THE CHEMICAL NEWS, VOL. CXXXI. No. 3431. MERTON HOUSE, SALISBURY SQUARE, LONDON, E.C.4. TELEPHONES : AN INVESTIGATION ON THE GREEN Philadelphia, Penna, U.S.A. Much has been said concerning the green colour tungstic oxide assumes on ignition. Roscoe attributes the discoloration to the presence of small traces of sodium salts, which are almost impossible to remove by washing. Von J. v. Leimpt* attributes the discolouration to the action of dust particles in the air, this action being greatly influenced by light. Others give reducing (Continued top of Column 2.) Wt. after heating in O, 4-hour TABLE The object of this investigation was to determine the extent of error produced by the discoloration of the yellow WO Determinations were first made to see if the ignited green oxide could be oxidised in a current of oxygen to the yellow oxide. Several samples of tungstic oxide were obtained, varying from a light green colour to a dark green colour, by precipitating the tungstic acid from hot solutions of normal sodium tungstate with hydrochloric or nitric acids and washing out the sodium. salts as much as possible. No change in weight or colour was perceptible after heating in the current of oxygen. The oxide was placed in a boat and heated in a silica tube at the highest temperature obtainable from a Teclu burner. The following table gives the results obtained : 1. Wt. after 2 In weight In colour v. Leimpt states that the unignited green oxide can be oxidised in a current of oxygen, but on strong ignition the oxide sinters together and forms large particles, thereby preventing access of oxygen to the green oxide. During the ignition of the oxide it was noticed that a certain amount of sintering did occur. Is this not more likely to be due to the presence of sodium salts than to the sintering of the much more difficultly-fusible substance, tungstic oxide? These sintered particles always deep green in colour and the impurities appeared to be concentrated in these particles. Samples of the oxide were *Z. Anorg Chem. 119-310 (1921). were examined under a microscope and no distinction could be made between the green and the yellow oxides. The particles appeared to have sharp and angular edges. They were magnified 228X. Impurities such as molybdenum, vanadium and manganese were looked for and were found to be absent. The sodium tungstate used, however, contained traces of iron. All the samples of green oxide shewed the presence of sodium salts in varying quantities, as indicated by the flame coloration. The yellow oxide thus far obtained contained only traces of sodium salts. Dilute solutions of ferrous sulphate and manganous sulphate were prepared and two or three drops of these solutions were added to the yellow oxide, and these failed to produce the green oxide on ignition. Since it has been shown that sodium salts are invariably present when the oxide is green, the next step was to prepare a yellow oxide free from sodium salts from a green oxide. Ten grams of sodium tungstate were dissolved in 200 c.c. of water, and heated to boiling. 50 c.c. of concentrated nitric acid were added and the precipitated tungstic acid allowed to settle. The tungstic acid was washed by decantation with 2,600 c.c. of water and finally baked at 120° C. The baked oxide was then divided into two portions. The first portion was strongly ignited in a porcelain crucible and it changed to a deep-green colour. The second portion was dissolved in ammonium hydroxide. It was necessary to warm the solution to get all the baked oxide in solution. In this solution there is only a very small amount of sodium salts. The solution was diluted to 1,600 c.c., and the tungstic acid precipitated by hydro chloric acid. The tungstic acid was allowed to settle, the solution decanted and washed once with hot water containing hydrochloric acid. The tungstic acid was then dried and any ammonium salts remaining were expelled by ignition. The oxide remaining had a bright lemon-yellow colour, and gave no test for sodium salts in the flame. Hence, it is shown definitely that the deep green colour of the tungstic oxide. is caused by the presence of sodium salts. During these precipitations of tungstic acid from sodium tungstate solutions by acids, it was observed that varying shades of green could be obtained by using varying concentrations of sodium tungstate solutions and acid. It was found that the white dihydrated tungstic oxide had a greater tendency to cling to sodium salts than the yellow tungstic oxide. The following table shows the concentrations of acid and sodium tungstate used. In all precipitations ten grams of normal sodium tungstate were used. of Previous to this it had been a matter of chance to obtain a dark green or a fairly good yellow tungstic oxide. By using