change of pressure being proportional to the pressure for any given voltage, and the velocity coefficient increases directly as the voltage. A direct measure of the electron current outside the cathode-ray tube was made by collecting the electrons passing through the window in a Faraday cylinder placed close to the window and measuring the current with a micro
ammeter. It is found by calculation that the number of molecules of acetylene reacting is of the same order as the number of ions formed by each electron passing through the window, and the observed facts are explained on the hypothesis that the reaction takes place between molecules of which some have been ionised by the cathode rays. L. L. Bi r c h m s h a w.
Salt-like compounds of sodium and their change into intermetallic phases. E. Zi n t l
(Naturwiss., 1929, 17, 782— 783).— Compounds of sodium with a number of metallic and non-metallic elements were prepared and their decomposition was investigated. For following the changes quantit
atively an electrometric titration method was used.
All elements of the long periods of the periodic system occupying places 1— 4 away from a rare gas, and forming liquid hydrides, will combine in liquid ammonia with sodium to give polysulphide-like compounds which have the characteristics of salts, and to which the name “ polyanionic ” salts is given.
By treating sodium with lead iodide compounds of the type Na4Pb7 and Na4Pb9 can be obtained. Their constitution is probably of the form N a/[Pb4~(Pb)8].
These polyanionic salts are soluble in ammonia, the solutions being deeply coloured and capable of being electrolysed, tho sodium going to the cathode, and other metals to the anode. Elements which occupy positions in tho periodic table more than four places away from a rare gas will not form these polyanionic salts, but form instead typical intermetallic phases of another structural type, insoluble in liquid ammonia.
Polyanionic salts may also be obtained by extracting alloys of the elements concerned with liquid ammonia.
The solution contains negatively-charged sub-microns.
The salts crystallise from ammonia in the form [Na(NH3)j,]i[Xn- ( X ) z]. X -R ay experiments show that Na4Pb7,yNH3 and Na4Pb9,yNH3 cannot exist in the poly anionic state in the ammonia-free form.
The latter changes into an intermetallic phase with a cubic structure and four atoms in the elementary cell, of which the homogeneity range extends from 2 8% to 35% of sodium; it contains no chemical com
pound. The substance Na4Pb7,yNH3 decomposes into two intermetallic phases. The introduction of a little sodium into the lead lattice causes a con
traction, whereas an expansion would bo expected;
the contraction is due to the distorted symmetry of the charge distribution between neighbouring
particles. A. J. Me e.
1250 B R IT IS H CHEM ICAL A B STR A C T S.— A .
Double carbonates of alkalis and alkaline earths. W . E i t e l and W . S k a lik s (Z. anorg.
Chem., 1929, 183, 263— 286).— B y heating together the simple- carbonates in a bom b, in which it was possible to obtain a pressure o f 1200— 1300 atm. at 800— 900°, the following double carbonates hare been prepared : Na«,Ca(C03) „ m. p. 812°; K 2Ca(C03)2, m. p. 813°; Na2M g(C 03)2, m. p. 677° ( p = 1240 k g ./
cm .3) ; K 2M g(C 03)2 ; N aL iC 03, m. p. 514°; K L iC 0 3, m. p. 515°. W hen calcium and lithium carbonates are heated together in equimolecular proportions an eutectic m ixture only is obtained. The magnesium salts h ad -n ot previously been produced in the dry way, because o f the high dissociation pressure of magnesium carbonate. The sodium lithium carbonate had not been prepared before. A ll the com pounds are optically negative and belong to the hexagonal or trigonal crystal system. The potassium magnesium carbonate can be super-cooled and obtained in the vitreous form at the ordinary temperature, suggesting a relationship between carbon and silicon such as m ight be expected from , their neighbouring positions in the periodic table. The refractive indices and specific volumes o f the com pounds are not additive.
