INORGANIC CHEi’vTI WRY. Inorganic Chemistry. ii. 39 The Hydrogen Ion Concentration of Sea-water. SVEN PALITZSCII (Biochenz. Zeitsch. 19 11 37 11 6-1 30).-By means of Sorensen’s indicator method with the use of phenolphthalein and a-naphtholphthalein the concentration of the hydrogen ions was determined in sea-water taken in different parts of the North Sea Mediterranean and Black Sea during a voyage of investigation. The variations (with the exception of certain measurements made with water from the Black Sea) lay between pri = 7.95 and p = 8.35. HANS KAPPELER (Ber. 1911 44 3496-3501).-The author has prepared the compounds usually denoted as IloOIn and I6OI3 according to the details given by their discoverers (Millon J. pr. Chem. 1845 34 321 ; Kammerer ibid. 1861 83 65 72) and finds that they are identical with basic iodine iodate 1204.Tho analysis was carried out by decomposing them with water into iodine and iodic acid and determining the ratio of the iodine t o the iodic acid. By the action of concentrated nitric acid (D=1.52) on finely powdered iodine at the ordinary temperature a voluminous yellow powder is produced which can be collected and dried on a porous plate. It rapidly decomposes on exposure to light or moisture. Analysis by decomposition with water and estimation of the free iodine and iodic acid showed it to be an iodine nitrate either I(N03)3 or IO(N0,); the decision between these two formulae could not be made. If in the oxidation of iodine or in the reduction of iodic acid water is excluded as much as possible the reaction ceases as soon as the stage I,O is reached and salts of tervalent iodine are produced; thus oxidation of iodine (1) with ozone gives I(1O3) (Fichter and Rohner Abstr.1909 ii 991) (2) with nitric acid gives iodine nitrate ; reduc- tion of iodic acid by heating with concentrated sulphuric acid gives an iodine sulphate ( 10),S04,$H20 whilst reduction with sulphur dioxide gives a similar sulphate or else a sulphite. Exposure of the iodine salts of volatile or deliquescent acids to moist air brings about slow decomposition. Hydrolysis takos place and the residual oxide I,O or the hydroxide I(OH) decomposes into hydriodic acid and iodic acid or into iodine and iodic acid 15I(OH) = 5H1+ 10HI0 + 15H,O = 31 + 9HI0 + 18H,O. This decomposition takes place so slowly with the sulphate nitrate and sulphite that the resulting iodic acid can react to some extent with part of the undecomposed salt forming insoluble basic iodine iodate I,O for example according to the equation (IO),SO + 2HI0 = 2(IO)IO3 + H2S04 ; the free iodine volatilises.With iodine acetate (Schiitzenberger Compt. rend. 1861 52 135; 1862 54 1026) and the iodine iodate I,O the decomposi- tion with water is too rapid for the formation of the basic iodine iodate to take place. S. B. S. The Iodine Oxides 16013 and IloO19 and Iodine Nitrate. T. S. P.ii. 40 ABSTRACTS OF CHEMJCAL PAPERS. Action of Hydrogen Peroxide on Iodine Compounds Con- taining Oxygen. VICTOR AUGER (Compt. ?-end. 19 1 1 153 1005-1007. Compare Abstr. 1911 ii 386 ; Tanatar Abstr. 1899 ii 414 ; PQchnrd i3id. ii 477).-The salt Na2H,T0 is very slowly decomposed by hydrogen peroxide with formation of sodium iodate and liberation of a larger amount of oxygen than corresporrds with the equation this being due t o catdytio decomposition of the peroxide.With periodic acid the results vary con*ideral~ly accm ding to conditions. I n dilute solutions the acid is rapidly and completely reduced to iodic acid only a trace of iodine being liberated whilst i n concentrated sdutions the reaction is incomplete and much iodine is formed. Cold solutions of hydrogen peroxide slon ly decompose iodic acid in solutions containing less than Oms% iodine being set free. If the concentration of acid exceeds O%% however the solution remains colourless owing to the fact that the reaction J 2 + 5H,O = 2H10 + 4H,O proceeds more rapidly than the reaction ZHIO + 5H,o = I + 6H,O + 50,.w. 0. w. Production of Solid Oxygen by the Evaporation of the Liquid. SIR JAME~ DEWAR (Proc. Roy. Soc. 1911 85 A 589-597).- By the use of charcoal as a gaseous condensing agent a t low temperatures combined with the employment of suitable vacuum vessels the change from liquid oxygen into the solid can be easily effected. Pare liquid oxygen contained in a properly isoIated vessel is subjected to the exhaust produced by a qnantity of charcoal kept at about the temperature of boiling oxygen ; the pressure is thereby lowered sufficiently to produce solidification to a transparent jelly. The pressure a t which Solidification took place was determined by connecting a McLeod gauge to the vessel containing the solid oxygen and from the results obtained and the application of the Rankine or Willard Gibbs vapour-tension equation the melting point is calciilated to be 53-55'abs.a t a vapour pressure of 1.12 mm. Actual measure- ment with a hydrogen gas thermometer gave the value 54' abs. (compare Onnes aiid Crommelin Abstr. 