In the summer of 1885, a rich silver ore of unusual appearance was found at Himmelsfurst Fundgrube near Freiburg; A. Weisbach recognized it as a new mineral species and called it "argyrodite." Th. Richter subjected the mineral to a preliminary investigation with the blowpipe and found as its main components sulphur and silver, but he also stated the presence of a small quantity of mercury, which is surprising and interesting since this metal has never shown up before in the lodes of Freiburg.
The analysis of the mineral undertaken by me showed that the mercury content is not more than 0.21 per cent; besides, depending on the purity of the investigated material, 73 to 75 per cent silver, 17 to 18 per cent sulphur, very small quantities of iron, and traces of arsenic were found. However, the most careful analysis always concluded with a loss of 6 to 7 per cent, and it did not seem possible to discover the missing part by the usual procedure of qualitative analysis.
After several weeks of painstaking search, I can now state definitely: argyrodite contains a new element, very similar to antimony and yet sharply distinguished from it, to which the name "germanium" shall be given. Its discovery was connected with much difficulty and painful doubt because the minerals that accompanied argyrodite contained arsenic and antimony; their great resemblance to germanium and the absence of methods for separation were extremely disturbing.
By its appearance alone it was beyond question that argyrodite was a sulpho salt. Therefore, it was probable that the component which so persistently escaped the analysis was similar to arsenic and antimony in other related silver ores. In that case, it was obvious to assume that this component would form a watersoluble sulpho salt when argyrodite was melted together with sulphur and sodium carbonate, and it should then be precipitated from the solution as the sulphide upon addition of acid. Actually, however, operating in this manner always gave only a white precipitate which, after washing on the filter, proved to be pure sulphur, leaving no residue after burning and dissolving in carbon disulphide. In the filtrate, no precipitable substance could be detected either by hydrogen sulphide or by other reagents. I realized only later that germanium really was present in the sodium sulphide solution but precipitated as the sulphide only after adding much acid, because it is considerably soluble in acid-free liquids and thus, once precipitated, could not be washed with water except with great losses or even complete disappearance (by dissolution). The presence of much free sulphur in the precipitate caused additional difficulties. They could be overcome by treating the precipitate, without preliminary washing, by dilute ammonia. Leaving the sulphur behind, germanium sulphide is dissolved, and when the solution is then strongly acidified by hydrochloric acid, a white, voluminous precipitate is obtained that cannot be mistaken for sulphur. By adding the hydrochloric acid slowly in drops, first antimony sulphide, then arsenious sulphide are separated; after filtration, a solution is obtained from which snow-white germanium sulphide precipitates upon further, ample addition of hydrochloric acid. Thus, by way of fractional precipitation, I succeeded in removing the interfering companions and in demonstrating the existence of the new element beyond all doubt . . . .
It is now established that germanium combines with sulphur in two proportions, that it forms a brown-red lower sulphide and a higher white sulphide. The sublimate obtained by heating argyrodite in a current of hydrogen is the colored sulphide which dissolves unchanged in potassium hydroxide but goes over into the other sulphide upon teatment with ammonium sulphide. This explains why the addition of acid forms a red precipitate in the first solution, a white one in the other.
The similarity in the behavior of argyrodite and antimony minerals . . . led to the assumption that the new element in argyrodite ought to be close to antimony in chemical character, that presumably its place in the periodic system should be between antimony and bismuth; in short, that it should be Mendeleev's eka-antimony.
It was definitely premature when I expressed such an assumption in my first notice concerning germanium; at least there was no basis for its proof. Nor would I have ventured at first to assume argyrodite to be a sulpho salt with a quadrivalent acid radical, because there were no analogies at all for such an assumption. Thus the present case shows very clearly how treacherous it can be to build upon analogies; the quadrivalency of germanium has by now become an incontrovertible fact, and there can be no longer any doubt that the new element is no other than the eka-silicium prognosticated fifteen years ago by Mendeleev.
Lothar Meyer early declared germanium to be eka-silicium and added that according to his curve of the atomic volumes it should be low-melting and perhaps easy to evaporate. At that time, germanium had not yet been produced in crystallized form; it is so much more remarkable that, as will be shown, Lothar Meyer's prediction has come true to a certain degree.
The specific gravity at 20.4° was found by P. Mann = 5.469, quite in agreement with Mendeleev's calculation, according to which it should be about 5.5.
The tetrachloride is a colorless, thin liquid of specific gravity 1.887 at 18.0° as measured by P. Mann. It does not solidify at -20°. Its boiling point is around 80° . . . From its composition GeCl4 and its specific gravity, its (molecular) volume is calculated as 113.3; Mendeleev prognosticated the specific gravity of EsCl4 to be about 1.9, the volume 113, the boiling point probably a little below 100°.