The following heats of formation from the simple carbonates are given : N aLiC 03, - f O-54 kg.-cal. per m o l.; Na2Ca(C03)2, —2-07 kg.-cal. per m o l.;
K X 'a (C 0 3)2, + 0 -7 2 kg.-cal. per mol. The lattice structure o f the sodium compounds and of the potassium calcium carbonate has been examined.
Substitution o f potassium for sodium in the calcium com pound has a marked m orphotropic effect in the direction to be expected from the respective ionic
dimensions. M. S. Bu r r.
, Preparation of pure cupric sulphide. K . Fi s c h b e c k and O. Do r n e r (Z. anorg. Chem., 1929, 1 8 2 , 228— 234).— Pure cupric sulphide has been prepared b y . heating under pressure finely-divided copper, obtained by the reduction of copper oxalate with hydrogen at 220— 260°, and sulphur precipitated from solution in carbon disulphide b y means of light petroleum. The copper was covered with carbon disulphide, and then carbon disulphide containing sulphur in excess o f that required to form the lower sulphide was added gradually with constant stirring.
A fter transference to the bom b-tube, double the quantity o f sulphur required to form the higher sulphide was added, and the tube was filled with carbon disulphide and heated, w ith 'rota tion , in a current o f steam for 4 hrs. The excess o f carbon disulphide was removed and the product dried at 90—
100° at a pressure o f 0-1— 1 nun. The dark blue cupric sulphide is soluble in potassium cyanide and is a good electrical conductor. The form ation of a higher sulphide b y this m ethod is improbable.
L . S . Th e o b a l d. N ew antipyrine co-ordination compounds of m etal perchlorates. E . Wi l k e- Do r e u r t and O . Sc h l i e p h a k e (Z. anorg. Chem., 1929, 183, 301—
3 1 0 ; cf. A ., 1928, 494).— The general m ethod of preparation o f the antipyrine compounds o f the metal perchlorates is to m ix together, at the ordinary temperature, the corresponding m etal salt solution, chloride, nitrate, or sulphate, and solutions of
antipyrine and pure ammonium perchlorate. In th is : w ay hexa-antipyrine s perchlorates of magnesium, calcium, strontium, - -zinc;- cadmium', lead,: manganese, iron, cobalt, and nickel, of the general formula [M (C0CiSH 12N 2)6](G104)2, have been obtained. The barium salt can be obtained only b y the use of barium perchlorate itself. Hecca-antipyrine perchlorates of aluminium, iron, and chromium, general formula [MtCOC^H^aN^ujiClCh).,, and of quadrivalent thorium, as well as penta-anlipyrine perchlorates of- copper and urnnijl, and silver tri-antipyrinc perchlorate,, have also been prepared. Solubilities', m. p., and densities are recorded in all cases, as well as the densities of the com pounds previously obtained (loc. cit.). In every case :the thermal stability of the perchlorate :is increased b y the introduction of: antipyrine into the
molecule. : , M. S. Bu r r.:
Displacement of cadm ium from its solutions by alum inium. P . G. Po p o v (Ukraine Chem. J., 1929, 4, 281— 284).— Aluminium displaces cadmium from solutions of cadmium sulphate or nitrate also in the. absence of cobalt or chromium nitrate, contrary to the observations of Goldschmidt (A., 1906, ii, 581).
The solution requires j however, prolonged heating and the aluminium should preferably be in the form o f a powder or dust. The reaction also, proceeds with aluminium powder in presence of sodium or potassium
hydroxide. A, Fr e i m a n.
Action of gaseous am m onia on mercuric brom ide and mercuric chloride. M. Fr a n c o i s
(Bull. Soc. chim., 1929, [iv], 45, 616— 621).— Contrary to statements b y earlier experimenters, mercuric chloride and mercuric bromide combine with dry gaseous ammonia to give the com pounds HgCl2,2NH3 arid HgBr2,2NH3; respectively, and therefore behave similarly to mercuric iodide. The above tw o com
pounds appear to be additive compounds. They are fairly stable and do not lose ammonia appreciably at the ordinary temperature. O. J. Wa l k e r.
