost exactly the
same proportions, but that there were differences in
the quantities of impurities present. Further investiga-
tion showed that very minute quantities of some elements
had a very marked effect on the friction, some increasing
and others diminishing it, and further that those ele-
ments of low atomic volume* increased the frictional
resistance, whilst those of high atomic volume decreased
it, provided that they were present in small proportions.
The addition of 0-1 per cent, of aluminium, which has
an atomic volume of 10-6, increased the frictional
resistance by about 30 per cent., whilst the addition of
bismuth, which has an atomic volume of 21 • 1, immediately
reduced the friction. An amount of bismuth equal to only
0-025 per cent, was sufficient to perceptibly reduce the
frictional resistance, and it was further reduced by addi-
tions of bismuth till the amount reached 0-25 per cent.,
after which further additions produced increased resistance.
A small quantity of bismuth is now added to many anti-
friction metals. It is difficult to say if this is any real

* See note p. 11.

WHITE ANTI-FRICTION ALLOYS.

187

advantage, because, as already remarked, in practice
lubricants are used to keep the metallic surfaces apart.

Many anti-friction metals are on the market, some of
them known by very high-sounding names, and for some
of them most absurd claims are made. A consideration of
the principles laid down above, and a study of the alloys
themselves, will enable any one to understand the
qualities which are required in a good anti-friction metal.

Babbitt's Metal — This is one of the best-known of
the bearing metals. The original formula was " to
melt separately 4 parts copper, 12 parts tin, and 8 parts
antimony, then after fusion to add 12 parts of tin, this
mixture constituting the hardening. For use this
hardening was melted with more tin in the proportion of
1 part hardening to 2 of tin, so that the composition
of the alloy would work out to copper 3-5 per cent.,
antimony 7-4 per cent., and tin 88 '9 per cent. Babbitt
metal, as now made, however, cannot be considered as

Number of
Alloy.

Composition.

Load
corresponding
to a
Compression
of 7'5 mm.

Tm.

Copper.

Antimony.

Broken

2000

1550

Broken

2258

Broken

72-5

12-5

8-5

2550

11-5

5-5

2750

2475

—

Broken

—

2600

—

2650

—

2150

188

WHITE ANTI-FRICTION ALLOYS.

having any definite composition, as each maker modifies it
to suit himself, and the name has become a general one to
indicate bearing metals consisting of tin, with a small
quantity of copper and antimony. The quantity of tin
is always over 50 per cent., usually over 80.

M. Charpy has examined the series of alloys of these
three metals, with special reference to their use as bearing

Tin.

Copper.

Antimony.

Quoted by Thurston. Rus-

sian railroad car bearings.

88-9

3-7

7-4

Normal Babbitt metal.

88-8

3-7

7-4

Quoted by Thurston and

Bolley as Karmarsch metal.

Best alloy. Charpy.

Quoted by Hiorns for bear-

ings heavily loaded.

Quoted by Ledebur and

Hiorns as Jacoby metal for

light pressures.

83-33

5-55

11-11

Car bearings.

Quoted by Ledebur. Used by

Berlin railroads.

Quoted by Ledebur. Used

by the Orleans and Western

Austrian railroads.

Bearings for valve rods and

eccentric collars. Com-

pagnie du Nord.

Quoted by Hiorns for very

hard bearings.

Quoted by Thurston. Used

by Swiss railways.

78-5

11-5

Quoted by Thurston. Used

by Russian railroads.

76-7

7-8

15-5

Quoted by Ledebur and

Thurston as English alloy.

Thurston standard white

metal.

Quoted by Thurston. Used

by the Great Western Rail-

way.

French State railroads.

WHITE ANTI-FRICTION ALLOYS.

189

metals, by compressive tests and by microscopic
examination. The table on page 187 gives the results
of some of his experiments.

FIG. 103.— ALLOY.

83*3 per cent. Tin. 11 '11 per cent. Antimony. 5*5 per cent. Copper (CharpA ).

V30 X

FIG. 104.— ALLOY.

70 per cent. Tin. 25 per cent. Antimony. 5 per cent. Copper (Charpy).
V30 X*

190 WHITE ANTI-FRICTION ALLOYS.

Nos. 4, 10, 13, and 18 showed internal cracks before
reaching a compression of 7-5 mm. Those which
broke and these numbers are therefore too hard for use
as bearing metals, and the one which seemed most
suitable for the purpose judged by this test was No. 14.*

Alloys of tin and antimony containing excess of tin
examined under the microscope show a ground mass of
tin with definite cubic crystals of an antimonide of tin
Sn Sb (in Fig. 98, p. 179), whilst alloys of tin and copper
containing excess of tin show needle-like crystals of Sn Cu3
(Fig. 73, p. 128). When both metals are present a
definite compound does not seem to be formed, but
the mass consists of tin with both sets of crystals scattered
through it. So that these bearing metals consist of a
ground mass of soft tin with hard crystals of Sb Cu3
and Sb Sn scattered through it. The addition of both
metals is an advantage, because to obtain the same number
of hard crystals with one metal only would require so
much of that metal that the alloy would be brittle.
As brittleness appears sooner in the copper series than
in the antimony series, the quantity of antimony should
be larger than the quantity of copper. The table on
page 188, abridged from M. Charpy's paper, gives the
composition of some of these alloys in actual use.

It will be seen therefore that the composition of the
alloys in use varies very much. According to Mr.
Clamer, the addition of a small quantity of lead to these
alloys is advantageous.

Lead and Antimony — Lead and antimony seem to
alloy very readily and under some circumstances the
alloy may be used as a bearing metal. When the anti-
mony rises above 13 per cent, it crystallises out, so that
the alloy consists of a ground mass of lead with hard
crystals of antimony. M. Charpy says that alloys con-
taining 15 to 25 per cent, of antimony are most suitable
for bearing metals, but Mr. Clamer states that alloys
with less than 13 per cent, are often very useful, and that
he has seen " many instances in service where alloys
containing between 15 and 20 per cent, were greatly

* The test pieces used were 15 mm. long, so that this corresponds to a
compression to half the original length.

WHITE ANTI-FRICTION ALLOYS. 191

inferior to alloys containing between Sand 12 per cent.

owing to their frequent renewal due to wear."

A lead-antimony alloy is probably the cheapest white

metal, as there is no costly tin in it. Mystic metal contains :

Lead 88-7

Antimony . . . . . . 10-8

Iron . . . . . . . . Trace

Bismuth . . Nil

FIG. 105.— MAGNOLIA METAL.
V x 100.

The best known of the alloys of this group is probably
Magnolia metal, which has been very extensively adver-
tised. It contains : — a) <2) (3>

Lead 78 79-41 78-27

Antimony .. ..21 20'15 17'81

Iron . . . . . : 1 Trace

Bismuth . . . . Trace

Tin .. ;>.;..:: .. 3-88

Copper ...... -04

No. 1 is from Hiorn's mixed metals. No. 2 an
analysis made in the author's laboratory. No. 3 from
"Metallurgie," Vol. 3, p. 607.

192

WHITE ANTI-FRICTION ALLOYS.

Alloys Consisting of Lead and Tin and Antimony. —
Lead and tin alloys, usually with the addition of an-
timony, have often been used for bearing metals.
Lead and tin alloy readily, there being usually

FIG. 106.— ALLOY.

76 per cent. Lead. 14 per cent. Tin. 10 per cent. Aatimony (Charpy).
V X 200.

FTG. 107.— SMALL PORTION OF ALLOY, FIG. 106 (Charpy).
V X 500.

WHITE ANTI-FRICTION ALLOYS.

193

crystals of one or other of the metals embedded in the
eutectic alloy. As lead and tin are both very soft, a
mixture of the two would not form a good bearing metal.
The addition of antimony, however, gives the necessary
hardness, and at the same time greatly increases the com-
pressive strength, the increase, according to M. Charpy,
being about the same for all the alloys except those
very rich in lead, in which case the influence of the anti-
mony is much less marked.

The following table is given by M. Charpy as the
result of his experiments : —

No.

Lead.

Tin.

Antimony.

Load corresponding
to a compression
of 7*5 mm.

100

1060

1750

1475

1400

1150

2700

2200

1825

1700

1775

The alloys 7, 8, 9 were badly cracked by the com-
pression test.

To avoid brittleness, the percentage of lead should
not exceed 15 to 18 per cent., and to obtain the best
results the alloy should contain over 10 per cent, of tin,
but it is not necessary to exceed 20 per cent.

Under the microscope, the alloys, if containing over
10 per cent, of antimony, are seen to consist of hard
crystals embedded in the tin-lead eutectic. " In alloys
of lead and antimony these grains are composed of pure
antimony ; in alloys of tin and antimony they are com-
posed of the compound Sn Sb, and it is probable that in
remaining alloys they are made up of a solid solution of
Sn Sb and Sb. It is, indeed, known that pure antimony
and the compound Sb Sn are capable of crystallising
together in alloys of tin and antimony containing more

194

WHITE ANTI-FRICTION ALLOYS.

than 50 per cent, of antimony."* The alloys have a
constitution much like the binary alloys of lead and
antimony. " The tin, however, intervenes — 1st, as a
constituent of the hard grains, diminishing their hard-
ness, but also their brittleness ; 2nd, as a constituent of
the eu tec tic alloy,- increasing its compressive strength."

Another well-known alloy, Jacana metal, contains : —

Lead 70-33

Antimony . . . . . . 18-99

Tin 10-11

Bismuth -01

FIG. 108.— JACANA METAL.
V x 100.

Hoyle's alloy contains about tin 46 per cent.,
antimony 12 per cent., lead 42 per cent.

Alloys of Zinc, Tin, and Antimony. — The structure of
these alloys is uncertain. Tin and zinc do not alloy
well, but the alloys consist of a eutectic containing about
10 per cent, of tin with crystals of the excess metal.

* Charpy, " The Metallograph^t," Vol. II., page 43.

WHITE ANTI-FRICTION ALLOYS.

195

" Alloys of zinc and antimony contain one definite
compound at least which is very hard and forms with
zinc a eutectic alloy containing about 3 per cent, of zinc."
Alloys of the three metals do not seem to form true
ternary alloys, but to consist of mixtures of the several
constituents due to the metals being united in pairs ;
some of them show a high compressive strength. Alloys
which contain free zinc have not been found to be very

FIG. 109.— ALLOY.
80 per cent. Zinc. 10 per cent. Antimony. 10 per cent. Tin. V 30 X .

satisfactory, the zinc having a great tendency to adhere
to iron when slightly heated, and the alloys are brittle
when heated.

Many other alloys have been suggested for bearing
purposes, but they are usually inferior to those already
described.

The following table, abridged from that compiled by
M. Charpy, gives some of the more important alloys,
excluding the brasses and the bronzes which have been
already considered.

196

WHITE ANTI-FRICTlto ALLOYS.

Copper.

Tin.

Lead.

Zinc.

Iron.

Anti-
mony.

f Bearings for
\ Locomotives.

—

f Quoted by
(. Ledebur.

8-3

7-6

83-3

3-8

f Beuquot White
\ Bronze used in
1 France for Naval
(.Construction.

5-6

17-5

0-7

76-2

—

C White Bronze
< used for Ship
(. Engines.

—

C Quoted by
\ Ledebur. Kneiss
(Metal.

4-01

9*91

115

85-57

—

C Salge Anti-fric-
(tion Metal.

5-5

17-5

—

{Quoted by
Ledebur for high-
speed shafts.

White Metal Patterns. — In the foundry patterns are often
made of white metal. Alloys of lead and antimony with
or without the admixture of other metals are usually
used. The following mixtures have been described as
being well suited for the purpose : —

(1) Lead . .
Antimony

(2) Lead . .
Antimony
Tin

(3) Lead . .
Antimony
Bismuth . .

Zinc, Cadmium, and Antimony. — Messrs. Siemens and
Halske have patented an alloy which they state is superior
to most anti-friction metals. It casts well, is hard, has
a low coefficient of friction, and machines well.

It consists of equal parts of zinc and cadmium with
from 5 to 10 per cent, of antimony. The antimony must not
exceed 10 per cent., and 5 per cent, is best, nor must the
relative proportions of zinc and cadmium be much varied.

Sflbs.

= 87-5 per cent.

IJlbs.

= 12-5

8lbs.

- 80-0

lib.

= 10.0

lib.

- 10-0

8lbs.

= 80-0

IJlbs.

= 12-5

ilb.

= 2-5

CHAPTER XII.
LIGHT ALLOYS AND FUSIBLE ALLOYS.

ALUMINIUM (sp.gr. 2-58) is the lightest of the metals
in common use, and it has been suggested for many
purposes where lightness is essential. The properties
of the metal, however, unfit it for purposes where it is
likely to be subjected to stress or to wear, as it is weak
and soft. It may, however, be strengthened and
hardened to some extent by the addition of foreign con-
stituents without its white colour being impaired or its
weight seriously increased. A large number of light
alloys, of which aluminium is the principal constituent,
are now upon the market. The purer the aluminium
the softer it is, and the metal now produced by the
electric processes is very pure, and therefore very soft.
Silicon, for instance, which may be present accidentally
or may be added, distinctly hardens the metal, and one
at least of the light alloys on the market is simply an
aluminium con taming a little silicon. When the amount
of silicon is over 2 per cent, the colour of the metal is
impaired, and it becomes less malleable ; but it is stated
that 5 per cent, may be present without interfering with
its use for castings. Silicon is readily taken up when
aluminium is melted in contact with silica or any
siliceous material, so that the presence of silicon may
sometimes be accidental.

The specific gravity of the alloy, of course, increases
with the percentage of copper, but the actual increase
is so small that under 10 per cent, of copper does not
seriously alter the weight of the metal, with 8 per cent,
its specific gravity being less than 2-9. The weight of
a cubic foot of the alloys is given by the Alloys Research
Committee in the table on the following page.

198

LIGHT ALLOYS AND FUSIBLE ALLOYS.

Copper.

Sand
Casting.

Chill
Casting.

Ro:led Bars,
J|in. diam.

Drawn Bars,
igin. diam.

o-oo

168

169

0-86

170

1-90

171

172

2-77

172

173

3-76

173

174

4-97

173

175

—

615

175

177

—

6-97

176

178

—

8-01

178

189

180

—

so that even with 8 per cent, of copper the alloy may
still be regarded as being a very light metal.

Aluminium-Copper Alloys. — The principal metal used
for hardening aluminium is copper, and alloys of the two
metals are largely used. Copper alloyed with a small

FIG. 110.— PROPERTIES OF ALUMINIUM ALLOYS (Chill Castings).
(Alloys Research Committee.)

proportion of aluminium (aluminium bronze) has already
been considered. The alloys with nearly equal quan-
tities of the two metals are of little use, so that only
those will be considered here which consist of aluminium
with 10 per cent, or less of copper. These have been

LIGHT ALLOYS AND FUSIBLE ALLOYS.

199

fully -investigated by the Alloys Research Committee,
and details of the work will be found in the report.

Although copper has a dark colour, it does not seem
to modify the colour of the aluminium in the slightest
when only present in small quantity ; indeed, up to 10
per cent, the alloys retain their white colour.

The tensile strength increases with the percentage of
copper, the increase in the case of chill castings being

T033

'-'0-£

FIG. ill.— PROPERTIES OF ALUMINIUM ALLOYS (Worked Bars).
(A. R. C.)

shown in Fig. 110, and for worked bars in Fig. Ill, whilst
the elongation rapidly decreases, the metal becoming
harder, less ductile, and more brittle.

With the worked bars the variation was much the
same, but the actual strengths obtained were greater,
a maximum strength of 17 tons being obtained with 3-76
per cent, of copper. The worked bars were also much
more ductile than the castings, an elongation of 17-2

200

LIGHT ALLOYS AND FUSIBLE ALLOYS.

per cent, on 2in. being given with 8 per cent, of copper.
The strength and ductility of the alloys are such that
they are not likely to be used for purposes where great
strength is required.

Aluminium is largely used for cooking utensils and
other purposes, and as the pure metal is soft it is
frequently alloyed with a small quantity of copper.
Aluminium is not attacked by pure water, but it is
attacked to some extent by salt, by weak acids, and
more strongly by alkalies, the surface becoming rough
and traces of the metal being dissolved. Experiments

FIG. 112.— ALUMINIUM ALLOY (Chill Casting) 8'08 per cent. Copper x 120.

on the corrosion of the alloys were made by the Alloys
Research Committee by boiling strips with dilute acids
for six hours. The strips were then weighed, and the
loss of weight calculated to the loss in pounds per square
foot per hour. The results are given in the following table.

Copper
per cent.

Water.

1 per cent.
Aqueous
Oxalic Acid.

1 per cent.
Aqueous
Acetic Acid.

2 per cent.
Aqueous
Citric Acid.

Nil

o-ooio

0-0002

o-oooi

1-57

0-0019

0-0002

2-36

0-0020

o-oooi

0-0002

3-74

0-0021

0-0002

4-74

0-0022

0-0002

5-34

•>•>

0-0023

0-0003

LIGHT ALLOYS AND FUSIBLE ALLOYS. 201

The amount dissolved in the case of the pure metal
is by oxalic acid 7 grains, by the aqueous acetic acid
1 • 4 grains, and by the 2 per cent, citric acid 0 • 7
grain per square foot per hour. In no case was any
trace of copper dissolved, but it is evident that addition
of copper so far from preventing corrosion rather increases
it.

Under the microscope the alloys show two con-
stituents, a ground mass of aluminium and threads of
a separated eutectic.

Aluminium and Magnesium. — Alloys of these metals under
various names have recently been put upon the market.
When either metal contains any considerable quantity of
the other the resulting alloys are brittle, but with only small
quantities of magnesium in aluminium alloys which have
useful properties are produced. The best known of these
alloys is that to which the name magnalium has been
commercially given. The alloy is lighter than aluminium,
its specific gravity varying from 2-4 to 2-57. Its
tensile strength is from 14 to 21 tons per square inch.
It casts well, is very ductile, and malleable, and there-
fore can be spun, drawn, forged, or rolled, the
forging or rolling being best carried out at a temperature
of about 300° to 350° C. It is not more corrodible than
aluminium. The alloys all contain under 2 per cent,
of magnesium, and according to the analysis of Mr.
Barrett* contain copper and tin. Three commercial
alloys, marked X, Y, Z, are made, and Mr. Barrett gives
samples of these as containing : —

X Z

Magnesium . . 1-60 .... 1-58

Copper .. .. 1-76

Nickel .. .. 1-16

Tin 3-15

Lead -72

Antimony \ Present not

Iron . . . . J estimated Present.

Also traces of titanium. The alloy Y he states to be
intermediate between X and Z.

» J. S. C. I., 1905, page 832.

202

LIGHT ALLOYS AND FUSIBLE ALLOYS.

In an article in the " Brass World," special attention
was called to these alloys, and it was stated that for
sand casting a good alloy was made by melting alu-
minium 9 parts with magnesium 1 part. Or for rolling,
aluminium 9-8 parts and magnesium '2 part.

Rapid cooling in water makes the alloys tough and
ductile, slow cooling hardens them.

The following details of tests of aluminium-magnesium
alloys are given by Mr. J. W. Richards, in a paper read
before the American Society for Testing Materials : —

2%
T.S.

«f.

4%
T.S.

ttg.
E.

6%]
T.S.

vig.

8%
T.S.

vig.
E.

10%
T.S.

Mg.
E.

Cast in sand...
„ „ chills...
Castings water
chilled
Annealed
sheet

17900
28600

40000

25600
41300

300
2-00

I -00

18-00

2-70

28600

28200
44900

2-60

8-00
2-10

57600

28100
44100

1-00

17-00
1-00

54900

1-60

21400
33600

61100

2-40
3-40

4-20

Hard sheet

—

Tensile strength in pounds per square inch.
Elongation per cent.

Aluminium Copper-zinc Alloys. — These alloys are made
commercially. An analysis of the Aluminium Company's
No. 6 alloy gave : —

Aluminium . . . . 87-7
Copper .. .. 2-8

Zinc 8-9

and it had a specific gravity of 2 • 96.

Aluminium and Nickel. — Nickel alloys with aluminium
and small quantities harden the metal very much, at the
same time making it brittle. Lejeal states that an
alloy he prepared containing 4-5 per cent, of nickel
had a coarsely-crystalline fracture, was easily worked,
rolled well, but gave poor mechanical tests, as follows : —

Lbs. per sq. in.

Elongation
per cent.

Forged cold and annealed
Forged at low red heat
Forged at weak cherry red ...

21,000

22,800
23,000

6
5-5
11*5

LIGHT ALLOYS AND FUSIBLE ALLOYS. 203

An alloy with from 7 to 10 per cent, of nickel is said
to last well, and to be tough and hard.

Richards states : " The Pittsburg Reduction Company
has recently commenced selling aluminium hardened by
a small percentage of nickel, made by adding nickel
oxide directly to the bath in which alumina is being
electrolysed. They claim for these alloys a tensile
strength in castings of 25,000lbs. to 30,000lbs. per square
inch, and in rolled bars or plates of 45,0001bs. to 50,000lbs.
A bar of this metal shown to the writer was certainly
very strong, and possessed of great elasticity, suggesting
its probable use for light wagon or carriage springs."*

These nickel alloys do not resist corrosion well.
Copper-nickel aluminium alloys are also sometimes made.
An alloy made by adding 2 to 3 per cent, of German
silver to aluminium is described by Richards as being of
a pure- white colour, strong, and quite elastic. " When
rolled hard its tensile strength exceeded 40,0001bs. with
an elongation from 3 to 5 per cent., while in casting its
strength is 22,000lbs. or 50 per cent, stronger than
aluminium. The white colour and elasticity of this alloy
commend it for many purposes where pure aluminium
is too soft and non-elastic."f It is said that an alloy of
aluminium with 4 per cent, of nickel falls to powder
at ordinary temperatures soon after being cast.

Aluminium and Iron. — Iron is a very constant im-
purity in aluminium, since any oxides that remain with
the alumina will be reduced. A small quantity of iron
darkens the colour of aluminium, hardens it considerably,
makes it less malleable, and causes it to crystallise more
readily. When the iron is present in considerable
quantity, say 10 per cent., some of the aluminium can be
liquated out. The presence of iron raises the melting
point of aluminium very considerably. With 5 per
cent, of iron Prof. Camelry found the melting point of
the alloy to be close on 700° C., whilst a specimen with
5 per cent, of iron commenced to fuse at 730° C. The
fused alloy is also much more pasty than the pure

* "Aluminium," p. 511. t "Aluminium," p. 538.

