Chapter XXII: , Note 35) respecting the combination of CuSO{4} (1)

with water and ammonia, we may add that Lachinoff (1893) showed
that CuSO_{4},5H_{2}O loses 4-3/4H_{2}O at 180°, that
CuSO_{4},5NH_{3} also loses 4-3/4NH_{3} at 320°, and that only
1/4H_{2}O and 1/4NH_{3} remain in combination with the CuSO_{4}.
The last 1/4H_{2}O can only be driven off by heating to 200°, and
the last 1/4NH_{3} by heating to 360°. Ammonia displaces water
from CuSO_{4},5H_{2}O, but water cannot displace the ammonia from
CuSO_{4},5NH_{3}. If hydrochloric acid gas be passed over
CuSO_{4},5H_{2}O at the ordinary temperature, it first forms
CuSO_{4},5H_{2}O,3HCl, and then CuSO_{4},2H_{2}O,2HCl. When air is
passed over the latter compound it passes into CuSO_{4}H_{2}O with
a small amount of HCl (about 1/8HCl). At 100° CuSO_{4},5H_{2}O in
a stream of hydrochloric acid gas gives CuSO_{4},1/4H_{2}O,2HCl,
and then CuSO_{4},1/4H_{2}O,HCl, whilst after prolonged heating
CuSO_{4} remains, which rapidly passes into CuSO_{4},5H_{2}O when
placed under a bell jar over water. Over sulphuric acid, however,
CuSO_{4},5H_{2}O only parts with 3H_{2}O, and if CuSO_{4},2H_{2}O
be placed over water it again forms CuSO_{4},5H_{2}O, and so on.

[11 bis] Commercial blue vitriol generally contains ferrous sulphate.
The salt is purified by converting the ferrous salt into a ferric
salt by heating the solution with chlorine or nitric acid. The
solution is then evaporated to dryness, and the unchanged cupric
sulphate extracted from the residue, which will contain the larger
portion of the ferric oxide. The remainder will be separated if
cupric hydroxide is added to the solution and boiled; the cupric
oxide, CuO, then precipitates the ferric oxide, Fe_{2}O_{3}, just
as it is itself precipitated by silver oxide. But the solution
will contain a small proportion of a basic salt of copper, and
therefore sulphuric acid must be added to the filtered solution,
and the salt allowed to crystallise. Acid salts are not formed,
and cupric sulphate itself has an acid reaction on litmus paper.

_The alloys of copper_ with certain metals, and especially with zinc and tin, are easily formed by directly melting the metals together. They are easily cast into moulds, forged, and worked like copper, whilst they are much more durable in the air, and are therefore frequently used in the arts. Even the ancients used exclusively alloys of copper, and not pure copper, but its alloys with tin or different kinds of bronze (Chapter XVIII., Note 35). The alloys of copper with zinc are called _brass_ or 'yellow metal.' Brass contains about 32 p.c. of zinc; generally, however, it does not contain more than 65 p.c. of copper. The remainder is composed of lead and tin, which usually occur, although in small quantities, in brass. Yellow metal contains about 40 p.c. of zinc.[12] The addition of zinc to copper changes the colour of the latter to a considerable degree; with a certain amount of zinc the colour of the copper becomes yellow, and with a still larger proportion of zinc an alloy is formed which has a greenish tint. In those alloys of zinc and copper which contain a larger amount of zinc than of copper, the yellow colour disappears and is replaced by a greyish colour. But when the amount of zinc is diminished to about 20 p.c., the alloy is red and hard, and is called 'tombac.' A contraction takes place in alloying copper with zinc, so that the volume of the alloy is less than that of either metal individually. The zinc volatilises on prolonged heating at a high temperature and the excess of metallic copper remains behind. When heated in the air, the zinc oxidises before the copper, so that all the zinc alloyed with copper may be removed from the copper by this means. An important property of brass containing about 30 p.c. of zinc is that it is soft and malleable in the cold, but becomes somewhat brittle when heated. We may also mention that ordinary copper coins contain, in order to render them hard, tin, zinc, and iron (Cu = 95 p.c.); that it is now customary to add a small amount of phosphorus to copper and bronze, for the same purpose; and also that copper is added to silver and gold in coining, &c. to render it hard; moreover, in Germany, Switzerland, and Belgium, and other countries, a silver-white alloy (melchior, German silver, &c.), for base coinage and other purposes, is prepared from brass and nickel (from 10 to 20 p.c. of nickel; 20 to 30 p.c. zinc: 50 to 70 p.c. copper), or directly from copper and nickel, or, more rarely, from an alloy containing silver, nickel, and copper.[12 bis]

[12] Among the alloys of copper resembling brass, _delta metal_,
invented by A. Dick (London) is largely used (since 1883). It
contains 55 p.c. Cu, and 41 p.c. Zn, the remaining 4 p.c. being
composed of iron (as much as 3-1/2 p.c., which is first alloyed
with zinc), or of cobalt, and manganese, and certain other metals.
The sp. gr. of delta metal is 8·4. It melts at 950°, and then
becomes so fluid that it fills up all the cavities in a mould and
forms excellent castings. It has a tensile strength of 70 kilos
per sq. mm. (gun metal about 20, phosphor bronze about 30). It is
very soft, especially when heated to 600°, but after forging and
rolling it becomes very hard; it is more difficultly acted upon by
air and water than other kinds of brass, and preserves its golden
yellow colour for any length of time, especially if well polished.
It is used for making bearings, screw propellers, valves, and many
other articles. In general the alloys of Cu and Zn containing
about 2/3 p.c. by weight of copper were for a long time almost
exclusively made in Sweden and England (Bristol, Birmingham).
These alloys for the most part are cheaper, harder, and more
fusible than copper alone, and form good castings. The alloys
containing 45-80 p.c. Cu crystallise in cubes if slowly cooled (Bi
also gives crystals). By washing the surface of brass with dilute
sulphuric acid, Zn is removed and the article acquires the colour
of copper. The alloys approaching Zn_{2}Cu_{3} in their
composition exhibit the greatest resistance (under other equal
conditions; of purity, forging, rolling, &c.) The addition of 3
p.c. Al, or 5 p.c. Sn, improves the quality of brass. Respecting
aluminium bronze _see_ Chapter XVII. p. 88.