Boron. I . Reaction of boron trifLuoride with am m onia and alkylamines. C. A . Kr a u s and E. H. Br o w (J. Amer. Chem. Soc., 1929, 5 1 , 2690—
2696).— Monoamminoboron tiifluoride, B F 3jNH3, is very conveniently prepared by saturating an ethereal solution of boron trifluoride with ammonia in the cold.
I t is a true com pound and forms a white solid which fuses and sublimes in a vacuum at about 180°
and is soluble in liquid ammonia, ethylamine, di
ethylamide, and triethylamine containing excess of ammonia-, ammonolysis occurring in the first two liquids. Alkyl derivatives of the monoammino-com- pound m ay be prepared as white solids b y treating a cooled ether solution of boron trifluoride with the corresponding alkylamine. Trieihylamminobbron tri- fluoride, N E t3,BF3, m. p. 29-5°, becomes brown on keeping, is soluble in benzene and ether but insoluble in water. Diethylamminoboron trifluoride, N H E t2,BF3, m- p- 150— 160°, is insoluble in benzene and ether b u t soluble in water. Ethylamminoboron trifluoride, N H 2E t,B F3, m. p. 89°, decomposes in air above its m . p., and is soluble in benzene and slightly soluble in ether. Other methods of preparing these compounds
are given. S. K . Tw e e d y.
g e n e r a l, p h y s i c a l, a n d i n o r g a n i c c h e m i s t r y. 1251
Ultramarine. J. Ho f f m a n n (Z. anorg. Chem., 1929, 1 8 3 , 37— 76).— Methods of formation of ultra- marine, and the reasons for the production of the colour and its variations; have been studied. The blue colour of boron ultramarines is primarily-duo to the presence o f sulphur and is independent of the presence of water, blit, physical factors are at least as important as chemical.. T he colour is partly dependent oh the nature o f the alkali metal, and its production depends on à change of the borate beginning at the stage R 2B 6O10—- > R 2B 10O 16. Replacement by boron trisulphide o f part of the boron trioxide which is in excess of that required for the presence of alkali tetraborate yields orange-red to brown substances, the blue colour appearing only when a sulphide of the alkali metal :is present ; thiosulphatc is not necessarily present in the blue substances, but may be formed as an intermediate substance in certain types o f -ultra- marine. The chemical and physical action of thé sulphur is discussed. Since ultramarine is formed much more slowly when sulphur is heated with anhydrous "borax than with alkali sulphide or poly
sulphide, the latter is probably formed as an inter
mediate; Conditions necessary for the formation of ultramarine are discussed, with particular reference to the part played b y auxiliary valencies' and the structure of the compounds formed. All thé blue polysulphide ultramarines contain a structure analogous to diborates bridged b y alkali monosulphide to the boric acid part of a higher borate which may range from tri- to octa-horate. The old idea that ultramarine synthesis depends on solubility pheno
mena is disproved. Ultramarines may be classified under five types : inorganic, including the ordinary ultramarines; organic, including a variety of organic dyes containing sulphur ; inorganic-organic, including certain hydrocarbon ultramarines and compounds such as ethyl • and butyl alumina ultramarine;
temporary inorganic, such as phosphate and certain borate varieties; temporary organic, such as cyano- ultramarine. T h e formation in nature of the mineral ultramarineë is discussed. H. F. Gi l l b e.
Behaviour of alkali fluoborates in tungsten- filament lam ps. J. H . Be Bo e r (Rec. trav. chim., 1929, 48, 979— 983).— Alkali fluoborates introduced into electric lamp bulbs can react with tungsten spattered from the filament during burning. In the case of the potassium salt a tungsten compound with à. W : F atom ic ratio of 1 : 3 is formed. The intensity of blackening o f the bulb wall with the introduction of other alkali fluoborates shows that the amount of reaction is in the decreasing order cæsium, rubidium, potassium. This is ; confirmed by the amount of tungsten extracted in these cases b y sodium hydroxide and b y hydrofluoric acid. R eaction is negligible with sodium fluoborate owing to the decomposition of this salt at temperatures above 320° during the evacuation of thé lamp. F. G. Tr y h o r n.