204 LIGHT ALLOYS AND FUSIBLE ALLOYS.

metal. With larger proportions of iron still the metal
becomes very crystalline. " Deville states that the
alloy containing 10 per cent, of iron has colour and
brittleness of native antimony sulphide." The only alloy
of aluminium and iron made commercially is ferro-
aluminium, which contains varying proportions of
aluminium, and which is used in steel making. The
alloy is yellowish-white, and is very hard and
brittle.

Aluminium and Tin. — Tin and aluminium alloy readily,
and when the quantity of aluminium is large and that of
tin small (about 3 per cent.) the resulting alloy is very
brittle, but with a larger quantity of tin the alloy seems
to become stronger.

" M. Bourbouze has recommended the use of an alu-
minium-tin alloy for the interior parts, especially, of
optical instruments in place of brass. The alloy formed of
100 parts aluminium to 10 of tin, or 9 per cent, tin, is recom-
mended as being the best for this purpose. It is white,
and has a specific gravity of 2-85, only slightly above
that of aluminium itself. It may therefore be used in
place of aluminium where great lightness is desired, and
it is further superior to aluminium itself in resisting
alterations better and being more easy to work, and,
finally, it can be soldered without any special apparatus
as easily as brass." An analysis of some of this metal
exhibited at the Paris Exhibition in 1889 gave

Aluminium . . . . 85-74

Tin 12-94

Silicon 1-32

100-00

A test of a similar alloy containing aluminium 88
per cent, tin 10 per cent., silicon 1-30 per cent., and
iron 0-65 per cent, gave a tensile strength of
14,000lbs. per square inch with an elongation of only
4-11 per cent. It is therefore no stronger than alu-
minium.

LIGHT ALLOYS AND FUSIBLE ALLOYS.

205

The following table of the melting points of the tin-
aluminium alloys is from Richards' "Aluminium," as
quoted from Minet.

Aluminium.

Tin.

Melting Point.

100

619° C

595°

575°

535°

575°

570°

530°

490°

100

233°

Aluminium and Zinc. — Zinc and aluminium • alloy
readily, the alloys being in general harder and more
fusible than aluminium. A very small quantity of zinc
is sufficient to make aluminium brittle. The alloys are
of no practical value. A small amount of aluminium
added to zinc is said to make the metal more fluid, and
thus in galvanising to increase the surface a given weight
of zinc will cover, and also to diminish oxidation. Alloys
with 25 to 33 per cent, of zinc are in use. That with 33
per cent, of zinc is said to be very hard and to have
a specific gravity of 3-8. With 25 per cent, zinc the
alloy gives good castings, is easily worked, and has a
specific gravity of 3-4. With about 18 per cent, of
zinc the alloys can be rolled or drawn. The use of
aluminium for this purpose has been patented in the
United States, and it is believed to be largely used.

An alloy commercially known as Ziskon contains
about 25 per cent, of zinc. It is white in colour, has a
specific gravity of 3-35, and is said to have a tensile
strength of 11 tons.

Tungsten and Aluminium — The addition of a small
quantity of tungsten to aluminium has been recommended
by Mannesmann, as improving its resistance to corrosion
and greatly increasing its strength.

206

LIGHT ALLOYS AND FUSIBLE ALLOYS.

Richards gives the following figures for an alloy with
5 per cent, of tungsten : —

T. S. Ibs. per sq. in.

Elongation
per cent.

Cast

22000

1-5

Rolled hard
Annealed
??

35,000
25,000
22,000

4-0

10-0

14-0

The alloy known as Wolframium belongs to this
class. An analysis by Mr. J. C. S. Jones gave aluminium
99 • 4 per cent., tungsten 0 • 1 per cent.

FUSIBLE ALLOYS.

As is well known, alloys usually melt at a temperature
which is a good deal below the mean melting point of
their constituents, and in some cases below the melting
point of their most fusible constituent. When easily-
fusible metals are used, very fusible alloys known as
fusible metals are produced.

The fusible metals are essentially alloys of bismuth
and tin, though generally other metals such as lead and
cadmium are also present.

Bismuth and tin alloy very readily, and seem to
form definite compounds. A small quantity of bismuth
added to tin makes it harder and more sonorous, and
increases its lustre and fusibility.

When the metals are in the proportions required for
the formula Sn3 Bi2 the alloy has only one solidifying
point at 143° C., and this may be taken as being the
melting point of that compound. With other alloys there
are always more solidifying points, the lowest being
143° C. Several definite compounds of tin and bismuth
are said to exist, -and Hiorns gives the following table : —

Formula.

Tin.

Bismuth.

Freezing Points.

Sn3Bi2

45-73

54-27

143°C.

Sn4Bi2

69-21

30-79

190°

Sn2Bi

52-91

47-09

160°

Sn3Bi3

27-25

72-72

170°

SnBi2

21-93

78-07

190°

LIGHT ALLOYS AND FUSIBLE ALLOYS. 207

When lead is added the alloys become more fusible,
and the triple alloy of the three metals is the base of
most of the fusible alloys.

The most fusible alloy is usually stated to contain
25 per cent, lead, 25 per cent, tin, and 50 per
cent, bismuth, and to melt at about 94° C., so that
it will melt in boiling water. The behaviour of the
alloy when heated is said to be very anomalous.
" It expands regularly from 32° C. to 95° C., and
then contracts gradually to 131° C. at which point it
occupies less bulk than it did at 32° C. ; it then expands
till it reaches 176° C.,and from that point the expansion
is uniform." This alloy is used for taking impressions
from dies, &c. It passes through a long pasty stage,
during which it is quite soft, and thus can be used like
sealing-wax for taking sharp impressions. This is
partially due to the separation of a solid portion of the
alloy which is retained in a still liquid eutectic.

Another curious phenomena in connection with this
alloy is an evolution of heat after solidification. If it is
cooled in cold water, and then left, it becomes hot again.
This is due to some change in the alloy which is attended
with evolution of heat. If the alloy be broken before
the evolution of heat the fracture is almost vitreous, but
afterwards becomes "grey, dull, and fine grained." In
the vitreous state the tensile strength is about 1 ton per
square inch, but after the molecular change it is about
2| tons per square inch. By a pressure of about 4 tons
on the square inch the thermal change is prevented.

M. Charpy gives the most fusible alloy as containing
32 per cent, lead, 16 per cent, tin, and 52 per cent, of
bismuth, which fuses at 96°. This is probably the alloy
of eutectic composition. An alloy of 40 per cent, bis-
muth, 20 per cent, lead, and 40 per cent, tin softens at
100° C., and can be kneaded between the fingers.

The following table, by Messrs. Parkes and Martin, of
alloys with definite melting points that can be used for
tempering steel, may be of interest. The results are
calculated into percentages from the published table : —

208

LIGHT ALLOYS AND FUSIBLE ALLOYS.

Bismuth.

Lead.

Tin.

M.P.
Deg. C.

Bismuth.

Lead.

Tin.

M.P.
Deg. C.

31.2

18-8

16.6

33-2

50-2

158

35.5

17-7

155

42-1

15-8

108

15-3

38-8

45-9

154

113

14-8

40-2

153

36-5

117

154

33-3 33-3

33-3

123

13-7

44-8

41-5

160

30-8 38-4

30-8

130

13-3

46-6

40-1

165

28-5 43

28-5

132

12-8

38-2

172

25 50

149

12.5

37.5

178

23.5

29-5

151

11-7

46-8

41.5

167

22-2

44-4

33-4

143

11.4

45-6

165

143

11-2

44.4

44-4

160

145

10-8

43-2

159

148

10.5

47.5

160

18.1

36-2

45-7

151

10-2

49-8

161

17-3

34-6

48-1

155

162

The addition of cadmium gives alloys of still lower
melting point.

The following table includes the best known of
these : —

Cadmium.

Lead.

Tin.

Bismuth.

Melting
Point.

Lipowitzs' alloy...

10-

26-6

13-3

50-1

70° C..

Fusible alloy

6-2

34-5

9-3

50-

77° C.

55 55

34-5

27-5

10-

27-5

75° C.

55 55

16-6

—

33*3

50-1

95° a

55 55

11-1

—

33-3

55-6

95° C.

55 55 * ' '

25-

—

25-

50'

95° C.

55 )5 ' ' '

12-5

25-

12-5

50-

65° C,

Woods' alloy

15-4

30-8

15-4

38-4

71° C.

Fusible alloy

25-

50-

—

86° C.

CHAPTER XIII.
NICKEL ALLOYS.

NICKEL is one of the comparatively rare metals, which
has only recently come into use, but its use has largely
increased of late, as it has been cheapened by the
discovery of new sources of supply and improvements
in the methods of production.

It is white — almost silver white — in colour, malleable
and ductile. In its properties it generally closely
resembles iron, but there is one difference of very great
practical importance — that is, it does not rust on exposure
to moist air ; indeed, apart from the noble metals, it is
the most stable metal in use. Nickel melts at about
1,450° C. (2,642° Fah.), and is therefore but little less
fusible than carbon free iron. Its high melting point
is a difficulty in its use for many purposes.

Nickel is largely used for the manufacture of small
articles, where resistance to oxidation is important. It
has been used for coinage, but the most of the so-called
nickel coins are alloys. It is frequently electrolytically
deposited to give a bright metallic surface, either for
ornamental purposes or to protect the under-lying metal
from rust.

Cube nickel is obtained by moulding nickel oxide
with some reducing agent into cubes, and then heating
these to a temperature at which reduction takes place,
but below the melting point of the nickel. The cubes are
more or less porous, and are often impure, containing
carbon and other disseminated impurities. For melting
purposes, however, the cubes are quite suitable, as the
intermixed impurities are separated.

Nickel does not cast well, as it absorbs gas (CO),
which is given out on cooling, causing holes in the
castings. This difficulty is now overcome by the
addition of a small quantity of magnesium. The amount

210

NICKEL ALLOYS.

originally added by Fleitman, who discovered its action,
was J per cent., but much less is frequently used. The
magnesium seems to have much the same effect that
aluminium has in the case of steel. A small quantity
of magnesium always remains in the metal. Some other
metals have a similar effect, and the use of aluminium
and phosphorus have been suggested in place of mag-
nesium.

The following analyses of commercial nickel will
indicate the impurities that are likely to be present,
and the degree of purity to be expected.

Copper

0-41

o-io

Iron

0-62

0-464

0-108

0-36

Sulphur

0-24

0-049

0-266

Silicon

0-303

0-130

0-06

Silica

1-41

Carbon

0*65

0-530

1-104

Magnesium
Nickel and Cobalt

0-11
99-93

(1) Nickel cube. (2) Cast nickel (Thurston). (3) Cast nickel
(Thurston). (4) Nickel cast with magnesium.

COPPEE-NICKEL ALLOYS.

Copper and nickel alloy readily in all proportions.
The alloys show under the microscope a crystalline
structure which varies very little with variations in
the composition of the alloy, at least until the
percentage of nickel approaches 80, after which the
structure changes. In all probability the metals are
soluble, the one in the other, in all proportions in the
solid condition, so that the alloys are non-eutectiferous.
At present the structure of these alloys has not been
fully investigated. As nickel is added to copper, the
colour is much more slowly destroyed than in the case
of some other metals. The alloy with 20 per cent, of
nickel is distinctly red, and with 30 per cent, the coppery
colour is still distinguishable, but the colour dis-
appears as the nickel approaches 40 per cent., and the
alloy becomes silver white, and continues unchanged in

NICKEL ALLOYS.

211

FIG. 113.— COPPER NICKEL ALLOY. Cu. 49 per cent. V x 50 diameters.

FIG. 114.— COPPER NICKEL ALLOY. Cu. 80 per cent. V x 50 diameters.

212 NICKEL ALLOYS.

colour till the nickel reaches about 80 per cent., after
which the colour darkens considerably.

Copper-nickel alloys are rarely used except in some
cases for coinage, but it is doubtful if such alloys have
any advantage over pure nickel. The United States
coinage contains 75 per cent, copper and 25 per cent,
nickel.

German Silver — Under the name, German silver,
are included a large number of alloys, containing copper,
nickel, and zinc, sometimes with the addition of other
metals which are extensively used for various purposes.
As might be inferred from the name, the alloys are
always white, and are sometimes used as imitations of
silver. Before the introduction of electro-plating, forks,
spoons, and other similar articles were made to resemble
silver as closely as possible by using a white nickel alloy,
and even now, when the articles are electro-plated, it is
important that the basis metal should be as white as
possible, so that the colour may not show conspicuously
when the plating wears off. German silver is
largely used for the manufacture of the so-called
nickel coins used in many parts of the world. A
large number of white alloys, used for the manufacture
of forks, spoons, &c., are on the market, but almost all
of these are simply German silver, the name given being
merely a trade designation, not implying any special
composition. In some cases the alloys contain other
metals in small quantity which are supposed to improve
the colour.

The properties required in a German silver are, of
course, very different from those required in an alloy used
for structural purposes. Generally, the alloys will be of
two classes ; the one to be used only for castings will
require to cast well ; the other, to be used for the manu-
facture of articles by working, stamping, spinning, or
otherwise, must be malleable and ductile and must flow
sufficiently readily to allow of ready shaping. Coinage
alloys must be of this class, as the coins are struck by
a die. When the alloy is to be subject to wear, as in
the case of coin 3, it should be as hard as is compatible

NICKEL ALLOYS. 213

with the necessary flow, and all the alloys must be
white.

Where a very white colour is required, the alloy
should contain at least 25 per cent, of nickel, but as
such an alloy is costly, the proportion is often much
less, and therefore when the articles are to be electro-
plated the basis metal is often distinctly yellow, in
fact, is often " little better than brass."

The addition of zinc to the alloy cheapens it, zinc
being much cheaper than either of the other metals, lowers
the melting point of the alloy, makes it whiter and enables
it to take sharper castings. At the same time, it tends
to harden the metal, and to make it more brittle ; but
when the constituents are in suitable proportions the
alloy will roll and work well. Mr. Hiorns states that about
30 per cent, of zinc, with less than double that amount of
copper, gives the best results as to malleability and
whiteness ; 32 per cent, of zinc makes the alloy more
brittle and requires more frequent annealing during
the rolling process.

Mr. Sperry has found that the addition of a small
quantity of aluminium to a German silver makes it much
more fluid, so that it casts better ; the castings are
sound, and do not adhere to sand, so that it can be readily
cast in sand moulds, and also that the alloy is whiter.
The quantity of aluminium used may be from -25 per
cent, upwards, the properties of the alloy varying with
the quantity of aluminium. It becomes tougher as the
aluminium is increased, i.e., it reaches 3J per cent, after
which the alloy tends to become brittle.

When the aluminium reaches about 3 per cent, an
alloy is obtained which Mr. Sperry says " is quite stiff and
strong, and will only bend slightly without breaking; casts
free from pinholes, blowholes, and other imperfections ;
gives castings true to pattern ; the cost of casting is not
more than that of brass ; is non-corrodible and compara-
tively non-tarnishing ; the colour is silver white, and it is
hard enough to take a high polish."

214 NICKEL ALLOYS.

The best composition for the alloy is

Copper . . . . 57 -00
Mckel .. ... 20-00

Zinc 20-00

Aluminium . . . . 3-00

If required to be very stiff, the aluminium may
be increased by J per cent. The nickel and copper,
which should be pure, are melted together under char-
coal, taking care to see that the surface is well covered;
the aluminium is then added. When it is melted,
stir vigorously. The temperature of the mixture rises
considerably. The mass is again stirred, best with a
plumbago stirrer, allowed to cool somewhat ; the zinc
is added as usual. The metal is then poured.

This alloy is sometimes called aluminium silver.

Of other metals that may be added :—

Iron in small quantity " makes the metal whiter,
increases the tenacity, but makes it harder." Mr. Hiorns
found " 1 to 2 per cent, of iron to have no deteriorating
effect, except with regard to hardness, and the colour
of an alloy containing 12 per cent, of nickel was equal
to one containing 16 per cent, when no iron was present,
the same quantity of zinc being used in each case."
Iron, therefore, may be regarded as not being deleterious
when the alloy is to be cast, but as being objectionable,
except in very small quantities, when a very malleable
alloy is required.

Tin is very injurious, giving the metal a decidedly
yellow colour, and tending to make it brittle.

Silver has been sometimes added, the idea being that
it would improve the colour of the alloy. This, however,
does not seem to be the case, but a little silver does not
impair its properties. Alloys of copper, nickel, and
silver, containing 20 to 30 per cent, of silver, are said to
be used in the manufacture of jewellery and to resemble
silver very closely.

Cobalt — This metal is very like nickel, but it seems
to darken the colour of the alloy.

NICKEL ALLOYS. 215

Tungsten is sometimes added to German silver to
form an alloy called platino d, which has a very high
resistance, and is used for electrical work. The amount
of tungsten added is 1 to 2 per cent.

Varieties of German Silver. — Mr. Hiorns gives the
following as being the composition of varieties of German
silver used by the best makers as used in Birmingham : —

Percentage Composition.
Name. Nickel. Copper. Zinc.

Extra White Metal. . 30 50 20

White Metal ... .. 24 54 22

Arguzoid 20-5 48-5 31

Best Best 21 50 29

Firsts of Best .. ..16 56 2s

Special Firsts .. ..17 56 27

Seconds 14 62 24

Thirds 12 56 32

Special Thirds .. .. 11 56-5 32-5

Fourths 10 55 35

Fifths for Plated Goods ..7 57 36

He also gives the following as being three qualities'
made by the same maker.

No. Nickel. Copper. Zinc. Iron.

1 8-2 66-0 25-3 5

2 16-0 59-2 23-8 1-0

3 20-7 55 23-3 1-0

As examples of the alloys used for coinage the
following analyses may be quoted : —

France. Belgium. Switzerland.

Nickel 96-5 26 26*3

Copper 74 73 -9

Zinc 3-5

But little has been published on the microstructure
of these alloys, but the microscope seems to give little
information. Alloys containing zinc differ very little
from those containing no zinc, at anyrate when the
percentage of nickel is fairly high, the structure showing
a dark network on a light ground. It seems as if the
alloy is simply a solid solution of the three metals, and
therefore that the zinc simply replaces a portion of the
nickel. When the alloys are worked the structure

216

NICKEL ALLOYS.

becomes highly crystalline, and the network structure
completely disappears.

Other Nickel Alloys — The most important alloy of
nickel is that with iron, which constitutes nickel steel,
but the iron alloys are not being considered in this
book.

FIG. 115.— COPPEB NICKEL ZINC ALLOY. Cu. 53, Zn. 19, Ni. 28. Vx50 diameters.

A few complex alloys are made, but these usually
consist of German silver, with the addition of small
quantities of foreign metals, antimony, tin, lead, and iron
being among the metals added. Alloys of nickel and
aluminium are described on page 202.

Soldering Nickel Alloys — German silver articles are

soldered by means of similar but more fusible alloys

than those to be united — that is, usually containing
more zinc.

CHAPTER XIV.

ALLOYS OF THE PRECIOUS METALS.
GOLD ALLOYS.

GOLD is the most valuable of the metals in common use,
its value depending partly on its properties and partly
on its scarcity. It cannot be called a rare metal, since
the annual output is considerable, but this does not
more than meet the demand.

The properties which render gold valuable are : Its
colour, which is unique among the metals ; its malleability,
which allows of its being hammered out into the thinnest
leaves; and its durability, that is, its power of resisting
the ordinary destructive agents to which the other metals
yield. It does not oxidise in air, wet or dry, at ordinary
temperatures, nor is it oxidised at a red heat, and it
resists most corroding agents. It is not dissolved by
any single acid, but is attacked by a mixture of nitric
and hydrochloric acid (hence called aqua-regia) on account
of the chlorine which is evolved. Its specific gravity
is about 19* 4. It melts at about 1060° C.. and is non-
volatile.

Gold is mainly used for two purposes — (1) for jewellery
and other ornamental purposes, including, of course,
the coating of articles of other material by gold in the
form of gold leaf, or by electro-deposition; and (2) for
coinage, gold now being the standard in most countries.

For either purpose pure gold is too soft, as it would
wear away too rapidly, and to overcome this difficulty
it is alloyed with some other metal in such proportions
that the colour and malleability will not be seriously
impaired, but the hardness will be considerably increased.
The alloying metal — technically called the alloy — is
almost always either copper or silver, generally the
former. It will be noticed that in speaking of gold
alloys the term alloy is somewhat ambiguous, as it may
mean either the resulting alloy of the two metals or

218 ALLOYS OF THE PRECIOUS METALS.

the foreign metal added to the gold. This, however,
will not lead to confusion if the double meaning be
always borne in mind.

The amount of gold in a gold alloy is not usually
expressed in percentages, but in parts in a thousand, this
being called the fineness of the gold. A gold alloy con-
taining 98-5 per cent, of gold would therefore be said
to be 985 fine.

In alloys to be used for jewellery or coinage still
another method of expression is used. Gold 1,000 fine
is said to be fine or pure gold, or 24 carat, and the value
of alloys is expressed by the number of parts in 24 which
are gold. Thus 24 carat is f f or 100 per cent. gold. In
22-carat gold f f or ij or 9T6 per cent, of the alloy is gold.

The gold alloys used for jewellery are : —

Carat. Gold. Per cenc. Fineness.

22 ff = 11 = 91-6 916-6
18 if = | = 75 750-0

16 if = f = 66-6 666-6

14 it = i72 = 58'3 583'3
9 ^ = | = 37 5 375-0

It will be noticed that the richness of the alloy is
stated as depending only on the quantity of gold, and as
being independent of the nature of the alloying metal.

It is, of course, impossible for a purchaser to have an
article of jewellery assayed, and so to avoid fraud, the
possibilities of which are, it will be seen, considerable,
assay offices are established in the various large towns
where jewellery is made, and the article can be assayed
and stamped before it is quite completed, so that it may
not be damaged by the removal of the necessary portion
for the assay. The mark, which is called the " Hall-
mark," is only put on articles of 9, 16, 18, or 22 carat in
this country, but in other countries alloys of other
values are stamped.

Twenty-two carat gold is called standard gold, and
it is the gold of which our gold coins are made. That
is, a sovereign contains 91-6 per cent, of gold, or it is
916 fine.

Since the sovereign is simply a weight of gold the
price of gold as measured in sovereigns can never fluctuate.

ALLOYS OF THE PRECIOUS METALS.

219

When it is said that the value of gold has gone up it
simply means that its purchasing power for other com-
modities has increased — that is, that prices have fallen.
This is in general the result of gold being scarce. When,
on the other hand, gold is very plentiful its price falls —
that is, its purchasing power diminishes or prices rise.

