dissolved off is at first increased considerably by each successive addition of potassium bichromate, then varies in arithmetical proportion, and finally is but little affected by the amount of potassium bichromate present.

XLI.-Experimental Researches on the Periodic Law. Part I. Tellurium.

By BOHUSLAV BRAUNER, Ph.D., F.C.S., late Berkeley Fellow of
Owens College.

On the Atomic Weight of Tellurium.

THE atomic weight of tellurium has been determined by Berzelius, who for this purpose oxidised tellurium with nitric acid, and weighed the tellurium dioxide left on ignition. He obtained the following numbers: 128.9 (in 1812), 128.9 (in 1818), and 128.3 (in 1832). In the year 1857, v. Hauer, by analysis of the potassium tellurium bromide, found Te = 127.9.* Up to the present time, the round number Te 128 has generally been accepted as the true atomic weight of tellurium. (See the atomic weight recalculations by Clarke, Washington, 1882; L. Meyer and Seubert, Leipsic, 1883; and Ostwald, Lehrbuch, Leipzig, 1884.)

In the periodic system of elements, however, tellurium lies in Series VI, between antimony = 122 and iodine 127, and it was therefore assumed by Mendeléeff, that the true atomic weight of tellurium lies between these numbers, and is about 125, as follows:

·

According to the recent determinations made by J. P. Cooke, as well as those of Stas, these elements have the following atomic weights, Sb 119.96 and I = 126.86, and we may therefore expect that the atomic weight of tellurium will be still smaller than the above, namely, about 123.5.

*All atomic weights used in this paper refer to the standard 0 = 16, for reasons published by me in the Chemical News, 58, 307, and Ber., 22, 1186, of May 27th, 1889. My arguments were, I may say, accepted as valid by Horstmann (Berl. Ber., 22, R. 85) and Ostwald (ibid., 1021), although not by Lothar Meyer and Seubert (ibid., 22, 872-879). The numbers used are O = 16, S = 32.07, Cu= 63.3, Br 79.963, Ag 107.938.

=

==

=

This circumstance caused Wills (Chem. Soc. Journ., Trans., 1879, 704-713) in 1879 to redetermine the atomic weight of tellurium. By the oxidation of tellurium with nitric acid (I), Wills obtained numbers varying between Te 126-63 and 129.66. Oxidation with aqua regia (II), gave Te = 128-09 and 128.30. Lastly, analysis of potassium tellurium bromide gave Te = 126 39-127.93. Wills concludes from his experiments "that the atomic weight of tellurium does not lie between those of iodine and antimony, but is greater than the former element, which must, therefore, precede tellurium in Mendeléeff's classification."

I may be allowed to remark that the differences between the maxima and minima of the numbers obtained by Wills as the atomic weight of tellurium differ in the Series I (see above) by 3.03 units, in Series II by 0.21, and in Series III by 157 units, those of the three series showing a distance of 3.27 units and, as 214 per cent. of Wills' numbers are lower than the atomic weight of iodine, it is certainly strange to see that from such discordant results a conclusion of such great theoretical importance should be drawn.

It has been shown lately by Basaroff (J. Russ. Chem. Soc., 1887, 61-73), that in the same manner as the elements follow each other in the periodic system, their atomic weights are found to increase, so that the regular variation of quotients obtained by dividing the atomic weight of an element by that of the next lower one may be represented graphically in a satisfactory manner, and it was therefore still more probable that the said regularity should hold good also in the case of tellurium, and that a lower value would be obtained for its atomic weight than that of iodine, I = 126.86.

The object of the following research was to undertake a thorough revision of the atomic weight of tellurium, and to that research I have devoted the greater part of my spare time during the past six years.

It was to be expected that tellurium would not remain an exception to the periodic law, for although nearly one-third of the known. elements would not have fitted into the periodic system with their originally accepted atomic weights, yet hitherto it has been found that these atomic weights are at fault, and not the system. These apparent exceptions are: Li, Be, Sc, Ti, V, Co, Ga, Y, Nb, Mo, Ru, In, Sb, Cs, La, Ce, Ta, Os, Ir, Pt, Bi, Au, Th, and U.

Preparation of Material.

Two kinds of material were used for the preparation of pure tellurium: Hungarian tellurium ores, nagyagite, sylvanite, and petzite and Hungarian crude tellurium. After bringing the tellurium. into solution in the form of chloride, it was precipitated in the usual

way with sulphur dioxide in the elementary state, and this precipitate, after drying, was fused with potassium cyanide in order to get rid of any selenium which might be present, although none could be detected in it,* and also of traces of the heavy metals which are carried down with the tellurium on precipitating it with sulphur dioxide. From the claret-coloured solution of potassium telluride, tellurium was precipitated by means of a current of air. Lastly, it was placed in little porcelain boats which were introduced into a wide Bohemian glass tube, surrounded by wire gauze, and then distilled in a current of hydrogen. Minute traces of tellurides of the heavy metals were left behind.

