1911 Encyclopædia Britannica/Gold

GOLD [symbol Au, atomic weight 195.7(H = 1), 197.2(O = 16)], a metallic chemical element, valued from the earliest ages on account of the permanency of its colour and lustre. Gold ornaments of great variety and elaborate workmanship have been discovered on sites belonging to the earliest known civilizations, Minoan, Egyptian, Assyrian, Etruscan (see Jewelry, Plate, Egypt, Crete, Aegean Civilization, Numismatics), and in ancient literature gold is the universal symbol of the highest purity and value (cf. passages in the Old Testament, e.g. Ps. xix. 10 “More to be desired are they than gold, yea, than much fine gold”). With regard to the history of the metallurgy of gold, it may be mentioned that, according to Pliny, mercury was employed in his time both as a means of separating the precious metals and for the purposes of gilding. Vitruvius also gives a detailed account of the means of recovering gold, by amalgamation, from cloth into which it had been woven.

Physical Properties.—Gold has a characteristic yellow colour, which is, however, notably affected by small quantities of other metals; thus the tint is sensibly lowered by small quantities of silver, and heightened by copper. When the gold is finely divided, as in “purple of Cassius,” or when it is precipitated from solutions, the colour is ruby-red, while in very thin leaves it transmits a greenish light. It is nearly as soft as lead and softer than silver. When pure, it is the most malleable of all metals (see Goldbeating). It is also extremely ductile; a single grain may be drawn into a wire 500 ft. in length, and an ounce of gold covering a silver wire is capable of being extended more than 1300 m. The presence of minute quantities of cadmium, lead, bismuth, antimony, arsenic, tin, tellurium and zinc renders gold brittle, 1/2000th part of one of the three metals first named being sufficient to produce that quality. Gold can be readily welded cold; the finely divided metal, in the state in which it is precipitated from solution, may be compressed between dies into disks or medals. The specific gravity of gold obtained by precipitation from solution by ferrous sulphate is from 19·55 to 20·72. The specific gravity of cast gold varies from 18·29 to 19·37, and by compression between dies the specific gravity may be raised from 19·37 to 19·41; by annealing, however, the previous density is to some extent recovered, as it is then found to be 19·40. The melting-point has been variously given, the early values ranging from 1425° C. to 1035° C. Using improved methods, C. T. Heycock and F. H. Neville determined it to be 1061·7° C.; Daniel Berthelot gives 1064° C., while Jaquerod and Perrot give 1066·1–1067·4° C. At still higher temperatures it volatilizes, forming a reddish vapour. Macquer and Lavoisier showed that when gold is strongly heated, fumes arise which gild a piece of silver held in them. Its volatility has also been studied by L. Eisner, and, in the presence of other metals, by Napier and others. The volatility is barely appreciable at 1075°; at 1250° it is four times as much as at 1100°. Copper and zinc increase the volatility far more than lead, while the greatest volatility is induced, according to T. Kirke Rose, by tellurium. It has also been shown that gold volatilizes when a gold-amalgam is distilled. Gold is dissipated by sending a powerful charge of electricity through it when in the form of leaf or thin wire. The electric conductivity is given by A. Matthiessen as 73 at 0° C., pure silver being 100; the value of this coefficient depends greatly on the purity of the metal, the presence of a few thousandths of silver lowering it by 10%. Its conductivity for heat has been variously given as 103 (C. M. Despretz), 98 (F. Crace-Calvert and R. Johnson), and 60 (G. H. Wiedemann and R. Franz), pure silver being 100. Its specific heat is between 0·0298 (Dulong and Petit) and 0·03244 (Regnault). Its coefficient of expansion for each degree between 0° and 100° C. is 0·000014661, or for gold which has been annealed 0·000015136 (Laplace and Lavoisier). The spark spectrum of gold has been mapped by A. Kirchhoff, R. Thalén, Sir William Huggins and H. Krüss; the brightest lines are 6277, 5960, 5955 and 5836 in the orange and yellow, and 5230 and 4792 in the green and blue.

Chemical Properties.—Gold is permanent in both dry and moist air at ordinary or high temperatures. It is insoluble in hydrochloric, nitric and sulphuric acids, but dissolves in aqua regia—a mixture of hydrochloric and nitric acids—and when very finely divided in a heated mixture of strong sulphuric acid and a little nitric acid; dilution with water, however, precipitates the metal as a violet or brown powder from this solution. The metal is soluble in solutions of chlorine, bromine, thiosulphates and cyanides; and also in solutions which generate chlorine, such as mixtures of hydrochloric acid with nitric acid, chromic acid, antimonious acid, peroxides and nitrates, and of nitric acid with a chloride. Gold is also attacked when strong sulphuric acid is submitted to electrolysis with a gold positive pole. W. Skey showed that in substances which contain small quantities of gold the precious metal may be removed by the solvent action of iodine or bromine in water. Filter paper soaked with the clear, solution is burnt, and the presence of gold is indicated by the purple colour of the ash. In solution minute quantities of gold may be detected by the formation of “purple of Cassius,” a bluish-purple precipitate thrown down by a mixture of ferric and stannous chlorides.

The atomic weight of gold was first determined with accuracy by Berzelius, who deduced the value 195·7 (H＝1) from the amount of mercury necessary to precipitate it from the chloride, and 195·2 from the ratio between gold and potassium chloride in potassium aurichloride, KAuCl₄. Later determinations were made by Sir T. E. Thorpe and A. P. Laurie, Krüss and J. W. Mallet. Thorpe and Laurie converted potassium auribromide into a mixture of metallic gold and potassium bromide by careful heating. The relation of the gold to the potassium bromide, as well as the amounts of silver and silver bromide which are equivalent to the potassium bromide, were determined. The mean value thus adduced was 195·86. Krüss worked with the same salt, and obtained the value 195·65; while Mallet, by analyses of gold chloride and bromide, and potassium auribromide, obtained the value 195·77.

Occlusion of Gas by Gold.—T. Graham showed that gold is capable of occluding by volume 0·48% of hydrogen, 0·20% of nitrogen, 0·29% of carbon monoxide, and 0·16% of carbon dioxide. Varrentrapp pointed out that “cornets” from the assay of gold may retain gas if they are not strongly heated.

Occurrence and Distribution.—Gold is found in nature chiefly in the metallic state, i.e. as “native gold,” and less frequently in combination with tellurium, lead and silver. These are the only certain examples of natural combinations of the metal, the minute, though economically valuable, quantity often found in pyrites and other sulphides being probably only present in mechanical suspension. The native metal crystallizes in the cubic system, the octahedron being the commonest form, but other and complex combinations have been observed. Owing to the softness of the metal, large crystals are rarely well defined, the points being commonly rounded. In the irregular crystalline aggregates branching and moss-like forms are most common, and in Transylvania thin plates or sheets with diagonal structures are found. More characteristic, however, than the crystallized are the irregular forms, which, when large, are known as “nuggets” or “pepites,” and when in pieces below 1/4 to 1/2 oz. weight as gold dust, the larger sizes being distinguished as coarse or nuggety gold, and the smaller as gold dust proper. Except in the larger nuggets, which may be more or less angular, or at times even masses of crystals, with or without associated quartz or other rock, gold is generally found bean-shaped or in some other flattened form, the smallest particles being scales of scarcely appreciable thickness, which, from their small bulk as compared with their surface, subside very slowly when suspended in water, and are therefore readily carried away by a rapid current. These form the “float gold” of the miner. The physical properties of native gold are generally similar to that of the melted metal.

Of the minerals containing gold the most important are sylvanite or graphic tellurium (Ag, Au) Te₂, with 24 to 26%; calaverite, AuTe₂, with 42%; nagyagite or foliate tellurium (Pb, Au)₁₆ Sb₃(S, Te)₂₄, with 5 to 9% of gold; petzite, (Ag, Au)₂Te, and white tellurium. These are confined to a few localities, the oldest and best known being those of Nagyag and Offenbanya in Transylvania; they have also been found at Red Cloud, Colorado, in Calaveras county, California, and at Perth and Boulder, West Australia. The minerals of the second class, usually spoken of as “auriferous,” are comparatively numerous. Prominent among these are galena and iron pyrites, the former being almost invariably gold-bearing. Iron pyrites, however, is of greater practical importance, being in some districts exceedingly rich, and, next to the native metal, is the most prolific source of gold. Magnetic pyrites, copper pyrites, zinc blende and arsenical pyrites are other and less important examples, the last constituting the gold ore formerly worked in Silesia. A native gold amalgam is found as a rarity in California, and bismuth from South America is sometimes rich in gold. Native arsenic and antimony are also very frequently found to contain gold and silver.