1911 ii 854). T. S. P. The Formation of Ozone. ANTON KAILAN (Zeitsch. Elektrochem. 1911 1'7 966-367).-In the residues obtained in the fractionation of a gas containing a very high percentage of ozone Harries found that there was a discrepancy between the iodometric and gravimetric estimations of the ozone and came to the conclusion tbat t h i s was caused by the presence of a constituent 0 (compare ibid.1'7 629). The author points that if tho molecule 0 contains two available oxygen &toms (0 = 0 + 0 -+ 0) as assumed by Harries both the iodometric and gravimetric estimations mill give the same result for a mixture of 0 and 0 so that the observations of Harries cannot be used in support of the existence of the molecule 0 ; they must be due to some as yet unknown cause. T. S. P. The SyBtem Sulphur. ANDEEAS SMITS and H. L. DE LEEUW (Proc. K. Akatl. Wetensch. Amsterdam 191 1 14 461-468. Compare Abstr. 1910 ii 400).-An account is given of experiments to deter-INORGANIC CHEMISTRY. ii. 41 mine the influence of the percentage of Sp present on (u) the transition point of monoclinic to rhombic sulphur; (6) the freezing point of sulphur with the formation of the monoclinic modi6cation.The tran- sition point may be lowered as mnch as lo' and in some cases freezing- point terriperatuies as low as 108' mere obtainetl care boing taken to avoid supercooling. The apparent separation of liquid sulphur into two ph:ises is brought about by t h e differences in temperature which arise in R column of the liquid owing to the bad conduction of heat Tn reality there is no formation of two phases the differences in colour being accounted for by the differences in temperature. T. S. P. The Electrolytic Preparation of Persulphates. A. BLUMER (Zeitsch. Elektvochern. 191 1 17 965-966).-A platinum mire anode is described which is beld together by a g1:iss frame of such shape that it forms a cage round the porous cell containing the cstholyte; the frame fits into a glass cylinder containing the anolyte.The whole apparatus is cooled by surrounding it with a jacket through which water flows continuously. The anode had a total length of 96 cm. the wire being 0.15 mm. in diameter and current densities of 9.5-4.9 amperes per sq cm. could be used,without the temperature of the anolyte rising above 28-29' (compare Levi Abstr. 1903 ii 474). Lead cathodes did not give Satisfactory results but with nickel cathodes keeping the electrolyte strongly acid by addition of sulphuric acid from time to time yields of persulphate varying from 51-81% were obtained without a diaphragm and 95-97&% yields with a diaphragm. (The particular persulphate obtained is not stated ; presumably it was ammonium persulphate.) The claims of the Ger.Pat. 205067-8 (Abstr. 1909 ii 312) with respect t o the effect of the addition of ferrocyanides on the yield of ammonium persulphate were verified yields as high as 93.5% being obtained. The addition of aluminium sulphate did not give such good results (50-62% yield); with cobalt sulphate a 60% yield wits obtained. T. s. P. The Oxidation of Nitrogen by the Electric Discharge in the Presence of Ozone. VIKTOR EHRLICR and FRANZ Russ (Monalsh. 1911 32 917-996).-By the action of the silent electric discharge on still mixtures of nitrogen and oxygen the authors have been able t o obt'ain high percentage oxidation of the nitrogen in accordance with the following table Original % (vol.) of oxygen in the blrLxiinum % (rol.) of nitric oxide in mixture .......................................'5 20-8 50 75 91.5% terms of the final volume of gas ...... 0.2 4 ' 1 13.2 18'0 21% The high percentage of nitric oxide thus obtained is not due to an equilibrium attained under the influence of the electrical discharge but to a conpling of electrical and chemical processes. Ozone in excess oxidises the nitric oxide formed to nitrogen pentoxide thus causing the further formation of nitric oxide. The formation of nitric oxide is oneii. 