The distilled tellurium forms a crystalline mass of silvery lustre which does not change its colour on exposure to the air. During distillation, part of the tellurium combines with hydrogen, forming tellurium hydride which, before escaping from the heated tube, is again decomposed with deposition of beautiful needle-shaped crystals of the element measuring 1 cm. or more in length; the hydrogen formed, however, always contains some gas of an extremely disagreeable odour which behaves like tellurium hydride, and has very poisonous properties.

By this costly process, large quantities of tellurium were prepared, the properties of which are those required of pure tellurium, according to the present state of our knowledge; it may be redistilled in hydrogen without leaving any residue, and no impurities can be detected in it by qualitative analysis.

Determination of the Atomic Weight.

It was my intention to determine the atomic weight of tellurium by as many independent methods as possible, but in the course of this investigation it was found to be very difficult to devise methods which would answer the purpose. In this, as in many other respects, tellurium approaches its next horizontal "atomanalogue," antimony, of which Berzelius said in 1812: "I have never worked with a material with which it was so extremely difficult to obtain constant results." (Schweiger, 6, 149.)

All the reagents used in the course of the present investigation, as hydrochloric, nitric, and sulphuric acids, were subjected to fractional distillation from platinum vessels, and the water employed was redistilled with alkaline permanganate, and the steam cooled in a platinum condenser; the set of weights used was corrected by the method of vibrations.

* Having been probably volatilised by the repeated evaporations with strong hydrochloric acid employed to get rid of nitric acid.

A. Conversion of Tellurium into the Dioxide.

Tellurium was oxidised (a) with nitric acid; (b) with aqua regia. (a.) Oxidation with Nitric Acid.-The finely powdered tellurium was treated in Experiments 1 and 2 in a platinum crucible with dilute nitric acid, in which it readily dissolves. The crucible was covered with a watch-glass, in order to avoid loss by spirting. After the reaction was over, the contents of the crucible were evaporated to dryness on a water-bath, and then heated slowly up to 400°. It was observed, however, that the basic tellurium nitrate formed loses its last trace of nitric acid at a temperature at which part of the tellurous anhydride contained in the crucible begins to volatilise, the watch-glass becoming covered with a white substance.

Experiment No. 3 was carried on in a flask of hard Bohemian glass (previously boiled with aqua regia) so that the volatile portions were condensed in its colder neck. This method was abandoned, as it did not give constant results.

(b.) Oxidation with Aqua Regia.-Weighed quantities of pure powdered tellurium were placed in a round-bottomed flask of hard Bohemian glass, Fig. 1 (previously boiled for many days with aqua regia), and, after introducing some hydrochloric acid, nitric acid was

added gradually. As, however, large quantities of the finely divided spray of the solution are carried over with the escaping gases, the neck of the flask was closed with a glass tube, b, containing bulbs in the middle, the wider end of which was tightly ground into the neck of the flask, the narrow end dipping into a small beaker containing nitric acid. The escaping gases are thus forced to pass through the nitric acid contained in the small bulbs and in the beaker, and in this way the escaping spray of tellurium solution is completely condensed.

The arrangement of this simple apparatus is seen from Fig. 1, and it was used several times during the course of the present investigation when weighed quantities of substance had to be dissolved. without any of the latter being carried off with the escaping gases.

On evaporating the solution of tellurium in aqua regia, some of the tellurium tetrachloride formed is volatilised-a circumstance which was overlooked by Wills-and, further, during the evaporation, organic substances contained in the air may be condensed in it, causing a subsequent reduction at a higher temperature. In order to avoid these sources of error, the solution of tellurium was evaporated in a current of pure, dry air in an apparatus the arrangement of which is seen from Fig. 2. Air from the open was aspirated through potash solution, e, a long layer of cotton-wool, f, over pumice-stone soaked

with sulphuric acid, g, and then passed through a tube, h, containing red hot platinum sponge. This tube ended in d, which is ground into the flask a containing the tellurium solution, and heated in an airbath, b, the bottom of which was protected from the direct action of the radiant heat by a screen, c, made of asbestos cardboard. The glass bulbs i and flask k serve for the condensation of the acids escaping from the flask a. The various single pieces of which this apparatus