The association and distribution of gold may be considered under two different heads, namely, as it occurs in mineral veins—“reef gold,” and in alluvial or other superficial deposits which are derived from the waste of the former—“alluvial gold.” Four distinct types of reef gold deposits may be distinguished: (1) Gold may occur disseminated through metalliferous veins, generally with sulphides and more particularly with pyrites. These deposits seem to be the primary sources of native gold. (2) More common are the auriferous quartz-reefs—veins or masses of quartz containing gold in flakes visible to the naked eye, or so finely divided as to be invisible. (3) The “banket” formation, which characterizes the goldfields of South Africa, consists of a quartzite conglomerate throughout which gold is very finely disseminated. (4) The siliceous sinter at Mount Morgan, Queensland, which is obviously associated with hydrothermal action, is also gold-bearing. The genesis of the last three types of deposit is generally assigned to the simultaneous percolation of solutions of gold and silica, the auriferous solution being formed during the disintegration of the gold-bearing metalliferous veins. But there is much uncertainty as to the mechanism of the process; some authors hold that the soluble chloride is first formed, while others postulate the intervention of a soluble aurate.

In the alluvial deposits the associated minerals are chiefly those of great density and hardness, such as platinum, osmiridium and other metals of the platinum group, tinstone, chromic, magnetic and brown iron ores, diamond, ruby and sapphire, zircon, topaz, garnet, &c. which represent the more durable original constituents of the rocks whose distintegration has furnished the detritus.

Statistics of Gold Production.—The supply of gold, and also its relation to the supply of silver, has, among civilized nations, always been of paramount importance in the economic questions concerning money (see Money and Bimetallism); in this article a summary of the modern gold-producing areas will be given, and for further details reference should be made to the articles on the localities named. The chief sources of the European supply during the middle ages were the mines of Saxony and Austria, while Spain also contributed. The supplies from Mexico and Brazil were important during the 16th and 17th centuries. Russia became prominent in 1823, and for fourteen years contributed the bulk of the supply. The United States (California) after 1848, and Australia after 1851, were responsible for enormous increases in the total production, which has been subsequently enhanced by discoveries in Canada, South Africa, India, China and other countries.

Table I.

Period.	Oz.	Period.	Oz.
1801–1810	590,750	1856–1860	6,350,180
1811–1820	380,300	1861–1865	5,951,770
1821–1830	472,400	1866–1870	6,169,660
1831–1840	674,200	1871–1875	5,487,400
1841–1850	1,819,600	1876–1880	5,729,300
1851–1855	6,350,180	—	—

The average annual world’s production for certain periods from 1801 to 1880 in ounces is given in Table I. The average production of the five years 1881–1885 was the smallest since the Australian and Californian mines began to be worked in 1848–1849; the minimum 4,614,588 oz., occurred in 1882. It was not until after 1885 that the annual output of the world began to expand. Of the total production in 1876, 5,016,488 oz., almost the whole was derived from the United States, Australasia and Russia. Since then the proportion furnished by these countries has been greatly lowered by the supplies from South Africa, Canada, India and China. The increase of production has not been uniform, the greater part having occurred most notably since 1895. Among the regions not previously important as gold-producers which now contribute to the annual output, the most remarkable are the goldfields of South Africa (Transvaal and Rhodesia, the former of which were discovered in 1885). India likewise has been added to the list, its active production having begun at about the same time as that of South Africa. The average annual product of India for the period 1886 to 1899 inclusive was £698,208, and its present annual product averages about 550,000 oz., or about £2,200,000, obtained almost wholly from the free-milling quartz veins of the Colar goldfields in Mysore, southern India. In 1900 the output was valued at £1,891,804, in 1905 at £2,450,536, and in 1908 at £2,270,000. Canada, too, assumed an important rank, having contributed in 1900 £5,583,300; but the output has since steadily declined to £1,973,000 in 1908. The great increase during the few years preceding 1899 was due to the development of the goldfields of the North-Western Territory, especially British Columbia. From the district of Yukon (Klondike, &c.) £2,800,000 was obtained in 1899, wholly from alluvial workings, but the progress made since has been slower than was expected by sanguine people. It is, however, probable that the North-Western Territory will continue to yield gold in important quantities for some time to come.

The output of the United States increased from £7,050,000 in 1881 to £16,085,567 in 1900, £17,916,000 in 1905, and to £20,065,000 in 1908. This increase was chiefly due to the exploitation of new goldfields. The fall in the price of silver stimulated the discovery and development of gold deposits, and many states formerly regarded as characteristically silver districts have become important as gold producers. Colorado is a case in point, its output having increased from about £600,000 in 1880 to £6,065,000 in 1900; it was £5,139,800 in 1905. Somewhat more than one-half of the Colorado gold is obtained from the Cripple Creek district. Other states also showed a largely augmented product. On the other hand, the output of California, which was producing over £3,000,000 per annum in 1876, has fallen off, the average annual output from 1876 to 1900 being £2,800,000; in 1905 the yield was £3,839,000. This decrease was largely caused by the practical suspension for many years of the hydraulic mining operations, in preparation for which millions of dollars had been expended in deep tunnels, flumes, &c., and the active continuance of which might have been expected to yield some £2,000,000 of gold annually. This interruption, due to the practical prohibition of the industry by the United States courts, on the ground that it was injuring, through the deposit of tailings, agricultural lands and navigable streams, was lessened, though not entirely removed, by compromises and regulations which permit, under certain restrictions, the renewed exploitation of the ancient river-beds by the hydraulic method. On the other hand, the progressive reduction of mining and metallurgical costs effected by improved transportation and machinery, and the use of high explosives, compressed air, electric-power transmission, &c., resulted in California (as elsewhere) in a notable revival of deep mining. This was especially the case on the “Mother Lode,” where highly promising results were obtained. Not only is vein-material formerly regarded as unremunerative now extracted at a profit, but in many instances increased gold-values have been encountered below zones of relative barrenness, and operators have been encouraged to make costly preparations for really deep mining—more than 3000 ft. below the surface. The gold product of California, therefore, may be fairly expected to maintain itself, and, indeed, to show an advance. Alaska appeared in the list of gold-producing countries in 1886, and gradually increased its annual output until 1897, when the country attracted much attention with a production valued at over £500,000; the opening up of new workings has increased this figure immensely, from about £1,400,000 in 1901 to £3,006,500 in 1905. The Alaska gold was derived almost wholly from the large low-grade quartz mines of Douglas Island prior to 1899, but in that year an important district was discovered at Cape Nome, on the north-western coast. The result of a few months’ working during that year was more than £500,000 of gold, and a very much larger annual output may reasonably be anticipated in the future; in 1905 it was about £900,000. The gold occurs in alluvial deposits designated as gulch-, bar-, beach-, tundra- and bench-placers. The tundra is a coastal plain, swampy and covered with undergrowth and underlaid by gravel. The most interesting and, thus far, the most productive are the beach deposits, similar to those on the coast of Northern California. These occur in a strip of comparatively fine gravel and sand, 150 yds. wide, extending along the shore. The gold is found in stratified layers, with “ruby” and black sand. The “ruby” sand consists chiefly of fine garnets and magnetites, with a few rose-quartz grains. Further exploration of the interior will probably result in the discovery of additional gold districts.

Mexico, from a gold production of £200,000 in 1891, advanced to about £1,881,800 in 1900 and to about £3,221,000 in 1905. Of this increase, a considerable part was derived from gold-quartz mining, though much was also obtained as a by-product in the working of the ores of other metals. The product of Colombia, Venezuela, the Guianas, Brazil, Uruguay, Argentina, Chile, Bolivia, Peru and Ecuador amounted in 1900 to £2,481,000 and to £2,046,000 in 1905.

In 1876 Australasia produced £7,364,000, of which Victoria contributed £3,084,000. The annual output of Victoria declined until the year 1892, when it began to increase rapidly, but not to its former level, the values for 1900 and 1905 being £3,142,000 and £3,138,000. There has been an important increase in Queensland, which advanced from £1,696,000 in 1876 to £2,843,000 in 1900, and subsequently declined to £2,489,000 in 1905. There has been no increase, and, indeed, no large fluctuation until quite recently in the output of New Zealand, which averaged £1,054,000 per annum from 1876 to 1898, but the production of the two years 1900 and 1905 rose to £1,425,459 and £2,070,407 respectively. By far the most important addition to the Australasian product has come from West Australia, which began its production in 1887—about the time of the inception of mining at Witwatersrand (“the Rand”) in South Africa—and by continuous increase, which assumed large proportions towards the close of the 19th century, was £6,426,000 in 1899, £6,179,000 in 1900, and £8,212,000 in 1905. The total Australasian production in 1908 was valued at £14,708,000.