The following table, slightly altered from Streeter's
" Gold," p. 138, gives the value of loz. of gold of any degree
of fineness, and the amount of foreign metal to be added
to pure gold to make 24 parts of the alloy : —

Quality.

Fineness.

Value.
£ Sterling per
Ounce.

Value.
Dollars per
Ounce.

Alloy
to be
Added.
Paits.

1000

£ s. d.

450

20-68

23
22

958-3
916-6

4 1 5J 19-82
3 17 11 18-95

21
20

875-0
833-3

3 14 4J
3 10 10

18-09
17-23

3
4

791 5

3 7 3}

16-37

750-0

339 15-51

17
16

708-3
666-6

3 0 2J

2 16 8

14-65
13-78

7
8

15
14

625-0
583-3

2 13 H

2 9 7~

12-92
12-06

9
10

13
12

541-6
500-0

2 6 OJ
226

11-2
10-34

11
12

11
10

458-3
416-6

1 18 11J
1 15 5

9-47
8-61

13
14

375-0

1 11 10£

7-84

333-3

1 8 4~

6-89

291-6

1 4 9*

5-83

250-0

1 1 3"

5-17

5
4

208 3
166 6

0 17 8J
0 14 2

4-3
3-44

19
20

3
2

125 0 ;
83 3

0 10 7J
0 7 1

2-58
1-72

21
22

41-6 03 6J

•86

The English sovereign weighs 123-27447 grains, and
remains legal tender till it is reduced to 122-5 grains.

220 ALLOYS OF THE PRECIOUS METALS.

These figures give the actual mint value of the gold in
the alloy. The price at which the gold can be purchased
will be always a little higher, up to Is. an ounce, according to
the amount purchased, and the alloying metal, especially
if it is silver, will be of some value, and this must be
allowed for.

In the reports of mines the value of the bullion
obtained is often stated in £ or S. It will be seen that
this at once gives the fineness, since the value is fixed.

Of course, in most articles of jewellery the value of
the gold used is small compared with the total value,
which is due to the labour put upon it in finishing it into
the required form.

The gold coinage of other countries is not of the
same standard as the British, so that the coins cannot
always be compared weight for weight.

Fineness. Carats.

Hungarian Ducats 989 . . 23 • 76

Austrian „ 986 ..23-6

Dutch „ 982 ... 23-75

English, Portuguese, Turkish, Brazilian 916-6 .. 22
German, French, Belgian, Italian,

Swiss, Spanish, Greek, United

States, and Chinese gold coins.. 900 .. 21-6

Old German coins (pistoles) .. 895 .. 21-5

Egyptian, Mexican, Spanish . . 875 . . 21

The alloying metal may either be copper or silver,
or a mixture of the two, which of the three is used being
a matter of small importance, as the value of the alloying
metal in any case is very small compared with the value
of the gold. When silver is used, as in some of the
Australian coins, the alloy is paler in colour than the British
coins, in which copper is used. In jewellery generally a
mixture of silver and copper in approximately equal
quantities is used.

GOLD AND SILVER ALLOYS.

Gold and silver alloy readily in all proportions, and
the alloys differ in some respects from those which have
already been considered.

Whatever be the composition of the alloy, it has only
one freezing point, or rather perhaps it should be said

ALLOYS OF THE PRECIOUS METALS. 221

freezing range. There is no second freezing point ;
that is, there is no definite eutectic separated. As silver
is added to gold the freezing point is lowered, very
little at first, then more rapidly, but with a continuous
more or less steady fall from the freezing point of gold
to that of silver. It does not, of course, follow that the
alloy will be perfectly homogeneous, for there still
may be selective freezing, the mother liquor being richer in
the one metal than the solidified portion ; but in these
alloys there seems to be little if any segregation, and
therefore, as Sir W. Roberts Austen pointed out, they
are specially well suited for making trial plates for mint
use.

Under the microscope these alloys are seen to be
highly crystalline ; but, as might be expected, show no
sign of the formation of a eutectic. They are, in fact,
non-eutectiferous through the whole range.

The electric conductivity curve has the characteristic
U form, falling to a minimum when the metals are
present in about equal quantities, then rising again ;
but there is no distinct break, such as would indicate a
critical point.

The colour of gold is rapidly destroyed by the
addition of silver — the colour becoming paler, the Austra-
lian gold coins, in which the alloying metal is silver,
being distinctly paler in colour than pure gold. Gold
containing silver is often spoken of as pale gold. The
change of colour begins apparently when there is about
5 per cent, of silver present ; the colour becomes paler
and assumes a greenish tint when the silver approaches
30 per cent., and when it reaches 50 per cent, the alloy
is white, and is sometimes called electrum, though the
alloy usually known by this name is simply a German
silver.

Alloys of gold and silver containing small quantities
of silver are not attacked by dilute nitric acid. As the per-
centage of silver increases this metal is partially dissolved
out, and when the percentage of gold falls to about 30
the silver is completely dissolved by warm nitric acid,
the gold being left insoluble. Advantage is taken of
this in the parting of gold and silver on the large scale
and in assaying. The ordinary jewellers' test for goldr

222

ALLOYS OF THE PRECIOUS METALS.

as is well known, is to treat its surface with a drop of
strong nitric acid, by which gold of high carat is not
attacked at all, poorer golds are slightly attacked, and
most of the gold-like alloys, brasses, &c., are rapidly
dissolved.

Silver hardens gold and makes it more sonorous,
but it does not interfere with its malleability, so that
it can still be used for the manufacture of coins and
other articles which have to be struck by a die.

GOLD AND COPPER.

Gold and copper also alloy very readily. The in-
fluence on the colour of the gold is, as might be expected,
much less than that of silver, the golds alloyed with
copper being redder than those alloyed with silver only,
so that the colour of an alloy of gold can be modified
by varying the alloying metal. In some respects the

1110

1080

1)1040

1020
lOOO

980
960

940
920

900

0 10 20 30 40 50 60 70 80 90 100

Percentage of Gold.
FIG. 116.— FREEZING POINT CURVE or COPPER-GOLD ALLOYS.

ALLOYS OF THE PRECIOUS METALS. 223

gold-copper alloys resemble those of gold and silver, and
it was at one time thought that the metals did not
segregate. Sir W. Roberts Austen has shown that this
is not the case, that whilst segregation is not as marked
as in many other alloys it certainly does take place to a
sufficient extent to render trial plates cast of such an
alloy unhomogeneous.

When copper is added to gold (melting point
1,063°) the melting point of the alloy rapidly falls till it
reaches a minimum, when the number of atoms of
gold is 59-69 per 100 of alloy, that is, when the alloy
contains 82-05 per cent, of gold, when it is 905° C.
Beyond that the temperature rises as the percentage
of copper is increased till the melting point of
copper, 1,083°, is reached. Copper apparently dissolves
in gold, but it is not soluble in the solid condition in all
proportions as is silver. When an alloy containing only
a small proportion of copper is examined, e.g., standard
gold, it is found to consist of definite crystals differing
from those of pure gold only in colour, and no second
constituent can be made out. This may therefore be
regarded as being a solid solution of gold in copper. As
the quantity of copper increases a second constituent
makes its appearance when there is about 27 per cent,
of gold present, and when about 82 per cent, of gold
is reached the alloy has a true eutectic structure. As
the quantity of copper is increased the structure shows
crystals of copper containing gold in solution embedded
in the eutectic.

In the case of the copper-gold series Sir W. Roberts
Austen points out that the eutectic is weak and brittle,
so that where strength is required the alloy must be
some distance on either side of it. Sir W. Roberts
Austen has also shown that the alloy with 82 per cent,
gold, i.e., the eutectic, is the only one that has a definite
freezing point, and in which therefore the temperature
remains constant till the solidification is complete. It
is impossible to say exactly how far the eutectic line
extends in his diagram. Prof. Roberts Austen only
shows it in the copper direction, and he says that the
eutectic makes its appearance before the gold reaches 27
per cent.

224 ALLOYS OF THE PRECIOUS METALS.

JEWELLERY GOLD.

Gold for jewellery consists of gold alloyed with silver
or copper, or more usually both, and the properties of
these alloys are therefore similar to those already
described. It will be quite obvious that a gold may have

FIG. 117.— STANDARD GOLD x 4'5 DIAMS. (ROBERTS AUSTEN).

a definite fineness, i.e., may contain a certain amount of
gold, and yet its properties, colour, &c., may vary as
the proportion of the alloying metals are varied. Mr.
Hiorns quotes the following table from Gee's " Goldsmiths*
Handbook " as the proportion of metals used in various
jewellers' alloys : —

Carat.

Copper.

Silver.

Gold.

-5

8-5

3-5

10-5

4-5

10-5

5-5

ALLOYS OF THE PRECIOUS METALS. 225

It is obvious that as the quantity of alloying metal
increases the value of the silver will increase so that it
will be of importance to reduce it, also if silver be used
alone the alloy will become paler and paler. A 10-carat
gold, for instance, in which the alloying metal was all
silver would be quite white.

The colour of a gold of given fineness may be varied
very much by varying the proportions. Thus, taking 15
carat as a type :— Gold. Silver. Copper.

Red gold contains ..15 6 8-4

Green gold contains ..15 9

Gold may be coloured in other ways, as, for instance,
by dissolving away some of the base metal by means of
solvents, preferably assisted by an electric current, thus

FIG. 118.— EUTECTIC (80 PER CENT. GOLD, 20 PER CENT. COPPKR) x 1,580 DIAMS.
(ROBERTS AUSTEN).

leaving a surface of a different composition and richer in
gold than the bulk of the alloy, or by heating the gold
with some substance which will have the same effect. If
the gold is too poor, say below 13 carat, the surface left
is too rough owing to the large amount of alloy dis-
solved away.

A little zinc is sometimes added to poor golds, the
zinc usually being added in the form of brass. The
colour of the alloy is darker than when silver is used.
Zinc-golds are brittle and difficult to manipulate, but
'occasionally as much as 15 per cent, of zinc is present.

226 ALLOYS OF THE PRECIOUS METALS.

OTHER ALLOYS.

Gold alloys well with other metals, but the alloys
formed are of no importance. Lead dissolves gold in all
proportions, and is used as a solvent for gold in certain
metallurgical operations.

Some metals, especially antimony, bismuth, and
arsenic, when present even in minute quantities, make
the alloys so brittle that they are useless for coinage
purposes.

Iron combines readily with gold, and is occasionally
added to gold for ornamental purposes, as it modifies
the colour of the alloy. *

The alloys of gold and aluminium are of no importance,
but they are of some interest. An alloy of gold and
aluminium having the formula Au A12, and containing
therefore 78' 48 per cent, of gold, was discovered by Sir
W. Roberts Austen, and has an intense red colour. A
small quantity of aluminium is said to greatly improve
the soundness of gold castings.

GOLD SOLDERS.

A solder must have as nearly as possible the colour
of the metal to be soldered. The solders are of two
kinds — soft solders which melt at a very low temperature,
and hard solders which melt at a temperature but little
below that of the metals being united.

The hard solders used for gold are gold alloys con-
taining either a little more of the more fusible constituent
or a more fusible metal, and the more fusible it is the
softer is the solder said to be.

Gee gives the following table of solders : —

Fine Gold. Fine Si'vjr. Copper.

Best solder .. ..12-5 45 3

Medium solder . . . . 10 6 4

Common solders . . 8-5 65 5

The solders for use are usually rolled out into thin
sheets, and cut into pieces of suitable size for use. f)

The table on the following page is from Mr. Hiorns'
" Mixed Metals," the figures having been calculated into
percentages.

ALLOYS OF THE PRECIOUS METALS.

227

Gold.

Silver.

Copper.

Zinc.

Hard solder for gold

62-5

31-2

6-3

Hard solder for 16-

carat gold . .

16-6

8-3

—

Easier

54-5

31-9

13-6

—

Solder for 14-ct. gold

33-3

16-6

—

55 55 55 55

66-8

16-6

—

„ for less than

14 carat

—

55 J5 55

33-35

66-65

—

55 55 55

33-35

—

66-65

—

Very easy solder . .

11-54

54-74

28-17

5-55

SILVER ALLOYS.

Silver is another of the metals usually regarded as a
precious metal, and which is used mainly for ornamental
purposes, and for coinage^. Silver has a pure silver-white
colour, and it does not oxidise either at ordinary tem-
peratures or at a red heat, but it tarnishes very rapidly
in presence of traces of sulphur compounds such as are
always present in the atmosphere of towns. Hence it
is not very suitable for ornamental purposes under such
conditions. It is much more abundant and much
cheaper than gold. It is largely used for the subsidiary
coinage in gold-using countries, and for the standard
coinage in many countries, such as India and most of the
South American Republics, where it has not been replaced
by gold. As the ratio of value between gold and silver
is constantly fluctuating, the standards of value between
gold-using and silver-using countries must vary from time
to time.

Pure silver is very soft, and is therefore always
alloyed with some other metal before use.

Silver-Copper Alloys — These are the most important of all
the silver alloys, and they are generally used for the
silver coinage of all countries.

Silver and copper alloy readily in all proportions,,
the alloy expanding in formation, so that the specific
gravity of the alloy is less than that calculated from the
proportions of its constituents.

228

ALLOYS OF THE PRECIOUS METALS.

When copper is added to silver, the freezing point
falls rapidly, and reaches a minimum when the alloy
has the composition Ag3 Cu2, and contains therefore

Temperature, degrees C.

>— ' i— i

O O O O O 0 O

*<z

-T> .

<Vy

Eutt

>ctic Alloy

Copper %0 10 20 30 40 50 60 70 80 90 100

Silver%100 90 80 70 60 50 40 30 20 10 0

FIG. 119. — CURVES OF FUSIBILITY OF SILVER AND COPPER ALLOYS.

71 893 per cent, of silver and 28 107 per cent, of copper,
and then the freezing point rises again rapidly. The
alloy of lowest freezing point is a true eutectic, and it is

FIG. 120. — COPPER, 28 PER CENT.; SILVER, 72 PER CENT. (CAST).
REHEATED TO A PURPLE COLOUR. MAGNIFIED 1,000 DIAMS.

ALLOYS OF THE PRECIOUS METALS.

229

the only member of the series which has a definite
melting point. It was discovered by Levol to solidify
without liquation, and is therefore known as Levol's
homogeneous alloy.

At each end of the series the alloy consists of a solid
solution of copper in silver or of silver in copper, but a&
the richness approaches the eutectic point there are two
distinct freezing points, so that the alloy is for a certain

FIG. 121. — COPPER, 15 PEK CENT.; SILVER, 85 PER CENT. (CAST).
REHEATED TO A PURPLE COLOUR. MAGNIFIED 600 DIAMS.

range at least truly eutectiferous, and all alloys except
the eutectic will be subject to more or less liquation.

The addition of copper hardly changes the colour
of the alloy until the quantity is considerable ; indeed,
the alloy retains its white colour till the copper reaches
50 per cent. The colour then becomes yellowish till it
is about 70 per cent., and then reddish. The addition
of copper to silver hardens the metal, and makes it more
sonorous. The hardest alloy is that which contains about
one-third its weight of silver.

The only alloys of any commercial importance are
those near the silver end of the series, which are used
for coinage and ornamental purposes.

230 ALLOYS OF THE PRECIOUS METALS.

Standard Silver — Silver coins and articles which are
to be hall-marked are in every country made of some
definite standard. In Great Britain the standard is
925, i.e., it contains 92-5 per cent, of silver, the alloying
metal being always copper. This is commonly called
sterling silver, and no other quality is hall-marked.
The colour is silver white, but it is harder than pure
silver.

An alloy of the standard fineness contains 11 oz.
2 dwts. of silver to the pound troy (J2 oz.). In actual
manufacture the makers take care to add a little more,
because if the alloy falls below the standard the article
is not only not marked, but is destroyed. When articles
are not to be hall-marked, a poorer alloy may, of course,
be used.

The following alloys are used for manufacturing
purposes : —

Silver. Copper.

1 90 ..10

2 80 . . 20

3 75 ..25

4 70 ..30

5 66-5 .. 33-5

6 65 . . 35

7 62-5 .. 36-5

8 60 . . 40

If the copper be increased beyond 40 per cent, the
alloy has a yellowish colour.

The value of silver alloys is often stated in penny-
weights per ounce. As there are 20 pennyweights in an
ounce troy the percentage divided by 5 will always give
the value in pennyweights per ounce. Thus 90 per cent,
silver is 90 -r 5 = 18 dwts. per ounce.

In Germany Mr. Hiorns states that four silver standards
are used.

Silverware . . .11 ozs. 8 dwts. per Ib. or 950 parts silver in 1,000
Coinage ...10 ozs. 16 dwts. ,, 900 ,, ,,

Silverware... 9 ozs. 12 dwts. ,, 800 ,, ,,

. 9oza. Odwts. 750

ALLOYS OF THE PRECIOUS METALS. 231

Coinage Alloys. — As in the case of gold alloys, the
standard used in different countries varies.

Countries. Fineness.

Netherlands . . . . . . . . . . 945

Great Britain, Australia, South Africa,
Canada, Newfoundland . . . . . . 925

East Indies, Burmah, Ceylon, Mauritius,

Brazil, Portugal 911-67

Mexico 902-7

France, Belgium, Switzerland, Italy, Greece,
Servia, Roumania, Austria, Hungary,
Spain, Argentine Republic, Bolivia, Chili,
Peru, Germany, Egypt, Persia, United
States, Japan . . . . . . . . 900

Russia 868-06

Bulgaria 835

Turkey 830

Denmark, Sweden, Norway . . . . . . 800

Silver coins are not kept as closely to the standard
as are gold coins, nor is it necessary, since the coins are
really only tokens, the actual value of the metal being
far less than the nominal or face value.

In a few cases other metals have been added in
addition to the copper, but the advantage of this is
very doubtful. Swiss coins are said to contain silver,
zinc, copper, and nickel, and Mr. Hiorns gives the following
table :—

20 Centimes. 10 Centimes. 5 Centimes,

Silver 15 . . 10 . . 5

Copper 20 . . 55 . . 60

Nickel 25 . . 25 . . 25

Zinc 10 .. 10 .. 10

and such alloys have been used for ornamental purposes.

Silver alloys readily with other metals, but the alloys
are of no commercial importance. Attempts have been
made to use aluminium-silver alloys, but they have not
so far been successful. The alloys are harder than silver,
white in colour, do not tarnish in air, and are malleable
provided the amount of aluminium does not exceed 10
per cent.

232 ALLOYS OF THE PRECIOUS METALS.

Colouring Silver and its Alloys. — The silver alloys
used commercially have a white colour, and take a high
polish, and on these properties their value to a large extent
depends, but for various purposes either a matte or a
coloured surface is required, and these may be imparted in
various ways.

To obtain a very bright surface, the alloy is, of course,
polished, the very finest polishing materials being used
so as to avoid the formation of scratches. A dead-white
surface is given by processes which practically consist
in roughening the surface, usually by dissolving away some
of the copper, and leaving a slightly rough surface of
silver. Various methods may be used. Mr. Hiorns
describes these : —

" An old method is to dip the work in a thick solution
of borax, then place in a copper annealing pan, sprinkle
it over with charcoal dust, and place the pan and its
contents upon a clear fire. Heat until red hot, then
withdraw and allow to cool. The work is next boiled
with dilute sulphuric acid, and if the right colour is
not obtained, the process is repeated one or more times.
The lower standards require five or six operations to
effect the proper degree of whiteness."

" Another plan is to dip the work in a mixture of
4 parts powdered charcoal and 1 part nitre well mixed
with water. The work is heated till the coating is
thoroughly dry, when it is removed from the fire, allowed
to cool, and boiled out in a solution of bisulphate of
potash. After two or three operations a beautiful
dead-white colour is the result. It is then washed in
soda and water containing a little soap, or scratched and
burnished if required bright. The process is completed
by drying in warm boxwood sawdust."

" Gee's method of whitening consists of making the
work red hot and boiling in dilute sulphuric acid (1 of
acid to 40 of water). The process is repeated if necessary
until the requisite colour is obtained."

It will be seen that all these methods depend on the
oxidation of the copper by heat, and then the solution
of the copper oxide by suitable solvents. The direct
action of solvents is not applicable, as the silver would be
dissolved as readily as the copper.

ALLOYS OF THE PRECIOUS METALS. 233

In the case of very poor alloys these oxidation and
solution processes cannot be used, as the amount of
copper dissolved would leave the surface too rough. In
such cases a layer of silver is usually deposited on its
surface by electro-deposition, or by simple immersion in a
solution of a silver salt.

A dark surface colour is sometimes imparted to
silver goods under the name of oxidised silver. The
name is not correct, because silver does not oxidise. The
dark surface may generally be produced in two ways : (1)
By treatment with a sulphide such as potassium, barium,
or ammonium sulphide, by which a very thin layer of
black silver sulphide is formed on the surface; or (2) by
depositing on the surface of the metal a layer of some
dark metal, preferably platinum, which is deposited from
its solution as chloride by the action of the silver alloy.

Silver Solders. — For soldering silver alloys, an alloy
more fusible than that being united must be used.
Obviously, if an alloy of silver and copper only is to be
used as a solder, it should be the most fusible member of
the series, i.e., the eutectic, as any increase of either
constituent will diminish the fusibility. Where this is
not fusible enough, some other metal which gives more
fusible alloys will be added, the metal usually used
being zinc.

Various formulae have been given. An alloy of 4
parts silver and 1 part copper is sometimes used, but
this is not the most fusible alloy, and is said to be too
infusible for ordinary work. A good solder often used
consists of 2 parts rilver and 1 part brass. Assuming
the brass to be a 50-50, this would give :—

Silver 66-68

Copper 16-66

Zinc 16-66

100-00

If copper and zinc are to be used together, a brass
must be selected which has been made from verj pure
metal, as the presence of lead is very objection-
able. For articles which have to be hall-marked, the
solder should be as near as possible to the standard,
because in taking the sample for assay the solder may be

234 ALLOYS OF THE PRECIOUS METALS.

included, and therefore if the solder used is too poor
the bulk of the alloy must be made richer than would
otherwise be necessary.

Solders of standard fineness can be made by replacing
part of the copper by zinc, but they are not easy to
work. The richest solders that answer well have a
fineness of about 800, the zinc and copper either being
present in equal quantities or the zinc being in larger
proportions. Mr. Hiorns found that a solder containing

Silver 80

Copper . . . . . . . . 2-5

Zinc 17-5

100-0

ran quite readily, and gave good results. Ordinarily, the
percentage of silver is considerably lower.