42 ABSTRACTS OF CHEMICAL PAPERS. of zero order and takes place so long as ozone is in excess; when all the ozone has been used up further discharge decomposes the nitrogen pentoxide with the formation of nitrogen tetroxide. This then further decomposes almost completely giving a new equilibrium which is con- ditioned by the nitric oxide equilibrium under the conditions of experi- ment and by the velocity of formittion of ozone.The available ozone increases with increasing concentrations of oxygen in the initial mixture and consequently also the maximum amount of nitric oxide formed until all the nitrogen has been oxidised. Above this limiting concentration a stationary condition is attained which is dependent on the ozone and nitrogen pentoxide. Under given electrical conditions the velocity of formation of nitric oxide is independent within wide limits of the composition of the initial mixture whereas the velocity of ozone formation increases with the oxygen content of the original mixture. However when the gaseoiis mixtures are poor in nitrogen there is some retardation in the velocity probably because of diminished conductivity.Variation in the electrical conditions for a given mixture has no effect on the course of reaction since the velocities of formation of nitric oxide and ozone are affected in the same direction ; there are however effects on the relative velocities of these reactiom so that the maximum amount of nitric oxide formed is altered. For example increase in the electrical energy or diminished gas pressure gives lower values for the maximum amount of nitric oxide produced. The increased velocity of decomposition of ozone caused by the presence of nitrogen pentoxide is independent of the concentration of the ozone within certain limits T. S. P. Thiophosphates and Thiophosphites.FRITZ EPHRAIN and REBECCA STEIN (Ber. 1911 44 3405-3413. Compare Glatzel Abstr. 1905 ii 31 8).-In aqueous solution the tetrathiophosphates are subject to a kind of hydrolysis whereby the sulphur is either com- pletely or partly replaced by oxygen. It is probable that sodium tetrathiopbosphate for example gives rise to the equilibrium 2Na,PS4 ZZ 3Na2S + P,S and that the phosphorus pentasulphide then undergoes decomposition with evolution of hydrogen sulphide whereby the equilibrium is disturbed. A consideration of this equilibrium equation shows that the tetrathiophosphates would probably exist in the presence of excess of the alkali sulphide and the authors have succeeded in preparing a number of such salts by the interaction of phosphorus pentasulphide and excess of alkali sulphide. When tetra- phosphorus trisulphide or heptasulphide is used in place of the pen tasulphide the thiophosphites are formed.Potussium tetrathiophosphate K3PS4,H,0 is prepared by heating an intimate mixture of 100 grams of crystallised potassium sulphide and 7.5 grams of phosphorus pentasulphide until the former compound melts in its water of crystallisation. Before fusion is complete 30 C.C. of water are added and after the whole mass has become liquid it is filtered through a hot water funnel. On cooling the filtrate deposits microscopic crystals of the required salt in the form of quadratic rods. The yield is only 0.4 gram the main quantity of the salt remainingINORGANIC CHEMISTRY. ii. 43 in solution from which it is impossible to obtain it free from potassium sulphide.The crystals are fairly stable in the air and can be pre- served unchanged for weeks over concentrated sulphuric acid excepb for gradual loss of water of crystallisation ; they undergo dehydration at 90’. The preparation of other thiophosphates is similar in principle to that of the potassium salt. forms thin rhombic leaflets or else thick prismg which can he dried in the air without loss of ammonia. The yield is better than that ob- tained with the potassium salt. Barium tetrathiophosphate Bn,(PS,),,aq. forms microscopic needles as also does strontium trithiophosphate Sr,H&PS,O),,aq. ; the sulphur content of the latter compound varies with small changes in the conditions in one case R compound of the composition Sr3H6(PS3.50,,.5)4 being obtained.Calcium and aluminium thiophosphates could not be prepared. The thiophosphites are prepared similarly to the thiophosphates as indicated above ; phosphorus hydrides are however evolved to some extent showing t h a t tbe reactmion is not purely one of addition of the components. Sodiuin trith,iopTtosphite Wn,PS,,aq. forms rectangular plates which are very soluble in water ; the crystals are stable in dry air and do not lose their glance. Barium trithiop?iospl&e Brig( Pd,)2,aq. forms white microscopic rhombic prisms whilst burium oxythio- phosphite Ba,(PS,,,0,,,)2,8H20 forms microscopic flat crystals. Amntoniurn tetrut~iio;uiios;lil~clts (NH,),P% T. 5. P. Arsenic Tri-iodide. ERW. RICHTER (Chem. Zentr. 19 11 82 ii 935 ; from Apotl~ Zeit.26 728-730 742-743).