Undoubtedly the greatest of the gold discoveries made in the latter half of the 19th century was that of the Witwatersrand district in the Transvaal. By reason of its unusual geological character and great economic importance this district deserves a more extended description. The gold occurs in conglomerate beds, locally known as “banket.” There are several series of parallel beds, interstratified with quartzite and schist, the most important being the “main reef” series. The gold in this conglomerate reef is partly of detrital origin and partly of the genetic character of ordinary vein-gold. The formation is noted for its regularity as regards both the thickness and the gold-tenor of the ore-bearing reefs, in which respect it is unparalleled in the geology of the auriferous formations. The gold carries, on an average, £2 per ton, and is worked by ordinary methods of gold-mining, stamp-milling and cyaniding. In 1899, 5762 stamps were in operation, crushing 7,331,446 tons of ore, and yielding £15,134,000, equivalent to 25.5% of the world’s production. Of this, 80% came from within 12 m. of Johannesburg. After September 1899 operations were suspended, almost entirely owing to the Boer War, but on the 2nd of May 1901 they were started again. In 1905 the yield was valued at £20,802,074, and in 1909 at £30,925,788. So certain is the ore-bearing formation that engineers in estimating its auriferous contents feel justified in assuming, as a factor in their calculations, a vertical extension limited only by the lowest depths at which mining is feasible. On such a basis they arrived at more than £600,000,000 as the available gold contained in the Witwatersrand conglomerates. This was a conservative estimate, and was made before the full extent of the reefs was known; in 1904 Lionel Phillips stated that the main reef series had been proved for 61 m., and he estimated the gold remaining to be mined to be worth £2,500,000,000. Deposits similar to the Witwatersrand banket occur in Zululand, and also on the Gold Coast of Africa. In Rhodesia, the country lying north of the Transvaal, where gold occurs in well-defined quartz-veins, there is unquestionable evidence of extensive ancient workings. The economic importance of the region generally has been fully proved. Rhodesia produced £386,148 in 1900 and £722,656 in 1901, in spite of the South African War; the product for 1905 was valued at £1,480,449, and for 1908 at £2,526,000.

The gold production of Russia has been remarkably constant, averaging £4,899,262 per annum; the gold is derived chiefly from placer workings in Siberia.

The gold production of China was estimated for 1899 at £1,328,238 and for 1900 at £860,000; it increased in 1901 to about £1,700,000, to fall to £340,000 in 1905; in 1906 and 1907 it recovered to about £1,000,000.

Table II.—Gold Production of Certain Countries, 1881–1908 (in oz.).

Year.	Australasia.	Africa.	Canada.	India.	Mexico.	Russia.	United States.	Totals.
1881	1,475,161	..	52,483	..	41,545	1,181,853	1,678,612	4,976,980
1882	1,438,067	..	52,000	..	45,289	1,154,613	1,572,187	4,825,794
1883	1,333,849	..	46,150	..	46,229	1,132,219	1,451,250	4,614,588
1884	1,352,761	..	46,000	..	57,227	1,055,642	1,489,950	4,902,889
1885	1,309,804	..	53,987	..	46,941	1,225,738	1,538,325	5,002,584
1886	1,257,670	..	66,061	..	29,702	922,226	1,693,125	5,044,363
1887	1,290,202	28,754	59,884	15,403	39,861	971,656	1,596,375	5,061,490
1888	1,344,002	240,266	53,150	35,034	47,117	1,030,151	1,604,841	5,175,623
1889	1,540,607	366,023	62,658	78,649	33,862	1,154,076	1,587,000	5,611,245
1890	1,453,172	497,817	55,625	107,273	37,104	1,134,590	1,588,880	5,726,966
1891	1,518,690	729,268	45,022	131,776	48,375	1,168,764	1,604,840	6,287,591
1892	1,638,238	1,210,869	43,905	164,141	54,625	1,199,809	1,597,098	7,102,172
1893	1,711,892	1,478,477	44,853	207,152	63,144	1,345,224	1,739,323	7,772,585
1894	2,020,180	2,024,164	50,411	210,412	217,688	1,167,455	1,910,813	8,813,848
1895	2,170,505	2,277,640	92,440	257,830	290,250	1,397,767	2,254,760	9,814,505
1896	2,185,872	2,280,892	136,274	323,501	314,437	1,041,794	2,568,132	9,950,861
1897	2,547,704	2,832,776	294,582	350,585	362,812	1,124,511	2,774,935	11,420,068
1898	3,137,644	3,876,216	669,445	376,431	411,187	1,231,791	3,118,398	13,877,806
1899	3,837,181	3,532,488	1,031,563	418,869	411,187	1,072,333	3,437,210	14,837,775
1900	3,555,506	419,503	1,348,720	456,444	435,375	974,537	3,829,897	12,315,135
1901	3,719,080	439,704	1,167,216	454,527	497,527	1,105,412	3,805,500	12,698,089
1902	3,946,374	1,887,773	1,003,355	463,824	491,156	1,090,053	3,870,000	14,313,660
1903	4,315,538	3,289,409	911,118	552,873	516,524	1,191,582	3,560,000	15,852,620
1904	4,245,744	4,156,084	793,350	556,097	609,781	1,199,857	3,892,480	16,790,351
1905	4,159,220	5,477,841	700,863	576,889	779,181	1,063,883	4,265,742	18,360,945
1906	3,984,538	6,449,749	581,709	525,527	896,615	1,087,056	4,565,333	19,620,272
1907	3,659,693	7,270,464	399,844	495,965	903,672	1,282,635	4,374,827	19,988,144
1908	3,557,705	7,983,348	462,467	504,309	1,182,445	1,497,076	4,659,360	21,529,300

Alloys.—Gold forms alloys with most metals, and of these many are of great importance in the arts. The alloy with mercury—gold amalgam—is so readily formed that mercury is one of the most powerful agents for extracting the precious metal. With 10% of gold present the amalgam is fluid, and with 12.5% pasty, while with 13% it consists of yellowish-white crystals. Gold readily alloys with silver and copper to form substances in use from remote times for money, jewelry and plate. Other metals which find application in the metallurgy of gold by virtue of their property of extracting the gold as an alloy are lead, which combines very readily when molten, and which can afterwards be separated by cupellation, and copper, which is separated from the gold by solution in acids or by electrolysis; molten lead also extracts gold from the copper-gold alloys. The relative amount of gold in an alloy is expressed in two ways: (1) as “fineness,” i.e. the amount of gold in 1000 parts of alloy; (2) as “carats,” i.e. the amount of gold in 24 parts of alloy. Thus, pure gold is 1000 “fine” or 24 carat. In England the following standards are used for plate and jewelry: 375, 500, 625, 750 and 916.6, corresponding to 9, 12, 15, 18 and 22 carats, the alloying metals being silver and copper in varying proportions. In France three alloys of the following standards are used for jewelry, 920, 840 and 750. A greenish alloy used by goldsmiths contains 70% of silver and 30% of gold. “Blue gold” is stated to contain 75% of gold and 25% of iron. The Japanese use for ornament an alloy of gold and silver, the standard of which varies from 350 to 500, the colour of the precious metal being developed by “pickling” in a mixture of plum-juice, vinegar and copper sulphate. They may be said to possess a series of bronzes, in which gold and silver replace tin and zinc, all these alloys being characterized by patina having a wonderful range of tint. The common alloy, Shi-ya-ku-Do, contains 70% of copper and 30% of gold; when exposed to air it becomes coated with a fine black patina, and is much used in Japan for sword ornaments. Gold wire may be drawn of any quality, but it is usual to add 5 to 9 dwts. of copper to the pound. The “solders” used for red gold contain 1 part of copper and 5 of gold; for light gold, 1 part of copper, 1 of silver and 4 of gold.

Gold and Silver.—Electrum is a natural alloy of gold and silver. Matthiessen observed that the density of alloys, the composition of which varies from AuAg₆ to Au₆Ag, is greater than that calculated from the densities of the constituent metals. These alloys are harder, more fusible and more sonorous than pure gold. The alloys of the formulae AuAg, AuAg₂, AuAg₄ and AuAg₂₀ are perfectly homogeneous, and have been studied by Levol. Molten alloys containing more than 80% of silver deposit on cooling the alloy AuAg₉, little gold remaining in the mother liquor.

Gold and Zinc.—When present in small quantities zinc renders gold brittle, but it may be added to gold in larger quantities without destroying the ductility of the precious metal; Péligot proved that a triple alloy of gold, copper and zinc, which contains 5.8% of the last-named, is perfectly ductile. The alloy of 11 parts gold and 1 part of zinc is, however, stated to be brittle.

Gold and Tin.—Alchorne showed that gold alloyed with 1/37th part of tin is sufficiently ductile to be rolled and stamped into coin, provided the metal is not annealed at a high temperature. The alloys of tin and gold are hard and brittle, and the combination of the metals is attended with contraction; thus the alloy SnAu has a density 14.243, instead of 14.828 indicated by calculation. Matthiessen and Bose obtained large crystals of the alloy Au₂Sn₅, having the colour of tin, which changed to a bronze tint by oxidation.

Gold and Iron.—Hatchett found that the alloy of 11 parts gold and 1 part of iron is easily rolled without annealing. In these proportions the density of the alloy is less than the mean of its constituent metals.

Gold and Palladium.—These metals are stated to alloy in all proportions. According to Chenevix, the alloy composed of equal parts of the two metals is grey, is less ductile than its constituent metals and has the specific gravity 11.08. The alloy of 4 parts of gold and 1 part of palladium is white, hard and ductile. Graham showed that a wire of palladium alloyed with from 24 to 25 parts of gold does not exhibit the remarkable retraction which, in pure palladium, attends its loss of occluded hydrogen.

Gold and Platinum.—Clarke states that the alloy of equal parts of the two metals is ductile, and has almost the colour of gold.

