INORGANIC CHEMISTRY 13 I n o r g a n i c Chemistry. Production of Ozone by the Decomposition of Water by Fluorine. By HENRI MOISSAN (Compt. Tend., 1899, 129, 570-573). -When fluorine is passed into water cooled at 0", the liberated oxygen contains from10 to12 per cent., or, under the most favourable conditions, as much as 14.39 per cent., of ozone, and is decidedly blue. The pro- portion of ozone is higher the more rapid the current of fluorine, provided that the water is kept at 0", and subject to the maximum limit of 14.39 per cent.; it is markedly less if the temperature of the water is allowed t o rise. By ARNOLD NABL (Monatsh., 189 9, 20, 679--684).-1f dry sulphur dioxide is passed into absolute alcohol in which metallic zinc is suspended, and contained in a flask from which a i r is excluded, a white precipitate is formed.This is left f o r a day, filtered, and the operation then repeated. The solution deposits small, microscopical, rhombic needles which, after drying in vacuum t o eliminate the alcohol of crystallisation, give figures on analysis agreeing with the formula ZnS20,. The other product, in- soluble in alcohol, appears to be a mixture. Isomerism of Salts of Ammonium, Hydroxylamine, and Hydrazine. By ALEXANDER P. SABANEEFF [with A. DSHEWACHOFF, (Clmu. Centr., 1899, ii, 32-33; from J . Rum. Ci~ern. Soc., 1899, 31, 375 -383. Compare Abstr., 1898, ii, 577).-The following pairs of C. H. B. Hyposulphurous Acid. R. H. P. M. EFROSS, z. GINSBURG, J. LEblKE, M. PROSIN, and A. WLASSOFF]14 ABSTRACTS OF CHEMICAL PAPERS.isomeric compounds have been prepared and their molecular weights determined by Raoult's method : (la) ammonium hydroxylamine hydrogen phosphite, (1 b) diammonium hydrogen phosphate ; (2a) hydroxylamine dihydrogen phosphite, (26) ammonium dihydrogen phosphate ; (3a) hydroxylamine formate, (36) ammonium hydrogen carbonate ; (4cc) hydroxylamine acetate, (4b) ammonium glycollate ; (5a) hydrazine oualate, (56) hydroxylamine oxamate ; (6a) hydrazine succinate, (6b) hydroxylamine succinama te ; (7c.5) hydroxylamine benz- oate, (7b) ammonium salicylate ; (8cc) hydroxylamine m a minobenzoate, ( 8 b ) hydrazine salicylate, QIIH,O>N,HG. (2cc) is a syrup which, with absolute alcohol, yields hydroxylamine phosphate. (3cr) crystallises in needles, and is easily soluble in water, forming an acid solution which decomposes on boiling, and does not then give the hydroxylamine reaction.( 4 b ) is prepared by mixing glycollic acid with the calculated quantity of ammonia and evaporat- ing in a vacuum; i f an excess of ammonia is used, diammonium glycollate is formed ; i t separates in needles, is easily soluble in water, giving an acid solution from which the acid salt is obtained on evap- orating. (5a) crystallises in needles, and is very slightly soluble in water. (5b) forms crystals, and is easily soluble in water. (6r.5) separates in very perfect crystals, and is soluble in water. (66) gradually changes into hydroxamic acid ;the same change is brought about by boiling the solution. (76) separates as the anhydrous salt when aqueous solutions of ammonia and salicylic acid are mixed and heated.(8a) is very slightly soluble in water, but more so in alcohol. (86) is prepared from salicylic acid and hydrazine hydr- oxide. (arc), (3a), ( 4 a ) , (5b), (6b), (?a), and (8a) are prepared from the barium salts of the corresponding acids and the sulphates of the nitrogen bases. Whilst measuring the electrical conductivity of hydrazine sulphate, its oxidation by platinum black was observed, and when oxygen is passed into an aqueous solution in which platinum black is suspended, the sulphate is completely oxidised, nitrogen, ammonium sulphate, sulphuric acid, and water being formed. Action of Potassammonium on Arsenic. By C. HUGOT (Compt. rend., 1899,129, 603-605).-The action of excess of potassammonium on arsenic yields a brick-red compound, AsK,,NH,, analogous to that previously obtained from sodarnmonium(Abstr., 1899, ii, 15l),and when this is heated a t 300" in a vacuum, it yields a black compound, ASK,.When the arsenic is in excess, the product is a n orange solid, K2As4,NH3, which, when heated a little below 300°, yields a cinnabar-red compound, K,As,. C. H. B. By HENRI GAUTIER (Compt. rend., 1899, 129, 595--598).-Boron sulphide, B,S,, prepared by the action of dry hydrogen sulphide on dry amorphous boron at a red heat, was decomposed by sodium hydroxide solution, oxidised with bromine water, and precipitated with barium chloride. Four experi- ments give B= 11.041 with a probable error +,0.017. Boron carbide prepared in the electric furnace by the action of c 0, (7a) is very slightly soluble in water.E. W. W. The Atomic Weight of Boron.INORGANIC CHEMISTRY. 