-The product obtained by keeping a mixture of arsenious acid (10 grams) with iodine (5 grams) is extracted with carbon disulphide and the solvent allowed t o evaporate when garnet-:to scarlet-red hexagonal crystals of arsenic tri-iodide eeparate and may be dried at 50-70°. This leaves 8-9% of residue when t.reated with water and rather more Then water containing alcohol is used. A 1% solution in water or in water containing 5-10% of alcohol is stable. ‘The product may be assayed by titration first with N/lO-potassium hydroxide and then with N/lO- iodine. Commercial preparatims should contain 97% of real iodide. T. A. H. Proper ties and Preparation of Boron. EZECHIEL WEINTRAUB (J. Ind. Bng. Chern. 1911 3 299-301).-h description of the pro- perties of boron when isolated in the pure fused condition an account OF the methods used in its preparation with a discussion on t h e agree- ment of these propert ics with its position in the periodic classification and the possibilities of its future industrial employment.Boron has a curved conchoidsl fracture is very hard scratching all known substances except diamond; the surface is very shiny black and takes a beautiful polish but as at present prepared it lacks tough- ness. The m. p. lies between 2000 and 2500° and i t is somewhat volatile exhibiting an appreciable vapour-tension at 1200O. The electrical properties are entirely unexpected when cold boron is a very poor conductor its specific resistance being about 1OI2 timesii. 44 ABSTRACTS OF CHEMICAL PAPERS.that of copper; this resistance drops as the temperature rises and between room temperature and about 400' this drop is in the approxi- mate ratio of 2 x lo6 to 1. Its bebaviour is more like that of a spark gap or arc than that of a solid conductor there existing a '' break down" voltage in the cnse of boron as in that of an air gap; this relationship between temperature and resistance has been studied up t o red heat. Two methods of preparation are employed in the first boric nnhydride is reduced hy means of magnesium and the impure boron so obtained then heated t o almost its melting point when magnesium and nitrogen volatilise in the elemental form and oxygen is eliminated as boric anhydride. The second method is based on the decomposition of boron chloride by hydrogen at a red heat and is carried out in two ways (u) in an arc discharge taking place between two boron or (water-cooled) copper electrodas iu an atmosphere of boron chloride and hydrogen ; ( b ) by deposition on a hot graphite tube heated by a current passing through it when if the conditions are properly adjusted there is practically no combination between the graphite and boron.Boron chloride is best obtained by passing chlorine over boron carbide which may be prepared in the electric furnace from boric anhydride and carbon. F. M. G. 3%. Silicon Hydride at Low Temperatures. KAROL ADWENTOWSKI and EDWARD DROZDOWSEI (Bull. Acad. Sci. Cmcow 19 11 A 330-344). -Pure silicon hydride can only be obtained after repeated fractionatiori of the gas prepared either by the method of Friedel and Ladenburg (Annulen 1867 143 123) or t h a t of Moissan and Smiles (Abstr.1902 ii 318). By the latter method (the action of conceutrated hydrochloric acid on magnesium silicide) liquid as well as gaseous fractions were obtained ; the liquid was not a pure substance but probably a mixture of several silicon hydrides. The pure gas is spontaneously inflammable in open vessels where the surface in contact with the air is large; the inflammability is not due to the presence of impurity One litre at Oo and 760 mm. weighs 1.4538 grams. The critical temperature is - 3 * 5 O the critical pressure 47.8 atmospheres and the boiling point - 11 6'/740 mm. The Melting of Carbon by means of the Joule Effect. M. LA ROSA (Ann. Physik 1911 [iv] 36 841-847.* Compare Abstr.1909 ii 399 and Ann. Physzk 1911 [iv] 34 95).-The changes observed by Watts and Mendenhall (Abstr. 191 1 ii 881) in carbon rods subjected t o a strong current are considered to be mainly due to impurities in the carbon. When rods purified by heating for some hours in a current of chlorine are employed deformation is not observed until much higher temperatures are reached. I n opposition to Watts and Mendenhall t h e author main- tains that the structuraI changes found in carbon after the passage of a strong current can only be accounted for satisfactorily on the assumption that a partial liquefaction takes place. * and Nuavo Cim. 1911 [vi] 2 ii 418-424. E. F. A. H. M. D.INORGANIC CHEMISTRY. ii. 45 The Deposition of Carbon in the Form of Diamond.WERNER VON BOLTON (Zeitsch. Elektrochem. 