Gold and Rhodium.—Gold alloyed with 1/4th or 1/5th of rhodium is, according to Wollaston, very ductile, infusible and of the colour of gold.

Gold and Iridium.—Small quantities of iridium do not destroy the ductility of gold, but this is probably because the metal is only disseminated through the mass, and not alloyed, as it falls to the bottom of the crucible in which the gold is fused.

Gold and Nickel.—Eleven parts of gold and 1 of nickel yield an alloy resembling brass.

Gold and Cobalt.—Eleven parts of gold and 1 of cobalt form a brittle alloy of a dull yellow colour.

Compounds.—Aurous oxide, Au₂O, is obtained by cautiously adding potash to a solution of aurous bromide, or by boiling mixed solutions of auric chloride and mercurous nitrate. It forms a dark-violet precipitate which dries to a greyish-violet powder. When freshly prepared it dissolves in cold water to form an indigo-coloured solution with a brownish fluorescence of colloidal aurous oxide; it is insoluble in hot water. This oxide is slightly basic. Auric oxide, Au₂O₃, is a brown powder, decomposed into its elements when heated to about 250° or on exposure to light. When a concentrated solution of auric chloride is treated with caustic potash, a brown precipitate of auric hydrate, Au(OH)₃, is obtained, which, on heating, loses water to form auryl hydrate, AuO(OH), and auric oxide, Au₂O₃. It functions chiefly as an acidic oxide, being less basic than aluminium oxide, and forming no stable oxy-salts. It dissolves in alkalis to form well-defined crystalline salts; potassium aurate, KAuO₂·3H₂O, is very soluble in water, and is used in electro-gilding. With concentrated ammonia auric oxide forms a black, highly explosive compound of the composition AuN₂H₃·3H₂O, named “fulminating gold”; this substance is generally considered to be Au(NH₂)NH·3H₂O, but it may be an ammine of the formula [Au(NH₃)₂(OH)₂]OH. Other oxides, e.g. Au₂O₂, have been described.

Aurous chloride, AuCl, is obtained as a lemon-yellow, amorphous powder, insoluble in water, by heating auric chloride to 185°. It begins to decompose into gold and chlorine at 185°, the decomposition being complete at 230°; water decomposes it into gold and auric chloride. Auric chloride, or gold trichloride, AuCl₃, is a dark ruby-red or reddish-brown, crystalline, deliquescent powder obtained by dissolving the metal in aqua regia. It is also obtained by carefully evaporating a solution of the metal in chlorine water. The gold chloride of commerce, which is used in photography, is really a hydrochloride, chlorauric or aurichloric acid, HAuCl₄·3H₂O, and is obtained in long yellow needles by crystallizing the acid solution. Corresponding to this acid, a series of salts, named chloraurates or aurichlorides, are known. The potassium salt is obtained by crystallizing equivalent quantities of potassium and auric chlorides. Light-yellow monoclinic needles of 2KAuCl₄·H₂O are deposited from warm, strongly acid solutions, and transparent rhombic tables of KAuCl₄·2H₂O from neutral solutions. By crystallizing an aqueous solution, red crystals of AuCl₃·2H₂O are obtained. Auric chloride combines with the hydrochlorides of many organic bases—amines, alkaloids, &c.—to form characteristic compounds. Gold dichloride, probably Au₂Cl₄, = Au·AuCl₄, aurous chloraurate, is said to be obtained as a dark-red mass by heating finely divided gold to 140°-170° in chlorine. Water decomposes it into gold and auric chloride. The bromides and iodides resemble the chlorides. Aurous bromide, AuBr, is a yellowish-green powder obtained by heating the tribromide to 140°; auric bromide, AuBr₃, forms reddish-black or scarlet-red leafy crystals, which dissolve in water to form a reddish-brown solution, and combines with bromides to form bromaurates corresponding to the chloraurates. Aurous iodide, AuI, is a light-yellow, sparingly soluble powder obtained, together with free iodine, by adding potassium iodide to auric chloride; auric iodide, AuI₃, is formed as a dark-green powder at the same time, but it readily decomposes to aurous iodide and iodine. Aurous iodide is also obtained as a green solid by acting upon gold with iodine. The iodaurates correspond to the chlor- and bromaurates; the potassium salt, KAuI₄, forms highly lustrous, intensely black, four-sided prisms.

Aurous cyanide, AuCN, forms yellow, microscopic, hexagonal tables, insoluble in water, and is obtained by the addition of hydrochloric acid to a solution of potassium aurocyanide, KAu(CN)₂. This salt is prepared by precipitating a solution of gold in aqua regia by ammonia, and then introducing the well-washed precipitate into a boiling solution of potassium cyanide. The solution is filtered and allowed to cool, when colourless rhombic pyramids of the aurocyanide separate. It is also obtained in the action of potassium cyanide on gold in the presence of air, a reaction utilized in the MacArthur-Forrest process of gold extraction (see below). Auric cyanide, Au(CN)₃, is not certainly known; its double salts, however, have been frequently described. Potassium auricyanide, 2KAu(CN)₄·3H₂O, is obtained as large, colourless, efflorescent tablets by crystallizing concentrated solutions of auric chloride and potassium cyanide. The acid, auricyanic acid, 2HAu(CN)₄·3H₂O, is obtained by treating the silver salt (obtained by precipitating the potassium salt with silver nitrate) with hydrochloric acid; it forms tabular crystals, readily soluble in water, alcohol and ether.

Gold forms three sulphides corresponding to the oxides; they readily decompose on heating. Aurous sulphide, Au₂S, is a brownish-black powder formed by passing sulphuretted hydrogen into a solution of potassium aurocyanide and then acidifying. Sodium aurosulphide, NaAuS·4H₂O, is prepared by fusing gold with sodium sulphide and sulphur, the melt being extracted with water, filtered in an atmosphere of nitrogen, and evaporated in a vacuum over sulphuric acid. It forms colourless, monoclinic prisms, which turn brown on exposure to air. This method of bringing gold into solution is mentioned by Stahl in his Observationes Chymico-Physico-Medicae; he there remarks that Moses probably destroyed the golden calf by burning it with sulphur and alkali (Ex. xxxii. 20). Auric sulphide, Au₂S₃, is an amorphous powder formed when lithium aurichloride is treated with dry sulphuretted hydrogen at −10°. It is very unstable, decomposing into gold and sulphur at 200°.

Oxy-salts of gold are almost unknown, but the sulphite and thiosulphate form double salts. Thus by adding acid sodium sulphite to, or by passing sulphur dioxide at 50° into, a solution of sodium aurate, the salt, 3Na₂SO₃·Au₂SO₃·3H₂O is obtained, which, when precipitated from its aqueous solution by alcohol, forms a purple powder, appearing yellow or green by reflected light. Sodium aurothiosulphate, 3Na₂S₂O₃·Au₂S₂O₃·4H₂O, forms colourless needles; it is obtained in the direct action of sodium thiosulphate on gold in the presence of an oxidizing agent, or by the addition of a dilute solution of auric chloride to a sodium thiosulphate solution.

Mining and Metallurgy.

The various deposits of gold may be divided into two classes—“veins” and “placers.” The vein mining of gold does not greatly differ from that of similar deposits of metals (see Mineral Deposits). In the placer or alluvial deposits, the precious metal is found usually in a water-worn condition imbedded in earthy matter, and the method of working all such deposits is based on the disintegration of the earthy matter by the action of a stream of water, which washes away the lighter portions and leaves the denser gold. In alluvial deposits the richest ground is usually found in contact with the “bed rock”; and, when the overlying cover of gravel is very thick, or, as sometimes happens, when the older gravel is covered with a flow of basalt, regular mining by shafts and levels, as in what are known as tunnel-claims, may be required to reach the auriferous ground.

The extraction of gold may be effected by several methods; we may distinguish the following leading types:

1. By simple washing, i.e. dressing auriferous sands, gravels, &c.;

2. By amalgamation, i.e. forming a gold amalgam, afterwards removing the mercury by distillation;

3. By chlorination, i.e. forming the soluble gold chloride and then precipitating the metal;

4. By the cyanide process, i.e. dissolving the gold in potassium cyanide solution, and then precipitating the metal;

5. Electrolytically, generally applied to the solutions obtained in processes (3) and (4).

1. Extraction of Gold by Washing.—In the early days of gold-washing in California and Australia, when rich alluvial deposits were common at the surface, the most simple appliances sufficed. The most characteristic is the “pan,” a circular dish of sheet-iron or “tin,” with sloping sides about 13 or 14 in. in diameter. The pan, about two-thirds filled with the “pay dirt” to be washed, is held in the stream or in a hole filled with water. The larger stones having been removed by hand, gyratory motion is given to the pan by a combination of shaking and twisting movements so as to keep its contents suspended in the stream of water, which carries away the bulk of the lighter material, leaving the heavy minerals, together with any gold which may have been present. The washing is repeated until enough of the enriched sand is collected, when the gold is finally recovered by careful washing or “panning out” in a smaller pan. In Mexico and South America, instead of the pan, a wooden dish or trough, known as “batea,” is used.