The following composition is quoted by Hiorns as a
French solder used for soldering wares of a 980 standard : —
Silver .. .. .. .. 66-6

Copper 23-3

Zinc 10

100-0

In preparing solders it is always well to alloy the
zinc and copper, making, of course, due allowance for
loss of zinc, and then to use the resulting brass for alloying
with the silver. The use of brass gives uncertain results,
as the percentage of copper may vary.

Silver-Tin Alloys — These are of little importance, but
they have been studied by several workers. The alloys
are white, and when the quantity of silver is large seem
to be homogeneous in structure, probably being solid solu-
tions. With less than 60 per cent, of silver the structure
is seen to be duplex, the white silver crystallising out in a
dark ground mass of eutectic. The eutectic contains about
5 per cent, of silver. Several definite compounds are
said to have been detected. The hardest alloys contain
from 60 to 85 per cent, of silver.

Silver-Antimony Alloys — Silver and antimony alloy
readily, the most fusible, i.e., the eutectic alloy, containing
50 per cent, of each metal, but the cooling curves also
show breaks at 20 per cent., and 80 per cent, of silver.
The alloys rich in antimony are very hard.

ALLOYS OF THE PRECIOUS METALS.

235

FIG. 122. — ALLOY OF SILVER AND ANTIMONY.
SILVER, 16 PER CENT.; ANTIMONY, 84 PER CENT. V 30 X.

Silver-Nickel — Alloys of these metals have been
prepared, but do not seem to have been investigated.
Berthier prepared one containing 13-5 per cent, of
nickel, which was white, took a high polish, rolled well,
and was very tough.

Silver- Lead Alloys — Silver and lead alloy in all pro-
portions, but the alloys are of no importance. The
metals dissolve in all proportions in the liquid condition,
but liquation takes place on solidification.

Scorification and Cupellation. — Silver (and gold) may be
recovered from substances containing them by melting
with lead and scorifying and cupelling the resulting
lead-silver alloy. The silver (and gold) dissolves in the
lead, thus separating it from earthy and other impuri-
ties. When the lead-silver alloy is heated to redness in
an oxidising atmosphere, i.e., with free access of air, the
lead is oxidised to litharge, and the silver remains.

Scorification, which is used to reduce the quantity
of lead, is usually carried out in a fireclay dish (a scorifier)
and is continued till the surface of the lead is covered
with a layer of litharge, when the molten mass is poured
into a mould. When solid, the litharge is broken away,
and the lead if necessary is returned to the scorifier.

236 ALLOYS OF THE PRECIOUS METALS.

Cupellation is carried out in a dish (the cupel) made
of bone ash, which absorbs the litharge as fast as it is
formed, so that when the lead is all removed, a bead or
prill of pure silver is left.

The behaviour of other metals when alloyed with
silver is of some interest.

Gold and platinum, being unoxidisable, are entirely
left with the silver.

Cadmium and bismuth are oxidised and entirely
removed with the litharge.

Copper is oxidised and removed, provided the
quantity of lead is sufficiently large, the cupel at the
same time being stained a dark olive green. The larger
the quantity of copper the more lead is required, and
the greater will be the loss of silver.

Antimony is carried away with the lead, and usually
causes the cupel to crack, if present in large quantity, and
also causes loss of silver by volatilisation.

Tin is carried away with the lead if the quantity of
lead be large ; if not, it is left as an infusible oxide.

As a rule, when foreign metals are present in large
quantity, they are best removed by scorification before
cupellation. When a silver or gold alloy is heated in
air, the base metal present often oxidises — thus standard
silver becomes black from the formation of copper oxide,,
which can be dissolved away, as already remarked,
leaving a white surface of silver. With standard gold
oxidation does not take place, but with poorer golds it.
frequently does.

PLATINUM ALLOYS.

PLATINUM is one of the most valuable of the metals,
indeed of late it has been more valuable than gold, but
its alloys are of little importance. It is white, almost
silver white, in colour, and is both malleable and ductile.
It is not acted on by any single acid, but is converted
into a chloride (Pt C14) by the action of the mixture of
nitric and hydrochloric acids known as aqua regia. It
does not oxidise in air at any temperature. It is very
difficultly fusible, its melting point being about 1,775° C.,
which is a higher temperature than is attainable in any

ALLOYS OF THE PRECIOUS METALS. 237

ordinary furnace fired with either gaseous or solid fuel ;
but it can be melted in the heat of the electric arc or the
oxy-hydrogen flame. It must be melted in vessels free
from silica, as this might be decomposed at the very high
temperature, and the silicon pass into the platinum,
making it brittle. Platinum is also said to take up carbon
at high temperatures. Platinum is the heaviest of the
metals in common use, its specific gravity being about 21-5,
which is only slightly less than that of the heaviest known
metals which are closely related to it, and which are
members of what is called the platinum group of metals.
Platinum alloys readily with most metals, but its alloys
are of little technical importance, though they are used
for some purposes.

Platinum and Copper. — These metals alloy readily in
all proportions. An alloy containing equal proportions
of the two metals is said to be yellow in colour, to have the
same specific gravity as gold, and to be easily worked.
With a larger proportion of platinum the alloys are white.

The following alloys are said by Mr. Hiorns to have a
golden-yellow colour, and No. 4, which is known as
Cooper's gold, is malleable and ductile, and closely
resembles 18-carat gold.

Platinum

i
18-2

29-3

18'

•7

(

•7

'1 19

Copper
Zinc

45-5

66'7
4

81-

•5

•8

29
4

•1
•2

66-
4-

7 81
2

Silver

9-5

Brass

18'3

Nickel...

9-0

The brass is, of course, an alloy of copper and zinc,
but the quality of the brass to be used is not stated. Mr.
Hiorns also mentions the following alloys : —

Cooper's Mirror Metal.

Platinum 9 49

Copper . . . . . . . . . . 57 • 85

Zinc.. ..... 3-51

Tin . . ;-.,'• 27 49

Arsenic . . . . . . . . . . 1 • 66

The inventor claims that this alloy is indifferent to
the weather, takes a beautiful polish, and is suitable for
pens.

238 ALLOYS OF THE PRECIOUS METALS.

Cooper's Pen A\etal.

Platinum . . 50

Copper . . . . . . . . 13

Silver 36

Both alloys are hard, non-corrosive, and could be
used for many purposes.

Platinum and Iridium. — An alloy of platinum with
10 per cent, of the rare metal iridium is hard, elastic, as
infusible as platinum, quite unalterable in the air, and
is not attacked by acids, even by aqua regia. It has been
used for the manufacture of standard weights and
measures for the Commission on the International Metric
System, as being the most unchangeable alloy that could
be found. Iridium is heavier than platinum (sp. gr.
22-421), and the specific gravity of the alloy is 21-615.

Platinum vessels for laboratory use are said to usually
contain iridium, which makes the alloy harder, and does
not in any way detract from its valuable properties.

Platinum and Rhodium. — An alloy of platinum with
10 per cent, of the rare metal rhodium has recently
acquired some importance as forming with platinum the
thermo couple used in the Le Chatelier pyrometer. It is
as infusible as platinum, resists corrosion as powerfully,
and is sufficiently ductile to be drawn into wire.

Platinum and Silver. — Platinum and silver unite, forming
alloys which are white in colour, and when the metals
are in some proportions are malleable and ductile, the
fusibility decreasing and the malleability increasing with
the quantity of platinum. The alloys high in platinum
do not tarnish, and are largely used in dentistry.

The behaviour of platinum-silver alloys with solvents
is somewhat peculiar and varies with the amount of
platinum present. When the alloy is rich in platinum
it is unacted upon by nitric acid ; when the platinum is
about 25 per cent, the silver is dissolved by nitric acid,
leaving the platinum almost completely ; but when the
percentage of platinum is reduced to about 5 per cent.,
the whole of it dissolves with the silver. For this reason
platinum cannot be parted from silver by nitric acid
as gold can.

ALLOYS OF THE PRECIOUS METALS. 239

Messrs. Johnson, Matthey, & Co. prepare commercially
an alloy of silver 2 parts, platinum 1 part, which is very
ductile, and is therefore easily drawn into wire, and which
is used as a standard for electric resistance.

Other Alloys. — The other alloys are of little or no
importance commercially. Their properties are not such
as to make them useful, and the presence of platinum
makes them costly. The alloys with fusible metals, such
as lead, in excess are very fusible, so that the platinum
metals are rapidly corroded by contact with such
metals. For that reason metallic oxides should never
be heated in platinum vessels. Even oxides of iron
and copper are to a small extent reduced, probably by
means of reducing gases which penetrate the platinum,
and the reduced metals alloy with the platinum. This is
one of the limitations to the use of platinum vessels.

AMALGAMS.

Mercury is the only metal liquid at ordinary tem-
peratures. It is silver- white in colour, and from the way
in which globules of it run over a dry surface without
wetting it, it was called quick, i.e., living, silver, a name
which is still frequently used. Its specific gravity is
13-59. It freezes at 38-5° C. and boils at 360° C., but
at much lower temperatures, even, indeed, at the ordinary
temperature of the air, it gives off vapour.

Alloys of mercury differ in many respects from those
of the other metals which have been considered, not
that there is any essential difference, but that the alloys
have to be used and studied at a temperature above
the melting point of one of the constituent metals. The
alloys, which are called amalgams, may be either
liquid, solid, or pasty, according to the quantity of
mercury which is present, and the nature of the other
metal. Mercury alloys with or dissolves almost all
metals.

Gold Amalgam. — Gold and mercury alloy very readily,
forming a white amalgam, the smallest trace either
of the liquid metal or its vapour being sufficient to
whiten a gold surface. Mercury will dissolve a
quantity of gold, the alloy being liquid at first, then

240 ALLOYS OF THE PRECIOUS METALS.

becoming pasty and ultimately waxy. A definite
amalgam seems to be formed which remains suspended in
its excess of mercury, from which it can be separated by
filtering through chamois leather or some very fine
fabric. As it can be separated by filtration, it is evidently
not in solution. The pasty amalgam separated in this
way contains about two-thirds its weight of mercury
and one- third gold.

On heating to above the boiling point of mercury, the
mercury is expelled and gold is left which does not retain
more than a trace of that metal.

A definite solid amalgam, Au Hg, containing 49-9 per
cent, of gold, which crystallises in four-sided prisms and
retains its lustre in air, has been obtained by heating the
pasty amalgam with dilute nitric acid. It is not soluble
in nitric acid, and its melting point is above the tempera-
ture at which the mercury is expelled. As pasty masses
very often consist of an intimate mixture of a solid and a
liquid, the pasty amalgam may consist of this solid
amalgam disseminated through excess of mercury.

Advantage is taken of the solubility of gold in
mercury in the metallurgy of gold. Crushed rock con-
taining gold is mixed with mercury which dissolves out
the gold, and the gold is recovered from the amalgam
thus formed by straining and distillation. The process
is called amalgamation, and it is the most common
method of treating gold ores.

The old method of fire-gilding metallic articles, which
is still used to a small extent, is carried out by means of
an amalgam containing about 66-6 per cent, of gold.
The article is thoroughly cleaned and is then dipped in a
solution of a mercury salt by which a layer of mercury
is deposited on the surface ; this is then rubbed over with
some of the amalgam, which adheres, and on heating to
redness the mercury is expelled, and a layer of gold in a
fine state of division is left, and this is firially burnished
to make it bright. Some amalgams of gold have been
found native.

Silver Amalgam. — Silver dissolves in mercury, forming
amalgams which closely resemble those of gold. It
dissolves less readily than gold in the cold, but very

ALLOYS OF THE PRECIOUS METALS. 241

readily when heated. A definite solid amalgam having
the formula Ag Hg.2 is found in nature, and crystallises in
the cubic system. It is known mineralogically as
amalgam.

Silver amalgam can be separated from excess of
mercury by filtration, exactly as in the case of gold.
Mercury is to some extent used for the extraction of silver
from its ores, but not very largely, because the silver is
usually present in combination in compounds which are
not decomposed by mercury. Silver amalgam was at one
time used for dry silvering just as gold amalgam was used
for dry gilding.

Copper Amalgam. — Copper and mercury unite when
finely divided copper is mixed with mercury, but the
amalgams are more generally prepared by decomposing
solutions of mercury salts by means of metallic copper.
If the copper be dissolved in mercury the excess of
mercury can be removed by filtration exactly as in the
case of the amalgams of gold and silver.

An amalgam containing from 25 to 30 per cent, of
copper may be obtained as a plastic mass of about the
consistency of clay by continued pounding or kneading
in a warm mortar ; on being left to itself for a few hours
it becomes crystalline and hard, so hard, indeed, that it
can be broken up in a mortar. On warming and well
triturating in a warm mortar it returns to its plastic
condition, and this change can be repeated any number of
times. The soft and hard forms have the same density,
so that the change is not attended either with contraction
or expansion. This alloy is used to some extent for stopping
teeth, but it is not suitable for the purpose on account
of the nature of the metals of which it is composed. It
may be used, however, for other purposes where the
change from the plastic to the solid condition would be
of use. Watt's dictionary gives the following method of
making this amalgam : —

Finely-divided copper is prepared by precipitating
copper from copper-sulphate solution by means of iron.
Ten grammes of mercury is heated with 10 grammes of
sulphuric acid and the copper obtained from 23-5 grammes
of copper sulphate is added. The materials are tritu-

242 ALLOYS OF THE PKECIOUS METALS.

rated together under hot water for from 20 to 30
minutes, the water is then poured off and the process
is repeated with fresh quantities of water until the
water shows no blue colour. The amalgam is then
dried again, triturated, well kneaded, and formed into
small cakes, which become quite hard in from 36 to 48
hours. The amalgam produced contains about 3 parts of
copper to 7 parts of mercury. Another method of
preparing it is to moisten the copper with a solution
of mercurous nitrate, then pour hot water upon it, add
the required quantity of mercury, and triturate under hot
water.

Tin Amalgam. — Mercury and tin unite readily at
ordinary temperatures, but more readily on heating.
Mercury applied to the surface of a rod of tin penetrates
it very rapidly, and makes it so brittle that it breaks short
off, and tin immersed in mercury splits up from the
expansion of the amalgam which is formed. The
amalgam is white. Solid and crystallised amalgams
have been obtained ; they may be definite compounds,
but this is uncertain ; while various formulae such as
Sn Hg, Sn Hg2, Sn3 Hg2, have been assigned to them.
The silvering on the back of mirrors is an amalgam of
tin and mercury, which being soft can be pressed into
optical contact with the glass.

A tin amalgam is said to be used by dentists for filling
teeth. One part of finely-divided tin is rubbed
in a mortar with 4 parts of mercury. The excess of
mercury is then removed by squeezing through a bag
of chamois leather. A plastic mass is left which hardens
in a few days. Another alloy is formed of 2 parts tin,
1 part cadmium, and excess of mercury ; the tin and
cadmium are melted together and mercury added. The
whole is poured into an iron mortar and well stirred
with a wooden pestle till it acquires a soft buttery con-
sistency. The excess of mercury is then squeezed off.
The amalgam is soft and plastic when kneaded in the
hand.

An amalgam of tin, silver, and gold is said to be used
as a cement for teeth. It is prepared by melting together
1 part gold, 3 parts silver, and adding 2 parts tin to the

ALLOYS OF THE PBECIOTJS METALS. 243

melted mass, pulverising the resulting alloy and kneading
it together with an equal weight of mercury.*

Sodium-Amalgam. — Sodium combines very readily with
mercury at ordinary temperatures and more rapidly
on warming. Heat is evolved and vivid combustion
takes place, some of the sodium burning away. The
two metals maybe triturated in a dry mortar provided with
a cover, and the amalgam should be covered with
petroleum as soon as the combustion is over. The amal-
gam is liquid or solid, according to the proportions of the
metals. With about 3 • 5 per cent, of sodium it is solid,
crystalline, and can be filed ; with 2 • 5 per cent, of sodium
it is still solid, but softer ; with 1 • 5 per cent, it forms a
thick paste ; with 1 per cent, it is viscid, and consists of a
solid and a liquid portion.

Sodium amalgam decomposes water, liberating hydro-
gen and separating mercury. On this account it is
largely used in chemistry as a reducing and hydrogenating
agent, the liberated hydrogen being in the nascent con-
dition. Amalgams are often more readily made by the
action of sodium amalgam than of mercury itself, and
many can be readily prepared by treating solutions of
salts of the metal with sodium amalgam. On exposure
to the air the sodium oxidises readily, so that the
amalgam must be kept under oil of some kind. Sodium
amalgam is sometimes added to the mercury in gold
amalgamation.

Most metals combine with mercury more or less
readily. In addition to those mentioned, bismuth,
cadmium, zinc, lead, antimony, magnesium, and the
alkaline metals form amalgams readily by direct union of
the metals in the cold, or by gentle heating. Nickel,
cobalt, iron, manganese, and platinum amalgams are
not readily prepared, but can be obtained by the
action of sodium amalgam on salts of the metals.

* Watt's "Dictionary of Chemistry," first edition, vol. iii., p. 891.

CHAPTER XV.

PREPARATION OF ALLOYS.

ALLOYS are almost invariably prepared by melting together
the constituent metals. As the fusibility and volatility
of the metals varies widely, the details of the methods
must be different in different cases.

The object aimed at is to secure a perfectly homo-
geneous mixture with as little loss of metal as possible.
When metals are melted they will as a rule mix perfectly,
like all miscible liquids, and in due time uniformity
would be brought about by diffusion without stirring,
even if the metals were very different in specific gravity,
but diffusion is so slow that other means of ensuring
mixture must be adopted.

If the metals have melting points not very different,
a mixture will to some extent take place even if
the metals are merely melted together, and the more
finely divided the metals are and the more intimately
mixed they are in the solid condition, the more perfect is
the mixture likely to be ; but if one of the metals is
much more fusible than the other, then the more fusible
metal will melt first, and either fall to the bottom or rise
to the top of the crucible, according to its density, and
though it will certainly dissolve some of the less fusible
metals there will be a tendency to form two layers which
can only be brought together by mechanical stirring.

When one of the metals has a much higher melting
point than the other, it is usual to melt it, and then to
add the metal of lower melting point, very often holding
the portion in the tongs and stirring it in so that it may
dissolve gradually. This must be done with care, the
cold metal only being added in small portions at a time,
so as not to chill the less fusible metal below its solidifying
point. As the melting point of the alloy falls as the
more fusible metal is added, the danger of solidification
is greatest when the first portions are added. When the

PREPARATION OP ALLOYS. 245

amount of the less-fusible metal is very large, and the
melting points of the metals are very different, it is often
advisable to make two alloys, and then to melt the less
fusible of these and to stir in the more fusible. Examples
of this practice will be mentioned later. Where one of
the constituents is a volatile metal such as zinc, the
temperature must be very carefully regulated.

To ensure homogeneity several methods may be used,
the usual one being vigorous stirring by means of an iron
rod. The stirring must be vigorous and long continued
to ensure uniformity, except in the case of metals which
diffuse readily, so that complete mixture by stirring is not
always easy. When an iron rod is used, this may be
attacked, and iron may pass into the alloy ; indeed
the iron almost invariably present in alloys is largely
derived from the stirring rods. Rods of fireclay or
graphite are therefore much better, but they are too
fragile for ordinary use.

The pouring of the metal from the crucible does some-
thing towards ensuring mixture, and this may be made
much more efficacious, where the alloy is of sufficiently
low melting point, by a double pouring — that is, by
pouring the metal into another crucible and thence into
the moulds. The same principle is applied in the case of
alloys of high melting point, by casting into small
ingots, then breaking these ingots and remelting them. In
some cases alloys are said to be very much improved
by remelting, and this improvement is mainly if not
entirely due to the greater homogeneity which is produced.

In the case of alloys of low specific gravity, such as
those containing a large proportion of tin, the stirring
has an additional advantage. Such alloys are very apt
to retain scattered through them particles of oxide,
which do not rise readily to the surface, but which are
brought up by stirring. Stirring with a stick of wood is
often of great advantage in such cases, as the evolved
gases tend to carry up with them the oxides, &c., and
these form a scum which can be skimmed off, or left in
the crucible when the alloy is poured. In some cases, also,
stirring tends to facilitate the escape of occluded gas.

When metals are melted, there is always a tendency to
oxidise on exposure to the air ; to avoid this, the molten

246 PREPARATION OF ALLOYS.

metal should always be covered with a layer of powdered
charcoal, or some similar material. Oxide formed may
simply form a scum on the surface, or it may in certain
cases be dissolved and impair the qualities of the
alloy.

Dissolved oxides can often be removed by the addition
of some metal or non-metal which will decompose the oxide,
and form an oxide which, not being soluble, will float up,
and can be removed. To this action is due the great
improvement in the properties of certain alloys produced
by the addition of minute quantities of aluminium,
manganese, phosphorus, or other easily oxidisable
element, some examples of which have already been
discussed. As a rule a molten metal will dissolve its
own oxide to a small extent, but will not dissolve oxides
of other metals.

In the case of alloys, one metal will usually oxidise
more readily than the other, so that oxidation will tend
to alter the composition of the alloy.

As a rule, an alloy should be cast immediately after
stirring, as there is sometimes a tendency to segregation,
even whilst in the liquid condition.

Preparation of Alloys of Low Melting Point.— Alloys of this
class, consisting mainly of lead and tin, are very easily
prepared by melting the metals together in the required
proportions under charcoal. A crucible may be used, or
even an iron ladle heated over a fire. As both the metals-
oxidise readily, the covering layer of charcoal is important.
When the metals are melted they should be vigorously
stirred with a stick, and then should be poured into a,
mould, the scum being kept back by means of an iron
rod or a stick. The addition of various elements has
been suggested for the removal of impurities likely to be
present, such as a little sulphur, but this is never advisable ;
the stirring with a stick will do all that can be done in
this direction, and if the metals are too impure they
should not be used.

Where other less-fusible metals have to be added, the
methods are modified. The metal of higher melting
point may be melted first, and the less-fusible metal then
added, the whole being well stirred after fusion.

PREPARATION OF ALLOYS. 247

In the manufacture of pewter, which is an alloy of
tin with a small quantity of copper, some copper is first
melted and then its own weight of tin is added, and this
alloy is cast into ingots. When it is required to make
the pewter, a portion of this alloy is added to the
required amount of tin, and the whole is melted. A little
zinc is often added, it being claimed that the zinc carries
the oxides to the surface as a scum, and also that by its
oxidation it saves the tin.