[12 bis] Ball (also Kamensky), 1888, by investigating the electrical
conductivity of the alloys of antimony and copper with lead, came
to the conclusion that only two definite compounds of antimony and
copper exist, whilst the other alloys are either alloys of these
two together or with antimony or with copper. These compounds are
Cu_{2}Sb and Cu_{4}Sb--one corresponds with the maximum, and the
other with the minimum, electrical resistance. In general, the
resistance offered to an electrical current forms one of the
methods by which the composition of definite alloys (for example,
Pb_{2}Zn_{7}) is often established, whilst the electromotive force
of alloys affords (Laurie, 1888) a still more accurate method--for
instance, several definite compounds were discovered by this
method among the alloys of copper with zinc and tin; but we will
not enter into any details of this subject, because we avoid all
references to electricity, although the reader is recommended to
make himself acquainted with this branch of science, which has
many points in common with chemistry. The study of alloys regarded
as solid solutions should, in my opinion, throw much light upon
the question of solutions, which is still obscure in many aspects
and in many branches of chemistry.

Copper, in its cuprous compounds, is so analogous to _silver_, that were there no cupric compounds, or if silver gave stable compounds of the higher oxide, AgO, the resemblance would be as close as that between chlorine and bromine or zinc and cadmium; but silver compounds corresponding to AgO are quite unknown. Although silver peroxide--which was regarded as AgO, but which Berthelot (1880) recognised as the sesquioxide Ag_{2}O_{3}--is known, still it does not form any true salts, and consequently cannot be placed along with cupric oxide. In distinction to copper, silver as a metal does not oxidise under the influence of heat; and its oxides, Ag_{2}O and Ag_{2}O_{3}, easily lose oxygen (_see_ Note 8 tri). Silver does _not oxidise_ in air at the ordinary pressure, and is therefore classed among the so-called _noble metals_. It has a white colour, which is much purer than that of any other known metal, especially when the metal is chemically pure. In the arts silver is always used alloyed, because chemically-pure silver is so soft that it wears exceedingly easily, whilst when fused with a small amount of copper, it becomes very hard, without losing its colour.[13]

[13] There are not many soft metals; lead, tin, copper, silver, iron,
and gold are somewhat soft, and potassium and sodium very soft.
The metals of the alkaline earths are sonorous and hard, and many
other metals are even brittle, especially bismuth and antimony.
But the very slight significance which these properties have in
determining the fundamental chemical properties of substances
(although, however, of immense importance in the practical
applications of metals) is seen from the example shown by zinc,
which is hard at the ordinary temperature, soft at 100°, and
brittle at 200°.

As the value of silver depends exclusively on its purity, and as
there is no possibility of telling the amount of impurities
alloyed with it from its external appearance, it is customary in
most countries to mark an article with the amount of pure silver
it contains after an accurately-made analysis known as the assay
of the silver. In France the assay of silver shows the amount of
pure silver in 1,000 parts by weight; in Russia the amount of pure
silver in 96 parts--that is, the assay shows the number of
zolotniks (4·26 grams) of pure silver in one pound (410 grams) of
alloyed silver. Russian silver is generally 84 assay--that is,
contains 84 parts by weight of pure silver and 12 parts of copper
and other metals. French money contains 90 p.c. (in the Russian
system this will be 86·4 assay) by weight of silver [English coins
and jewellery contain 92·5 p.c. of silver]; the silver rouble is
of 83-1/3 assay--that is, it contains 86·8 p.c. of silver--and the
smaller Russian silver coinage is of 48 assay, and therefore
contains 50 p.c. of silver. Silver ornaments and articles are
usually made in Russia of 84 and 72 assay. As the alloys of silver
and copper, especially after being subjected to the action of
heat, are not so white as pure silver, they generally undergo a
process known as 'blanching' (or 'pickling') after being worked
up. This consists in removing the copper from the surface of the
article by subjecting it to a dark-red heat and then immersing it
in dilute acid. During the calcination the copper on the surface
is oxidised, whilst the silver remains unchanged; the dilute acid
then dissolves the copper oxides formed, and pure silver is left
on the surface. The surface is dull after this treatment, owing to
the removal of a portion of the metal by the acid. After being
polished the article acquires the desired lustre and colour, so as
to be indistinguishable from a pure silver object. In order to
test a silver article, a portion of its mass must be taken, not
from the surface, but to a certain depth. The methods of assay
used in practice are very varied. The commonest and most often
used is that known as _cupellation_. It is based on the difference
in the oxidisability of copper, lead, and silver. The cupel is a
porous cup with thick sides, made by compressing bone ash. The
porous mass of bone ash absorbs the fused oxides, especially the
lead oxide, which is easily fusible, but it does not absorb the
unoxidised metal. The latter collects into a globule under the
action of a strong heat in the cupel, and on cooling solidifies
into a button, which may then be weighed. Several cupels are
placed in a muffle. A muffle is a semi-cylindrical clay vessel,
shown in the accompanying drawing. The sides of the muffle are
pierced with several orifices, which allow the access of air into
it. The muffle is placed in a furnace, where it is strongly
heated. Under the action of the air entering the muffle the copper
of the silver alloy is oxidised, but as the oxide of copper is
infusible, or, more strictly speaking, difficultly fusible, a
certain quantity of lead is added to the alloy; the lead is also
oxidised by the air at the high temperature of the muffle, and
gives the very fusible lead oxide. The copper oxide then fuses
with the lead oxide, and is absorbed by the cupel, whilst the
silver remains as a bright white globule. If the weight of the
alloy taken and of the silver left on the cupel be determined, it
is possible to calculate the composition of the alloy. Thus the
essence of cupellation consists in the separation of the
oxidisable metals from silver, which does not oxidise under the
action of heat. A more accurate method, based on the precipitation
of silver from its solutions in the form of silver chloride, is
described in detail in works on analytical chemistry.