15 sugar-charcoal on amorphous boron in presence of copper, was decom- posed by chlorine, and the residual carbon weighed as such, and also in the form of carbon dioxide. Two experiments give B = 10.997. C. H. B. Atomic Weight of Boron. By H E N R r GAUTIER (Coinpt. rend., 1899, 129, 678-681. Compare Abrahall, Trans., 1892, 61, 650; Ramsay and Aston, Trans., 1893, 63, 207 ; Rimbacb, Abstr., 1893, ii, 207).-Boron tribromide, obtained by passing pure bromine over pure boron heated to a dull red heat, was fractionally distilled alone, and then finally over reduced silver, when it distilled quite coiistantly a t 90.5O. A weighed quantity of the boron haloid was decomposed by water, and the halogen determined as silver haloid.The mean value obtained for the atomic weight of boron by nine experiments on the decomposition of the bromide was 11.021, with a probable error of +, 0.006, and from five experiments on the chloride, 11.011, with a probable error of 0.008, The mean of these values is 11.016, a number which the author adopts for the atomic weight of boron in preference to the vdues obtained by analysis of boron sulphide and boron carbide (com- pare preceding abstract). Graphite. By LUDWIG STAUDENMAIER (Be?.., 1899, 32, 2834-2834. Compare Abstr., 1898, ii, 96 ; 1899, ii, 481).-When graphitic acid is heated with sulphuric acid and water a t lSOo, it gives an insoluble pyrographitic acid, which retains the form and appearance of the original graphite.When oxidised with nitric acid, this gives a soluble product containing mellitic acid, a red substance soluble in ether, a dark brownish-red substance soluble in alcohol, and a black residue insoluble in alcohol or ether. When treated with a mixture of fuming nitric acid and potassium chlorate, unlike ordinary pyrographitic acid, i t gives a transparent, yellowish-green pseudo- gmphitic acid ; this resembles graphitic acid but is more soluble, less stable, and more readily oxidised ; it gives a black reduction product with stannous chloride. Treatment of Lepidolite. By JULIUS FORMANEK (Chem. Centy., lS99, ii, 11-12; from 0estes.r. Chern. Zeit., 2, 309--312).-1n order to obtain cEsium and rubidium compounds from lepidolite, the finely powdered mineral is decomposed with concentrated sulphuric acid in an iron pan, and the solution evaporated in a lead vessel.The residue is taken up with water, and the potassium, c;lesium, and rubidium alums which separate are crystallised from water several times, and then treated with barium hydroxide. The excess of the baryta and most of the alumina are removed by means of carbon dioxide, and the solu- tion, after neutralising with oxalic acid, is evaporated until crystals begin t o form. Most of the potassium salt separates on cooling, and the rest is removed from the residue by fractional crystallisation, the czesium and rubidium salts being afterwards separated by the same means. E. W. ?V. Commercial Calcium Carbide. By HENRI MOISSAN (Bull.Soc. Chim., 1899, [ iii], 21, 865-871).-Theoretically, 1 gram Pure boron trichloride was obtained in a similar manner. H. R. LE S. T. M. L.16 ALlSTJtACTS OF C BEMICAL YAPEltS. of calcium carbide should yield 349 C.C. of acetylene, the amounts obtained from seven commercial samples varied from 292.8-318.7 C.C. ; three inferior specimens, which were grey and porous, instead of having a fused, crystalline structure, gave only 228.6, 250.4, and 260.3 C.C. respectively. The gas sometimes contains notable quantities of ammonia, and several specimens of carbide yielded a little hydrogen phosphide. To facilitate the study of the insoluble residue, the carbide was decomposed with a n aqueous solution of sugar, whereby the lime produced is kept in solution. The residue consists principally of the silicides of carbon, calcium, and iron, sometimes mixed with a little graphite and calcium sulphide ; dilute (10 per cent,.) hydrochloric acid extracts iron, lime, and small quantities of aluminium and phosphorus ; the concentrated acid dissolves further quantities of lime and silica, whilst carbon silicide and graphite remain unat tacked.The various forms in which these impurities exist were recognised by microscopical examination. The silicon occurs chiefly a3 carbon silicide, but small quantities of calcium silicide, silica, and a compound containing iron, carbon, and silicon are also formed. Silicon