In the manufacture of metal for casting shot, a little
arsenic is added. This is almost always added in the
form of white arsenic (arsenious oxide). Lead is melted in
an iron pot covered with charcoal, and the temperature
is raised considerably above the melting point of lead;
the white arsenic, usually wrapped in a sheet of lead, is
then put into the lead and pressed down, and the whole
is vigorously stirred with a wooden pole. The mass is
kept melted for some hours to ensure complete reduction,
and the alloy is ladled into moulds for use. As there is
considerable loss of arsenic, the lead pot must be covered
with a suitable hood. In some works enough arsenic is
added to make an alloy containing about 2 • 0 per cent, of
arsenic, which is added to the shot lead in the required
proportions; in others the white arsenic is added directly
to the shot lead. The amount of arsenic present in the
shot lead is very minute. The quantity of white arsenic
added, when it is added directly to the shot lead, varies
from about 3 to 6 parts per 1,000 parts of lead. Brannt
states that the lead should never contain more than 1
per cent, of arsenic ; it actually always contains very
much less.

Preparation of Bronze. — In the case of bronze, the two
metals have very different melting points and specific
gravities. As a rule, the copper is melted first under
charcoal either in a crucible or reverberatory furnace,
according to the amount being melted, and when it is
completely melted the tin is added, or very frequently
the tin, being very fusible, is placed in the ladle in the
form of ingot, and the molten copper is poured
upon it. The tin being much lighter than the
copper will tend to float, so that vigorous stirring
is necessary in order to ensure complete mixture.

248 PREPARATION OF ALLOYS.

The tin is also much more easily oxidisable than the
copper, so that it tends to oxidise, and as oxide of tin seems
to be soluble in bronze, some of this may be retained.
The addition of a little phosphor copper, or other easily
oxidisable metal which will decompose the tin oxide,
is therefore often advisable.

Sometimes the tin and copper are melted together,
but owing to the oxidisable character of the tin this is
never advisable. The casting temperature is a matter
of very great importance ; too low a temperature is
particularly to be avoided.

Preparation of Brass.— In the case of brass not only
have the two metals very different melting points, but the
zinc is very volatile, and this complicates the process of
manufacture very considerably. Also the quantity of
zinc to be used is as a rule very much larger than the
quantity of tin in bronzes. Brass may be made from
copper and zinc, but almost always in practice a con-
siderable quantity of scrap will be used.

The copper is always melted first under charcoal,
either in a crucible or a reverberatory furnace, as the
case may be. When the copper is melted and has
reached a temperature somewhat above its melting point,
but not too hot, the cover of the crucible is removed and
the zinc is added. The zinc is in the form of fragments
of cakes of spelter. These are taken one by one in the
tongs and are carefully stirred into the molten copper, so
that they may dissolve gradually. This addition requires
great care. Usually the spelter is warmed by leaving
it on the furnace top for some time before it is
added. The temperature of the copper must not be
too high, since the higher the temperature the greater
will be the loss of zinc. The cold zinc at once tends to
chill the copper, and it must be kept moving, so that no
copper is solidified. If the lump of spelter be let fall into
the crucible it may cause some of the copper to solidify,
and this cannot be remelted without undue loss of zinc.
As soon as the zinc is all added, the whole is well stirred,
and the metal is poured without delay into the moulds.

As soon as the zinc touches the copper some of it is
volatilised, and the vapour coming into the air at once

PREPARATION OF ALLOYS. 249

burns, forming oxide of zinc, which is deposited in white,
woolly flakes — called philosopher's wool. The oxide of
zinc is extremely irritating, but its formation cannot be
avoided. All the time the zinc is being added, and whilst
the brass is kept melted, zinc is being given off, so that
there will always be a considerable loss of zinc, and a loss
which will vary very much according to the conditions
of working. To minimise the loss the temperature of the
copper must not be too high, and all the operations must
be performed rapidly. As the brass will usually be too
cool for casting after the copper has been added, the
temperature must be rapidly brought up to the required
point.

In most cases scrap will be added as well as the zinc,
and where accurate composition is required it is obvious
that the composition of the scrap must be known. The
way in which the scrap will be added will depend on
circumstances. The usual method is to place the copper
on the bottom of the crucible or furnace, then put the
scrap on the top and melt the two together ; or, if in
large pieces, it may be stirred in in the same way as the
zinc, if in smaller pieces it may be added in small
portions at a time, the whole being well stirred after
each addition ; but in this case the copper must be very
hot before the addition is made. It may be added cold,
as is usual when a small quantity is being added, OF it
may be heated to redness if a large quantity is being
used.

When brass is required to a specified composition,
great care must be taken hi the manufacture. The com-
position of the scrap used must be known, and allowances
must be made for the loss of zinc, but the allowance must
not be too great. It is only with the richer brasses, say
those with about 70 per cent, of copper, that is, with brasses
which have to be worked, that accuracy of composition is
usually required ; with ordinary cast brass a few per cent,
either way is of little or no moment. It is quite im-
possible to make a brass with a perfectly definite com-
position, but the variation should be within 1 per cent,
of copper above or below the specified percentage. As
the copper is the more costly constituent a minimum
copper is usually specified, no notice being taken of

250 PREPARATION OF ALLOYS.

variations in the other direction. For a 70/30 brass the
amount of allowance for loss in manufacture will
usually be 2 per cent, on the whole weight (that is,
nearly 7 per cent, of the zinc used), or for lOOlbs. of brass
the total weight of mixture will be 102lbs., thus : —

Copper 70

Zinc 32

102

When scrap is to be used, allowance must, of course,
be made for the zinc contained in it. Suppose it be
required to cast lOOlbs. of 70/30 brass, using 201bs. of scrap
containing 68 • 5 per cent, of copper. The scrap will con-
tain 13-71bs. of copper and 6-31bs. of zinc. The mixture
will be : —

Copper 70-13.7=56-3

Zinc 32- 6-3=25.7

Scrap 20

102-0

It will be noticed that the less the loss of zinc the
poorer will be the resulting alloy, so that very careful
work may lead to saving in zinc, and thus to a reduction
in the value of the alloy. When the percentage of copper
is very closely specified, it is better only to allow for
1 per cent, loss, as it sometimes happens that the actual
loss is below 2 per cent.

One of the great causes of the uncertainty of the
composition of brass is the use of scrap of uncertain
composition, and often intermixed with bronze and other
scrap. Scrap should always be melted and cast into
ingots, the composition of which can be determined before
it is used in the foundry for any except the commonest
castings.

The remelting of brass always entails a considerable
loss of zinc, so that remelted brass is always richer in
copper than that before remelting. The loss may vary
from 1 to 5 per cent., according to the conditions of
melting.

The addition of phosphorus, or other deoxidising
agent, is not so necessary in the case of brass making

PREPARATION OF ALLOYS. 251

as for bronze, but the addition of manganese is often
advantageous, as manganese oxide is fusible. The zinc
is so readily oxidised that it decomposes oxides of copper
that may be present, and oxide of zinc does not seem
to be soluble in brass, but it is infusible and may remain
disseminated through the brass, rising slowly to the
top and forming a scum. If there is much copper oxide
in the molten copper, as when it has not been properly
protected during melting, a large quantity of zinc oxide
will be formed, and this may take a long time to rise ;
or if a very large quantity be present a mixture of zinc
oxide and metal, called a " salamander," which will not
melt, may be formed.

Even when the quantity of zinc oxide is not large
enough to bring this about, it may be quite large enough
to reduce the fluidity of the metal and make it pour
badly. Such a brass will not give clear, sharp castings,
and the castings produced will be poor and weak, and
will probably crack if the metal has to be rolled.

Sometimes a brass is improved by remelting, because
a better chance is given to the entangled oxide to rise.
When there is no entangled oxide remelting does not
seem to improve the quality of the brass.

Mr. Sperry states that melting the brass under com-
mon salt improves the quality of the brass by removing,
or rather preventing, the formation of copper oxide.
The explanation which he gives is that salt dissociates,
and that the sodium removes the oxygen from the oxide.
Admitting the fact, the explanation is probably not correct.
It is more likely that in presence of the molten salt
oxychlorides of copper are formed, and that these yield
fusible oxychlorides of zinc which rise readily.

Molten alloys, as indeed all molten metals, tend to
absorb gases, and the evolution of these on solidification
causes blowholes. The gases are probably mainly carbon-
monoxide and nitrogen, and perhaps sulphur dioxide.
Though but little is known as to the absorption of gas
by brass, that it is sometimes absorbed or given out
is proved by the formation of blowholes in brass castings.
The higher the temperature, and the longer the metal is
exposed to the gas, the more likely is gas to be absorbed.

252 PREPARATION OF ALLOYS.

Mr. Sperry gives the following directions for brass
melting :—

" Place a small amount of scrap in the bottom of the
crucible. This serves as a cushion for the copper ingots,
and the small amount of zinc which is present also has a
reducing action on the oxide of copper that is formed. Over
this scrap place the copper ingots, but do not pack them so
tightly that the crucible will be cracked when they expand
with the heat. Also see that too many ingots are not
placed in the crucible. Ingots which cannot be covered
with charcoal are oxidised in melting. Excellent results
are obtained by cutting the copper so that the pieces pack
well in the crucible and leave none projecting above the
top. In this manner the whole may be covered well
with charcoal.

" When the ingots have been placed in the crucible
on top of the scrap, some charcoal is put in around them
and the melting begun. A fire that is too fierce is objec-
tionable, as it is apt to ' burn ' the top of the copper
ingots before the bottom is melted. A fire that allows
the metal to melt uniformly is the best. Forced draught
must be carefully regulated so that the metal is not
rapidly ' burnt.' Watch the metal carefully, and when
the first signs of melting are seen add one or two pounds
of common salt (to a No. 60 crucible). Now add some
more charcoal so that no part of the copper is exposed.
The charcoal should be granulated and not in large pieces,
as the latter do not cover the metal well. When the
copper begins to melt add some more scrap. The
addition of scrap will cause the copper to melt more
quickly. When the scrap has melted add the remainder
of the copper, and then another dose of salt. Stir the
salt into the metal, and then add more charcoal if the
surface is not well covered. The best results will be
obtained when no part of the copper is exposed.

" Do not attempt to add the spelter when the copper is
just melted. The cold spelter will cause the copper to
chill in the bottom of the crucible. The zinc will then
float on the top of the copper, until finally the rise in
temperature will cause the zinc to volatilise almost
instantaneously with a flash and its complete loss. On
the other hand, it is not conducive to the best results to

PREPARATION OF ALLOYS. 253

overheat the copper so that the spelter will ' sing '
when it is introduced. There seems to be a widespread
notion among brass casters that good brass cannot be
made unless the spelter ' sings ' when it is added to
the copper. This idea is false, as the best brass is
produced when the temperature of the copper is as low
as it can be and yet melt the spelter. The heating of the
spelter so that it is not brittle is also helpful, as it does
not then chill the copper to such an extent.

" When the copper has arrived at the right heat —
experience only can tell this point — the spelter is added.
If the ' heat ' of the copper is right, the spelter will
' sizzle ' somewhat, but will not ' sing.' The whole is
then carefully stirred and more scrap added if necessary.
The spelter has now cooled the brass to such an extent
that to pour it would mean an imperfect casting. The
caster is now brought face to face with two evils : First,
the pouring of the brass at too low a temperature or
' heat,' and thus saving the excessive spelter loss.
Second, pouring the brass at a good heat so that it smokes
freely, and thus losing considerable spelter. Of the two
evils choose the less, and the ' less ' in this instance is
the loss of spelter. Brass which is poured too cold does
not produce good castings, particularly in chill moulds
for the casting of rolling-mill plates. Unless the heat
of the brass is sufficiently high the oil which is used on
the moulds and which burns at the mouth of the mould
will not reduce the film of oxide that envelops the stream
of metal as it is poured. For this reason a dirty casting
results.

" In order that the oil shall reduce the film of oxide
which forms on the stream of oxide as it enters the mould,
it is necessary to pour at a suitable temperature. This
' suitable temperature ' is determined by the eye.
Pyrometers have not yet proved advantageous for it.
The temperature for pouring must be high enough, so
that the brass smokes freely. At this ' heat ' the stream
of metal, provided the oil burns at the mouth of the
mould, is clear and free from oxidation.

" While the foregoing directions are somewhat general
in their scope, there are a few very important rules to be
strictly adhered to. These are of such importance that

254 PREPARATION OF ALLOYS.

they may really be called axioms. They are three in
number, and the whole subject of brass melting may be
said to hinge upon them :—

" (1) Do not overheat the copper. More brass is
ruined by not following this rule than anything else.
Brass may be overheated with less danger than the
copper. Heat the copper to the required point and then
immediately add the spelter.

" (2) Do not ' soak ' the copper or the brass in the fire.
The longer metals are allowed to remain in the fire the
greater the oxidation and the more gas is absorbed. As
soon as metal is ready for pouring, it should be at once
removed from the fire and poured. Leaving metal in
the fire after the right temperature for pouring has been
reached is injurious.

" (3) Do not pour the metal at too low a temperature.
Even if the brass smokes it does not indicate that the loss
is enormous. It is better to lose the spelter than the
casting."

The brass may take up sulphur from the fuel if a
sulphurous fuel be used, and though the actual quantity
is small, it is enough to impair the quality of the alloy.

Calaminc Brass — Brass was made before zinc was
known in the separate condition. It was then made by
melting copper with zinc oxide — roasted calamine — and
carbon, and was therefore known as calamine brass.
The copper was always finely divided, usually in the form
of granulated copper, the zinc oxide was reduced, and the
zinc liberated, the reduction taking place at a tempera-
ture below the melting point of copper. The reduced
zinc was volatilised, and the vapour attacked and com-
bined with the copper. The loss of zinc was always
very high, and the brass of very uncertain composition.
As the oxide of zinc was never pure, a considerable
quantity of slag was produced which was often very
infusible. This process is not now used.

Aluminium Alloys. — Aluminium alloys may be pre-
pared in two ways : (1) The ordinary foundry method, by
melting together the component metals ; and (2) the
Cowles process, by reducing the aluminium by carbon in

PREPARATION OF ALLOYS. 255

the electric arc, in presence of the metal to be alloyed
with it.

The second method was largely used before the
electrolytic methods enabled metallic aluminium to be
produced almost pure and at a cheap rate. It is now
but little used, probably because the bronze produced is
apt to be impure, owing to the conditions of production.
When it is used, a copper-aluminium alloy rich in copper
is first made, and this is diluted by the addition of copper.
It is almost always more satisfactory for users to buy the
materials and prepare their own alloys.

As aluminium bronze is hardened by the presence of
iron, silicon, and other impurities, the metals used should
be as pure as possible.

The following instructions are given by the Magnesium
and Aluminium Fabric of Hemelingen for the preparation
of the aluminium bronzes : " Melt the copper in a
plumbago crucible, and heat it somewhat hotter than
its melting point. When quite fluid and surface clean,
sticks of aluminium of suitable size are taken in tongs
and pushed down under the surface, thus protecting the
aluminium from oxidising. The first effect is necessarily
to chill the copper more or less in contact with the
aluminium, but if the copper was at a good heat to start
with the chilled part is speedily dissolved and the alu-
minium attacked. The chemical action of the aluminium
is then shown by a rise of temperature, which may even
reach a white heat ; considerable commotion may take
place at first, but this gradually subsides. When the
required amount of aluminium has been introduced the
bronze is let alone for a few minutes, and then well
stirred, taking care not to rub or scrape the sides of the
crucible. By the stirring, the slag which commenced to
rise even during the alloying is brought almost entirely
to the surface. The crucible is then taken out of the
furnace, the slag removed from the surface with a skimmer,
the melt again stirred to bring up what little slag may
still remain in it, and it is then ready for casting. It is
very injurious to leave it longer in the fire than is absolutely
necessary; also, any flux is unnecessary, the . bronze
needing only to be covered with charcoal powder. The

256 PBEPARATION OF ALLOYS.

particular point to be attended to in melting these
bronzes is to handle as gently as possible when once
melted."

When the bronze is to be made by diluting a high-
aluminium alloy, either the one constituent or the other
may be melted first, usually the one largest in quantity,
and the other is then stirred in, or the two may be melted
together. Aluminium itself should never be melted in
clay crucibles, or in contact with siliceous materials, as
it may decompose them and take up silica and other
impurities ; plumbago (graphite) crucibles are generally
used ; these, however, also contain enough clay to
impart silicon to the metal.

The difficulties in the way of making alloys rich in
aluminium are generally two-fold.

If the mixture is too hot or is kept too long melted,
the aluminium will attack the crucibles in which the
melting takes place, reducing silicon and iron, which pass
into the metal and make it brittle. As aluminium alloys
are very light, the oxide scum rises comparatively slowly,
and therefore sufficient time must always be allowed
before pouring.

The other difficulty is that known as burning. Molten
aluminium seems to absorb gases, probably nitrogen
and carbon-monoxide, and these being given out on
solidification make the metal porous and brittle.

Mr. J. E. S. Jones says, speaking of aluminium : —

" The ingot structure of ' burnt ' metal, i.e., its
appearance to the naked eye, is often quite distinctive.
We have : —

"(1) A great profusion of crystals on the top
surface of the ingot, like those on galvanised ironware
which have been dipped in spelter containing tin. The
crystals are beautifully filicoid or fern-shaped, and occur
on the good metal as well ; but in that case there are only
a few, and their size is nothing like that attained in a
really ' well-burnt ' sample. The opinion is conse-
quently suggested that these crystals are always indicative
of ' burning,' incipient or pronounced, and that per-

PREPARATION OF ALLOYS. 257

fectly good aluminium should show no crystals at all on
the ingot surface.

" (2) The centre of the top surface is usually not only
sunk in like every cast ingot, but large cracks have
appeared traversing the crystals. In my opinion, and
also in that of other men who are qualified to say, these
cracks are absolutely indicative of poor metal. I have
never seen them in undoubtedly good ingots, and as a
rough-and-ready test to differentiate (before subsequent
examination) good from bad aluminium, the presence or
absence of cracks is a good guide.

" (3) The top ingot surface is nearly always also
covered with numerous parallel streaks, mainly at the
edges. These streaks in appearance are exactly like
that of a piece of skin, not caught evenly at the edges,
which is pulled tight and ' ruckles 'up. I imagine
that it is the skin of oxide on the metallic surface which
does the same, and that the streaks are produced as a con-
current effect from the same cause which gives the central
cracks. We also get fine streaks sometimes on the good
metal, but in nothing like the size and quantity that the
' burnt ' material shows.

" The above tests, therefore, will give the aluminium
f oundryman a good idea as to whether his metal is good or
bad when it arrives, but, of course, none are absolute
proofs. For this recourse must be had to the more
scientific examination by physical, microscopical, and
chemical means."

Aluminium oxidises very readily, and thus removes
every trace of oxygen from the copper ; the oxide slag
formed is no doubt partly produced in this way and partly
by the direct oxidation of the aluminium.

The metals copper and aluminium diffuse readily.
Many workers think that to obtain a homogeneous alloy
remelting is necessary, this sometimes being repeated two
or three times. This, however, is quite unnecessary if
ordinary care be used, though it may be advisable when
casting very small heats. At each remelting there is
some loss of aluminium, but if the surface be kept covered
with charcoal this will be very small.

258 PREPARATION OF ALLOYS.

When zinc is to be added, it is added just before
casting, as in the manufacture of brass.

Nickel Alloys. — The preparation of nickel alloys
presents some difficulties, owing to the very high melting
point of nickel. Where only copper is to be alloyed, the
difficulties are not serious, since neither of the metals is
volatile. The two metals, preferably in a fine state of
division, are mixed and melted under charcoal ; the copper
melts and dissolves the nickel, or the nickel may be
melted first, and the copper in ingot form is heated red-
hot and then added to the nickel.

When, however, zinc is to be added, the volatility of
the zinc introduces a difficulty, and the method is usually
modified.

The three metals to be used are made into alloys,
each containing two metals. Thus part of the copper is
alloyed with the nickel by fusing the metals together so
as to form an alloy which may contain from 1 part copper
to 1 part nickel to 2 parts copper to 1 part nickel. The
zinc is alloyed with copper, so as to form a brass, either
equal quantities of copper and zinc, or 1 part copper
to 2 parts zinc, as the case may be. The two alloys are
cast into ingots of suitable form, and the zinc alloy may be
broken up into pieces. The required amount of the less-
fusible, i.e., the copper-nickel, alloy is melted in a
graphite crucible, and the more fusible alloy is then stirred
in, exactly in the same way as the zinc is added in making
brass. By this method of working the loss of zinc is much
reduced, because the temperature is much below themelting
point of nickel. The alloy is then cast, and very frequently
is remelted for use so as to make it more homogeneous.

Platinum Alloys. — Platinum is so infusible that it
can only be melted in the electric furnace, in the oxy-
hydrogen flame. The latter is usually used. The plati-
num is melted in lime crucible before the oxy-hydrogen
flame, and the metal to be added is stirred in. If a small
quantity of platinum only is to be added to an alloy, the
alloying metal may be melted and the platinum stirred
in till it is dissolved.

Amalgams. — These alloys are usually easily prepared
by grinding the finely-divided metal with mercury, or in

PREPARATION OF ALLOYS. 259

some cases by liberating the metal by chemical means
from its compounds in presence of mercury.

FURNACES.

Since alloys are always prepared by fusion, the con-
struction of furnaces for the fusion is a matter of great
importance.

The furnaces are in general of two kinds : —

(1) Crucible furnaces ;

(2) Reverberatory or air furnaces.

In the former the metal to be melted is contained in
a crucible which is heated by contact with the fuel or
by the products of combustion; in the latter the metal is
heated on the hearth of the furnace by the products of
combustion and by radiation from the hot masonry of the
furnace.

Cupola furnaces — such as are used in iron founding,
in which the metal is heated by contact with the solid
fuel — are rarely used in the manufacture of alloys.

Crucible Furnaces. — These are almost always used for
dealing with small quantities of metal, up to about
1601bs., but larger quantities are difficult to deal with
owing to the large size of the crucibles needed, and the
difficulty of handling them. They have the great advan-
tage that the metal is protected from the fuel and the
products of combustion, and therefore is much less likely
to take up deleterious impurities such as sulphur, and as
it is quite easily kept covered with a layer of charcoal,
oxidation can be reduced to a minimum. Such furnaces
are, however, very wasteful of heat, and the crucibles
are always costly.

Crucible furnaces may be fired either with solid fuel,
gas, or oil.