Silver occurs in _nature_, both in a native state and in certain compounds. Native silver, however, is of rather rare occurrence. A far greater quantity of silver occurs in combination with sulphur, and especially in the form of _silver sulphide_, Ag_{2}S, with lead sulphide or copper sulphide, or the ores of various other metals. The largest amount of silver is extracted from the lead in which it occurs. If this lead be calcined in the presence of air, it oxidises, and the resultant lead oxide, PbO ('litharge' or 'silberglätte,' as it is called), melts into a mobile liquid, which is easily removed. The silver remains in an unoxidised metallic state.[14] This process is called _cupellation_.

[14] In America, whence the largest amount of silver is now obtained,
ores are worked containing not more than 1/5 p.c. of silver,
whilst at 1/2 p.c. its extraction is very profitable. Moreover,
the extraction of silver from ores containing not more than 0·01
p.c. of this metal is sometimes profitable. The majority of the
lead smelted from galena contains silver, which is extracted from
it. Thus near Arras, in France, an ore is worked which contains
about 65 parts of lead and 0·088 part of silver in 100 parts of
ore, which corresponds with 136 parts of silver in 100,000 parts
of lead. At Freiberg, in Saxony, the ore used (enriched by
mechanical dressing) contains about 0·9 of silver, 160 of lead,
and 2 of copper in 10,000 parts. In every case the lead is first
extracted in the manner described in Chapter XVIII., and this lead
will contain all the silver. Not unfrequently other ores of silver
are mixed with lead ores, in order to obtain an argentiferous lead
as the product. The extraction of small quantities of silver from
lead is facilitated by the fact (Pattinson's process) that molten
argentiferous lead in cooling first deposits crystals of pure
lead, which fall to the bottom of the cooling vessel, whilst the
proportion of silver in the unsolidified mass increases owing to
the removal of the crystals of lead. The lead is enriched in this
manner until it contains 1/400 part of silver, and is then
subjected to cupellation on a larger scale. According to Park's
process, zinc is added to the molten argentiferous lead, and the
alloy of Pb and Zn, which first separates out on cooling, is
collected. This alloy is found to contain all the silver
previously contained in the lead. The addition of 0·5 p.c. of
aluminium to the zinc (Rossler and Edelman) facilitates the
extraction of the Ag from the resultant alloy besides preventing
oxidation; for, after re-melting, nearly all the lead easily runs
off (remains fluid), and leaves an alloy containing about 30 p.c.
Ag and about 70 p.c. Zn. This alloy may be used as an anode in a
solution of ZnCl_{2}, when the Zn is deposited on the cathode,
leaving the silver with a small amount of Pb, &c. behind. The
silver can be easily obtained pure by treating it with dilute
acids and cupelling.

The ores of silver which contain a larger amount of it are: silver
glance, Ag_{2}S (sp. gr. 7·2); argentiferous-copper glance, CuAgS;
horn silver or chloride of silver, AgCl; argentiferous grey copper
ore; polybasite, M_{9}RS_{6} (where M = Ag, Cu, and R = Sb, As),
and argentiferous gold. The latter is the usual form in which gold
is found in alluvial deposits and ores. The crystals of gold from
the Berezoffsky mines in the Urals contain 90 to 95 of gold and 5
to 9 of silver, and the Altai gold contains 50 to 65 of gold and
36 to 38 of silver. The proportion of silver in native gold varies
between these limits in other localities. Silver ores, which
generally occur in veins, usually contain native silver and
various sulphur compounds. The most famous mines in Europe are in
Saxony (Freiberg), which has a yearly output of as much as 26 tons
of silver, Hungary, and Bohemia (41 tons). In Russia, silver is
extracted in the Altai and at Nerchinsk (17 tons). The richest
silver mines known are in America, especially in Chili (as much as
70 tons), Mexico (200 tons), and more particularly in the Western
States of North America. The richness of these mines may be judged
from the fact that one mine in the State of Nevada (Comstock, near
Washoe and the cities of Gold Hill and Virginia), which was
discovered in 1859, gave an output of 400 tons in 1866. In place
of cupellation, chlorination may also be employed for extracting
silver from its ores. The method of chlorination consists in
converting the silver in an ore into silver chloride. This is
either done by a wet or by a dry method, roasting the ore with
NaCl. When the silver chloride is formed, the extraction of the
metal is also done by two methods. The first consists in the
silver chloride being reduced to metal by means of iron in
rotating barrels, with the subsequent addition of mercury which
dissolves the silver, but does not act on the other metals. The
mercury holding the silver in solution is distilled, when the
silver remains behind. This method is called _amalgamation_. The
other method is less frequently used, and consists in dissolving
the silver chloride in sodium chloride or in sodium thiosulphate,
and then precipitating the silver from the solution. The
amalgamation is then carried on in rotating barrels containing the
roasted ore mixed with water, iron, and mercury. The iron reduces
the silver chloride by taking up the chlorine from it. The
technical details of these processes are described in works on
metallurgy. The extraction of AgCl by the wet method is carried on
(Patera's process) by means of a solution of hyposulphite of
sodium which dissolves AgCl (_see_ Note 23), or by lixiviating
with a 2 p.c. solution of a double hyposulphite of Na and Cu
(obtained by adding CuSO_{4} to Na_{2}S_{2}O_{3}). The resultant
solution of AgCl is first treated with soda to precipitate
PbCO_{3}, and then with Na_{2}S, which precipitates the Ag and Au.
The process should be carried on rapidly to prevent the
precipitation of Cu_{2}S from the solution of CuSO_{4} and
Na_{2}S_{2}O_{3}.