hydride, from the decom- position of calcium silicide, is often evolved in the treatment with con- centrated hydrochloric acid. The total sulphur in three samples of carbide was found t o be 0.37, 0.43, and 0.74 per ccnt.; i t exists as calcium sulphide and aluminium sulphide. Hydrogen sulphide is not liberated when impure calcium carbide is decomposed by water, since it is retained by the calcium hydroxide formed in the reaction ; traces of a volatile organic compound containing sulphur seem, however, to be formed in some cases, since the gas, after being mashed with potash and lead acetate solution, yields a small quantity of sulphuric acid when burnt. Iron is found as silicide and carbosilicide. Phosphorus occurs chiefly as calcium phosphide, but is also found combined with iron and silicon. Carbon is sometimes found as graphite retaining calcium and silicon ; none mas detected in the form of diamond.N. L. Action of Magnesium on Saline Solutions. By DONATO TOM- MASI (Bull. SOC. Chirn., 1899, [iii], 21, S85-887).-When magnesinm wire is placed in a solution of potassium chloride, t h e metal is con- verted into hydroxide, with liberation of hydrogen ; the potassium chloride simply favours the oxidation of the magnesium, and is not itself decomposed. With ammonium chloride solution, a vigorous reaction occurs, hydrogen is evolved, and ammonium magnesium chloride formed. A 30 per cent. solution of calcium chloride attacks mag- nesium, slightlyin the cold and more readily on heating, with the formation of magnesium hydroxide and hydrogen, but a saturated solution is without action unless the metal is finely divided and the liquid heated. A 30 per cent.solution of magnesium chloride has very little action on magnesium, but the boiling solution is easily decom- posed by a magnesium-platinum couple, with t h e formation of mag- nesium hydroxide and oxychloride. The action on magnesium of solutions of various other salts was examined, and the products of the reactions, in addition t o hydrogen, were found to be as follows. Sodium and lithium chlorides : magnesium hydroxide, Barium and strontiumINORQANIC CHEMISTRY. 17 chlorides : scarcely any action on magnesium, Cupric chloride ! cuprous chloride, copper oxychloride, and magnesium chloride. Cadmium chloride : cadmium oxychloride, metallic cadmium, and magnesium chloride. Cobalt chloride : cobalt hydroxide and magnesium chloride. Lead chloride : lead oxychloride, metallic lead, and magnesium chloride, Mercuric chloride : mercurous chloride, mercury oxide, and magnesium chloride.Ferric chloride : ferric hydroxide, ferric oxychloride, and magnesium chloride ; ferric chloride is not reduced by magnesium. Chromic chloride : chromium hydroxide and magnesium chloride, Platinic chloride : metallic platinum and magnesium hydroxide. Gold chloride : metallic gold, magnesium hydroxide, and magnesium chloride. Copper sulphate : metallic copper, cuprous hydroxide, basic copper sulphate, and magnesium sulphate ; at 0", cuprous hydroxide only is obtained. Zinc sulphate : metallic zim, zinc hydroxide, basic zinc sulphate, and magnesium sulphate. Ferrous ' sulphate : ferrous hydroxide, and magnesium sulphate. Mangmese sulphate : manganese hydroxide and magnesium sulphate.N. L. Anhydrous Magnesium Carbonate. By RODOLPHE ENGEL (Conzpt. rend., 1S99, 129, 59S--600).-When magnesium ammonium carbonate is gradually heated in a current of dry air a t a temperature not exceeding 1 30-140°, it yields anhydrous magnesium carbonate which differs from the natural mineral, and from the carbonate pre- pared by Senarmont, but resembles that obtained by the action of heat on magnesium potassiumsesquicarbonate (Abstr., 1886,821). It retains the crystalline form of the double carbonate, is very hygroscopic, sets like plaster when mixed with water, and absorbs almost instantly about 100 times its own volume of ammonia gas. Attempts to prepare an ammonium magnesium seaquicarbonate gave negative results, and the compound, MgCO,,NH,HCO, + 4H,O, ascribed to Pavre, was never described by him.C. H. B, Dicarbonyl Cuprous Chloride. By WILLIAM APP JONES (Arne?*. Chenz. J., 1899, 22, 287-31 l).-Carbonic oxide is not absorbed by dry cuprous chloride, but in presence of sufficient hydrochloric acid to form a thin paste, an amount of gas is absorbed at 0' which corre- sponds with the formation of dicur6onyZ cup?*ous chZo?