Solid Fuel Crucible Furnaces. — A furnace of this
type consists of a fireplace with firebars at the bottom,
and a suitable cover at the top, and with an opening by
which the products of combustion can be drawn away.
Beneath the firebars is an ashpit to receive the ashes
from the fuel.

The size of the furnace will vary with the size of the
crucibles to be used, and this will in its turn depend on
the amount of metal to be melted at one time. For pots

260

PREPARATION OF ALLOYS.

up to 1501bs. capacity, the furnace must be about 18in.
square. As the pot must not come in contact with the
bars or the bottom would be chilled, a fireclay support,
often half a brick, is placed on the bars to support it.
The furnace must be sufficiently deep for the top of the
pot to be 6in. or Sin. below the flue, otherwise the air
entering it at the furnace top
and passing to the flue may
chill the surface of the metal.
The furnace must be large
enough to contain a layer
of ignited fuel all round the
pot, but this layer need not
be very thick; 3in. or 4in. is
quite sufficient. It must be
remembered that the crucible
will only be heated by the
coke which is in contact with
it, as the hot coke is quite
opaque to radiation from the
external layers of the fuel.
It will be obvious, therefore,
that for a single pot a cir-
cular furnace will be better
than a square one, because
the fuel in the corners of

the latter will be of little FIG. 123.— SOLID FUEL CRUCIBLE

use. Similarly for two pots FURNACE.

an elliptical furnace will be

(CROSS- SECTION).

better than one which is rectangular, but furnaces
holding two crucibles are never to be recommended for
making or melting alloys.

The chimney must be of sufficiently large area and
sufficiently high to produce a good draught.

The body of the furnace may be built of ordinary
brick, but the lining must always be of a refractory
material, firebrick, ganister, or similar material. When
a circular furnace is used, the bricks should be made to
fit the curve, so that as they are "cut " away there will be
no gaping joints. For square or rectangular furnaces
ordinary firebricks may be used. In any case they must
be set in good fireclay mortar, and the joints should be

PREPARATION OF ALLOYS. 261

as thin as possible; as the wear is always very much
greater with thick joints than with thin ones. Very
durable linings may be made with ganister or some
similar material. A wooden core is made the size
and form which the ulterior of the furnace is to
have. This is placed in position, and ganister, mixed
with enough water to make it plastic, is rammed round.
This is then dried gently. For circular furnaces this is
often better than a brick lining, and is more easily made.
It is the method usually used when lining the Sheffield
steel-melting furnaces, which have to stand a very high
temperature.

The masonry of the furnace must be held together
by iron stays and ties, the ties passing through the
masonry between the furnace openings.

The position of the furnace will vary. For very small
work the whole structure may be above the floor level, the
furnace top being about 2ft. Gin. above the floor, but for
larger work the furnaces must be below the floor, so that
the working floor is on a level with the top of the furnace.
This allows the workman to work from above, and gives
him a good position over the crucibles. For heavy
charges the crucible is always lifted by means of a crane.

The arrangement of the bars is a matter of importance.
The masonry of the furnace will be carried on iron
bars, and the firebars will be supported on cross-
bars below. The firebars may be of the ordinary
firebar shape, but they are better simply rectangular
bars of malleable iron, so that they can be moved and
turned when necessary, and they may be fixed with a
diagonal vertical by resting them in V grooves in the
supporting bars. It is very important that the bars
should fit close up to the furnace masonry, so that all
the air that finds its way in passes between the bars,
and none is admitted round the edges of the furnace.
The gases find a much easier passage up the wall
than through the mass of the fuel, and if the air
can pass up in this way the combustion at the outer
edge of the fire is very rapid. This means loss of
heat, because as already pointed out the heat evolved
there cannot be utilised, and at the same time the
temperature close to the wall being very high the brick-

262

PREPARATION OF ALLOYS.

work is apt to be rapidly destroyed. An alteration in the
arrangement of the firebars of a furnace has often greatly
improved its working. The air spaces between the
bars should be as large as possible, so that a maximum
of air can be admitted when required.

The furnace should always be provided with dampers
for regulating the draught, and preferably there should be

FIG. 124.— DETAIL SECTIONAL VIEWS OF CRUCIBLE BRASS FURNACE.

two, the one in the chimney to control the outflow of
the products of combustion, and the other at the ashpit to
control the inflow of air. The careful worker will keep
the rate of combustion under strict control by means
of the dampers.

An excellent form of furnace for alloy melting used
in the United States is shown in Fig. 124. The
furnace consists of two cast-iron cylinders one within
the other. The inner cylinder is lined with firebrick

PREPARATION OF ALLOYS.

263

in the usual way. The outer cylinder is closed at the
bottom by a circular casting provided with legs resting
on the supporting beams and has a circular opening for
the insertion of the grate. The lower portion of the
furnace is closed by a bell-shaped casting swinging on
hinges and operated by a chain wound round a shaft
which is held in position by a ratchet and pawl. The
surfaces of the castings are machined so as to form a
reasonably tight joint, and when the bottom is up it is
secured in place by an iron rod. The bell-shaped receiver
has a curved piece secured to its inner side by which
the grate is raised into place when it is closed. The air
is supplied by a fan, and passes into the annular space
between the cylinder and thence to the furnace.

Crucible furnaces may, of course, be modified in
many ways. One of the best known modifications is
that of Mr. Carr, of Birmingham, and his furnaces are
largely used. The furnaces are self-contained, the
sides of the furnace are supported independently,
the grate being a little below the bottom, so as to leave a
free air space all round. The ashpit is provided with a
damper. It is stated by the makers that with these
furnaces lOOlbs. of brass can be melted in three-quarters
of an hour with 361bs. of coke.

FIG. 125.— CABR'S FURNACE.

264 PREPARATION OF ALLOYS.

A very simple modification of the ordinary crucible
furnace has recently been introduced by Messrs. Weir, of
Cathcart, Glasgow, whereby a great improvement in
efficiency has been obtained. The ordinary firebars are
replaced by a perforated truncated cone of specially
worked out dimensions. In this arrangement the crucible
sits directly on the cone, and does not sink down as the
fuel is consumed. The combustion of the fuel is very
perfect, and there is practically no carbon-monoxide
in the escaping gases.

The resulting advantages are :—

(1) Greatly increased speed of melting ;

(2) Consumption of coke greatly reduced ;

(3) Labour cost considerably lowered.

A careful trial has shown that the quantity of coke per
cwt. of gun-metal melted has been reduced from 1071bs.
to 44lbs. Also it is found that five furnaces can now do
as much as 14 could formerly do under the old con-
ditions. The loss of metal is also found to be considerably
less than before, while there is practically no danger of
overheating the metal. No forced draught of any kind is
required, and the cost of the alteration is very slight.

FUEL.

The fuel used in crucible furnaces is always coke, and
for good work a good coke must be selected. In selecting
a coke, three points must be attended to: ( 1 ) The percentage
of fixed carbon, because on this the heating power of the
coke depends. It is only the fixed carbon that is of
any use, any volatile carbon simply escapes with the
waste gases, or, if it burns, is burnt at the top of the
furnace, where it is of no use. (2) The percentage and
quality of the ash left when the coke is burned. The
percentage of ash will usually vary inversely as the
quantity of fixed carbon, because a good coke will consist
essentially of fixed carbon and ash. The quality of the
ash may, however, vary very considerably. When
a furnace has been in use a little time, it becomes much'
enlarged by the fluxing away of the bricks. The tem-
perature is, of course, never up to the melting point of the
bricks, and the corrosion is due to the action of the

PREPARATION OF ALLOYS. 265

coke ash. The ash of some cokes is much more corro-
sive than that of others. In general, the less ash in a
coke and the less basic material in the ash, the better it
will be.

The efficiency of crucible furnaces is always very low.
In round numbers about lOOlbs. of coke will be required
to melt lOOlbs. of brass. The amount of heat theoretically
required can be calculated. Assume, for convenience, that
it is copper which is to be melted.

The melting point of copper is 1,085° C.

The specific heat of copper is, according to Frazer and
Richards, 0-0939 +0-00001778^, so that the heat required
to raise lib. of copper from 0° to its melting point, 1,085° C. ,
will be 1085 x (0-0939 + 0*00001778 x 1085) = 1085 x
(0-0939+01923) =1085 xl!31 -122-71, say, 122 units.
Taking the latent heat of fusion to be 45, the heat required
to raise lib. of copper to its melting point and to melt it
would be about 167 units, or for lOOlbs., 16,700. As coke
may be taken as having a calorific power of, say, 6,400
units, 2 6lbs. of coke would be sufficient to melt lOOlbs.
of copper. Taking the amount of fuel given above, the
efficiency of the furnace will be 2 • 6 per cent.

It is easy to see the sources of the loss : (1) The coke
is not completely burnt to carbon-dioxide, but a con-
siderable portion escapes as carbon-monoxide ; (2) the
products of combustion must leave the furnace at a
high temperature in order to produce a draught. These
sources of loss cannot be avoided, but owing to faulty
furnace construction, the actual loss is often much higher
than it need be.

Mr. J. F. Buchanan gives the following table of
fuel consumption in certain cases in making bronze : —

FUEL.

Losses Melting

Metal Kind of Quantity in Ratio

No. Melted. Method. Fuel. Used. Melting, perlb.

Lbs. Lbs. Per cent, of Fuel.

1 400 Crucibles N.O Charcoal 318 -89 1-25

2 400 „ „ Prepared Coke... 300 1-22 133

3 400 ,, F.I) Coke 348 2 18 1-12

4 400 ,, N.D Coke 325 1 04 1 20

5 1750 Cupola Coke 2181 7 93 7'91

'6 2240 Reverberatory Coal 1768 3 57 1'26

266 PREPARATION OF ALLOYS.

Tilting Furnaces. — With all ordinary forms of crucible
furnace, the lifting of the crucibles out of the hot fire into
the cold air is a source of danger, and seriously diminishes
the life of the pot. Not only so, but the handling of the pot
weakens it, and there is always the danger that the
pot may break in the tongs. The use of graphite
(plumbago) crucibles reduces these dangers, as such
crucibles will stand alternations of temperature and
handling much better than clay crucibles. Still, they
would last much better if the exposure to the cold air
could be avoided.

This difficulty is overcome in the Piat oscillating
furnace. This consists of a circular iron shell lined with
firebrick, which is provided with a grate, and which
stands over an air chamber into which the air is supplied
under pressure. The crucible stands on a block within
the furnace, and is provided with a spout passing through
the casing, by which the metal can be poured off. When
the charge is to be poured, the chimney is disconnected,
the whole furnace is lifted by means of a crane, and
the metal is poured into the moulds. As there is no
cooling, once the furnace is hot, the melting is very quick,
and the life of the crucibles is much prolonged.

Many other types of tilting furnace have been designed,
but in spite of their advantages none of them have come
largely into use.

One of the latest, and probably the best, of the
tilting furnaces for solid fuel, is that recently introduced
by the Morgan Crucible Company. It is made in several
types and sizes, and with a melting capacity of from
400lbs. to l,000lbs. It is made either in the fixed form
here described, in which the metal is poured into movable
moulds or into a ladle, or in a movable form in which
the body of the furnace can be carried by an overhead
crane to the moulds.

In the fixed type, the body is an octagonal steel
casing hinged to a solid framework in such a way that
the spout is the centre of rotation. In the movable
type the body is carried on trunnions in the usual way.
The casing is lined with refractory firebrick, so as
to make a cylindrical fuel chamber, and the crucible

FIG. 12(5.— MORGAN CRUCIBLE TILTING FURNACE.

FIG. 127.— MORGAN CRUCIBLE TILTING FUR

268 PREPARATION OF ALLOYS.

is, of course, fixed in this so that it is not removed until
it is worn out. At the bottom of the casing is fixed the
grate. A double casing is also provided round the
lower portion of the furnace, from which air holes
communicate with the interior. When at work the
body rests on an iron base forming an air chamber, into
which the air is blown, and thence finds its way through
the bars and through the openings in the furnace wall.
The furnace is covered with a movable hood, and the
products of combustion are carried away by an iron pipe.
This pipe is surrounded by another pipe, and the air to
be supplied is passed through the annular space between
the two pipes, thus becoming heated to 300° C. before
entering the furnace.

The combustion is very complete, as much as 17 per
cent, of carbon dioxide having been found in the products
of combustion, and therefore a very high temperature
can be obtained. The air should be supplied at a pressure
of IJin. to 2in. water gauge, and about 250ft. of air is
required per minute.

The following figures are quoted by the Morgan
Company as examples of the work that can be done.

Two days' work, starting cold, and working on gun
metal, five heats being worked each day : —

. First Day. Second Day.

Total pounds of metal melted .. 2191 .. 2102
coke used . . 400 .. 386J

Time under blast . . 6 hrs. 52min. 6hr. 21min.

Average pounds of metal per

pound of coke (including

first heat) 5-48 5-44

Pounds of metal melted per

hour ..' ., .. 320 330

One day's work on cast iron starting cold, four heats
being worked : —
Total pounds metal melted .. .. 1360

,, ,, coke used . . . . . . 383

Time under blast (including first heat) 6 hrs. 10 mins.

Average pounds of metal per pound of coke . . 3-55
Pounds of metal melted per hour „ . . . ... 220

PREPARATION OF ALLOYS. 269

GAS AND LIQUID FUEL FOR CRUCIBLE FURNACES.

There are many objections to the use of solid fuel
where a high temperature is required : (1) The frequent
addition of cold fuel is necessary, so that it is impossible
to maintain a uniform temperature over any long period ;
(2) the presence of ash in the fuel constantly tends to
corrode the furnace and the crucibles ; (3) the loss of
fuel through the fire-bars by poking may be considerable.

To overcome these objections, the use^of gas and oil
fuel has frequently been suggested.

Gas-fired Furnaces. — Gas furnaces have never come largely
into use. As usually used, gas made in gas producers is not
economical for work on the small scale required for melting

FIG. 128. — MONARCH NON-TILTING CEUCIBLE FURNACE FOB USE WITH
OIL OR GAS.

alloys in crucibles, and though it has been tried in
Sheffield for melting crucible cast steel, which is done
on a much larger scale, even there it has never become
general.

Gas furnaces may be of the usual crucible type,
the gas and air being supplied into the melting
chamber, or they may be of much larger size,

270 PREPARATION OF ALLOYS.

capable of holding many crucibles, in which case
they are practically gas reverberatory furnaces of the
ordinary type, with holes in the roof, by which the cruci-
bles can be let down on to the hearth. But little
attention seems as yet to have been given to the use of
gaseous fuel for foundry work ; probably when more
attention is given to it, it may be found to be not only
practicable but economical to use gaseous fuel in large
foundries. Of course, it is only producer gas of some form
that can be used economically. Coal gas, whilst an
excellent fuel, is far too expensive.

Oil Furnaces. — The use of oil as a fuel has many
advantages. It is cleaner than solid fuel, does not yield
any ash to corrode the furnace and crucibles, and it has
the advantage over gas that it does not need a plant to
be continually going for its production. The oils used
are always heavy oils, which are comparatively cheap,
and they are blown into the furnace in the form of spray
by means of a blast of air, so that air under pressure is
necessary. Oil has been little used for crucible furnaces,
but the recently introduced Steele-Harvey furnace seems
to promise success, and in the few works in this country
where it has been introduced it has been successful.
It consists of a circular steel shell, lined with two layers
of firebrick ; the size of the inner cavity is such as to
leave a combustion space round the crucible.

The furnaces are of two types. In the fixed type, the
crucible is supported on a block of refractory material,
and the furnace is worked exactly in the ordinary way.
In the tilting furnace the crucible rests on a block of
refractory material, and is so supported by side blocks
that it does not move when the furnace is tilted. The
casing is carried on trunnions, so that it can be tilted to
pour out the metal, or it may be arranged so that it can
be lifted away bodily by means of a crane and carried
to the moulds exactly as a large crucible might be carried.
As the crucibles are not handled, they can be made
of much larger size than when they have to be used in
ordinary furnaces. As in the case of other tilting furnaces,
neither the furnace nor the crucible need cool much
between the heats, and therefore the melting may be
rapid.

PREPARATION OF ALLOYS.

271

These furnaces are made up to a capacity of 750lbs.
per heat. The lining is said to stand about 500 heats,
and the crucible may stand 30 heats or more. As the air
is supplied under pressure no chimney draught is needed.

The air is required at a pressure of about 401bs. In a
paper read before the Pittsburgh Foundrymen's Asso-
ciation in 1905, Mr. T. W. Krause gave some details as to

FIG. 129. — STEELE-HARVEY FURNACE IN POURING POSITION.

experiments with this furnace at the works of the Mary-
land Steel Company.

Taking one day's work as an example, he gives : —

Charge. Time of Melting.

1st, 750lbs. ... 2hrs. 30 min.
2nd, 738lbs. ... 2 hrs. 30 min.

l,448lbs. 5 hrs. 20 min.

The loss in melting was 1 • 06 per cent.

Oil Consumed,

32 gals.

272 PREPARATION OF ALLOYS.

On a four days' test the figures were : —

Metal— Pounds. Oil— Gallons. Cost. Loss.

1st day 1,488 32 $0-80 1 06

2nd „ 2,252 56 1-40 1-19

3rd „ 2,579| 65 1-62 1-96

4th „ 2,534 62 1-55 1-03

8,853J 215 $5-37 I- 06 average %
Average cost of melting lOOlbs., including oil and

proportional part of cost of crucible, 1 34 cents.

Comparing this with a work in a coke furnace at the

same works, the cost was : —

1st day
2nd „
3rd „
4th

Pounds of Metal.

Cost of Coke.

Loss.

1,465J
975

1-98
1-24

2 7
2 8

1,547
534

1-89
1-29

2-8
2-0

4,521J $6-40 2. 80 average %

Melting lOOlbs. of metal cost 33-3 cents.

The life of the crucibles was on an average 24 heats ; in
one case Mr. Krause states that a crucible which had run
seven heats in the coke furnace, and which was put aside
as done with, was put into the oil furnace and lasted 22
heats.

The great question, of course, is that of cost. The
furnace for the oil is more costly and air under pressure has
to be supplied, but against this has to be set the larger out-
put and greater cleanliness and convenience, and the longer
life of the crucibles, always a large item of expense. The
principal point which will determine the economy will be
the cost of fuel.

The oil will have a specific gravity of about • 87, so that
a gallon will weigh about 7 • 271bs. It will have a calorific
power of about 20,000 B.Th.TL, so that the heat evolved
by the combustion of 1 gal. will be about 145,400 units. A
good coke will have a calorific power of about 13,000
B.Th.U., so that 1 gal. of oil will be equivalent in actual
heating power to about lllbs. of coke. It will be seen,
therefore, that the oil must be very cheap to compete
with coke. It must be remembered that the oil can be

PREPARATION OF ALLOYS.

273

much more economically burned than the coke, and the
incidental expenses are all in favour of the oil. It is
generally said that in this country oil can compete with
solid fuel when it can be obtained for about Id. a gallon. Of
course many other points besides the actual cost of the
oil have to be taken into account.

REVERBERATORY FURNACES.

The reverberatory furnace is usually used when large
quantities of metal have to be cast, as it avoids the use of a
large number of crucibles. Such furnaces have advantages

FIG. 130. — REVERBERATORY FURNACE FOR THE PREPARATION OF ALLOYS.

and disadvantages. They are not economical in fuel, but
coal can be used which is much cheaper than good coke.
There is usually a much greater loss by oxidation, and in
general the reverberatory furnace is not suited for melting
easily oxidisable metals. As the metal is in contact with
the products of combustion, and these will usually contain
sulphur from the fuel, sulphur may be taken up by the
metal. The metal may be kept melted any required
time (in the case of brass it will be constantly losing zinc),

274

PREPARATION OF ALLOYS.

and it should be kept covered with a layer of charcoal to
prevent oxidation, or sand may be added, so as to form
with the oxides produced a layer of fusible slag, which is

FIG. 131. — SMALL REVERBERATORY FURNACE FOR PREPARATION OF ALLOYS,
TO MELT 1J CWTS. SCALE ABOUT ^. (LONGITUDINAL SECTION).

a much better protection against oxidation. The rever-
beratory furnace is very convenient for melting down
scrap and casting it into ingots, so as to obtain ingots of
definite and fairly uniform composition.

Solid Fuel Furnaces. —
The furnaces used in
the brass foundry are
usually fed with solid
fuel, and are simply
ordinary reverbera-
tory furnaces. The
bed must slope to-
wards a tap hole, so
that the metal may
accumulate ready for
tapping ; the hearth
may slope downwards
from the fire bridge
to the flue, or it may
dip in the centre and
rise again. The bed

FIG. 132.— SMALL REVERBERATORY FURNACE . 11 * j n

(HALF CROSS-SECTION AND HALF ELEVATION.) IS Usually Ot sand Well

PREPARATION OF ALLOYS. 275

rammed, laid, over a layer of firebrick set on edge, or
sand or firebrick alone may be used. As a high tempera-
ture is required, the area of the fireplace must be large,
usually about one-fourth or one-fifth of the area of the
hearth.

The furnace should be cased with iron, and held by
vertical stays tied by cross ties, which can be loosened
as the furnace is heated, and tightened up when it cools,
so as to allow for the expansion and contraction of the
roof. The charging door is usually near the fire bridge, and
the tap hole may be either at the side or at the end. This
is stopped with clay when the charge is melting, and is
opened by means of an iron rod for tapping. Whilst as
a rule the air furnaces are of large size, in some works
small furnaces are used quite successfully. The amount of
coal consumed will vary with the size of the furnace.
As a rule, the larger the furnace the more economical
it will be. About 50lbs. of coal will be required to melt
lOOlbs. of metal, but with a five-ton furnace working at
its full capacity the amount may be reduced to 331bs.
The melting down is much more rapid than in a crucible
furnace.

Gas Furnaces. — For large reverberatory furnaces gas
may be satisfactorily used.

Oil Furnaces. — Many oil-fired reverberatory furnaces
have been introduced. These consist as a rule of iron
cylinders lined with refractory material, and carried
on trunnions so that they can be tilted to pour out the
metal, the oil being sprayed in by means of an air blast
at one end. The best known of these is the Rockwell
furnace, though there are many others on the market.
This furnace is frequently built double, consisting
then of two independent chambers, the oil being blown
into them alternately. It is obvious that two different
alloys could be melted at the same time and the furnace
could be worked continuously in one direction if preferred.
The furnace body is lined with ganister, which is rammed
round a core.

The shell is made in two halves, which are hinged
together. These furnaces may be made of any size,

276

PREPARATION OF ALLOYS.

but the usual capacity is about 5001bs. to l,0001bs. for
each furnace.