Commercial silver generally contains copper, and, more rarely, other metallic impurities also. Chemically _pure silver_ is obtained either by cupellation or by subjecting ordinary silver to the following treatment. The silver is first dissolved in nitric acid, which converts it and the copper into nitrates, Cu(NO_{3})_{2} and AgNO_{3}; hydrochloric acid is then added to the resultant solution (green, owing to the presence of the cupric salt), which is considerably diluted with water in order to retain the lead chloride in solution if the silver contained lead. The copper and many other metals remain in solution, whilst the silver is precipitated as silver chloride. The precipitate is allowed to settle, and the liquid is decanted off; the precipitate is then washed and fused with sodium carbonate. A double decomposition then takes place, sodium chloride and silver carbonate being formed; but the latter decomposes into metallic silver, because the silver oxide is decomposed by heat: Ag_{2}CO_{3} = Ag_{2} + O + CO_{2}. The silver chloride may also be mixed with metallic zinc, sulphuric acid, and water, and left for some time, when the zinc removes the chlorine from the silver chloride and precipitates the silver as a powder. This finely-divided silver is called 'molecular silver.'[15]

[15] There is another practical method which is also suitable for
separating the silver from the solutions obtained in photography,
and consists in precipitating the silver by oxalic acid. In this
case the amount of silver in the solution must be known, and 23
grams of oxalic acid dissolved in 400 grams of water must be added
for every 60 grams of silver in solution in a litre of water. A
precipitate of silver oxalate, Ag_{2}C_{2}O_{4}, is then obtained,
which is insoluble in water but soluble in acids. Hence, if the
liquid contain any free acid it must be previously freed from it
by the addition of sodium carbonate. The resultant precipitate of
silver oxalate is dried, mixed with an equal weight of dry sodium
carbonate, and thrown into a gently-heated crucible. The
separation of the silver then proceeds without an explosion,
whilst the silver oxalate if heated alone decomposes with
explosion.

According to Stas, the best method for obtaining silver from its
solutions is by the reduction of silver chloride dissolved in
ammonia by means of an ammoniacal solution of cuprous
thiosulphate; the silver is then precipitated in a crystalline
form. A solution of ammonium sulphite may be used instead of the
cuprous salt.

Chemically-pure silver has an exceeding pure white colour, and a specific gravity of 10·5. Solid silver is lighter than the molten metal, and therefore a piece of silver floats on the latter. The fusing-point of silver is about 950° C., and at the high temperature attained by the combustion of detonating gas it volatilises.[16] By employing silver reduced from silver chloride by milk sugar and caustic potash, and distilling it, Stas obtained silver purer than that obtained by any other means; in fact, this was perfectly pure silver. The vapour of silver has a very beautiful green colour, which is seen when a silver wire is placed in an oxyhydrogen flame.[17]

[16] Silver is very malleable and ductile; it may be beaten into leaves
0·002 mm. in thickness. Silver wire may be made so fine that 1
gram is drawn into a wire 2-1/2 kilometres long. In this respect
silver is second only to gold. A wire of 2 mm. diameter breaks
under a strain of 20 kilograms.

[17] In melting, silver absorbs a considerable amount of oxygen, which
is disengaged on solidifying. One volume of molten silver absorbs
as much as 22 volumes of oxygen. In solidifying, the silver forms
cavities like the craters of a volcano, and throws off metal,
owing to the evolution of the gas; all these phenomena recall a
volcano on a miniature scale (Dumas). Silver which contains a
small quantity of copper or gold, &c., does not show this property
of dissolving oxygen.

The absorption of oxygen by molten silver is, however, an
oxidation, but it is at the same time a phenomenon of solution.
One cubic centimetre of molten silver can dissolve twenty-two
cubic centimetres of oxygen, which, even at 0°, only weighs 0·03
gram, whilst 1 cubic centimetre of silver weighs at least 10
grams, and therefore it is impossible to suppose that the
absorption of the oxygen is attended by the formation of any
definite compound (rich in oxygen) of silver and oxygen (about 45
atoms of silver to 1 of oxygen) in any other but a dissociated
form, and this is the state in which substances in solution must
be regarded (Chapter I.)

Le Chatelier showed that at 300° and 15 atmospheres pressure
silver absorbs so much oxygen that it may be regarded as having
formed the compound Ag_{4}O, or a mixture of Ag_{2} and Ag_{2}O.
Moreover, silver oxide, Ag_{2}O, only decomposes at 300° under low
pressures, whilst at pressures above 10 atmospheres there is no
decomposition at 300° but only at 400°.

Stas showed that silver is oxidised by air in the presence of
acids. V. d. Pfordten confirmed this, and showed that an acidified
solution of potassium permanganate rapidly dissolves silver in the
presence of air.

It has long been known (Wöhler) that when nitrate of silver, AgNO_{3}, reacts as an oxidising agent upon citrates and tartrates, it is able under certain conditions to give either a salt of suboxide of silver (see Note 19) or a red solution, or to give a precipitate of metallic silver reduced at the expense of the organic substances. In 1889 Carey Lea, in his researches on this class of reactions, showed that _soluble silver_ is here formed, which he called _allotropic silver_. It may be obtained by taking 200 c.c. of a 10 per cent. solution of AgNO_{3} and quickly adding a mixture (neutralised with NaHO) of 200 c.c. of a 30 per cent. solution of FeSO_{4} and 200 c.c. of a 40 per cent. solution of sodium citrate. A lilac precipitate is obtained, which is collected on a filter (the precipitate becomes blue) and washed with a solution of NH_{4}NO_{3}. It then becomes soluble in pure water, forming a red perfectly transparent solution from which the dissolved silver is precipitated on the addition of many soluble foreign bodies. Some of the latter--for instance, NH_{4}NO_{3}, alkaline sulphates, nitrates, and citrates--give a precipitate which redissolves in pure water, whilst others--for instance, MgSO_{4}, FeSO_{4}, K_{2}Cr_{2}O_{7}, AgNO_{3}, Ba(NO_{3})_{2} and many others--convert the precipitated silver into a new variety, which, although no longer soluble in water, regains its solubility in a solution of borax and is soluble in ammonia. Both the soluble and insoluble silver are rapidly converted into the ordinary grey-metallic variety by sulphuric acid, although nothing is given off in the reaction; the same change takes place on ignition, but in this case CO_{2} is disengaged; the latter is formed from the organic substances which remain (to the amount of 3 per cent.) in the modified silver (they are not removed by soaking in alcohol or water). If the precipitated silver be slightly washed and laid in a smooth thin layer on paper or glass, it is seen that the soluble variety is red when moist and a fine blue colour when dry, whilst the insoluble variety has a blue reflex. Besides these, under special conditions[18] a golden yellow variety may be obtained, which gives a brilliant golden yellow coating on glass; but it is easily converted into the ordinary grey-metallic state by friction or trituration. There is no doubt[18 bis] that there is the same relation between ordinary silver which is perfectly insoluble in water and the varieties of silver obtained by Carey Lea[18 tri] as there is between quartz and soluble silica or between CuS and As_{2}S_{2} in their ordinary insoluble forms and in the state of the colloid solution of their hydrosols (_see_ Chapter I., Note 57, and Chapter XVII., Note 25 bis). Here, however, an important step in advance has been made in this respect, that we are dealing with the solution of a simple body, and moreover of a metal--_i.e._ of a particularly characteristic state of matter. And as boron, gold, and certain other simple bodies have already been obtained in a soluble (colloid) form, and as numerous organic compounds (albuminous substances, gum, cellulose, starch, &c.) and inorganic substances are also known in this form, it might be said that the colloid state (of hydrogels and hydrosols) can be acquired, if not by every substance, at all events by substances of most varied chemical character under particular conditions of formation from solutions. And this being the case, we may hope that a further study of soluble colloid compounds, which apparently present various transitions towards emulsions, may throw a new light upon the complex question of solutions, which forms one of the problems of the present epoch of chemical science. Moreover, we may remark that Spring (1890) clearly proved the colloid state of soluble silver by means of dialysis as it did not pass through the membrane.