The “cradle” is a simple appliance for treating somewhat larger quantities, and consists essentially of a box, mounted on rockers, and provided with a perforated bottom of sheet iron in which the “pay dirt” is placed. Water is poured on the dirt, and the rocking motion imparted to the cradle causes the finer particles to pass through the perforated bottom on to a canvas screen, and thence to the base of the cradle, where the auriferous particles accumulate on transverse bars of wood, called “riffles.”

The “tom” is a sort of cradle with an extended sluice placed on an incline of about 1 in 12. The upper end contains a perforated riddle plate which is placed directly over the riffle box, and under certain circumstances mercury may be placed behind the riffles. Copper plates amalgamated with mercury are also used when the gold is very fine, and in some instances amalgamated silver coins have been used for the same purpose. Sometimes the stuff is disintegrated with water in a “puddling machine,” which was used, especially in Australia, when the earthy matters are tenacious and water scarce. The machine frequently resembles a brickmaker’s wash-mill, and is worked by horse or steam power.

In workings on a larger scale, where the supply of water is abundant, as in California, sluices were generally employed. They are shallow troughs about 12 ft. long, about 16 to 20 in. wide and 1 ft. in depth. The troughs taper slightly so that they can be joined in series, the total length often reaching several hundred feet. The incline of the sluice varies with the conformation of the ground and the tenacity of the stuff to be washed, from 1 in 16 to 1 in 8. A rectangular trough of boards, whose dimensions depend chiefly on the size of the planks available, is set up on the higher part of the ground at one side of the claim to be worked, upon trestles or piers of rough stone-work, at such an inclination that the stream may carry off all but the largest stones, which are kept back by a grating of boards about 2 in. apart. The gravel is dug by hand and thrown in at the upper end, the stones kept back being removed at intervals by two men with four-pronged steel forks. The floor of the sluice is laid with riffles made of strips of wood 2 in. square laid parallel to the direction of the current, and at other points with boards having transverse notches filled with mercury. These were known originally as Hungarian riffles.

In larger plant the upper ends of the sluices are often cut in rock or lined with stone blocks, the grating stopping the larger stones being known as a “grizzly.” In order to save very fine and especially rusty particles of gold, so-called “under-current sluices” are used; these are shallow wooden tanks, 50 sq. yds. and upwards in area, which are placed somewhat below the main sluice, and communicate with it above and below, the entry being protected by a grating so that only the finer material is admitted. These are paved with stone blocks or lined with mercury riffles, so that from the greatly reduced velocity of flow, due to the sudden increase of surface, the finer particles of gold may collect. In order to save finely divided gold, amalgamated copper plates are sometimes placed in a nearly level position, at a considerable distance from the head of the sluice, the gold which is retained in it being removed from time to time. Sluices are often made double, and they are usually cleaned up—that is, the deposit rich in gold is removed from them—once a week.

The “pan” is now only used by prospectors, while the “cradle” and “tom” are practically confined to the Chinese; the sluice is considered to be the best contrivance for washing gold gravels.

2. The Amalgamation Process.—This method is employed to extract gold from both alluvial and reef deposits: in the first case it is combined with “hydraulic mining,” i.e. disintegrating auriferous gravels by powerful jets of water, and the sluice system described above; in the second case the vein stuff is prepared by crushing and the amalgamation is carried out in mills.

Hydraulic mining has for the most part been confined to the country of its invention, California, and the western territories of America, where the conditions favourable for its use are more fully developed than elsewhere—notably the presence of thick banks of gravel that cannot be utilized by other methods, and abundance of water, even though considerable work may be required at times to make it available. The general conditions to be observed in such workings may be briefly stated as follows: (1) The whole of the auriferous gravel, down to the “bed rock,” must be removed,—that is, no selection of rich or poor parts is possible; (2) this must be accomplished by the aid of water alone, or at times by water supplemented by blasting; (3) the conglomerate must be mechanically disintegrated without interrupting the whole system; (4) the gold must be saved without interrupting the continuous flow of water; and (5) arrangements must be made for disposing of the vast masses of impoverished gravel.

The water is brought from a ditch on the high ground, and through a line of pipes to the distributing box, whence the branch pipes supplying the jets diverge. The stream issues through a nozzle, termed a “monitor” or “giant,” which is fitted with a ball and socket joint, so that the direction of the jet may be varied through considerable angles by simply moving a handle. The material of the bank being loosened by blasting and the cutting action of the water, crumbles into holes, and the superincumbent mass, often with large trees and stones, falls into the lower ground. The stream, laden with stones and gravel, passes into the sluices, where the gold is recovered in the manner already described. Under the most advantageous conditions the loss of gold may be estimated at 15 or 20%, the amount recovered representing a value of about two shillings per ton of gravel treated. The loss of mercury is about the same, from 5 to 6 cwt. being in constant use per mile of sluice.

In working auriferous river-beds, dredges have been used with considerable success in certain parts of New Zealand and on the Pacific slope in America. The dredges used in California are almost exclusively of the endless-chain bucket or steam-shovel pattern. Some dredges have a capacity under favourable conditions of over 2000 cub. yds. of gravel daily. The gravel is excavated as in the ordinary form of endless-chain bucket dredge and dumped on to the deck of the dredge. It then passes through screens and grizzlies to retain the coarse gravel, the finer material passing on to sluice boxes provided with riffles, supplied with mercury. There are belt conveyers for discharging the gravel and tailings at the end of the vessel remote from the buckets. The water necessary to the process is pumped from the river; as much as 2000 gallons per minute is used on the larger dredges.

The dressing or mechanical preparation of vein stuff containing gold is generally similar to that of other ores (see Ore-dressing), except that the precious metal should be removed from the waste substances as quickly as possible, even although other minerals of value that are subsequently recovered may be present. In all cases the quartz or other vein stuff must be reduced to a very fine powder as a preliminary to further operations. This may be done in several ways, e.g. either (1) by the Mexican crusher or arrastra, in which the grinding is effected upon a bed of stone, over which heavy blocks of stone attached to cross arms are dragged by the rotation of the arms about a central spindle, or (2) by the Chilean mill or trapiche, also known as the edge-runner, where the grinding stones roll upon the floor, at the same time turning about a central upright—contrivances which are mainly used for the preparation of silver ores; but by far the largest proportion of the gold quartz of California, Australia and Africa is reduced by (3) the stamp mill, which is similar in principle to that used in Europe for the preparation of tin and other ores.

The stamp mill was first used in California, and its use has since spread over the whole world. In the mills of the Californian type the stamp is a cylindrical iron pestle faced with a chilled cast iron shoe, removable so that it can be renewed when necessary, attached to a round iron rod or lifter, the whole weighing from 600 to 900 ℔; stamps weighing 1320 ℔ are in use in the Transvaal. The lift is effected by cams acting on the under surface of tappets, and formed by cylindrical boxes keyed on to the stems of the lifter about one-fourth of their length from the top. As, however, the cams, unlike those of European stamp mills, are placed to one side of the stamp, the latter is not only lifted but turned partly round on its own axis, whereby the shoes are worn down uniformly. The height of lift may be between 4 and 18 in., and the number of blows from 30 to over 100 per minute. The stamps are usually arranged in batteries of five; the order of working is usually 1, 4, 2, 5, 3, but other arrangements, e.g. 1, 3, 5, 2, 4, and 1, 5, 2, 4, 3, are common. The stuff, previously broken to about 2-in. lumps in a rock-breaker, is fed in through an aperture at the back of the “battery box,” a constant supply of water is admitted from above, and mercury in a finely divided state is added at frequent intervals. The discharge of the comminuted material takes place through an aperture, which is covered by a thin steel plate perforated with numerous slits about 1/50th in. broad and 1/2 in. long, a certain volume being discharged at every blow and carried forward by the flushing water over an apron or table in front, covered by copper plates filled with mercury. Similar plates are often used to catch any particles of gold that may be thrown back, while the main operation is so conducted that the bulk of the gold may be reduced to the state of amalgam by bringing the two metals into intimate contact under the stamp head, and remain in the battery. The tables in front are laid at an incline of about 8° and are about 13 ft. long; they collect from 10 to 15% of the whole gold; a further quantity is recovered by leading the sands through a gutter about 16 in. broad and 120 ft. long, also lined with amalgamated copper plates, after the pyritic and other heavy minerals have been separated by depositing in catch pits and other similar contrivances.

When the ore does not contain any considerable amount of free gold mercury is not, as a rule, used during the crushing, but the amalgamation is carried out in a separate plant. Contrivances of the most diverse constructions have been employed. The most primitive is the rubbing together of the concentrated crushings with mercury in iron mortars. Barrel amalgamation, i.e. mixing the crushings with mercury in rotating barrels, is rarely used, the process being wasteful, since the mercury is specially apt to be “floured” (see below).