191 1 17 971-972)- I n 1910 the author found that hydrocarbons are decomposed by amalgams for example by sodium amalgam with deposition of carbon which is amorphous for the most part but contains some crystalline particles which are probably diamonds. The amount of these crystals obtained hitherto has been too small for analysis and the author has endeavoured to increase the quantity as follows a 14% sodium amalgam was put in a test-tube and covered with a dilute solution of sodium silicate containing very fine diamond dust in suspension. The test-tube was heated a t 100' for four weeks moist coal gas being passed continually over the surface of the amalgam by means of a tube passing through the sodium silicate solution.The diamond dust used showed a few crystalline particles under a mag- nification of 68 but after the end of the experiment t h e number of such particles had greatly increassd. These particles had the properties of diamonds disappearing completely on heating in a current of oxygen and dissolving in fused sodium carbonate; their amount (less than 1 mg.) way however too small for analysis. The results show that the crystalline particles in the diamond dust act as nuclei for the deposition of carbon in the form of diamond from the hydrocarbons in the coal gas. T. s. P. The Formation of Carbamide by HeatiDg Ammonium Carbamate. FRITZ FICHTER and BERNHARD BECILER (Bela. 191 1 44 3473-3450).-A systematic investigation of the dehydration of ammonium carbarnate to carbamide (compare Basaroff this Joum.1868 21 194 and Bourgeois Abstr. 1898 i 564). Ammonium carbarnate was prepared by the interaction of gaseoiis ammonia and carbon dioxide in a wide glass tube enclosing a water- cooled narrower tube ; the salt deposited on the latter in hard crusts. The results of earlier workers appear to have been vitiated by the presence of traces of alcohol introduced by the method of preparation involving the use of this liquid. For the experiments the substance was enclosed in a small steel bomb lined with tin ; after heating the bomb was reopened and the yield of carbamide estimated. 'l'he results of a series of tests show that under otherwise similar conditions the yield of carbamide rises rapidly from 115' to 135O and then slowly falls again. Experiments performed at 135' indicate that the yield is improved considerably by a closer packing of the substance ; under these optimum conditions 40% of the carbarnate is converted into carbamide in the course of four days whilst the final yield is approached even a t the end of t wenty-f our hours.The investmigation indicates that it is the unvaporised ammonium car bamate which undergoes dehydration to carbamide the diminution in the yield above 135' being due to the larger proportion vaporised. A calculation elicits the fact that at 135O whatever the amount of carbarnate taken an approximately constant percentage (42%) of the unvaporised portion is converted into carbamide. The change is an syuilibrium one and a final mixture of the same compositionii.46 ABSTRACTS OF CHEMICAL PAPERS. is reached whether for example one starts with ammonium carbamate or with a corresponding amount of an equimolecular mixture of carbamide and water. Ammonium rarbouate and hydrogen carbonate likewise give an incomplete change into carbamide the results indicating that of the compounds of ammonia and carbon dioxide only ammonium carbnmabe is stable under pressure at 135". D. F. T. The Gaseous Condensable Compound ExploRive at Low Temperatures Produced from Carbon Disulphide Vapour by the Action of the Silent Electric Discharge. 11. SIR JANES DEWAR and HUMPHREY 0. JONES (Proc. IZog. Xoc. 1911 85 A 574-588).-The explosive properties of the product previously described (now called the " ozoniser product ") (Abutr.1910 ii 408) are not due to the presence of impurities in the carbon disulphide used. The ozoniser product is formed whether the ozonisor is cooled to -SOo or heated t o 220" but the quautity obtained is less than under the ordinary conditions of temperature. No difference in its hehaviour has been traced to variations in the amount of carbon disulphide condensed with it. The interposition of a cotton-wool filter between the ozoniser and the condenser for collecting the ozonised product diminishes the intensity of the phenomena observed in the condenser. to a very con- siderable extent and at the same time increases the rate a t which brown solid is deposited in the ozoniser and leading tubes. At - 210° the ozoniser product collected is perfectly white a t first but even at this temperature it becomes brown after about fifteen minutes.Examinn- tion of the gas after it left the ozoniser by means of a beam of light did not reveal the presence of solid particles indicating that the transformation into the brown solid takes place only in contact with solid surfaces. Tho spectrum of the flash produced during the trans- formation showed the presence of sulphur bands (wave-lengths 384-392) hydrocarbon bands especially that of wave-length 436-443 