[18] When solutions of AgNO_{3}, FeSO_{4}, sodium citrate, and NaHO are
mixed together in the manner described above, they throw down a
precipitate of a beautiful lilac colour; when transferred to a
filter paper the precipitate soon changes colour, and becomes dark
blue. To obtain the substance as pure as possible it is washed
with a 5-10 p.c. solution of ammonium nitrate; the liquid is
decanted, and 150 c.c. of water poured over the precipitate. It
then dissolves entirely in the water. A small quantity of a
saturated solution of ammonium nitrate is added to the solution,
and the silver in solution again separates out as a precipitate.
These alternate solutions and precipitations are repeated seven or
eight times, after which the precipitate is transferred to a
filter and washed with 95 p.c. alcohol until the filtrate gives no
residue on evaporation. An analysis of the substance so obtained
showed that it contained from 97·18 p.c. to 97·31 p.c. of metallic
silver. It remained to discover what the remaining 2-3 p.c. were
composed of. Are they merely impurities, or is the substance some
compound of silver with oxygen or hydrogen, or does it contain
citric acid in combination which might account for its solubility?
The first supposition is set aside by the fact that no gases are
disengaged by the precipitate of silver, either under the action
of gases or when heated. The second supposition is shown to be
impossible by the fact that there is no definite relation between
the silver and citric acid. A determination of the amount of
silver in solution showed that the amount of citric acid varies
greatly for one and the same amount of silver, and there is no
simple ratio between them. Among other methods of preparing
soluble silver given by Carey Lea, we may mention the method
published by him in 1891. AgNO_{3} is added to a solution of
dextrine in caustic soda or potash; at first a precipitate of
brown oxide of silver is thrown down, but the brown colour then
changes into a reddish chocolate, owing to the reduction of the
silver by the dextrine, and the solution turns a deep red. A few
drops of this solution turn water bright red, and give a perfectly
transparent liquid. The dextrine solution is prepared by
dissolving 40 grams of caustic soda and the same amount of
ordinary brown dextrine in two litres of water. To this solution
is gradually added 28 grams of AgNO_{3} dissolved in a small
quantity of water.

The insoluble allotropic silver is obtained, as was mentioned
above, from a solution of silver prepared in the manner described,
by the addition of sulphate of copper, iron, barium, magnesium,
&c. In one experiment Lea succeeded in obtaining the insoluble
allotropic Ag in a crystalline form. The red solution, described
above, after standing several weeks, deposits crystals
spontaneously in the form of short black needles and thin prisms,
the liquid becoming colourless. This insoluble variety, when
rubbed upon paper, has the appearance of bright shining green
flakes, which polarise light.

The gold variety is obtained in a different manner to the two
other varieties. A solution is prepared containing 200 c.c. of a
10 p.c. solution of nitrate of silver, 200 c.c. of a 20 p.c.
solution of Rochelle salt, and 800 c.c. of water. Just as in the
previous case the reaction consisted in the reduction of the
citrate of silver, so in this case it consists in the reduction of
the tartrate, which here first forms a red, and then a black
precipitate of allotropic Ag, which, when transferred to the
filter, appears of a beautiful bronze colour. After washing and
drying, this precipitate acquires the lustre and colour peculiar
to polished gold, and this is especially remarked where the
precipitate comes into contact with glass or china. An analysis of
the golden variety gave a percentage composition of 98·750 to
98·749 Ag. Both the insoluble varieties (the blue and gold) have a
different specific gravity from ordinary silver. Whilst that of
fused silver is 10·50, and of finely-divided silver 10·62, the
specific gravity of the blue insoluble variety is 9·58, and of the
gold variety 8·51. The gold variety passes into ordinary Ag with
great ease. This transition may even be remarked on the filter in
those places which have accidentally not been moistened with
water. A simple shock, and therefore friction of one particle upon
another, is enough to convert the gold variety into normal white
silver. Carey Lea sent samples of the gold variety for a long
distance by rail packed in three tubes, in which the silver
occupied about the quarter of their volume; in one tube only he
filled up this space with cotton-wool. It was afterwards found
that the shaking of the particles of Ag had completely converted
it into ordinary white silver, and that only the tube containing
the cotton-wool had preserved the golden variety intact.

The soluble variety of Ag also passes into the ordinary state with
great ease, the heat of conversion being, as Prange showed in
1890, about +60 calories.

[18 bis] The opinion of the nature of soluble silver given below was
first enunciated in the _Journal of the Russian Chemical Society_,
February 1, 1890, Vol. XXII., Note 73. This view is, at the
present time, generally accepted, and this silver is frequently
known as the 'colloid' variety. I may add that Carey Lea observed
the solution of ordinary molecular silver in ammonia without the
access of air.