procedure (4) of the preceding table, a deep green oxide could always be obtained. In all these precipitations the tungstic acid was washed with 2,500 to 3,000 c.c. water by decantation. In procedure (3) and (4) it was not possible to change the creamy-white colored tungstic acid to the deep yellow tungstic acid by heat or by the addition of more acid. It was also observed that the baked oxide of procedure (4) on ignition evolved nitric oxide fumes; also the baked oxide of procedure (3) on ignition evolved hydrochloric fumes. This would seem to indicate that tungstic acid was replacing nitric acid or hydrochloric acid from their corresponding salts. Το boiling Pptd. from hot solution Pptd. from hot solution test this out, two or three drops of a sodium chloride solution were added to a yellow sample of tungstic oxide which had previously been ignited. On strong ignition, acid fumes were detected and the yellow oxide changed to the green oxide. The next step was to determine the amount of sodium salts producing these varying colorations. The following procedure was carried out in all determinations. The ignited tungstic oxide was weighed in a boat and heated in a current of hydrogen at the lowest temperature possible to affect a reduction. It was not necessary to reduce it all to metal. The hydrogen apparatus was then disconnected and a chlorine generator attached. The tungsten was volatilised at a low temperature as oxy-chlorides and collected in the v. Leimpt has stated that light has a discolorising effect upon yellow tungstic oxide. Experiments were next carried out to determine whether the production of the green oxide was a photochemical reaction greatly accelerated by the presence of sodium salts or if it was due to the presence of sodium salts alone. A yellow oxide was prepared according to procedure (1) of Table II. The oxide contained only traces. of sodium salts. It was exposed to diffused sunlight, and only turned a very pale Beneath the surgreenish-yellow colour. face, the tungstic oxide was still yellow. Tungstic oxide was next prepared free from sodium salts by the ammonium hydroxide method previously described. This, on exposure to direct sunlight, only received a superficial pale greenish-yellow coating. No change in weight was noticed with this slight color change. Green tungstic oxide was next prepared according to procedure (3) of Table II., and divided into two portions. The first portion was ignited in diffused sunlight. On ignition, the tungstic oxide assumed a deep green colour. The second portion was ignited in a dark room and allowed to cool in the dark. The oxide was also dark green in colour, and could not be distinguished from the oxide obtained by igniting the first portion. Hence, light has no appreciable effect upon the tungstic acid, and the material change of the tungstic oxide is due to the presence of sodium salts. As an item of interest, during the evaporation of a solution containing an ammonium tungstate, ammonium chloride, am monium hydroxide, and sodium chloride, a shiny crystalline solid separated out. This was recrystallised and an analysis, corresponded closely to a sodium ammonium complex prepared by Gibbs*, having the complex formula 10 WO,.4 Na,O+ 4(10 WO,.4 (NH),0) + 50 H2O. The salt, which was separated out, was dried at 105° C., and the anhydrous salt analysed. Calculated to anhydrous salt: WO, 91.50%, Na,O 1.95%, and (NH)2O 6.55%; found: WO, 91.33%, Na,O 2.22%, and (NH)2O 6.45% by difference. This salt, on gentle ignition, loses ammonia, and assumes a dirty yellow color, from which it is not possible to extract all the sodium salts. On strong ignition it This assumes an intense green colour. gives conclusive proof that sodium salts produce the green oxide. SUMMARY. 1. Ignited green tungstic oxide cannot be oxidised in a current of oxygen to a yellow oxide. 2. It has been shown definitely that the deep green color of the tungstic oxide is caused by the presence of sodium salts. 3. It has been shown that light produces very little change in color of the oxide and the slight discoloration is only superficial. 4. The magnitude of error has been deof termined for varying discolorations tungstic oxide. THE THEORY OF THE TELEPHONE W. H. INGRAM, in Journal of the The equations connecting the winding current and diaphragm motion of a telephone receiver under impressed simple harmonic electromotive and mechanical forces are formulated under: (1) The Maxwell flux theory; (2) the magnetising-winding theory; (3) the Ampèrean resistanceless fixed molecular circuit theory of permanent magnetism; and (4) a general molecular orbital theory of ferromagnetism. On the latter theory it is found that the coupling coefficients M and N in the equations in *Amer. Chem. Jour., Vol. I., 229 (18791, |