the cyanogen line (3885) and various other bands. Tnere is practically no electric effect even with the brightest flashes during the transformation. By the passage of the ozoniser product through a U-tube a t - 120" and then into one a t - 1 8 5 O it was found possible t o wpuate it into two parts.A white crystalline deposit collects at - 120° which does not molt uutil the temperature has risen almost to 0'; the hubstance which gives rise to the brown solid is not condensd a t - 120° but is collected at - 186O. The ozoniser product is absorbed destroyed or caused to poIymerise by cocoanut charcoal either at - 7 8 O or ah ZSO" or at the ordinary temperature. When passed over finely shredded rubber some of the carbon disulphide is removed from the gas. The brown solid product and also the material deposited in the ozoniser after removal of free sulphur had a composition agreeing approximately with the formula CS. Tho ozoniser product is destroyed by finely-divided platinum nickel or silver ; ferric oxide (not ignited) yellow mercuric oxide,INORGANIC CHEMISTRY.ii. 47 silver oxide and barium peroxide react with it giving sulphides. Solid potacjsium hydroxide silver nitrate lead acetate and sodium peroxide have little or no action. Nitric acid and concentrated sulphuric acid bring about decomposition. Mixture of carbon disulphide vapour with hydrogen nitrogen ether methyl iodide or chloroform has no effect on the phenomena observed. The vapours of thiocarbonyl chloride ethyl thiocarbonate ethyl dithiocarbonate ethyl trithiocarbonate and thiophen do not give rise to carbon monosulphide. Comparison of t lie properties of the brown substance produced as above and the polymeric form of carbon monosulphide obtained from thiocarbonyl chloride (Abstr. 1910 ii 408) shows that they possibly differ in degree of polymerisation.The product obtained by Sidot (this Journ. 1875 28 1236) is also a polymeric form of carbon monosulphide. The specific heat between 15" and - 1 8 5 O of the polymeric form of carbon monosulphide obtained from thiocarbonyl chloride is 0.1 27 the molecular heat being 5.59 as compared with the calculated value 5.556. The heat of combustion is 178,050 cals. and the heat of formation - 10,000 cals. T. S. P. Ratios of the Rare Gases to One Another and to Nitrogen in Mine Gases. CHARLES MOUREU and ADOLPHE LEPAPE (Corrzpt. rend. 1911 153 1043-1045. Compare Abstr. 1913 ii 808 1087 1134).-The results of new analyses are given. The ratios of the rare gases to one another confirm the conclusions set forth in previous communications. There is evidence that coal exerts a selective action towards the heavier rare gases thus tending to reduce slightly below normal the ratio between argon and the other gases in mixtures occurring in coal mines.w. 0. w. The Thermal Analysis of Binary Mixtures of chlorides of Elements of the Same Valency. CARLO SANDONNINI (Atti R. Accad. Lincei 1911 [v] 20 ii 503-510).-The binary mixtures of chlorides which have been studied by means of thermal analysis are examined in the light of the rules given by Tammann for metallic alloys (Abstr. 1906 ii 346; 1907 ii 857) and by Herrmann for salts (Abstr. 1911 ii 801) and t h e results confirm the latter's conclusions. Of the univalent metals a close similarity occurs between potassium rubidium and thallous chlorides whilst lithium sodium cuproue and silver chlorides also form a group.The bivalent metals are also compared the three sub-groups being calcium strontium and barium chlorides ; cadmium and mercuric chlorides; and stannous lead and manganous chlorides. C. H. D. A Remarkable Case of Isopolymorphism with Salts of the Alkali Metals. FRANB M. JAEGER (Yroo. K. Akad. Wetemch. Amsterdam 19 11 14 35+369).-The author points out that the conclusions drawn by Tuttm from his researches on the sulphates selenates double sulphaters and double Eielenates of the alkali metals are too general since they are not supported by the results obtainedii. 48 ABSTRACTS OF CHEMICAL PAPERS. with other salts of these metals. As a further contribution to the subject the full crystallographic data for various acid trichloro- acetates are given ; these have the general formula CCI 3* CO,M,CCI,* CO,H where M represents a univalent metal and a comparison of them is given in the following table Molec.Salt. Habit. D1"m. vol. Axial ratio. Potassium Tetragonal-trapezohedric 2.005 182.04 a c = 1 0.7808 Ammonium > 7 1.775 193'80 a c = l 0.7678 Rubidium Moiioclinic-prismatic .. 2.150 191.21 a b c=1'4649 1 3.1596 Monoclinic-prismatic a:b:c=1.0434:1 :O 9706 Czsium { (pseudo-tetragonal) } 2'143 214'18{~ b ~ = 0 9 9 3 8 1 0.2538) Thallium Tetiagonal-trapezohedric 2.822 187.74 n c = l 0.7672 Thallium* Monoclinic-prismatic ... 