Columbium was discovered early in the nineteenth century in the mineral columbite from Connecticut by Hatchett, an Englishman, who did not, however, obtain the pure oxide. It was afterwards obtained by Rose who named it niobium. Both names for the element are in use, but the former has priority. Attention was called to this fact by an article in the Journal by Connell, an Englishman (18, 392, 1854).
The Platinum Group of Metals.—In 1854 a new member of the platinum group of metals, ruthenium, was discovered by Claus. Platinum had been discovered about the middle of the eighteenth century, while its other rarer associates, iridium, osmium, palladium, and rhodium, had been recognized in the very early years of the nineteenth century. It was during the latter period that platinum ware began to be employed to a considerable extent in chemical operations, and this use was greatly extended as time went on. The discovery was made by Phillips in 1831 that finely divided platinum by contact would bring about the combination of sulphur dioxide with atmospheric oxygen, and this application during the past 20 years has become enormously important in the sulphuric acid industry, while other important applications of platinum as a “catalytic agent” have also been made. Wolcott Gibbs and Carey Lea have contributed perhaps more than any other recent chemists to a knowledge of the platinum metals. Carey Lea (38, 81, 248, 1864) dealt chiefly with the separation of the metals from each other, while Gibbs’s work (31, 63, 1861; 34, 341, 1862) included investigations of many of the compounds.
It may be mentioned that while platinum and its associates were formerly known only in the uncombined condition in nature, the arsenide sperrylite, PtAs2, was described by the late S. L. Penfield, and the senior writer of this chapter, in articles published in the Journal (37, 67, 71, 1889).
Applications of the Spectroscope.—The discovery in certain mineral waters of the rare alkali-metals rubidium and cæsium by Bunsen and Kirchoff in 1861 was in consequence of the application of spectroscopy by these same scientists a short time previously to the identification of elements imparting colors to the flame. Since that time the employment of the spectroscope for chemical purposes has been much extended, as it has been used in the examination of light from electric sparks and arcs, as well as from Geissler tube discharges and from colored solutions.
The metals rubidium and cæsium are interesting in being closely analogous to potassium and in standing at the extreme electro-positive end of the series of known metals. It should be noticed here that Johnson and Allen of our Sheffield Laboratory, having obtained a good supply of rubidium and cæsium material from the lepidolite of Hebron, Maine, made some important researches upon these elements, accounts of which were published in the Journal (34, 367, 1862; 35, 94, 1863). They established the atomic weight of cæsium, thus correcting Bunsen’s determination which was unsatisfactory on account of the small quantity and impurity of his material. Pollucite, a mineral rich in cæsium, which had been found in very small amount on the Island of Elba, has more recently been obtained in large quantities—hundreds of pounds—at Paris, Maine, and its vicinity. This American pollucite was first analyzed and identified by the senior writer of this article (41, 213, 1891), and later (43, 17, 1892 et seq.) the results of many investigations on cæsium and rubidium compounds, in which the junior writer played an important part, carried out in Sheffield Laboratory, were published in the Journal.
The application of the spectroscope led to the discovery of thallium in 1861 by Crookes of England, and to that of indium in 1863 by Reich and Richter in Germany. Both of these metals are extremely rare, but they are of considerable theoretical interest. Thallium is particularly remarkable in showing resemblances in its different compounds to several groups of metals.
The spectroscope was employed again in connection with the discovery of gallium in 1875 by Boisbaudran. It is in the same periodic group as thallium and indium, and it has a remarkably low melting point, just above ordinary room-temperature. It has been among the rarest of the rare elements, but within two or three years a source of it has been found in the United States in certain residues from the refining of commercial zinc. The recent issues of the Journal (41, 351, 1916; 42, 389, 1916) show that Browning and Uhler of Yale have availed themselves of this new material in order to make important chemical and physical researches upon this metal.
Germanium.—The discovery of germanium in the mineral argyrodite in 1886 by Winkler revealed a curious metal which gives a white sulphide that may be easily mistaken for sulphur and which is volatilized completely when its hydrochloric acid solution is evaporated, so that it is evasive in analytical operations. This element had been predicted with much accuracy by Mendeléeff, and it is rather closely related to tin.
A few years after the discovery of germanium, Penfield published in the Journal (46, 107, 1893; 47, 451, 1894) some analyses of argyrodite, correcting the formula given by Winkler to the mineral; also he described canfieldite, an analogous mineral from Bolivia, in which a large part of the germanium was replaced by tin.
The Rare Earths.—Before the year 1818 two rare earths, the oxides of yttrium and cerium, were known in an impure condition. Since that time about fourteen others have been discovered as associates of the first two. The rare earths are peculiar from the fact that many of them are always found mixed together in the minerals containing them, and also from the circumstance that most of them are remarkably similar in their chemical reactions and consequently exceedingly difficult to separate from each other. In many cases multitudes of fractional precipitations or crystallizations are needed to obtain pure salts of a number of these metals. The solutions of the salts of several of these elements give characteristic absorption bands when examined spectroscopically by the use of transmitted light.