Early Investigations. CHAPTER III. ALLOYS. As many valuable mechanical properties are conferred upon metals by associating them with each other, it seldom happens that metals are used in a state of purity when they are intended for industrial purposes, and this fact was discovered at a very early period of metallurgical history. The word alloy originally comes, in all probability, from the Latin adligo (alligo), "to bind to," and not, as Sir John Pettus thought, from the Teutonic linderen, "to lessen," suggestive as it is of the fact that a precious metal is lessened in value by the addition of a base one. The distinguished chemist, Dumas, eloquently pleaded, many years ago, against leaving alloys in the oblivion to which moderr chemists consigned them, and there still seems to be a prevalen impression that our knowledge of the phenomena which attend the union of metals is very imperfect, and that it rests upon a slender experimental basis. We are apt to forget the extent and complexity of the subject, and Lupton has opportunely directed attention to the number of alloys which await examination. He says:- "Hatchett recommended that a systematic examination of all possible alloys of all the metals should be undertaken. He forgot to remind any one who should attempt to follow his advice that if only one proportion of each of the thirty common metals were considered, the number of binary alloys would be 435, of ternary 4060, and of the quarternary 27,405. If four multiples of each of the 30 metals be taken, the binary compounds are 5655, ternary 247,660, and quarternary 1,013,985." Nevertheless, if the properties of many alloys have yet to be investigated, the study of alloys generally has not been neglected. The modern bibliography relating to them is much more extensive than it is usually supposed to be, and the older writings are very full, and contain the results of far more accurate observation Nature, vol. xxxvii., Jan. 5, 1888, p. 238. than they are credited with. In the early days of chemistry, as its history abundantly proves, alloys received much attention, and although the early chemists often failed to distinguish alloys from simple metals, or used them in unsuitable ways, they left an experimental record, the value of which is sadly unappreciated. From this record it is, incidentally, evident that the development of the art of separating metals from their ores, and from each other, was quickly followed by the acquisition of the knowledge that metals possess peculiar properties when re-united in certain proportions, and are thereby rendered more useful than they were in the pure state. In early times some metals were used unalloyed, although at the present day they have no industrial application except in union with other metals. Antimony, for instance, now only employed as a constituent of certain alloys, was formerly cast and fashioned into ornaments, as is proved by the analyses of Virchow, and by a fragment of a very ancient Chaldean vase, which fragment, when examined by Berthelot, proved to be of pure antimony.* The implements and ornaments discovered by Schliemann abundantly show that the early Greeks were familiar with alloys of silver and gold, copper and tin, lead and silver, and with many others, all artificially prepared. Throughout the Middle Ages there seems to have been a belief that the action of metals on gold and silver was, on the whole, corrupting; and Biringuccio, in 1540, possibly seeing that this was the prevailing view, carefully defined such alloys as being "nothing but amicable associations of metals with each other"; and he further pointed out that metals must be mixed by weight, and not at random. Views as to the Constitution of Alloys.-Passing from the sixteenth to the eighteenth century, we find four writers whose names deserve to be specially mentioned, because they seem to have been the first to indicate the direction in which modern investigation has been conducted. These are Réaumur, Gellert, Musschenbroek, and Achard, who respectively studied 1st (Réaumur), molecular change produced in a metal by heat; 2nd (Gellert), the relation of fluid metals to each other considered as solvents; 3rd (Musschenbroek), the cohesion of alloys as shown by certain mechanical properties; and 4th (Achard), the electrical behaviour of metals and alloys. It is interesting to trace the connection between the older work and the new. Réaumur,† in explaining the hardening of steel by rapid cooling from an elevated * Ann. de Chim. et de Phys., vol. xii., 1887, p. 135. temperature, comes very near the modern view that a metal may, under certain conditions, pass from one allotropic state to another, for he distinctly contemplates the possibility of molecular change produced by the expulsion by heat of "sulphurs and salts" from the molecules into interstitial spaces between