3.923 203'16 a b :c=1*5515 1 3.3007 * Acid thallium tribromoacetatc CBr3*CO2T1 CBr;CO,H. The acid trichloroacetates of the alkali metals thus give an isopolymorphous series ; in all probability it is an isotetramorphous series with three monoclinic and one tetragon:d-trapezohedric modification.The thallium and ammonium salts probably give a continuous series of mixed crystals ; they are undoubtedly isomorphous. From mixtiires of potassium and rubidium salts the first fractions are crystals analogous t o those of the rubidium salt whilst the last fractions contain both monoclinic and tetragonal double pyramids. Mixed crystals of the pot'assium and cesium salts exhibit the form of the rubidium salt. The potassium and ammonium salts give mixed crystal& possessing solely the tetragonal-bipyramidal form. From solutions contaiuiug the rubidium and thallium salts mixed crystals of the form of the rubidium salt alone are obtained.Mixtures of caesium and tballium salt of low thallium content give mixed cryhtals having the form of the rubidium salt ; when the concentration OC the czsium salt is very great there are also found mixed crystals exhibit- ing the form of the first caesium salt. Mixed crystals of the rubidium type appear first from mixtures of the caesium and rubidium salts; with very high concentration of the caesium salt mixed crystals of the casium type are also deposited. The Capacity of Potassium Halides for Forming Solid Solutions in Relation to Temperature. MAHIO AMADOR1 and G. PAMPANINI (Atti N. Accad. Lincei 1911 [v] 20 ii 473-480)- I n the present paper the miscibility of pairs of t h e solid potassium halides at 25' is studied as the first part of a n investigation of the change of miscibility with temperature.The solubility in water is determined at 25" fifteen to twenty days being allowed for the attainment of equilibrium and curves are drawn showing the ratio between the two salts in the crystals and in the solution. Potaasium chloride and potassium bromide are completely miscible in the solid state at 25'. The curve for potassium bromide and iodide is discoutinuous indicating a gap in the miscibility from 25 to 94 mol. % KBr and the curve for potassium chloride and iodide shows a similar gap extending from 7 to 99 mol. % of chloride. T. S. P. C. H. D.INORGANIC CHEMISTRY. ii. 49 Occurrence of Potassium N i t r a t e in Western America. ROBERT STEWART (J. Arner. Chem. Xoc. 1911 33 1952-1954).- Crude potassium nitrate containing some calcium sul phate has been found in sandstone in the State of Utah. It is suggested that the existence of such deposits may have some bearing on the occurrence of excessive quantities of nitrates in some of the agricultural soils of the Western States.E. G. Experiments with Sodium Silicate. W. J. CLUNIES Ross (J. Proc. Roy. Xoc. New South TPules 1910 44 583-592).-Crystals of various salts are dropped into a solution of sodium silicate D 1.08 contained in test-tubes or cylinders. Growths resembling plant shoots are obtained from salts of silver mercury lead copper iron aluminium cobalt nickel manganese zinc cadmium calcium zirconium yttrium and cerium A ntimony uranium tin and bismuth do not yield shoots. The growths are coloured when coloured salts are used but the silicate solution surrounding them does not become coloured except in the case of manganese.The form of the growths is generally character- istic of the metal cadmium giving hair-like filaments and nickel thick fungoid forms. The growths m e probably tubular and act as semi-permeable mem branes. C. H. D. Dimorphism of Rubidium Dichromate. JOSEPH A. LEBEL (Compt. rend. 1911 153 1081).-From acid solutions of rubidium dichromate the red variety alone crystallises whilst in presence of a little alkali only the yellow modification separates. It has usually been supposed that the dimorphism is conditioned by temperature alone and that the solubility of the two forms is the same. This is negatived by the observation that a red crystal placed in a solution from which yellow ones are separating goes completely into solution.w. 0. w. Solubility of Silver Chloride in Chloride Solutions and the Existence of Complex Argentichloride Ions. GEORGE S. FORBES (J. Ame7*. Chem. Xoc. 1911 33 1937-1946).