*ide, Cu2C1,,2C0, + 4H,O ; on adding ice-water, white plates having this composition separate which rapidly decompose in the air. The compound prepared by passing carbonic oxide into cuprous chloride in presence of water or dilute hydrochloric acid (1HC1: 2H,O) is not decomposed by diminishing the pressure until the latter is reduced to 135-125 mm., when complete decomposition occurs ; in presence of concentrated hydrochloric acid, decomposition begins at 410 mm.pressure and gradually increases as the pressure is reduced to 160 mm., when it takes place more rapidly and is colnplete a t 130 mm. pressure. Curves are given showing these results. On passing carbonic oxide into a solution of cuprous chloride in pyridine at Oo, the compourtd 2Cu2C1,,3C0 appears to be formed, but could not be isolated ; on raising the temperature, decomposition occurs VOL. LXXVIII. ii. 218 ABSTRACTS OF CHEMICAL PAPERS. regularly, until a t looo it is nearly complete. The resnlts are plotted in the form of a curve. When oxygen is passed into dicarbonyl cuprous chloride a t Oo, the latter is decomposed, giving rise to a small quantity of carbon dioxide ; the formation of this cannot be due to increased activity of the oxygen brought about by the cuprous chloride undergoing oxidation, and thus splitting the molecular into atomic oxygen, for when a mixture of oxygen and carbonic oxide is passed through ferrous sulphate or chloride these compounds are oxidised, but no cgrbon dioxide is formed.I n discussing the decomposition of dicarbonyl cuprous chloride by reducing the pressure or raising the temperature, it is pointed out that this takes place similarly to that of compounds the ‘‘ atomic” nature of which cannot be doubted; its properties are compared with those of other inorganic compounds containing carbonic oxide, and the conclusion is reached that i t must be regarded as a true chemical compound.The behaviour of dicnrbonyl cuprous chloride with oxygen, chlorine, hydrogen, and nitrogen shows that, a t the moment of its liberation from the compound, carbonic oxide possesses no increased chemical activity. W, A. D. Spectroscopic Analysis of Neodymium and of Praseo- dymium. By WILHELM MUTHMANN and L. STUTZEL (Bey., 1899, 32, 2653-2677, Compare Abstr., 1898, ii, 518 ; von Scheele, Abstr., 1898, ii, 519 ; 1899, ii, 291 ; Bettendorf, Ahstr., 1890, 851).-Accord- ing to Crookes’ ‘‘ one band one element ” theory, praseodymium must contain 37 different elements, whereas according to von Scheele the sub- stance is homogeneous. The authors themselves consider that the composite nature of praseodymium is probable, although not established, and that even if true the separation of the constituents, on account of the similarity in their chemical behaviour, presents an almost im- possible task.Specimens of praseodymium from Bastnas cerite and from Australian and Brazilian monazite have been compared with Shnpleigh’s preparations and no differences have been detected, The composite nature of neodymium is even still more in doubt, as, so far, pure neodymium has not been obtained ; the product obtained by the sulphate method of Reparation contains 2 per cent. of praseodymium. A new separation by the aid of the chromaies is quicker, and it is hoped will give better results. Neodymium has the most complex absorption spectrum of all the earths, the number of bands in neutral chloride solution being 24. All these are riot necessarily visible a t the same time; in certain con- centrated solutions, the six yellow bands become one, similarly for the green ; on dilution, several groups disappear completely, particularly the narrow bands at the red end of the spectrum.Numerous instancesof change in the configuration of bands of the rare earths are known, but the causes have not been determined. A notable instance is found in the case of praseodymium; ordinary didymium gives the blue line X 469 of praseodymium as a well-defined, not over broad, absorption band ; when the earth is fractionated by Welsbach’aINORGANIC CHEMISTRY. 