At Schemnitz, Kerpenyes, Kreuzberg and other localities in Hungary, quartz vein stuff containing a little gold, partly free and partly associated with pyrites and galena, is, after stamping in mills, similar to those described above, but without rotating stamps, passed through the so-called “Hungarian gold mill” or “quick-mill.” This consists of a cast-iron pan having a shallow cylindrical bottom holding mercury, in which a wooden muller, nearly of the same shape as the inside of the pan, and armed below with several projecting blades, is made to revolve by gearing wheels. The stuff from the stamps is conveyed to the middle of the muller, and is distributed over the mercury, when the gold subsides, while the quartz and lighter materials are guided by the blades to the circumference and are discharged, usually into a second similar mill, and subsequently pass over blanket tables, i.e. boards covered with canvas or sacking, the gold and heavier particles becoming entangled in the fibres. The action of this mill is really more nearly analogous to that of a centrifugal pump, as no grinding action takes place in it. The amalgam is cleaned out periodically—fortnightly or monthly—and after filtering through linen bags to remove the excess of mercury, it is transferred to retorts for distillation (see below).

Many other forms of pan-amalgamators have been devised. The Laszlo is an improved Hungarian mill, while the Piccard is of the same type. In the Knox and Boss mills, which are also employed for the amalgamation of silver ores, the grinding is effected between flat horizontal surfaces instead of conical or curved surfaces as in the previously described forms.

One of the greatest difficulties in the treatment of gold by amalgamation, and more particularly in the treatment of pyrites, arises from the so-called “sickening” or “flouring” of the mercury; that is, the particles, losing their bright metallic surfaces, are no longer capable of coalescing with or taking up other metals. Of the numerous remedies proposed the most efficacious is perhaps sodium amalgam. It appears that amalgamation is often impeded by the tarnish found on the surface of the gold when it is associated with sulphur, arsenic, bismuth, antimony or tellurium. Henry Wurtz in America (1864) and Sir William Crookes in England (1865) made independently the discovery that, by the addition of a small quantity of sodium to the mercury, the operation is much facilitated. It is also stated that sodium prevents both the “sickening” and the “flouring” of the mercury which is produced by certain associated minerals. The addition of potassium cyanide has been suggested to assist the amalgamation and to prevent “flouring,” but Skey has shown that its use is attended with loss of gold.

Separation of Gold from the Amalgam.—The amalgam is first pressed in wetted canvas or buckskin in order to remove excess of mercury. Lumps of the solid amalgam, about 2 in. in diameter, are introduced into an iron vessel provided with an iron tube that leads into a condenser containing water. The distillation is then effected by heating to dull redness. The amalgam yields about 30 to 40% of gold. Horizontal cylindrical retorts, holding from 200 to 1200 ℔ of amalgam, are used in the larger Californian mills, pot retorts being used in the smaller mills. The bullion left in the retorts is then melted in black-lead crucibles, with the addition of small quantities of suitable fluxes, e.g. nitre, sodium carbonate, &c.

The extraction of gold from auriferous minerals by fusion, except as an incident in their treatment for other metals, is very rarely practised. It was at one time proposed to treat the concentrated black iron obtained in the Ural gold washings, which consists chiefly of magnetite, as an iron ore, by smelting it with charcoal for auriferous pig-iron, the latter metal possessing the property of dissolving gold in considerable quantity. By subsequent treatment with sulphuric acid the gold could be recovered. Experiments on this point were made by Anossow in 1835, but they have never been followed in practice.

Gold in galena or other lead ores is invariably recovered in the refining or treatment of the lead and silver obtained. Pyritic ores containing copper are treated by methods analogous to those of the copper smelter. In Colorado the pyritic ores containing gold and silver in association with copper are smelted in reverberatory furnaces for regulus, which, when desilverized by Ziervogel’s method, leaves a residue containing 20 or 30 oz. of gold per ton. This is smelted with rich gold ores, notably those containing tellurium, for white metal or regulus; and by a following process of partial reduction analogous to that of selecting in copper smelting, “bottoms” of impure copper are obtained in which practically all the gold is concentrated. By continuing the treatment of these in the ordinary way of refining, poling and granulating, all the foreign matters other than gold, copper and silver are removed, and, by exposing the granulated metal to a high oxidizing heat for a considerable time the copper may be completely oxidized while the precious metals are unaltered. Subsequent treatment with sulphuric acid renders the copper soluble in water as sulphate, and the final residue contains only gold and silver, which is parted or refined in the ordinary way. This method of separating gold from copper, by converting the latter into oxide and sulphate, is also used at Oker in the Harz.

Extraction by Means of Aqueous Solutions.—Many processes have been suggested in which the gold of auriferous deposits is converted into products soluble in water, from which solutions the gold may be precipitated. Of these processes, two only are of special importance, viz. the chlorination or Plattner process, in which the metal is converted into the chloride, and the cyanide or MacArthur-Forrest process, in which it is converted into potassium aurocyanide.

(3) Chlorination or Plattner Process.—In this process moistened gold ores are treated with chlorine gas, the resulting gold chloride dissolved out with water, and the gold precipitated with ferrous sulphate, charcoal, sulphuretted hydrogen or otherwise. The process originated in 1848 with C. F. Plattner, who suggested that the residues from certain mines at Reichenstein, in Silesia, should be treated with chlorine after the arsenical products had been extracted by roasting. It must be noticed, however, that Percy independently made the same discovery, and stated his results at the meeting of the British Association (at Swansea) in 1849, but the Report was not published until 1852. The process was introduced in 1858 by Deetken at Grass Valley, California, where the waste minerals, principally pyrites from tailings, had been worked for a considerable time by amalgamation. The process is rarely applied to ores direct; free-milling ores are generally amalgamated, and the tailings and slimes, after concentration, operated upon. Three stages in the process are to be distinguished: (i) calcination, to convert all the metals, except gold and silver, into oxides, which are unacted upon by chlorine; (ii.) chlorinating the gold and lixiviating the product; (iii.) precipitating the gold.

The calcination, or roasting, is conducted at a low temperature in some form of reverberatory furnace. Salt is added in the roasting to convert any lime, magnesia or lead which may be present, into the corresponding chlorides. The auric chloride is, however, decomposed at the elevated temperature into finely divided metallic gold, which is then readily attacked by the chlorine gas. The high volatility of gold in the presence of certain metals must also be considered. According to Egleston the loss may be from 40 to 90% of the total gold present in cupriferous ores according to the temperature and duration of calcination. The roasted mineral, slightly moistened, is introduced into a vat made of stoneware or pitched planks, and furnished with a double bottom. Chlorine, generally prepared by the interaction of pyrolusite, salt and sulphuric acid, is led from a suitable generator beneath the false bottom, and rises through the moistened ore, which rests on a bed of broken quartz; the gold is thus converted into a soluble chloride, which is afterwards removed by washing with water. Both fixed and rotating vats are employed, the chlorination proceeding more rapidly in the latter case; rotating barrels are sometimes used. There have also been introduced processes in which the chlorine is generated in the chloridizing vat, the reagents used being dilute solutions of bleaching powder and an acid. Munktell’s process is of this type. In the Thies process, used in many districts in the United States, the vats are rotating barrels made, in the later forms, of iron lined with lead, and provided with a filter formed of a finely perforated leaden grating running from one end of the barrel to the other, and rigidly held in place by wooden frames. Chlorine is generated within the barrel from sulphuric acid and chloride of lime. After charging, the barrel is rotated, and when the chlorination is complete the contents are emptied on a filter of quartz or some similar material, and the filtrate led to settling tanks.

After settling the solution is run into the precipitating tanks. The precipitants in use are: ferrous sulphate, charcoal and sulphuretted hydrogen, either alone or mixed with sulphur dioxide; the use of copper and iron sulphides has been suggested, but apparently these substances have achieved no success.

In the case of ferrous sulphate, prepared by dissolving iron in dilute sulphuric acid, the reaction follows the equation AuCl₃ + 3FeSO₄ = FeCl₃ + Fe₂(SO₄)₃ + Au. At the same time any lead, calcium, barium and strontium present are precipitated as sulphates; it is therefore advantageous to remove these metals by the preliminary addition of sulphuric acid, which also serves to keep any basic iron salts in solution. The precipitation is carried out in tanks or vats made with wooden sides and a cement bottom. The solutions are well mixed by stirring with wooden poles, and the gold allowed to settle, the time allowed varying from 12 to 72 hours. The supernatant liquid is led into settling tanks, where a further amount of gold is deposited, and is then filtered through sawdust or sand, the sawdust being afterwards burnt and the gold separated from the ashes and the sand treated in the chloridizing vat. The precipitated gold is washed, treated with salt and sulphuric acid to remove iron salts, roughly dried by pressing in cloths or on filter paper, and then melted with salt, borax and nitre in graphite crucibles. Thus prepared it has a fineness of 800-960, the chief impurities usually being iron and lead.

Charcoal is used as the precipitant at Mount Morgan, Australia. Its use was proposed as early as 1818 and 1819 by Hare and Henry; Percy advocated it in 1869, and Davis adopted it on the large scale at a works in Carolina in 1880. The action is not properly understood; it may be due to the reducing gases (hydrogen, hydrocarbons, &c.) which are invariably present in wood charcoal. The process consists essentially in running the solution over layers of charcoal, the charcoal being afterwards burned. It has been found that the reaction proceeds faster when the solution is heated.