[18 tri] It is, however, noteworthy that ordinary metallic lead has
long been considered soluble in water, that boron has been
repeatedly obtained in a brown solution, and that observations
upon the development of certain bacteria have shown that the
latter die in water which has been for some time in contact with
metals. This seems to indicate the passage of small quantities of
metals into water (however, the formation of peroxide of hydrogen
may be supposed to have some influence in these cases).

As regards the capacity of silver for chemical reactions, it is remarkable for its small capacity for combination with oxygen and for its considerable energy of combination with sulphur, iodine, and certain kindred non-metals. _Silver does not oxidise_ at any temperature, and its oxide, Ag_{2}O, is decomposed by heat. It is also a very important fact that silver is not oxidised by oxygen either in the presence of alkalis, even at exceedingly high temperatures, or in the presence of acids--at least, of dilute acids--which properties render it a very important metal in chemical industry for the fusion of alkalis, and also for many purposes in everyday life; for instance, for making spoons, salt-cellars, &c. Ozone, however, oxidises it. Of all acids nitric acid has the greatest action on silver. The reaction is accompanied by the formation of oxides of nitrogen and silver nitrate, AgNO_{3}, which dissolves in water and does not, therefore, hinder the further action of the acid on the metal. The halogen acids, especially hydriodic acid, act on silver, hydrogen being evolved; but this action soon stops, owing to the halogen compounds of silver being insoluble in water and only very slightly soluble in acids; they therefore preserve the remaining mass of metal from the further action of the acid; in consequence of this the action of the halogen acids is only distinctly seen with finely-divided silver. Sulphuric acid acts on silver in the same manner that it does on copper, only it must be concentrated and at a higher temperature. Sulphurous anhydride, and not hydrogen, is then evolved, but there is no action at the ordinary temperature, even in the presence of air. Among the various salts, sodium chloride (in the presence of moisture, air, and carbonic acid) and potassium cyanide (in the presence of air) act on silver more decidedly than any others, converting it respectively into silver chloride and a double cyanide.

Although silver does not directly combine with oxygen, still three different grades of combination with oxygen may be obtained indirectly from the salts of silver. They are all, however, unstable, and decompose into oxygen and metallic silver when ignited. These three oxides of silver have the following composition: _silver suboxide_, Ag_{4}O,[19] corresponding with the (little investigated) suboxides of the alkali metals; _silver oxide_, Ag_{2}O, corresponding with the oxides of the alkali metals and the ordinary salts of silver, AgX; and _silver peroxide_, AgO,[19 bis] or, judging from Berthelot's researches, Ag_{2}O_{3}. _Silver oxide_ is obtained as a brown precipitate (which when dried does not contain water) by adding potassium hydroxide to a solution of a silver salt--for example, of silver nitrate. The precipitate formed seems, however, to be an hydroxide, AgHO, _i.e._ AgNO_{3} + KHO = KNO_{3} + AgHO, and the formation of the anhydrous oxide, 2AgHO = Ag_{2}O + H_{2}O, may be compared with the formation of the anhydrous cupric oxide by the action of potassium hydroxide on hot cupric solutions. Silver hydroxide decomposes into water and silver oxide, even at low temperatures; at least, the hydroxide no longer exists at 60°, but forms the anhydrous oxide, Ag_{2}O.[19 tri] Silver oxide is almost insoluble in water; but, nevertheless, it is undoubtedly a rather powerful basic oxide, because it displaces the oxides of many metals from their soluble salts, and saturates such acids as nitric acid, forming with them neutral salts, which do not act on litmus paper.[20] Undoubtedly water dissolves a small quantity of silver oxide, which explains the possibility of its action on solutions of salts--for example, on solutions of cupric salts. Water in which silver oxide is shaken up has a distinctly alkaline reaction. The oxide is distinguished by its great instability when heated, so that it loses all its oxygen when slightly heated. Hydrogen reduces it at about 80°.[20 bis] The feebleness of the affinity of silver for oxygen is shown by the fact that silver oxide decomposes under the action of light, so that it must be kept in opaque vessels. The silver _salts_ are colourless and decompose when heated, leaving metallic silver if the elements of the acid are volatile.[20 tri] They have a peculiar metallic taste, and are exceedingly poisonous; the majority of them are acted on by light, especially in the presence of organic substances, which are then oxidised. The alkaline carbonates give a white precipitate of silver carbonate, Ag_{2}CO_{3}, which is insoluble in water, but soluble in ammonia and ammonium carbonate. Aqueous ammonia, added to a solution of a normal silver salt, first acts like potassium hydroxide, but the precipitate dissolves in an excess of the reagent, like the precipitate of cupric hydroxide.[21] Silver oxalate and the halogen compounds of silver are insoluble in water; hydrochloric acid and soluble chlorides give, as already repeatedly observed, a white precipitate of silver chloride in solutions of silver salts. Potassium iodide gives a yellowish precipitate of silver iodide. Zinc separates all the silver in a metallic form from solutions of silver salts. Many other metals and reducing agents--for example, organic substances--also reduce silver from the solutions of its salts.

[19] Silver suboxide (Ag_{4}O) or argentous oxide is obtained from
argentic citrate by heating it to 100° in a stream of hydrogen.
Water and argentous citrate are then formed, and the latter,
although but slightly soluble in water, gives a reddish-brown
solution of colloid silver (Note 18), and when boiled this
solution becomes colourless and deposits metallic silver, the
argentic salt being again formed. Wöhler, who discovered this
oxide, obtained it as a black precipitate by adding potassium
hydroxide to the above solution of argentous citrate. With
hydrochloric acid the suboxide gives a brown compound, Ag_{2}Cl.
Since the discovery of soluble silver the above data cannot be
regarded as perfectly trustworthy; it is probable that a mixture
of Ag_{2} and Ag_{2}O was being dealt with, so that the actual
existence of Ag_{4}O is now doubtful, but there can be no doubt as
to the formation of a subchloride, Ag_{2}Cl (_see_ Note 25),
corresponding to the suboxide. The same compound is obtained by
the action of light on the higher chloride. Other acids do not
combine with silver suboxide, but convert it into an argentic salt
and metallic silver. In this respect cuprous oxide presents a
certain resemblance to these suboxides. But copper forms a
suboxide of the composition Cu_{4}O, which is obtained by the
action of an alkaline solution of stannous oxide on cupric
hydroxide, and is decomposed by acids into cupric salts and
metallic copper. The problems offered by the suboxides, as well as
by the peroxides, cannot be considered as fully solved.