19 method, the line loses in intensity, broadens and moves towards the violet end of the spectrum ; on further fractionation, the band becomes smaller and more intense, but even in the purest specimens is never so sharp as in the crude didymium.Bunsen (Annalen, 1866, 128, 190) pointed out that the bands in the spectra of didyminm salts vary considerably, according to the nature of the acid constituent. Further variations have been found in the case of both praseodymium and neodymium. Neodymium nitrate and chloride give different absorption spectra, especially in the green part of the spectrum ; the nitrate in dilute nitric acid gives a single pale band about A = 522, whereas in the chloride there are two bands, 525.5 and 521.5, and a pale band 520.5; the yellow bands in the chloride are sharper than in the nitrate. Still greater differences are observed when the spectra of salts derived from carboxylic acids are examined. A solution of carbonate, probably containing the metal as the hydrogen carbonate, has not the characteristic dark red colour of the nitrate, but a fairly intense blue, due to the fact that the yellow absorption band has increased consider- ably in intensity, whilst the violet band X 432-424 has completely disappeared ; the green bands are more pronounced than in the nitrate, and all bands are some 7.5 h nearer the red end of the spectrum, and, i n addition, a new orange band, X = 600.5, has made its appearance.Somewhat similar differences have been observed in the case of praseodymium, not merely does the relative luminosity of the bands vary with different salts, but also the order in which the lines dis- appear on diluting the solutions. It is obvious that the absorption spectrum is not simply de- pendent on the molecular weight of the anbydrous salt, as Bunsen suggested.The authors have employed a spectroscopic method for the estimation of praseodymium and neodymium, although, according t o Schottlander (Bey., 1892, 25, 569), this method is inapplicable. The observations were ma.de with a special Kriiss apparatus, and in calculating the results, Vierordt’s equation, A = c/ - log J, was em- ployed, where A = constant, c = concentration, and J= intensity of light after passing through a 10 mm. layer of the solution when the original intcnsit,y of the light = 1. In the following minerals-orthite from Miask, cerite from Riddar- hyttan, and orthite (allsnite) from Llano Go., Texas, it was found that the percentage of neodymium was practically twice that of praseo- d y miu m.J. J. S. Radio-active Barium Salts and Polonium. By FRITZ GIESEL (Ann. Phyls. Chem., 1899, [ii], 69, 91--94).-The author has independently obtained from uranium ores other than pitchblende a substance, consisting chiefly of barium sulphate, which emits Becquerel rays. The substance is similar t o that obtained by P. and S. Curie, according t o whom it contains a n active element radium, Freshly crystallised barium salts containing radium are only slightly active, but in the course of a few days or weeks the activity increases to a maximum ; the portions which crystallise first are more active than 2-220 ABSTRACTS OF CHEMICAL PAPERS. those subsequently obtained from the mother liquor. The chloride, bromide, and iodide phosphoresce without any previous illumination, especially when anhydrous; in moist air, they gradually lose this property, but renewed heating restores it to them.The stronger the phosphorescence of any particular specimen, the feebler is the emission of Becquerel rays. Barium platinocyanide, prepared from active barium chloride and potassium platinocyanide, shows strong, spontaneous phosphorescence, which grows less intense as time goes on, the green colour of the double salt changing to yellow and finally t o brown. By dissolving and crystallising the brown product, the green salt can again be obtained, The active constituent of the barium salts has not been isolated, and it seems impossible to effect a separation by fractional cry stallisation alone. Strongly active products containing polonium are also being investigated. Hydrogen sulphide produces in them a precipitate which surpasses in activity the best barium salt preparation. The chloride prepared from this sulphur compound is equally effective, as is also the metal deposited from the chloride solution by metallic zinc or an electric current. The penetrating power of polonium raysis much less than that of radium rays, and consequently the shadow produced by a hand or metallic object is much sharper and deeper with the former than with the latter. J. C. P. A New Radio-active Substance. By A. DEBIEERNE (Compt. rend., 1899, 129, 593--595).-The constituents of pitchblende which are not precipitated by hydrogen sulphide from an acid solution, but are precipitated by ammonia or ammonium sulphide, include a small quantity of a substance which emits radiations capable of acting on a photographic plate, making barium platinocyanide phosphorescent and accelerating the discharge of electrified bodies. Apart from its radio-activity, which seems to be about 100,000 times tls great as t h a t of uranium, it resembles titanium in general properties, It differs from radium in not being luminescent. C. H. B.