Attack of aluminium bÿ ammoniacal solutions.
J. Ca l v e t (Compt. rend., 1929, 1 8 9 , 485— 186).—
The action o f ammonia solutions of different con
centrations on aluminium o f different degrees of purity has been investigated. V ery pure aluminium (99-96%) is attacked just as readily as less pure
specimens; Dilute solutions attack- the m etal to a greater extent than concentrated ones, -f ,
C. W . Gi b b y. Complex scandium oxalato-compounds. J.
St e r b a- Bo h mand J. Sk r a m o v s k y (Casopis Ceskoslov.
Lek., 1928, 8, 211— 215; Chem. Zentr., 1929, i, 2399).— S6andium oxalate crystallises + 6 H 20 . The compounds HSc(C20 4)2, H 4Sc2(C20 4)B, and H 3Sc(C20 4)3 have been prepared, and their relative stability has been observed. A , A. El d r i d g e.
Oxidisability of silicon as a function of its state of division. A . Sa n f o u r c h e (Compt. rend,, 1929, 1 8 9 , 533— 535).— A reply to Bedel’s criticism (of. this v o l.,: 997, 1030). The author’s experiments on pyrophoric silicon m ay be reproduced if the silicon is in a sufficiently fine state of division, e.g., if it is obtained from the aluminium-silicon alloy “ alpax.”
The presence of impurities exerts only a secondary
influence. J. Gr a n t.
Hydrates and hydrogels. X II. M ono- and di-silicic acids. R . Wi l l s t a t t e r, H . Kr a u t, and K . Lo b i n g e r (Ber., 1929, 6 2 , [B], 2027—203 4; cf. A . , 1926, 36; this vol., 39).— B y shortening the time of action and exercising greater precautions with regard to the optimal acidity (0-002N— 0-001N) of the solution, it has been found possible to prepare monosilicic acid ( 4 7 = 6 2 ; calc, for S i0 2, 4 7 = 6 0 ) in greater purity. The stability o f the acid and the subsequent members o f the series depends greatly on the acidity of the solution. The mono- and di-acids
"are m ost stable in feebly acid solution and condense most rapidly in nearly neutral or incipiently alkaline solution. The variability of condensation with acidity causes silicic acids of mean mol. wt. 200— 300 to vary irregularly in their properties, as shown b y their behaviour towards egg-albumin. The gradual condensation of monosilicic acid in solutions of differing p sl is illustrated b y a series of tim e-m ol. wt.
graphs, the most remarkable feature o f which is the indication of the existence of a structural change in disilicic acid during the process without alteration of mol. wt. The condensation of silicic acid is delayed b y addition of small amounts of ethylene glycol, glycerol, and other polyhydric alcohols, whereas ethyl alcohol is inactive.
The volatility o f silicic acid during distillation of its solutions has been re-investigated, using m ono
silicic acid and the apparatus of Kraut, Lobinger, and Pollitzer (this vol., 1261). I t has not been found possible to increase the mere trace of acid volatilised or to conduct the distillation without such volatilis
ation. Under the same conditions, distillation o f a 0-4% solution of boric acid from 400 to 70 c.c. at 11°
gave no trace of boric acid in the distillate.
H . Wr e n. Action of the alkali carbonates on lead b ro m ide, iodide, and nitrate in aq[ueous solutions.
(Mm e.) N. De m a s s i e u x (Compt. rend., 1929, 1 8 9 , 428— 430).— These reactions have been follow ed by chemical analysis o f the precipitates and b y conduc
tivity measurements. The bromide and iodide behave analogously to lead chloride (this vol., 1154), whilst the nitrate gives only lead carbonate.
J. Gr a n t.
1252 B R IT IS H CH EM IC AL A B STR A C T S. A .
Action of the alkali oxalates on the halogen salts of lead in aqueous solution. (Mm e.) N.