Precipitation with sulphur dioxide and sulphuretted hydrogen proceeds much more rapidly, and has been adopted at many works. Sulphur dioxide, generated by burning sulphur, is forced into the solution under pressure, where it interacts with any free chlorine present to form hydrochloric and sulphuric acids. Sulphuretted hydrogen, obtained by treating iron sulphide or a coarse matte with dilute sulphuric acid, is forced in similarly. The gold is precipitated as the sulphide, together with any arsenic, antimony, copper, silver and lead which may be present. The precipitate is collected in a filter-press, and then roasted in muffle furnaces with nitre, borax and sodium carbonate. The fineness of the gold so obtained is 900 to 950.

4. Cyanide Process.—This process depends upon the solubility of gold in a dilute solution of potassium cyanide in the presence of air (or some other oxidizing agent), and the subsequent precipitation of the gold by metallic zinc or by electrolysis. The solubility of gold in cyanide solutions was known to K. W. Scheele in 1782; and M. Faraday applied it to the preparation of extremely thin films of the metal. L. Eisner recognized, in 1846, the part played by the atmosphere, and in 1879 Dixon showed that bleaching powder, manganese dioxide, and other oxidizing agents, facilitated the solution. S. B. Christy (Trans. A.I.M.E., 1896, vol. 26) has shown that the solution is hastened by many oxidizing agents, especially sodium and manganese dioxides and potassium ferricyanide. According to G. Bodländer (Zeit. f. angew. Chem., 1896, vol. 19) the rate of solution in potassium cyanide depends upon the subdivision of the gold—the finer the subdivision the quicker the solution,—and on the concentration of the solution—the rate increasing until the solution contains 0.25% of cyanide, and remaining fairly stationary with increasing concentration. The action proceeds in two stages; in the first hydrogen peroxide and potassium aurocyanide are formed, and in the second the hydrogen peroxide oxidizes a further quantity of gold and potassium cyanide to aurocyanide, thus (1) 2Au + 4KCN + O₂ + 2H₂O = 2KAu(CN)₂ + 4KOH + H₂O₂; (2) 2Au + 4KCN + 2H₂O₂ = 2KAu(CN)₂ + 4KOH. The end reaction may be written 4Au + 8KCN + 2H₂O + O₂ = 4KAu(CN)₂ + 4KOH.

The commercial process was patented in 1890 by MacArthur and Forrest, and is now in use all over the world. It is best adapted for free-milling ores, especially after the bulk of the gold has been removed by amalgamation. It has been especially successful in the Transvaal. In the Witwatersrand the ore, which contains about 9 dwts. of gold to the metric ton (2000 ℔), is stamped and amalgamated, and the slimes and tailings, containing about 31/2 dwts. per ton, are cyanided, about 2 dwts. more being thus extracted. The total cost per ton of ore treated is about 6s., of which the cyaniding costs from 2s. to 4s.

The process embraces three operations: (1) Solution of the gold; (2) precipitation of the gold; (3) treatment of the precipitate.

The ores, having been broken and ground, generally in tube mills, until they pass a 150 to 200-mesh sieve, are transferred to the leaching vats, which are constructed of wood, iron or masonry; steel vats, coated inside and out with pitch, of circular section and holding up to 1000 tons, have come into use. The diameter is generally 26 ft., but may be greater; the best depth is considered to be a quarter of the diameter. The vats are fitted with filters made of coco-nut matting and jute cloth supported on wooden frames. The leaching is generally carried out with a strong, medium, and with a weak liquor, in the order given; sometimes there is a preliminary leaching with a weak liquor. The strengths employed depend also upon the mode of precipitation adopted, stronger solutions (up to 0.25% KCN) being used when zinc is the precipitant. For electrolytic precipitation the solution may contain up to 0.1% KCN. The liquors are run off from the vats to the electrolysing baths or precipitating tanks, and the leached ores are removed by means of doors in the sides of the vats into wagons. In the Transvaal the operation occupies 31/2 to 4 days for fine sands, and up to 14 days for coarse sands; the quantity of cyanide per ton of tailings varies from 0.26 to 0.28 ℔, for electrolytic precipitation, and 0.5 ℔ for zinc precipitation.

The precipitation is effected by zinc in the form of bright turnings, or coated with lead, or by electrolysis. According to Christy, the precipitation with zinc follows equations 1 or 2 according as potassium cyanide is present or not:

(1) 4KAu(CN)₂ + 4Zn + 2H₂O = 2Zn(CN)₂ + K₂Zn(CN)₄ + Zn(OK)₂ + 4H + 4Au;

(2) 2KAu(CN)₂ + 3Zn + 4KCN + 2H₂O = 2K₂Zn(CN)₄ + Zn(OK)₂ + 4H + 2Au;

one part of zinc precipitating 3.1 parts of gold in the first case, and 2.06 in the second. It may be noticed that the potassium zinc cyanide is useless in gold extraction, for it neither dissolves gold nor can potassium cyanide be regenerated from it.

The precipitating boxes, generally made of wood but sometimes of steel, and set on an incline, are divided by partitions into alternately wide and narrow compartments, so that the liquor travels upwards in its passage through the wide divisions and downwards through the narrow divisions. In the wider compartments are placed sieves having sixteen holes to the square inch and bearing zinc turnings. The gold and other metals are precipitated on the under surfaces of the turnings and fall to the bottom of the compartment as a black slime. The slime is cleaned out fortnightly or monthly, the zinc turnings being cleaned by rubbing and the supernatant liquor allowed to settle in the precipitating boxes or in separate vessels. The slime so obtained consists of finely divided gold and silver (5-50%), zinc (30-60%), lead (10%), carbon (10%), together with tin, copper, antimony, arsenic and other impurities of the zinc and ores. After well washing with water, the slimes are roughly dried in bag-filters or filter-presses, and then treated with dilute sulphuric acid, the solution being heated by steam. This dissolves out the zinc. Lime is added to bring down the gold, and the sediment, after washing and drying, is fused in graphite crucibles.

5. Electrolytic Processes.—The electrolytic separation of the gold from cyanide solutions was first practised in the Transvaal. The process, as elaborated by Messrs. Siemens and Halske, essentially consists in the electrolysis of weak solutions with iron or steel plate anodes, and lead cathodes, the latter, when coated with gold, being fused and cupelled. Its advantages over the zinc process are that the deposited gold is purer and more readily extracted, and that weaker solutions can be employed, thereby effecting an economy in cyanide.

In the process employed at the Worcester Works in the Transvaal, the liquors, containing about 150 grains of gold per ton and from 0.08 to 0.01% of cyanide, are treated in rectangular vats in which is placed a series of iron and leaden plates at intervals of 1 in. The cathodes, which are sheets of thin lead foil weighing 11/2 ℔ to the sq. yd., are removed monthly, their gold content being from 0.5 to 10%, and after folding are melted in reverberatory furnaces to ingots containing 2 to 4% of gold. Cupellation brings up the gold to about 900 fine. Many variations of the electrolytic process as above outlined have been suggested. S. Cowper Coles has suggested aluminium cathodes; Andreoli has recommended cathodes of iron and anodes of lead coated with lead peroxide, the gold being removed from the iron cathodes by a brief immersion in molten lead; in the Pelatan-Cerici process the gold is amalgamated at a mercury cathode (see also below).

Refining or Parting of Gold.—Gold is almost always silver-bearing, and it may be also noticed that silver generally contains some gold. Consequently the separation of these two metals Is one of the most important metallurgical processes. In addition to the separation of the silver the operation extends to the elimination of the last traces of lead, tin, arsenic, &c. which have resisted the preceding cupellation.

The “parting” of gold and silver is of considerable antiquity. Thus Strabo states that in his time a process was employed for refining and purifying gold in large quantities by cementing or burning it with an aluminous earth, which, by destroying the silver, left the gold in a state of purity. Pliny shows that for this purpose the gold was placed on the fire in an earthen vessel with treble its weight of salt, and that it was afterwards again exposed to the fire with two parts of salt and one of argillaceous rock, which, in the presence of moisture, effected the decomposition of the salt; by this means the silver became converted into chloride.

The methods of parting can be classified into “dry,” “wet” and electrolytic methods. In the “dry” methods the silver is converted into sulphide or chloride, the gold remaining unaltered; in the “wet” methods the silver is dissolved by nitric acid or boiling sulphuric acid; and in the electrolytic processes advantage is taken of the fact that under certain current densities and other circumstances silver passes from an anode composed of a gold-silver alloy to the cathode more readily than gold. Of the dry methods only F. B. Miller’s chlorine process is of any importance, this method, and the wet process of refining by sulphuric acid, together with the electrolytic process, being the only ones now practised.

The conversion of silver into the sulphide may be effected by heating with antimony sulphide, litharge and sulphur, pyrites, or with sulphur alone. The antimony, or Guss und Fluss, method was practised up till 1846 at the Dresden mint; it is only applicable to alloys containing more than 50% of gold. The fusion results in the formation of a gold-antimony alloy, from which the antimony is removed by an oxidizing fusion with nitre. The sulphur and litharge, or Pfannenschmied, process was used to concentrate the gold in an alloy in order to make it amenable to “quartation,” or parting with nitric acid. Fusion with sulphur was used for the same purpose as the Pfannenschmied process. It was employed in 1797 at the St Petersburg mint.