Some experiments were made by Mr. W. S. Quigley,
and are described in a paper read before the Pittsburg

FIG. 133. — KOCKWELL OIL FURNACE.

FIG. 134. — VIEW SHOWING POURING POSITION.

FIG. 135.— VIEW SHOWING FUBNACE OPENED

FOB RELINING.
ROCKWELL OIL FUBNACE.

Foundrymen's Association in 1905, and the following
figures were published : — .

PREPARATION OF ALLOYS.

277

Metal charged

Oil used in melting, including that

used in heating up
Oil used per lOOlbs. of metal melted
Time required to heat up furnace,

starting cold

Oil consumed in heating up
Actual time furnace was in blast,

including heating up . .
Time per lOOlbs. of metal made . .
Weight of metal per minute
Average time per heat of 5001bs.

7,0001bs.

93 gals.
1-3 gals.

27 mins.
8 gals.

7 hours 58 mins.
6-8 mins.
14-6lbs.
34 mins.

FIG. 136. — SECTION, LUNKENHEIMEB FURNACE.

Another excellent furnace is the Lunkenheimer,
made by the Lunkenheimer Foundry Company. The
furnace consists of a steel drum lined with firebrick.
There are two openings through the shell, only one of
which is in use at a time, the other being closed with a
brick. The cylinder is carried on trunnions so that it
can be turned over to pour the metal, and the oil for
combustion is supplied through one of these trunnions.
When the pouring hole is badly worn it is closed by
means of a tile and clay ; the furnace is reversed and

278

PREPARATION OF ALLOYS.

the brick is removed from the other opening in the
shell.

From six to seven heats per day of ten hours are
said to be made in these furnaces (each heat 550 Ibs.),
with a consumption of 2 gals, to 2J gals, of oil per lOOlbs.
of metal melted. The linings are said to last 300 or
400 meltings.

Cupolas Cupolas are rarely used for melting alloys.

Mr. J. F. Buchanan says, " In the cupola the fuel is in

FlG. 137. — LUNKENHEIMEB FURNACE.

contact with the bronze, and gases and impurities are
absorbed by the molten metal from the waste products
of combustion." " To obtain satisfactory results, the
pressure of the blast must be lowered, and the more
fusible metals — tin, lead, zinc — must be mixed in the
ladle instead of passing through the cupola to form the
alloy." The alloy is therefore not likely to be as
homogeneous as when it is melted in crucibles.

OTHER METHODS FOR THE PREPARATION OF ALLOYS.

ALLOYS are always made on the large scale by the methods
already described, but they can be prepared by other
methods which are of interest or of practical utility.

PREPARATION OF ALLOYS. 279

Preparation of Alloys by Pressure. — As early as 1878 Prof.
Spring, of Liege, succeeded in alloying metals by pressure.
The metals, in a fine state of division, were mixed, and
then were subjected to a pressure of some thousands
of atmospheres. Under a pressure of about 2,000
atmospheres, or 13 tons on the square inch, lead is
compressed into a solid block, whilst bismuth, though
a very brittle metal, unites under a pressure of 6,000
atmospheres. Under similar pressures the metals can
be made to unite to form alloys which have all the pro-
perties of alloys formed in the usual way by fusion. For
instance, finely-divided bismuth, lead, tin, and cadmium,
mixed in the proper proportions and subjected to pressure,
produced a fusible alloy, the melting point of which was
below 100° C. The alloys have to be broken up, and
again subjected to pressure in order to ensure uniformity.

Only the metals which flow at a moderately low pressure,
such as tin, lead, cadmium, bismuth, &c., have as yet been
alloyed by pressure.

i Preparation of Alloys by Cementation. — In some cases metals
when heated in contact to temperatures far below their
melting point will unite, the one metal slowly diffusing
into the other. This process,1 however, is always very
slow.

Production of Alloys by Electro-deposition. — It is well known
that when an electric current is passed through a fused
metallic salt, or the solution of a metallic salt in water, the
salt is broken up into two portions or ions, the one, the
metal, being deposited on the electro-negative plate or
cathode, whilst the other, which may either be an element
or group of elements, is liberated at the electro-positive
plate or anode. This constituent may escape, if it is an
element and the anode is not attacked by it, or if it is a
complex group it may be broken up.

Thus, for instance, if a solution of copper chloride be
electrolysed, copper is deposited at the cathode and
chlorine is liberated at the anode. If a solution of copper
sulphate is used, copper is still liberated at the cathode,
but the group S04 which is liberated at the anode cannot
exist in the free condition, but at once combines with
water, forming sulphuric acid and liberating oxygen.

280 PREPARATION OF ALLOYS.

When the anode is the same metal as that which is being
deposited, it is attacked and dissolved, so that the
solution retains its strength practically unchanged, and
the energy given out by the dissolving anode exactly
compensates for that used in depositing the metal at the
cathode. Advantage is taken of these facts in the electro-
deposition of the metals, which is now carried on on a very
large scale, gold, silver, nickel, and copper being the
principal metals deposited.

The metal may be thrown down in the form of a very
thin layer for ornamental or protective purposes, as in
the case of gold in electro-gilding, silver in electro-plating,
nickel in electro-nickelling, or other metals ; or in larger
quantities when articles of small size, such as medals,
medallions, plaques of silver or copper are prepared by
electro- deposition.

Whether a metal will be deposited or not depends on
the solvent present. If the solvent be of such a nature
that the liberated metal would dissolve in it, it is obvious
that the metal will not be precipitated unless the pre-
cipitation be much more rapid than the solution, and then
only a small portion of the metal would be obtained.

Sir H. Davy obtained the alkali metals by decom-
posing fused chlorides by means of an electric current,
but the metals cannot be obtained from aqueous solutions
of the salts, because the liberated metal is at once
dissolved in the water, and hydrogen is liberated ; so
similarly zinc is not deposited in an acid solution because
it is dissolved by the acid, hydrogen being evolved.

In order to obtain a pure metal by electrolysis,
the solution or electrolyte is kept as pure as possible,
and of such a character that any impurities present
are not likely to be thrown down. In the case of
electro-deposition, it is usually quite easy by using a
pure solution to start with and an anode of a pure metal, to
have a solution in which there is no metal except that
to be precipitated. Even in this case, however, the
current must not be too strong, or the water present may
be decomposed and hydrogen be liberated with the metal,
thus tending to make the deposit unsound, and at the
same time to waste energy.

PREPARATION OF ALLOYS. 281

In the electro-refining of copper, where an impure
blister copper is used as the anode, the electrolyte is
always a solution of copper sulphate acidified with
sulphuric acid ; in this some of the constituents of the
blister copper, such as gold and silver, will not dissolve,
and others, such as zinc and iron, which do dissolve, are
not precipitated, so that though there are impurities
present in the blister copper, the electro deposited
copper is nearly pure.

When a current is passed through a solution which
contains two or more metals which can be deposited
under the conditions of the experiment, all the
metals may be thrown down, but the proportions
in which they will be deposited will depend on
at least three conditions : (1) The proportions in
which the metals are present in the solution, (2) the
chemical character of the metals, and (3) the strength of
the current. So that when it is desired to deposit two or
more metals at the same time, all these conditions must
be carefully considered and arranged. When two metals
which alloy easily are thrown down together they tend
to form alloys.

The principal alloys prepared electrolytically are
brass, bronze, German silver, and gold alloys.

Electro-deposition of Brass. — Of all the alloys, brass
is that most largely deposited. Articles of zinc, white
metal, &c., are frequently coated with brass to give
them the appearance of brass articles. Such articles are
much more easily made, and are therefore much cheaper
than those of real brass.

The electrolyte solution must, of course, contain both
copper and zinc. It must be of such a character that
both metals can be precipitated together, and it must con-
tain the metals in the proportions in which they are to
be deposited, otherwise its composition would vary as
the deposition goes on. Obviously, the solution must not
be acid, or at least not strongly so, or the deposition of the
zinc would be prevented. Very many solutions have
been suggested, but that which is always used in
practice is a double cyanide of the metals, which
may be made in various ways, and which always contains

282 PREPARATION OF ALLOYS.

an excess of cyanide. It is usually used cold. The solu-
tion may vary in strength by irregular deposition, and its
strength is brought up to the required point by the
addition of copper or zinc as the case may be.

The anode is usually of brass of as nearly as possible
the composition it is intended to deposit, or in rare cases
it is composed of plates of copper and zinc. The former
is, however, much preferable.

The deposit depends much on the strength of the
current. As a rule, copper is much more readily deposited
than zinc, and this difference becomes greater the weaker
the current, so that with a very weak current nearly
pure copper might be deposited, whilst the stronger the
current the more nearly will the rate of deposition of the
zinc approach that of the copper. As the deposition
goes on the colour of the deposit is carefully watched. If
it becomes too red either the current is made stronger or
more zinc is added to the solution. If it becomes too
white, either the current is reduced or more copper is
added to the solution. The limits between which the
current can be varied is not very large, because it is
essential to produce a close adherent deposit, which can only
be done by a current which has approximately definite
strength. As variations in the strength of the solution
might cause change in the colour of the deposit, and if the
solution were left at rest, local changes might take place
which would be only slowly rectified by diffusion, the
solution is kept continually stirred or circulated.

Electro-bronzing. — Electro-bronzing is not very fre-
quently used, it being usually easier to deposit a layer of
copper and then to colour it to imitate bronze. Bronze
is usually precipitated from alkaline solutions.

German Silver. — German silver may be precipitated from
a cyanide solution of copper and nickel, with or without
the addition of zinc. It is, however, rarely used, since
nickel itself is much more easily precipitated, and answers
most purposes quite as well. It is recommended by Watt
for certain purposes, as the colour is more pleasing than
the silver white of pure nickel. This is, of course, a
matter of taste.

PREPARATION OF ALLOYS. 283

Coloured Gold. — Pure gold has too pale a colour for
many tastes, and the gold used in gilding is therefore
often coloured by the addition of a little copper ; that
is, in place of pure gold, an alloy of gold and copper is
thrown down. The deposit can be made of any com-
position so as to imitate gold of any carat required.
Even gold articles are frequently electro-gilded to give the
surface the desired shade.

The alloys thrown down by an electric current do
not seem to differ essentially from those prepared by
fusion, but they are usually highly crystalline and porous.

BIBLIOGRAPHY.

THE number of books dealing with Alloys is not large.

THE METALLIC ALLOYS (KRUPP & WILDBEROER).

Translated and edited by WILLIAM T. BRANNT.
Philadelphia: H. Carey Baird. 1889.

MIXED METALS OR METALLIC ALLOYS

ARTHUR H. HIORNS. Second edition. 1901.
Macmillan & Co.

This is a very valuable book.

A good account of Alloys, especially from the engineering stand-
point, is to be found in most of the books on the Materials of
Engineering.

THE MATERIALS OF ENGINEERING.

(Prof. THURSTON), Vol. II., (The Non-ferrous Metals), may be
specially mentioned.

THE REPORTS OF THE ALLOYS RESEARCH COMMITTEE
OF THE INSTITUTION OF MECHANICAL ENGINEERS

are invaluable* Eight reports have now been issued, of which the
fifth and sixth treat of Iron and Steel Alloys

These reports should be studied by all interested in the subject.
The eighth report, which deals with the Alloys of Copper and Alumi-
nium, is a model of how such work ought to be done.

Reference is made in the text to various papers to which reference
may be made, but as the number of these is constantly and rapidly
increasing, a list of them would be of little value.

INDEX.

PAGE

Admiralty Brass 105

" Ajax " Plastic Metal 153

— Standard Metal 153

Alloys, Colour of 8

— Micro-structure of 61

— Nature of 1, 3, 7

— Preparation of 244, 279

Allotrimorphic Crystals 57

Aluminium 85, 197

- Alloys, Preparation of 254, 257

- Bronze 155, 168

- Alternating Stress Test 163

- Colour 155

— Comparison with Steel 167

- Containing Other Elements. . 168

- Cooling Phenomena 163, 164

— Corrosion of 166

- Ductility 161, 162

- Hardness 156

- Heat Treatment 165

- Microstructuie 1C6

- Specific Gravity 155, 156

- Tensile Strength 157-160

Torsional Strength 163

— Uses of 168

- Copper Alloys 198, 199

Zinc Alloys 202

- Iron Alloys 205, 204

— Magnesium Alloys 201

- Nickel Alloys 202

-Tin Alloys 181, 204

— Tungsten Alloys 205

- Zinc Alloys 205

- in Bearing Metal 186

— in German Silver 212, 213

Amalgams 239-243

— Preparation of 258

Analyses of " Ajax " Metals 153

— — Aluminium, Copper, Zinc
Alloys 202

Aluminium, Tin, Silicon

I Alloys 204

Arsenical Bronzes .153, 154

- " Babbitt's " Metal 187

Bearing Metals 154

PAGE

Analyses of Bell Metals 138

-Brasses 90-93, 104-106

- Britannia Metal 181

- Bronzes 123, 124, 139

- Cooper's Gold, &c 237, 238

— Commercial Antimony 84

Bismuth 85

- Copper 81

- Lead 84

- Nickel 210

- Tin 83

- Fusible Alloys 206, 208

- German Silver 214, 215

Gold Solders 227

- Hensler's Magnetic Alloy. . . . 171

- Hoyle's Alloy 194

- Jacana Metal 194

— Machinery Brasses 148

- Plastic Metals 152, 153

- Silver Solders 233, 234

- Specula Metals 140

- Type Metal 183

- White Bearing Metals 188

- White Metal Patterns 196

Annealing 68, 69

— Temperature of 70

Anti-friction Metals, White 184

Antimony 84

— Copper Alloys 170

— Lead Alloys 181, 190

— Silver Alloys 234

— Tin Alloys 177

Atomic Volume, Definition of ... .^ 11

Babbitt's Metal "187

Banca Tin ''. 81

Bearing Metal 148

— Bronze 149

Bell Metal 138

Benedick's Formula for Hardness. . 157
Bismuth 85

— in Bearing Metal 186

Blowholes 39

— Prevention of 40

11.

INDEX,

PAGE

Brasses, Aluminium in 107

— Antimony in 109

— Arsenic in 110

— Bismuth in 112

— Calamine 254

- Classification of 103-106

— Colour of the 90

— Extensibility of the 95

— Foreign Constituents in 106

- Fracture of . .90, 102, 109, 111, 113

— Freezing Point Curves 96, 103

— Fusibility 95

- Hardness 95

- Impurities in 87

— Iron in 113

- Lead in 114

— Machinery 147

— Manganese in 117

— Microstructure of 98, 99, 100

— Naval 118

— Nickel 118

— Oxygen in 118

— Phosphorus in 118

— Preparation of 248

- Properties of the 89

— Range in Composition of 88

— Specific Gravity of the 91

— Sulphur in 87

- Tenacity of 92

- Tin in 118

Brazing Metal 104, 106

Brinell's Test for Hardness 156, 157

Britannia Metal 180

Bronze 120

— Aluminium 155-168

— Bearings 148, 149

- Castings : . . . 147

— Coinage 138

— Colour of 123

— Ductility of 125

— Electric Conductivity of 126

— Freezing Point Curves. .121, 122, 132

— French (Brass) 106

- Friction of Lead 151

— Heat Treatment of 133-137

— Lead in 150-153

— Manganese 145, 146

— Medal 138

- Mirror 123, 140

— Microstructure of 126-134

- Nickel 152

— Oxides in 140, 141

- Phosphor 141, 143, 144

— Preparation of 247

- Quenching of 135-137

— Range of Composition 120

PAGE

! Bronze, Rate of Solidification 131

1 — Silicon 145

- Specific Gravity of 123, 124

- Statuary 139

- Tenacity of 124

- Varieties of 137-140

- Wear of Lead 151, 152

- Wear of Plastic 153

Burning 74

Carr's Crucible Furnace 263

Castings, Bronze 147

— Under Pressure 41

Cementation, Preparation of Alloys

by 279

Changes During Cooling 64

Chemical Analysis, Limits of 4

Chemical Compounds, Alloys of

Metals Forming 31

— of Copper and Zinc 97

of Copper and Tin 131

- Nature of 2

Coinage, Bronze 138

— Gold 220

— Silver 231

Colour of Alloys 8

— Aluminium Bronze 155

— of Brass 90, 102

- of Bronze 123

- of Gold-silver Alloys 221

— of Nickel-copper Alloys 210

— of Silver-copper Alloys 229

Colouring Silver 232

Comparison of Aluminium Bronze

with Steel 167

Composite Cooling Curves (see

Freezing Point Curves).
Composition, of the Brasses, Range

of 88

- Bronzes 120

Compression Strength of Tin-lead

Alloys 175

— Tests on Bearing Metals 187, 193

Conductivity, Electrical 12-15, 126

— Heat, of Bronzes 126

Cooling, Effect of Slow 72

— Internal Changes During 64

— Phenomena of Aluminium-Bronze

160-164
Cooling Curves of Water 21

— Water and Salt 24

Cooper's Metals 237, 238

Copper 75

— Admiralty Test for 79

INDEX.

111.

PAGE

Copper Aluminium 155

— Amalgam 241

— Analysis of Commercial 78

— Antimony Alloys 170

— Chili Bar 80

— Commercial 78

— Gold Alloys 222

— Iron Alloys 170

— Lead Alloys 168

— Magnetic Alloys 171

— Manganese Alloys 170

— Nickel Alloys 210

— Platinum Alloys 237

— Properties of 75

— Silver Alloys 227

- Tin Alloys (see Bronze) 120

- Zinc Alloys (see Brass) 87

— Zinc -aluminium Alloys 202

Corrosion of Aluminium Bronze. ... 166

— Aluminium-copper Alloys 200

Coulomb's Law of Friction 184

Crucible Furnaces 259-273

Crystals, Growth of 22, 35

at High Temperature 68

Cupellation 236

Cylindrical Crucible Furnace 262

" Diamantine " for Washing 47

Diffusion 71

Ductility 11

— of Aluminium Bronze 161

— of Brass 95

— of Bronze 125

Dutch Metal . . 104

Effect of Slow Cooling 72

- Slow Solidification 72

— Work 73

Electrical Properties of Alloys 12

— Conductivity of Alloys 12

— of Bronzes 126

Electro-bronzing 282

— Deposition of Alloys 279, 280

— Brass 281, 282

— Bronze 282

— German Silver 282

Electromotive Force 16

Curves 17, 18

Electrotype Copper 78

Electrolysis, Nature of 16

Electrum 221

Elongation (see Ductility) 11

PAGE
Elongation Curves ..108, 116, 117,

118, 125, 158,159

— (see also Extensibility) —

161, 162, 175, 198, 199

Emery for Polishing 47

Etching of Micro-sample 47

Eutectic Alloys of Copper Tin

121, 127-129

- Copper Zinc 96, 97, 99

- Gold Copper 223-225

- Iron Phosphorus 61, 62

- Lead Antimony 181

- Lead Tin 173

- Silver Copper 228

— Silver Antimony 234

- Tin Antimony 177

Eutectic, Definition of 26

— Stead's Classification of 60

- Under Microscope 58

Expansion by Heat 12

Extensibility Curves : —

Aluminium Brass 108

- Bronze 158-162

- Copper 198, 199

Brass 94

Bronze 125

Lead Brass 116

Manganese Brass 117

Tin Brass 119

Tin Lead 175

Fracture of Metals 43

Freezing Point Curves of Brass. .96, 103

- Bronze 127, 132

- Copper Gold 222

- Copper Silver 228

- Copper Tin 121, 132

- Copper Zinc 96, 103

- Lead Tin 31, 173

Salt Solution 25

— Silver Gold 29

- Typical 30, 32

French Bronze (Brass) 106

- Gold (Brass) 104

Friction 184

Frictional Resistance 186

Fuel for Crucible Furnaces 264

- Gas for Crucible Furnaces 269

— Liquid for Crucible Furnaces .... 270
Furnaces 259-278

— Carr's 263

— Crucible 259-273

— Cupola 259, 278

— Gas Fired 269

IV.

INDEX.

PAGE

Furnaces, Lunkenheimer 277, 278

- Monarch Non-tilting 269

— Morgan's Tilting 267

— Oil Fired 271,272

— Piat Oscillating 266

— Reverberatory 273-277

— Rockwell 275

— Steel-Harvey 270-272

— Tilting 266

— Weir's 264

Fusible Alloys 206-208

Fusibility 12

— of Brasses 95

— of Bronzes 121

Gas-Fired Crucible Furnaces 269

— Reverberatory Furnaces 275

Gases, Evolution in Metals, Result of 39

— Solution of, in Metals 38

German Silver 212, 213

Iron in 214

Varieties of 215

Gold 217

— Alloys (minor) 226

— Amalgam 239

- Coinage 220

- Colouring 283

- Electro Deposition of 283

— Fineness of 218

— Silver Alloys 220, 221

— Solders 226, 227

Gun Metal 137, 138

Hardness of Aluminium Bronze. . . . 156

— of Brasses 95

— Brinell's Test for 156

Heat Conductivity of Bronze 126

Heat, Expansion of Alloys by 12

— Treatment of Alloys 64

of Aluminium Bronze. . 165, 166

- Brass 65-69

Bronze 133-137

Hensler's Magnetic Alloy 171

Homogeneity, Conditions of 33

Illumination 51

- Kinds of 52

Illuminators, Vertical 50

Impurities in Commercial Antimony 84
Bismuth . 85

PAGE

Impurities in Commercial Brass . . 87

Copper 78

Lead 84

Tin 83

Zinc 82

insolubility, Mutual, of Metals 27

Investigation, Methods of 4

Iron and Aluminium 203

— in German Silver 214

Jacana Metal 194

Jewellery Gold 224

Kalchoids . . .147

Lead 84

- Alloys (Minor) 181

- Antimony Alloys 182, 190

— Arsenic Alloys 183

-in Bronze 151, 152

- Copper Alloys 168, 169

- Silver Alloys 235

— Tin Alloys 172-177

- Tin, Antimony Alloys 182, 190

Light Alloys 197

Liquation 37

Liquid Fuel 269, 270

Lunkenheimer Furnace 277, 278

Machinery, Brasses 147

— Bronzes 147

Lead in 151

Magnesium-aluminium Alloys 201, 202

Magnetic Alloys 171

Magnolia Metal 191

Manganese Bronze . . . . : 145, 146

- Copper Alloys 170, 171

Mechanical Mixtures 2

Medal Bronze 138

Melting Points of Copper-aluminium

Alloys 164

- Lead-tin Alloys 174

— Tin-aluminium Alloys 205

Mercury 239

Metals, Crystalline surface of 55

— Evolution of Gas from Cast 39

— Forming Definite Chemical Com-
pounds 31

- Impurities in 78, 82-87

INDEX.