De m a s s i e u x (Compt. rend., 1929, 1 8 9 , 535— 536).—
Further experiments (cf. preceding abstract) have shown that analogous results are obtained b y the use of the alkali oxalates. A lead chloro- or bromo- oxalate is first produced which is subsequently transformed b y an excess of reagent into lead oxalate (cf. following abstract). J. Gr a n t.
X -R a y study of som e halogen salts. Ma t h i e u
(Compt. rend., 1929, 1 8 9 , 536— 537; cf. preceding abstract).— X -R a y measurements of the dimensions o f the lattice confirm the conclusion that the salts obtained b y Demassieux contain the PbCl or PbB r ion. In general, isomeric similarity between salts is much closer when the substituted element enters a com plex group than when it plays an ionic role.
J. Gr a n t. Union of nitrogen and sulphur under the influence of electrical discharges. W . Mo l d e n- h a u e r [with A . Zi m m e r m a n n] (Ber., 1929, 62, [2?], 2390— 2392).— Under the influence of the dark electrical discharge, pure nitrogen at 10-—1 mm.
readily combines with sulphur at 80— 100°, yield
ing a dark-coloured product from which extraction with ether and carbon disulphide yields nitrogen pentasulphide and nitrogen tetrasulphide. The residue left after treatment with these solvents is heated at 90— 93°/vac. and then exhaustively treated with the same media, whereby the bluish-black nitrogen disulpliide, (NS2b , is isolated. The com pound decomposes above 100° into nitrogen penta
sulphide and tetrasulphide. It is decomposed by alkali hydroxide with evolution of ammonia and b y concentrated hydrochloric acid into ammonium chloride and sulphur. H . Wr e n.
Constitution of nitrogen sulphide, N 4S 4. A.
Me u w s e n (Ber., 1929, 6 2 , [ j? ] , 1959— 1969).—
Nitrogen sulphide in benzene is converted by stannous chloride in 96% alcohol into the com pound (HSN)4 (cf. W olbling, A ., 1908, ii, 272), decomp, about 145°
after becoming discoloured at about 100°. The product is moderately stable towards dilute acids, but com pletely hydrolysed by hot, concentrated alkali hydroxide mainly with formation of thiosulphate and com plete evolution of nitrogen as ammonia, thus showing the absence of the N -N linking. The presence of the -S H group is indicated by the yellow colour which is observed when the colourless solution o f (HSN)4 ih acetone is treated with neutral, alcoholic ethyl nitrite. Further, when the substance (HSN)4 is dissolved in boiling formaldehyde, preferably in the presence of alkali, the compound (NS-CH2,OH)4 is obtained. I t follows, therefore, that all the hydrogen atoms in (HSN)4 are united to sulphur atoms, but they cannot be present as simple sul- phvdryl, SH, groups, since this would involve the presence of N~N linkings. It is therefore probable that the sulphur is quadrivalent, and this hypothesis is confirmed by the oxidative fission with bromine which requires 4 x 4 atoms of the halogen instead of 4 x 6 atoms required b y the ~SH group. The con
stitution is therefore deduced
and the com pound is regarded as the inorganic analogue of hydrocyanic acid. Treatment of nitrogen sulphide with bromine shows that the four sulphur atoms exert twelve valencies towards nitrogen without indicating the number of valencies of the sulphur atoms to one another. H ydrolysis of nitrogen sulphide b y alkali occurs almost quantitatively according to the scheme S4N4+ 6N aO H -f-3H 20 = Na2S20 3-|-2Na2S 0 3-f4 N H 3. The reaction thus closely resembles the hydrolysis of sodium hyposulphite, strengthening the conception that nitrogen sulphide is the cyclic nitrile o f 2 mols. of hyposulphurous acid, N < ^ ~ ^ ^^>N. The constitution of the compounds P bN 2S2,NH3 and H gN 2S,NH3, described by Ruff and Geisel (A.i 1904, ii, 396), is discussed. H . Wr e n.