[19 bis] _Silver peroxide_, AgO or Ag_{2}O_{3}, is obtained by the
decomposition of a dilute (10 p.c.) solution of silver nitrate by
the action of a galvanic current (Ritter). On the positive pole,
where oxygen is usually evolved in the decomposition of salts,
brittle grey needles with a metallic lustre, which occasionally
attain a somewhat considerable size, are then formed. They are
insoluble in water, and decompose with the evolution of oxygen
when they are dried and heated, especially up to 150°, and, like
lead dioxide, barium peroxide, &c., their action is strongly
oxidising. When treated with acids, oxygen is evolved and a salt
of the oxide formed. Silver peroxide absorbs sulphurous anhydride
and forms silver sulphate. Hydrochloric acid evolves chlorine;
ammonia reduces the silver, and is itself oxidised, forming water
and gaseous nitrogen. Analyses of the above-mentioned crystals
show that they contain silver nitrate, peroxide, and water.
According to Fisher, they have the composition
4AgO,AgNO_{3},H_{2}O, and, according to Berthelot,
4Ag_{2}O_{5},2AgNO_{3},H_{2}O.

[19 tri] According to Carey Lea, however, oxide of silver still retains
water even at 100°, and only parts with it together with the
oxygen. Oxide of silver is used for colouring glass yellow.

[20] The reaction of Pb(OH)_{2} upon AgHO in the presence of NaHO leads
to the formation of a compound of both oxides, PbO_n_Ag_{2}O, from
which the oxide of lead cannot be removed by alkalies (Wöhler,
Leton). Wöhler, Welch, and others obtained crystalline double
salts, R_{2}AgX_{3}, by the action of strong solutions of RX of
the halogen salts of the alkaline metals upon AgX, where R = Cs,
Rb, K.

[20 bis] According to Müller, ferric oxide is reduced by hydrogen
(_see_ Chapter XXII., Note 5) at 295° (into what ?), cupric oxide
at 140°, Ni_{2}O_{3} at 150°; nickelous oxide, NiO, is reduced to
the suboxide, Ni_{2}O, at 195°, and to nickel at 270°; zinc oxide
requires so high a temperature for its reduction that the glass
tube in which Müller conducted the experiment did not stand the
heat; antimony oxide requires a temperature of 215° for its
reduction; yellow mercuric oxide is reduced at 130° and the red
oxide at 230°; silver oxide at 85°, and platinum oxide even at the
ordinary temperature.

[20 tri] A silica compound, Ag_{2}OSiO_{2} is obtained by fusing
AgNO_{3} with silica; this salt is able to decompose with the
evolution of oxygen, leaving Ag + SiO_{2}.

[21] If a solution of a silver salt be precipitated by sodium
hydroxide, and aqueous ammonia is added drop by drop until the
precipitate is completely dissolved, the liquid when evaporated
deposits a violet mass of crystalline silver oxide. If moist
silver oxide be left in a strong solution of ammonia it gives a
black mass, which easily decomposes with a loud explosion,
especially when struck. This black substance is called fulminating
silver. Probably this is a compound like the other compounds of
oxides with ammonia, and in exploding the oxygen of the silver
oxide forms water with the hydrogen of the ammonia, which is
naturally accompanied by the evolution of heat and formation of
gaseous nitrogen, or, as Raschig states, fulminating silver
contains NAg_{3} or one of the amides (for instance, NHAg_{2} =
NH_{3} + Ag_{2}O - H_{2}O). Fulminating silver is also formed when
potassium hydroxide is added to a solution of silver nitrate in
ammonia. The dangerous explosions which are produced by this
compound render it needful that great care be taken when salts of
silver come into contact with ammonia and alkalis (_see_ Chapter
XVI., Note 26).

_Silver nitrate_, AgNO_{3}, is known by the name of lunar caustic (or _lapis infernalis_); it is obtained by dissolving metallic silver in nitric acid. If the silver be impure, the resultant solution will contain a mixture of the nitrates of copper and silver. If this mixture be evaporated to dryness and the residue carefully fused at an incipient red heat, all the cupric nitrate is decomposed, whilst the greater part of the silver nitrate remains unchanged. On treating the fused mass with water the latter is dissolved, whilst the cupric oxide remains insoluble. If a certain amount of silver oxide be added to the solution containing the nitrates of silver and copper, it displaces all the cupric oxide. In this case it is of course not necessary to take pure silver oxide, but only to pour off some of the solution and to add potassium hydroxide to one portion, and to mix the resultant precipitate of the hydroxides, Cu(OH)_{2} and AgOH, with the remaining portion.[22] By these methods all the copper can be easily removed and pure silver nitrate obtained (its solution is colourless, while the presence of Cu renders it blue), which may be ultimately purified by crystallisation. It crystallises in colourless transparent prismatic plates, which are not acted on by air. They are anhydrous. Its sp. gr. is 4·34; it dissolves in half its weight of water at the ordinary temperature.[22 bis] The pure salt is not acted on by light, but it easily acts in an oxidising manner on the majority of organic substances, which it generally blackens. This is due to the fact that the organic substance is oxidised by the silver nitrate, which is reduced to metallic silver; the latter is thus obtained in a finely-divided state, which causes the black stain. This peculiarity is taken advantage of for marking linen. Silver nitrate is for the same reason used for _cauterising wounds_ and various growths on the body. Here again it acts by virtue of its oxidising capacity in destroying the organic matter, which it oxidises, as is seen from the separation of a coating of black metallic powdery silver from the part cauterised.[22 tri] From the description of the preparation of silver nitrate it will have been seen that this salt, which fuses at 218°, does not decompose at an incipient red heat; when cast into sticks it is usually employed for cauterising. On further heating, the fused salt undergoes decomposition, first forming silver nitrite and then metallic silver. With ammonia, silver nitrate forms, on evaporation of the solution, colourless crystals containing AgNO_{3},2HN_{3} (Marignac). In general the salts of silver, like cuprous, cupric, zinc, &c. salts, are able to give several compounds with ammonia; for example, silver nitrate in a dry state absorbs three molecules (Rose). The ammonia is generally easily expelled from these compounds by the action of heat.