The conversion of the silver into the chloride may be effected by means of salt—the “cementation” process—or other chlorides, or by free chlorine—Miller’s process. The first process consists essentially in heating the alloy with salt and brickdust; the latter absorbs the chloride formed, while the gold is recovered by washing. It is no longer employed. The second process depends upon the fact that, if chlorine be led into the molten alloy, the base metals and the silver are converted into chlorides. It was proposed in 1838 by Lewis Thompson, but it was only applied commercially after Miller’s improvements in 1867, when it was adopted at the Sydney mint. Sir W. C. Roberts-Austen introduced it at the London mint; and it has also been used at Pretoria. It is especially suitable to gold containing little silver and base metals—a character of Australian gold—but it yields to the sulphuric acid and electrolytic methods in point of

The separation of gold from silver in the wet way may be effected by nitric acid, sulphuric acid or by a mixture of sulphuric acid and aqua regia.

Parting by nitric acid is of considerable antiquity, being mentioned by Albertus Magnus (13th cent.), Biringuccio (1540) and Agricola (1556). It is now rarely practised, although in some refineries both the nitric acid and the sulphuric acid processes are combined, the alloy being first treated with nitric acid. It used to be called “quartation” or “inquartation,” from the fact that the alloy best suited for the operation of refining contained 3 parts of silver to 1 of gold. The operation may be conducted in vessels of glass or platinum, and each pound of granulated metal is treated with a pound and a quarter of nitric acid of specific gravity 1.32. The method is sometimes employed in the assay of gold.

Refining by sulphuric acid, the process usually adopted for separating gold from silver, was first employed on the large scale by d’Arcet in Paris in 1802, and was introduced into the Mint refinery, London, by Mathison in 1829. It is based upon the facts that concentrated hot sulphuric acid converts silver and copper into soluble sulphates without attacking the gold, the silver sulphate being subsequently reduced to the metallic state by copper plates with the formation of copper sulphate. It is applicable to any alloy, and is the best method for parting gold with the exception of the electrolytic method.

The process embraces four operations: (1) the preparation of an alloy suitable for parting; (2) the treatment with sulphuric acid; (3) the treatment of the residue for gold; (4) the treatment of the solution for silver.

It is necessary to remove as completely as possible any lead, tin, bismuth, antimony, arsenic and tellurium, impurities which impair the properties of gold and silver, by an oxidizing fusion, e.g. with nitre. Over 10% of copper makes the parting difficult; consequently in such alloys the percentage of copper is diminished by the addition of silver free from copper, or else the copper is removed by a chemical process. Other undesirable impurities are the platinum metals, special treatment being necessary when these substances are present. The alloy, after the preliminary refining, is granulated by being poured, while molten, in a thin stream into cold water which is kept well agitated.

The acid treatment is generally carried out in cast iron pots; platinum vessels used to be employed, while porcelain vessels are only used for small operations, e.g. for charges of 190 to 225 oz. as at Oker in the Harz. The pots, which are usually cylindrical with a hemispherical bottom, may hold as much as 13,000 to 16,000 oz. of alloy. They are provided with lids, made either of lead or of wood lined with lead, which have openings to serve for the introduction of the alloy and acid, and a vent tube to lead off the vapours evolved during the operation. The bullion with about twice its weight of sulphuric acid of 66° Bé is placed in the pot, and the whole gradually heated. Since the action is sometimes very violent, especially when the bullion is treated in the granulated form (it is steadier when thin plates are operated upon), it is found expedient to add the acid in several portions. The heating is continued for 4 to 12 hours according to the amount of silver present; the end of the reaction is known by the absence of any hissing. Generally the reaction mixture is allowed to cool, and the residue, which settles to the bottom of the pot, consists of gold together with copper, lead and iron sulphates, which are insoluble in strong sulphuric acid; silver sulphate may also separate if present in sufficient quantity and the solution be sufficiently cooled. The solution is removed by ladles or by siphons, and the residue is leached out with boiling water; this removes the sulphates. A certain amount of silver is still present and, according to M. Pettenkofer, it is impossible to remove all the silver by means of sulphuric acid. Several methods are in use for removing the silver. Fusion with an alkaline bisulphate converts the silver into the sulphate, which may be extracted by boiling with sulphuric acid and then with water. Another process consists in treating a mixture of the residue with one-quarter of its weight of calcined sodium sulphate with sulphuric acid, the residue being finally boiled with a large quantity of acid. Or the alloy is dissolved in aqua regia, the solution filtered from the insoluble silver chloride, and the gold precipitated by ferrous chloride.

The silver present in the solution obtained in the sulphuric acid boiling is recovered by a variety of processes. The solution may be directly precipitated with copper, the copper passing into solution as copper sulphate, and the silver separating as a mud, termed “cement silver.” Or the silver sulphate may be separated from the solution by cooling and dilution, and then mixed with iron clippings, the interaction being accompanied with a considerable evolution of heat. Or Gutzkow’s method of precipitating the metal with ferrous sulphate may be employed.

The electrolytic parting of gold and silver has been shown to be more economical and free from the objections—such as the poisonous fumes—of the sulphuric acid process. One process depends upon the fact that, with a suitable current density, if a very dilute solution of silver nitrate be electrolysed between an auriferous silver anode and a silver cathode, the silver of the anode is dissolved out and deposited at the cathode, the gold remaining at the anode. The silver is quite free from gold, and the gold after boiling with nitric acid has a fineness of over 999.

Gold is left in the anode slime when copper or silver are refined by the usual processes, but if the gold preponderate in the anode these processes are inapplicable. A cyanide bath, as used in electroplating, would dissolve the gold, but is not suitable for refining, because other metals (silver, copper, &c.) passing with gold into the solution would deposit with it. Bock, however, in 1880 (Berg- und hüttenmännische Zeitung, 1880, p. 411) described a process used at the North German Refinery in Hamburg for the refining of gold containing platinum with a small proportion of silver, lead or bismuth, and a subsequent patent specification (1896) and a paper by Wohlwill (Zeits. f. Elektrochem., 1898, pp. 379, 402, 421) have thrown more light upon the process. The electrolyte is gold chloride (2.5-3 parts of pure gold per 100 of solution) mixed with from 2 to 6% of the strongest hydrochloric acid to render the gold anodes readily soluble, which they are not in the neutral chloride solution. The bath is used at 65° to 70° C. (150° to 158° F.), and if free chlorine be evolved, which is known at once by its pungent smell, the temperature is raised, or more acid is added, to promote the solubility of the gold. The bath is used with a current-density of 100 ampères per sq. ft. at 1 volt (or higher), with electrodes about 1.2 in. apart. In this process all the anode metals pass into solution except iridium and other refractory metals of that group, which remain as metals, and silver, which is converted into insoluble chloride; lead and bismuth form chloride and oxychloride respectively, and these dissolve until the bath is saturated with them, and then precipitate with the silver in the tank. But if the gold-strength of the bath be maintained, only gold is deposited at the cathode—in a loose powdery condition from pure solutions, but in a smooth detachable deposit from impure liquors. Under good conditions the gold should contain 99.98% of the pure metal. The tank is of porcelain or glazed earthenware, the electrodes for impure solutions are 1/2 in. apart (or more with pure solutions), and are on the multiple system, and the potential difference at the terminals of the bath is 1 volt. A high current-density being employed, the turn-over of gold is rapid—an essential factor of success when the costliness of the metal is taken into account. Platinum and palladium dissolved from the anode accumulate in the solution, and are removed at intervals of, say, a few months by chemical precipitation. It is essential that the bath should not contain more than 5% of palladium, or some of this metal will deposit with the gold. The slimes are treated chemically for the separation of the metals contained in them.

Authorities.—Standard works on the metallurgy of gold are the treatises of T. Kirke Rose and of M. Eissler. The cyanide process is especially treated by M. Eissler, Cyanide Process for the Extraction of Gold, which pays particular attention to the Witwatersrand methods; Alfred James, Cyanide Practice; H. Forbes Julian and Edgar Smart, Cyaniding Gold and Silver Ores. Gold milling is treated by Henry Louis, A Handbook of Gold Milling; C. G. Warnford Lock, Gold Milling; T. A. Rickard, Stamp Milling of Gold Ores. Gold dredging is treated by Captain C. C. Longridge in Gold Dredging, and hydraulic mining is discussed by the same author in his Hydraulic Mining. For operations in special districts see J. M. Maclaren, Gold (1908); J. H. Curle, Gold Mines of the World; Africa: F. H. Hatch and J. A. Chalmers, Gold Mines of the Rand; S. J. Truscott, Witwatersrand Goldfields Banket and Mining Practice; Australasia: D. Clark, Australian Mining and Metallurgy; Karl Schmeisser, Goldfields of Australasia; A. G. Charleton, Gold Mining and Milling in Western Australia; India: F. H. Hatch, The Kolar Gold-Field.