[Glow produced by] oxidation of phosphorus vapour. E. J. Bo w e n and A . C. Ca v e l l (J.C.S., 1929, 1920— 1926).— Intensity measurements are made b y a photographic method incorporating special precautions. W ithin the range 60— 380° and for oxygen pressures between 40 and 350 mm. the intensity of the glow does not vary appreciably. This is in agreement with the “ chain ” mechanism of oxidation (Backstrom, A ., 1927, 1151). Ozone at low concentrations has no effect on the glow, but with more than 0-14% of ozone the intensity is increased linearly with concentration. Sulphur dioxide diminishes the glow and the effect is linear with respect to sulphur d io x id e : oxygen ratio.
Chlorine is a more efficient inhibitor and the intensity o f glow decreases linearly with increase of chlorine: oxygen ratio over a large range, until the glow is less than one fifth of the maximum intensity.
Further addition o f chlorine has not the same quench
ing effect. I t is suggested that chlorine can stimulate reaction chains involving oxygen. Ether vapour has no effect on the glow intensity even when present at a partial pressure fifty times that of the dry oxygen. J. G . A. Gr i f f i t h s.
Action of gaseous hydrogen chloride on phosphorus pentoxide. (M lle.) J. M. A. Hoef-
l a k e (Rec. trav. chim., 1929, 4 8 , 973— 978).— The curve for the absorption b y phosphorus pentoxide of hydrogen chloride dried b y distillation at low temper
atures is autocatalytic in type. The rapid absorption of further hydrogen chloride confirms the formation of a catalyst in the initial stages of the absorption.
A comparison of the effects of the compounds H P 0 3, POCl3, and o f mixtures of P 0 2C1, and P20 3C14 on the rate o f absorption indicates that one or both of the latter intermediate chlorides is the catalyst.
The initial period o f delay in the reaction between hydrogen chloride and phosphorus pentoxide may be prolonged to 7 days b y previously drying the gas with phosphorus pentoxide in two stages.
F. G . Tr y h o r n. Dichloropbospboric acid. H . Me e r w e i n and K . Bo d e n d o r f (Ber., 1929, 6 2 , [B], 1952— 1953; cf.
Lange, this vol., 662, 764).— If phosphoryl chloride or phosphorus pentachloride is treated with ice-water until dissolution is just complete the product requires almost exactly two or four equivalents o f barium hydroxide for neutralisation in the presence of
G E N E R A L , P H Y SIC A L , A N D IN O R G A N IC CH EM ISTR Y . 125
»
th y m olp h th alein , th u s sh o w in g th e p r o d u c tio n o f .dich loroph osph oric a c id as a re co g n isa b le in te rm e d i
ate. A n o ily p r o d u c t c o n s is tin g m a in ly of d ic h lo r o p h osph oric aoid rem a in s a fte r tr e a tm e n t o f p h o sp h o r y l chloride w ith th e re q u isite q u a n t it y o f w a te r in e t h e r ; a c ry sta llin e a m m o n iu m o r o x o n iu m sa lt co u ld riot b e
prepared. H . Wr e n.
T a n ta lu m . V . Sr i t z i n and L. Ka s c h t a n o v
(Z. anorg. Chem., 1926, 182, 207— 227).—-Tantalum containing not more than 5 % 0 as impurity has been prepared by the reduction of potassium fluotantalate with sodium in an atmosphere of hydrogen. The action of magnesium on tantalum pentoxide yields a product containing combined magnesium and having the formula T a 0 2Mg or Ta204Mg. The behaviour of the pure metal and some of its compounds on being heated in a stream of hydrogen chloride has
(Z. anorg. Chem., 1926, 182, 207— 227).—-Tantalum containing not more than 5 % 0 as impurity has been prepared by the reduction of potassium fluotantalate with sodium in an atmosphere of hydrogen. The action of magnesium on tantalum pentoxide yields a product containing combined magnesium and having the formula T a 0 2Mg or Ta204Mg. The behaviour of the pure metal and some of its compounds on being heated in a stream of hydrogen chloride has