PAGE

Metals Insoluble in One Another . . 28

— Partly Soluble in One Another . . 27

— Soluble in One Another 30

— Solution of Gases in 38

— Treatment for Sound Castings. 40, 41

- Used for Alloys . . . .75-85, 209, 236
Microscope, Aid to Investigation . . 4

— Chemical Compounds Under. ... 57

— Examination 51

— Eutectics Under the 60

— Metallurgical 49

— Preparation of Samples for .... 45

- Principle of Methods used for . . 44

- Pure Metals Under 56

Microstructure of Alloys 61-63

— Aluminium Bronze 165

— Aluminium -copper Alloys 201

— Brasses 98-100

— Bronzes 126-133

— Copper-nickel Alloys 210

— German Silver 215, 216

— Lead, Tin, Antimony Alloys. 192, 193

— Silver -antimony Alloys 235

— Tin-antimony Alloys 178

— Tin-lead Alloys 174

— White Bearing Metals 189

Mirror Metal 123, 140

- Cooper's » 237

Molecular Mobility 65, 71

Muntz Metal 105

Mystic Metal 191

Naval Brass (for Condenser Tubes). . 118

Nickel 209

— Alloys : Preparation of 258

— Aluminium Alloys 202

— in Bronze 152

— Commercial 210

— Copper Alloys 210

— Silver Alloys 235

Oil-fired Crucible Furnaces 270-272

— Reverberatory Furnaces . . 275-278
Overheating 68

Pewter 172, 176

Phosphor Bronze 141, 143-144

— Copper 141

Phosphor Tin 142

Photo-micrographs, Al. Bronze.. 163, 164

— Aluminium Copper 200

— Antifriction Alloys 189-195

— Bearing Bronze 149

— Brasses 65-69, 98-100

PAGE
Photo-micrographs, Bronzes . . 127-131

— Cementite 60

- Copper-gold Alloys 224, 225

— Copper-lead Alloy 169

- Nickel Alloy 211

- Zinc Alloy 216

- Silver Alloy 228, 229

- Ferrite 56

— German Silver 216

- Gold-copper Alloys 224, 225

- Graphite and Ferrite 61

- Iron Phosphide 61, 62

- Jacana Metal 194

- Lead-antimony Alloy 182

— Arsenic Alloy 183

- Magnolia Metal 191

- Manganese Bronze 146

- Pearlite 59

- Phosphor Copper 141

-Tin 142

- Silicon Ferrite 57

— Silver -antimony Alloy 235

- Copper Alloy 228, 229

- Lead Alloy 59

- Standard Gold Alloy 224

- Tin-antimony Alloys 178-180

Physical Properties of Alloys 5-12

Plastic Bronze 152, 153

Plasticity of Bearing Metal 185

Platinum 236

- Alloys (Minor) 239

Preparation of 258

- Copper Alloys 237

- Iridium Alloys 238

- Rhodium Alloys 238

— Silver Alloys 238

Polishing Agents 47

-Machines 43, 45,46

Potential Difference 76

Preparation of Alloys 244

- by Cementation 279

- by Electro-deposition, 279,280

- by Pressure 279

- of Low Melting Point 246

Aluminium Alloys '254-257

— Amalgams 258

- Brass 248, 254, 281, 282

- Bronze 247, 282

- Calamine Brass 254

- Coloured Golds 283

German Silver 282

Micro-sample 45

- Nickel Alloys 258

- Platinum Alloys 258

Pressure, Casting under 41

— Preparation of Alloys by 279

VI.

INDEX.

PAGE

Quenching of Bronzes 134-137

Reduction Area Curves. Al. Bronze 162

Bronze 134-136

Regulus of Venus 170

Research, Methods of 5

Reverberatory Furnaces 273, 274

Rockwell Furnace •'.... 275-277

Scorification 235

Segregation 36

— of Bronze 137

- Prevention of 37

Silicon Bronze 145

— in Aluminium Bronze 168

Silver Alloys 227

- Amalgam 243

- Coinage Alloys 231

- Colouring 232

— Copper Alloys 227, 228

- Lead Alloys 235

- Nickel Alloys 235

— Platinum Alloys 238

- Solders Alloys 233

— Standard Alloys 230

— Tin Alloys 234

Sodium Amalgam 243

Solder 176-7

— Brazing 105, 106

- Gold 226, 227

- Nickel 216

- Silver ' 233

Solid Diffusion 71

Solid Fuel for Furnaces. . . . 259-261, 274
Solidification, Conditions of 33-35

— Effect of Slow 72

— of Metallic Alloys 27

— Mode of 35

— Phenomena of 5, 20, 33

— of Salt Solutions 23

Solutions 22

Pure Substances 21

Solubility, Mutual, of Metals 28-33

Solutions, Character of 3

— Essentials of 20

— of Gases in Metals 38

- Solid 3

— — under the Microscope 58

Specific Gravity 9

— of Aluminium Bronze .... 155, 156

- Brass 91

- Bronze 123, 124

— Tin-lead Alloys 172

Speculum Metal 123, 146

PAGE

Standard English Brass 104

- Gold 224

- Silver 230

Statuary Bronze 139

Stead's Polishing Machine 46

Structure of Pure Metals ... 54

Tenacity 11

— of Aluminium Bronze .... 159-162

- Brass 92

- Bronze 124

Tensile Strength : Curves of —

- Aluminium Brass . , 107, 108

- Bronze 107, 158

- Copper 198, 199

- Brass 94

- Bronze 125

- Lead Brass 116

— Manganese Brass 117

— Tin Brass 119

- Lead 175

Ternary Alloys . .148, 191-196, 208, 212

Tin

— Aluminium Alloys 181, 204

Silicon Alloys 204

— Amalgam 242

— Antimony Alloy 177

- Lead Alloys 172

- Silver Alloys 234

Tombac 104

Tungsten Aluminium Alloys . .205, 206

Type Metal 183

Uses of Aluminium Bronze , . 168

Wear in Bearing Metals 151

Weir's Crucible Furnace Modification 265

White Alloys 172

— Antifriction Alloys 184

— Metal Patterns. 196

Wolframium ... .206

Yield Point Curves: Al. Bronze.. 158-162
Aluminium Copper. . 198, 199

Z
Zinc 81

— Aluminium Alloys 205

— Cadmium, Antimony Alloys 194, 195

— Copper, Aluminium Alloys . . , . 202

— Impurities in 82

— Tin Alloys 181

Ziskon 205

THE

PHOSPHOR BRONZE CO.,

Telegrams: "PHOSBRONZE, LONDON"; "PHOSPHOR, BIRMINGHAM."

87, SUMNER STREET, SOUTHWARK, LONDON, S.E.,
And at CHESTER STREET, ASTON, BIRMINGHAM.

SOLE MAKERS OF THE
Original

PHOSPHOR BRONZE.

"COG WHEEL BRAND11
and "VULCAN "

The Best and Most Durable Alloys for all kinds of

SLIDE VALVES, BEARINGS, BUSHES,

and other parts of Machinery exposed to pressure, friction and wear ;

Propellers, Pump Rods, Pumps, Piston Rings, Pinions, Worm Wheels, Motor Gearing, Ac.

Rolled and Drawn Phosphor and other Bronze Alloys.

CHILL CAST SOLID AND CORED BARS A SPECIALTY.

CASTINGS,

IN BRONZE, BUN METAL, AND ALUMINIUM ALLOYS.

" VULCAN " Brand Babbitt Metal, and Plastic White Metal
for Lining purposes,

"WHITE ANT" Metal, a superior Magnolia Metal,
cheaper than common Babbitt's.

" WHITE ANT1' Bronze, a superior quality of Fenton's Metal
for Car Bearings, do.

"PHOSPHOR" White Lining Metal, superior to Parson's
White Brass No, 2 for Marine Engines.

Phosphor Tin and Phosphor Copper,1' GOG WHEEL BRAND,"
the most reliable on the market.

Please specify the manufacture and Brands of the
Phosphor Bronze Co., Ltd., SOUTHWARK, LONDON, S.E.,
to prevent imposition.

SELLERS of REFINED METALS

AND

ALLOYS.

COPPER.

INGOT BRASS.

GUN METAL.

GERMAN SILVER.

NICKEL.

TINMEN'S and

PLUMBERS' SOLDER.

BAR LEAD.

SPELTER.

ALUMINIUM.

ANTI-FRICTION and

WHITE METALS.

MAGNOLIA and

BABBITT'S BEARING METALS.

BAR and CAKE TIN.

McKECHNIE BROTHERS,

Rotton Park Street,

BIRMINGHAM.

ALSO AT

WIDNES.

MANCHESTER, Trafford Park.
LONDON, 22, Emerson Street, S.E.
BL AYDON-ON-T Y NE.

Manufacturers of HAND, FOOT, and POWER-DRIVEN

METALLURGICAL TOOLS
and APPLIANCES.

We have been favoured with orders from—

Admiralty,

War
Department,

Universities,

Technical
Colleges.

CATALOGUE SENT POST FREE. Nat. Telephone 815.

J. CABLING & SON,

Gloucester St. and Orange Rd. B.9

(England).

"THE MECHANICAL ENGINEER"

Pocket Calculator

INVALUABLE TO

ENGINEERS, DRAUGHTSMEN, STUDENTS, AND
ALL WHO HAVE CALCULATIONS TO MAKE.

PRICE :

7/9 Post Free

UNITED KINGDOM.

9/- Post Free

An ACCURATE and

RELIABLE

INSTRUMENT.

ANYWHERE
ABROAD

" The Mechanical Engineer" Pocket Calculator. — Actual Size.

ENGINE DIVIDED.

FINGERS IN PERFECT ALIGNMENT.
SOLID NICKEL CASE.
LARGE DIAL, 2in. DIAM.
DIAL ALWAYS CLEAN & EASY TO READ.
COMPACT. Can be carried in the waist-
coat pocket.

Special

Features ;

~W~

SOLVES

Multiplication, division, square and cube roots, sines and cosines of

angles, areas of circles, logarithms, &c*
Write for full descriptive circular, post free.

THE SCIENTIFIC PUBLISHING COMPANY,

53, NEW BAILEY STREET, MANCHESTER.

BOOKS FOR ENGINEERS

FOWLER'S MECHANICAL ENGINEER'S POCKET BOOK.

This work is published annually, and comprises nearly 700 pages of rules,
tables, and data for mechanical engineers.

Price 1/9 cloth ; leather 2/9, gilt edges, post free.

FOWLER'S ELECTRICAL ENGINEER'S POCKET BOOK.

Tnis work is published annually, and comprises nearly 7CO pages of lules,
tables, and data for electrical engineers.

Price 1/9 cloth; 2/9 leathsr, gilt edges ; post free.

TESTING OF ENGINES, BOILERS, & AUXILIARY MACHINERY.

A treatise on the methods and instruments used in testing and experimenting
with engines, boilers, and auxiliary machinery. By W. W. F. PULLEN,
Wh. Sc., Assoc.M.Inst.C.E., &c. Price 15/6, post free.

TESTING AND STRENGTH OF MATERIALS OF CONSTRUCTION.

A treatise on the methods, machines, and instruments used in the testing of
materials of construction. By W. C. POPPLEWELL, M.Sc.

Price 10/10, post free.

COMMERCIAL MANAGEMENT OF ENGINEERING WORKS.

A comprehensive work dealing with the best methods of conducting the
commercial side of an engineering establishment. Second Edition, revised
and greatly enlarged By FRANCIS G. BURTON. Price 12/10, post free.

INDICATOR DIAGRAMS.

A treatise on the use of the indicator and its application to the steam engine.
By W. W. F. PULLEN, Wh. Sc., Assoc.M.Inst.C.E., &c. Price 6/4, post-free.

ENGINEERING TABLES AND DATA.

For the use of engineering students. Gives a large number of tables frequently
required by students in the laboratory in convenient form. By W. W. F.
PULLEN, Wh.Sc., A.M.Inst.C.E. Price 2/9, post free.

THE COMBUSTION OF FUEL.

With special reference to smoke prevention. Contains numerous illustrations
of mechanical stokers, forced draught furnaces, and refuse destructors. By
W. W. F. PULLEN, Wh. Sc., Assoc.M.Inst.C.E., &c. Price 3/8, post free.

PRACTICAL PATTERN MAKING.

A treatise for the use of pattern makers, foundrymen, apprentices, and
others. Profusely illustrated. By HERBERT AUGHTIE, ^.M.I.Mech.E.

Price 4/4, post free.

THE SLIDE RULE AND WATCH CALCULATOR.

A work dealing with the capacity and method of using the slide rule or its
later development, the pocket or watch form of calculator.

Price 1/1, post free.

STEAM ENGINEERING.

A treatise on boilers, steam, gas, and oil engines and supplementary machinery,
with instructions for carrying out numerous simple experiments. Second
edition, revised and greatly enlarged. By W. W. F. PULLEN, Wh.Sc.,
A.M.Inst.C.E. Price 4/5, post free.

STEAM BOILERS AND SUPPLEMENTARY APPLIANCES.

A practical treatise on their construction, equipment, and working. By
WILLIAM H. FOWLER, Wh.Sc., M.Inst.C.E., &c. Price 12/11, post free

THE SCIENTIFIC PUBLISHING COMPANY,

53, NEW BAILEY STREET, MANCHESTER.

BOOKS FOR ENGINEERS.

STATIONARY STEAM ENGINES.

A practical treatise on the design of stationary steam engines, illustrated
with numerous examples from actual practice. Edited by WILLIAM H.
FOWLER, Wh.Sc., M.Inst.C.E., &c. Price 12/11, post free.

METALLURGY OF IRON AND STEEL.

This work has been prepared to meet a need for a book which, in one volume
of moderate size, shall cover the whole field of the Metallurgy of Iron and
Steel. By A. HUMBOLDT SEXTON, P.I.C., F.C.S. Price 16/6, post free

LATHES, SCREW MACHINES, BORING AND TURNING MILLS.

A practical treatise on the design and construction of turning machines,
including lathes, automatic screw machines, boring and turning mills, p,nd
their accessories. By THOS. R. SHAW, A.M.I.Mech.E. Price 15/6, post free.

MACHINE TOOLS.

A treatise on the design and construction of planing, sb?,ping, slotting,
drilling, boring, milling, wheel-cutting machines, &c. By THOS. R. SHAW,
A.M.I.Mech.E. Price 15/6, post free.

MODERN LOCOMOTIVE PRACTICE.

A treatise on the design, construction, and working of steam locomotives. By
C. E. WOLFF, B.Sc., A.M.Inst.C.E. Price 10/10, post free.

PRACTICAL DYNAMO AND MOTOR TESTING.

A treatise on the methods and instruments used in the testing of dynamos

and motors. Second edition, revised, and greatly enlarged. By CHARLES F.

SMITH, Wh.Sc., A.M.I.E.E. Price 5/4, post free.

PRACTICAL ALTERNATING CURRENTS AND ALTERNATING CURRENT

TESTING.

Deals in an elementary and practical manner with the principles of alternating
currents and alternating current machinery. By CHARLES F. SMITH, Wh.Sc.,
A.M.I.E.E. Price 6/5, post frse.

COMPRESSED AIR.

A treatise on the theory and practice of pneumatic power transmission. By
W. C. POPPLEWELL, M.Sc.v A.M.Inst.C.E. Price 7/10, post free.

PRODUCER GAS.

A treatise on the properties, manufacture, and uses of gaseous fuel. By
A. HUMBOLDT SEXTON, F.I.C., F.C.S. Price 10/4, post free.

STEAM TURBINES AND TURBO COMPRESSORS.

A treatise on their design and construction. By FRANK FOSTER, M.Sc.

Price 10/11, post free.
CORROSION AND PROTECTION OF METALS.

With special reference to the preservation of engineering structures. Ey
A. HUMBOLDT SEXTON, F.I.C., F.C.S. Price 5/4, post free.

GAS AND OIL ENGINES.

A treatise on the design, construction, and working of internal-combustion
engines, operated by gas from town's mains, blast furnaces, pressure and
suction producers, as well as by oil and petrol. By HORACE ALLEN.

Price 12/11, post free.
HOW TO DESIGN A GAS ENGINE.

With full working drawings of a 7 B.H.P. gas engine. By HORACE ALLEN.

Price 2/9, post freo.

THE SCIENTIFIC PUBLISHING COMPANY,

53, NEW BAILEY STREET, MANCHESTER.

3d.

Edited by WILLIAM H. FOWLER, WEEKLY

Wh.Sc., M.Inst.C.E., &c.

A Practical Journal
for Practical Men.

Should be in the hands of all Engineers, Works Managers,
Foremen, Students, &c.

ANNUAL SUBSCRIPTIONS :

UNITED KINGDOM : 12/6 post free.

ANYWHERE ABROAD : Thick Paper Edition, 21/- post free.

Thin ,, ,,

Foreign Remittances by International Money Order only.

17/6

If owlet's "flDecbanical

To all direct Ammal Subscribers to the Journal a copy
of this Pocket Book will be sent post free.

Write for sample copy, post free.

The Scientific Publishing Co.,

53, New Bailey St., Manchester, England.

SPECIALLY DESIGNED for ENGINEERS.
"The Mechanical Engineer"

Non-Magnetic

Chronograph.

AS SUPPLIED TO H.M. GOVERNMENT.

in Solid Sllver

Post Free.
In Steel Oxydised Case.

• ISct.Gold

Case.

FRONT VIEW

ACTUAL SIZK.

Invaluable for Testing and Time- Recording Purposes.

A MARVEL FOR THE MONEY.

Will stand comparison with watches at double the price.

"The Mechanical Engineer" Keyless Lever

ph has been specially designed to meet the wants of Engineers
generally. For testing and time-recording purposes, and for Electrical Engineers
and others in charge of electrical machinery, it will be found invaluable. The
works, which are thoroughly non-magnetic, are of the latest design, full- jewelled,
and super-finished. It is a thoroughly reliable timekeeper.

THE SCIENTIFIC PUBLISHING COMPANY,

53, NEW BAILEY STREET, MANCHESTER.

The Metallurgy of Iron
and Steel.

This work has been prepared to meet a need for a book which in one
volume of moderate size shall cover the whole field of the Metallurgy of Iron
and Steel.

By A. HUMBOLDT SEXTON, F.I.C, F.C.S.,

Professor of Metallurgy in the Glasgow and West of Scotland
Technical College;

Past'President "West of Scotland Iron and Steel Institute.

Price I6s. 6d., post free.

Contents.

Sources of Iron — Pig Iron — Preparation of Materials for the Smelter —
Chemistry of the Blast Furnace — Thermal Phenomena of the Blast
Furnace — The Blast Furnace — The Air Supply — The Hot Blast —
Blast Furnace Slag — Calculating Charges — Construction of the Blast
Furnace — Blast Furnace Practice — Utilisation of By-products — History
of Pig Iron — The Foundry — Malleable Iron — Puddling — Other Methods of Pre-
paring Malleable Iron — The Forge and the Mill — Steel — Production of Steel
direct from the Ore and from Malleable Iron — Production of Steel by Partial
Decaiburisation of Pig Iron — The Bessemer Process — Chemistry of the Besse-
mer Process — Thermal Phenomena of the Bessemer Blow — Working the
Bessemer Process — Bessemer Plant — The Basic Bessemer Process — Plant for the
Basic Bessemer Process — Modifications of the Bessemer Process — Historical
Notes on the Bessemer Process — The Siemen's or Open Hearth Process — The
Siemen's Process: Plant — The Basic Open Hearth Process — Modification of the
Open Hearth Process — Appliances Applicable to all Processes — Working Mild
Steel — Casting Mild Steel — After Treatment of Iron and Steel — Special Steels
— Structure of Iron and Steel — Testing Iron and Steel — Rusting and Protection
of Iron and Steel — Additional Notes — Nomenclature of Metallography —
Bibliography.

Press Opinions.

" To undertake the compilation of a comprehensive treatise upon the metallurgy
of iron and steel, within the limits of a 'single portable volume, must impress anyone
acquainted with the extent and ramifications of that industry as a task requiring no
small measure of courage and literary skill. Undaunted, however, with the formidable
nature of the task, Prof. Sexton has tackled it with a degree of determination and
skill which deserves and commands our unstinted admiration. The volume may be
heartily commended to the student as a most excellent text book and a valuable guide
to the study of a great industry." — Iron and Coal Trades Review.

" Mr. Sexton's work is a very valuable contribution to the literature of iron and
steel, and will, we think, be welcomed by all who are scientifically and practically
interested in the many important developments arising out of both. The author
writes with a fullness of knowledge and earnestness of spirit, which will be appreciated
by fellow investigators."— Sheffield Telegraph.

" Prof. Sexton's book possesses features that make it a work of real value to the
student of metallurgical science."— Glasgow Herald.

"The book is admirably printed, and its general 'get up' is very pleasing; it will
meet a distinct need, and for its size it is certainly one of the best and most
complete." — Manchester Guardian.

THE SCIENTIFIC PUBLISHING CO.,

53, New Bailey St., Manchester, England.

Telegrams :

" Cuprum," Glasgow.
Cuprum," Manchester.

ESTABLISHED
1872.

Telephones :
Glasgow- Nat. 2679,

Bridgeton
., — P.O. Y 193.
Manchester- Nat. 4625.

PARK & PATERSON, Ltd.,

PARKHEAD, GLASGOW,

WITH WHICH IS INCORPORATED

R. W. MACKINLAY & SONS,

110, Gt. Wellington St., KINNING PARK.

Metal Merchants and Smelters.

Branch: 17, St. Simon Street, Salford, Manchester.

of IWGOT METALS
for all purposes : —

INGOT GUN-METALS of all grades.
„ BUSH METALS.
„ YELLOW METALS.
„ RED METALS.
„ WHITE METALS.

Specialists in Fine Gun-metal for Admiralty
Requirements, &c., &c.

GUN METAL SCRAP, COPPER SCRAP, YELLOW
METAL, BRONZE GUN-METAL BORINGS, YELLOW
METAL BORINGS, DUST, BRASSFOUNDERS AND
COPPERSMITHS' ASHES, LEAD, ASHES, TIN, ZINC,
PEWTER, ALUMINIUM, and every description of
Metallic Drosses Bought.

UNIVERSITY OF CALIFORNIA LIBRARY

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Alloys (non-ferrous)"

See other formats