[22] So that we here encounter the following phenomena: copper
displaces silver from the solutions of its salts, and silver oxide
displaces copper oxide from cupric salts. Guided by the
conceptions enunciated in Chapter XV., we can account for this in
the following manner: The atomic volume of silver = 10·3, and of
copper = 7·2, of silver oxide = 32, and of copper oxide = 13. A
greater contraction has taken place in the formation of cupric
oxide, CuO, than in the formation of silver oxide, Ag_{2}O, since
in the former (13 - 7 = 6) the volume after combination with the
oxygen has increased by very little, whilst the volume of silver
oxide is considerably greater than that of the metal it contains
[32 - (2 × 10·3) = 11·4]. Hence silver oxide is less compact than
cupric oxide, and is therefore less stable; but, on the other
hand, there are greater intervals between the atoms in silver
oxide than in cupric oxide, and therefore silver oxide is able to
give more stable compounds than those of copper oxide. This is
verified by the figures and data of their reactions. It is
impossible to calculate for cupric nitrate, because this salt has
not yet been obtained in an anhydrous state; but the sulphates of
both oxides are known. The specific gravity of copper sulphate in
an anhydrous state is 3·53, and of silver sulphate 5·36; the
molecular volume of the former is 45, and of the latter 58. The
group SO_{3} in the copper occupies, as it were, a volume 45 - 13
= 32, and in the silver salt a volume 58 - 32 = 26; hence a
smaller contraction has taken place in the formation of the copper
salt from the oxide than in the formation of the silver salt, and
consequently the latter should be more stable than the former.
Hence silver oxide is able to decompose the salt of copper oxide,
whilst with respect to the metals both salts have been formed with
an almost identical contraction, since 58 volumes of the silver
salt contain 21 volumes of metal (difference = 37), and 45 volumes
of the copper salt contain 7 volumes of copper (difference = 38).
Besides which, it must be observed that copper oxide displaces
iron oxide, just as silver oxide displaces copper oxide. Silver,
copper, and iron, in the form of oxides, displace each other in
the above order, but in the form of metals in a reverse order
(iron, copper, silver). The cause of this order of the
displacement of the oxides lies, amongst other things, in their
composition. They have the composition Ag_{2}O, Cu_{2}O_{2},
Fe_{2}O_{3}; the oxide containing a less proportion of oxygen
displaces that containing a larger proportion, because the basic
character diminishes with the increase of contained oxygen.

Copper also displaces mercury from its salts. It may here be
remarked that Spring (1888), on leaving a mixture of dry mercurous
chloride and copper for two hours, observed a distinct reduction,
which belongs to the category of those phenomena which demonstrate
the existence of a mobility of parts (_i.e._ atoms and molecules)
in solid substances.

[22 bis] The reaction of 1 part by weight of AgNO_{3} requires
(according to Kremers) the following amounts of water: at 0°, 0·82
part, at 19°·5, 0·41 part, at 54°, 0·20 part, at 110°, 0·09 part,
and, according to Tilden, at 125°, 0·0617 part, and at 133°,
0·0515 part.

[22 tri] It may be remarked that the black stain produced by the
reduction of metallic silver disappears under the action of a
solution of mercuric chloride or of potassium cyanide, because
these salts act on finely-divided silver.

Nitrate of silver easily forms double salts like AgNO_{3}2NaNO_{3} and AgNO_{3}KNO_{3}. Silver nitrate under the action of water and a halogen gives nitric acid (_see_ Vol. I. p. 280, formation of N_{2}O_{5}), a halogen salt of silver, and a silver salt of an oxygen acid of the halogen. Thus, for example, a solution of chlorine in water, when mixed with a solution of silver nitrate, gives silver chloride and chlorate. It is here evident that the reaction of the silver nitrate is identical with the reaction of the caustic alkalis, as the nitric acid is all set free and the silver oxide only reacts in exactly the same way in which aqueous potash acts on free chlorine. Hence the reaction may be expressed in the following manner: 6AgNO_{3} + 3Cl_{2} + 3H_{2}O = 5AgCl + AgClO_{3} + 6NHO_{3}.

Silver nitrate, like the nitrates of the alkalis, does not contain any water of crystallisation. Moreover the other salts of silver almost always separate out without any water of crystallisation. The silver salts are further characterised by the fact that they _give neither basic nor acid salts_, owing to which the formation of silver salts generally forms the means of determining the true composition of acids--thus, to any acid H_{n}X there corresponds a salt Ag_{n}X--for instance, Ag_{3}PO_{4} (Chapter XIX., Note 15).

_Silver_ gives insoluble and exceedingly stable _compounds with the halogens_. They are obtained by double decomposition with great facility whenever a silver salt comes in contact with halogen salts. Solutions of nitrate, sulphate, and all other kindred salts of silver give a precipitate of silver chloride or iodide in solutions of chlorides and iodides and of the halogen acids, because the halogen salts of silver are insoluble both in water[23] and in other acids. _Silver chloride_, AgCl, is then obtained as a white flocculent precipitate, silver bromide forms a yellowish precipitate, and silver iodide has a very distinct yellow colour. These halogen compounds sometimes occur in nature; they are formed by a dry method--by the action of halogen compounds on silver compounds, especially under the influence of heat. Silver chloride easily fuses at 451° on cooling from a molten state; it forms a somewhat soft horn-like mass which can be cut with a knife and is known as _horn silver_. It volatilises at a higher temperature. Its ammoniacal solution, on the evaporation of the ammonia, deposits crystalline chloride of silver, in octahedra. Bromide and iodide of silver also appear in forms of the regular system, so that in this respect the halogen salts of silver resemble the halogen salts of the alkali metals.[24]