them. He speaks of "molecules and elementary parts of molecules," like a modern writer, and tries to show that when hot steel is rapidly cooled, "sulphurs and salts" cannot return into the molecules, but remain in the interstitial spaces; and that, therefore, the physical properties of hard steel become quite different from those of soft. If it should be urged that the analogy between carburized iron and alloys is over-strained, it may be pointed out that, in 1867, Matthiessen said, after appealing to the fact that in certain alloys the constituent metals are present in allotropic states, "I have always made a comparison between iron and steel (and alloys). This has been done to show that the carbon iron alloys behave in an analogous manner to other alloys, which cannot be looked upon as chemical combinations."* Gellert makes the analogy of certain alloys to solutions very clear, and in his Metallurgic Chemistry he gives a table showing the relative solubilities of metals in each other, while in the observations which accompany it † he says, to cite one of the cases, he takes as an illustration, "Since copper and silver and copper and gold dissolve one another very readily, the copper cannot be parted from iron by means of gold or silver," probably having in mind a reaction which enables silver to be parted from gold by the action of sulphur and iron. He further clearly shows that, with regard to the solution of metals in a triple alloy, he understood the possibility of a division of a metal between two other metals acting as solvents. The mechanical properties of alloys were investigated by Musschenbroek, who, working in the early part of the eighteenth century, made some experiments on the tensile strength of metals and alloys. He writes of "the absolute cohesion by which a body resists fracture when acted upon by force drawing according to its length," and gives the tenacity of several metals, and the alloys, brass and pewter. He shows the importance of such work so clearly that it is remarkable how slowly the mechanical testing of metals developed since his time. Achard, whose researches were published in 1784, made a very *Journ. Chem. Soc., 1867, p. 220. + English translation of his work, London, 1776, p. 186. Elements of Philosophy, translated by John Colson, F.R. S., vol. i., 1744, p. 237. extended series of experiments on multiple alloys, as well as those of simple metals. He pointed out that the relative conductivities of substances for heat and for electricity are closely related.* He devised an appliance for the experimental verification of this fact, and, as he included alloys in his researches, it may fairly be claimed that he led the way for the important generalisation that alloys may be ranged in the same order as regards their power of conducting heat and electricity, which was made by Wiedemann and Franz in 1853-9. The necessity of metals being pure when added to each other was hardly recognised until the eighteenth century, and Duhamel, who contributed the article on alloys to the Encyclopédie Méthodique in 1792, appears to have been the first writer to insist on the necessity for making exact experiments upon alloys with metals which possess a high degree of purity and on effecting their union by heat in closed vessels. He further pointed out that up to his time no chemist had taken these precautions, and it is certain that in conducting some modern experiments they have been neglected. In the early part of the nineteenth century researches on alloys became more numerous; they were mainly directed to ascertaining the effect on the density of metals produced by alloying them and to determining the effects of slow cooling on alloys with low melting points. Of such a nature was the work of Ermann in 1827 and of Rudberg (1830-1). Ermann called attention to the peculiar behaviour of alloys of lead and tin when solid. Rudberg studied anomalies in these alloys when in the liquid state. Regnault showed that the specific heats of certain fusible alloys were greater near 100° than the mean specific heat of their constituents, and this fact appears, as Spring has shown,† to have induced Person to undertake researches on the latent heat of alloys and on these specific heats. Undoubtedly one of the greatest works on alloys of the present century was that of Matthiessen, who studied the electrical resistance of metals and alloys, and was led to the conclusion that in many cases metals are present as allotropic modifications-that is, in totally different forms from those in which we ordinarily know them. It is by no means easy to investigate the molecular constitution of alloys, but evidence may be gathered in the following ways: * Sammlung physikalischer und chemischer Abhandlungen, Berlin, 1784,. vol. i. + Bull. de l'Académie Royale de Belgique, 1886 (3), vol. xi. p. 355. |