-Determinations have been made of the solubility at 25’ of silver chloride in concentrated solutions of the chlorides of sodium calcium ammonium strontium potassium and barium and also of hydrochloric acid. The temperature- coefficient of the solubility was determined in 3N-potassium and calcium chlorides a t lo 25’ and 35’. The solubility is nearly doubled by an increase of the temperature from Oo to 254 the rate of increase above and below 25’ being nearly logarithmic. The solubility is closely proportional to integral powers of the chloride concentration through- out considerable ranges which is explained by assuming the existence of the complex anions SgCl,”! AgCl,“’ and possibly AgCl,””.No evidence was obtained of the existence of the ionAgC1;. It is shown that the extent of complex formation by mercurous and silver chlorides is not proportional to the activity of the chloride ion in concentrated solutions. The data obtained in this investigation lead to the con- clusion that silver chloride should be least soluble in N/lOO-chloride solutions a fact which is of importance in analytical work. E. a. VOL. CII ii. 4ii. 50 ABSTRACTS OF CHEMICAL PAPERS. Thermal Analysis of Binary Mixtures of Calcium Chloride with Chlorides of Other Bivalent Elements.CARLO SANDONNINL ( A t t i R. Accad. Lincei 1911 [v] 20 ii 496-503).-CaIcium chloride forms a continuous series of solid solutions with strontium cadmium and manganous chlorides the freezing-point curve in each case having a minimum which occurs at 646" 545O and 583' respectively. I n the case of mixtures of calcium chloride with strontium and man- ganous chlorides a developmeut of heat a t lower temperatures is also observed over a certain range of composition indicating a decompositiou of the solid solutions. Mixtures of calcium chloride with barium chloride and with lead chloride form simple eutectiferous series the eutectic points being at 600' and 35 mol. % BaCI and at 468" and 83 mol. % PbCI respectively. C. H. D. The Plasticity of Barium Sulphate. ALBERT ATTEHBEKG (Zeitsch.angew. Chem 1911 24 2355).-Polemical. A reply to Ehrenberg's criticism (Abstr. 1911 ii 972) of the author's paper (ibid. 605). C. H. 1). The Replacement of Metals from' Aqueous Solutions of their Salts by Hydrogen at High Temperatures and Pressures. WLADIMIR IPATIEFF (Be?.. 191 1 44 3452-3459. Compare Abstr. 1911 ii 716)-In the presence of copper N-nitric acid is reduced completely to ammonia by hydrogen under 100 atmos. pressure at 150-160O; in the absence of copper no reduction takes place. In solutions of 2N-copper nitrate a t temperatures of 100-180' and initial pressures of 100 atmos. hydrogen gives rise a t first t o the production of emerald-green crystals of the compound which is identical with the mineral gerhardtite ; nitrous acid is pro- duced at the same time in quantities varying with the experimental conditions.Prolonged reaction produces black hard crystals of cupric oxide and finally a mixture of cupric oxide and copper forms the only product. With N-copper nitrate the above-mentioned complex com- pound is produced only in traces even at 140-160" ; at 100-120' crystallised cuprous oxide is.formed together with traces of metallic copper. Above 120° cupric oxide and copper are deposited the quantity of copper increasing with the time of reaction. A t 155' after forty-eight hours pure copper is the final product a result which could not be obtained with 2N-copper nitrate. A t 1 5 5 O and a n initial hydrogen pressure of 100 atmos. a 2N- solution of cupric chloride gives crystals of cuprous chloride ; under the same conditions and even at 180° a 1CT/S-solutio~i gives rise to a mixture of cuprous chloride and copper.I n a quartz tube at 186-240" and initial pressures of 100 ntmos- pheres hydrogen causes precipitation of nickel from a N/5-solution of nickel sulphate but the precipitation is not complete even after thirty hours; when a glass tube is used complete precipitation of the nickel takes place. From a 2N-solution of nickel sulphate in quartz tubes the compound NiSO,,H,O is produced whether the tube contains CU( NqJ,,3Cu(OH)pINORGANIC CHEMISTRY. ii. 51 hydrogen nitrogen or air under pressure. This compound is green and crystalline insoluble in water and does not lose water at 270"; when hydrogen is used i t is always mixed with traces of nickel. I n the presence of reduced nickel N- and N/lO-eolutions of nitric acid are reduced completely to ammonia by hydrogen under pressure.I n a glass tube a t 200° and initial pressure of 100 atmospheres hydrogen causes the precipitation of a green hydrated nickelous oxide (61*04-61.95% Ni) from a N/5-solution of nickel nitrate. I n a quartz tube metallic nickel is also deposited (at 810°) and a t 230-250O a blackish-grey precipitate of a mixture of nickel and nickelous oxide is formed. A N/5-solution of nickel acetate at 120" gives a precipitate of hydrated nickelous oxide ; nickel is deposited at 168". From a X / 5 - solution of nickel chloride a t 230-240° small quantities of nickel are deposited but the reaction is far frcm complete. T. S. P. Preparation of Colloidal Copper.J. GAUBE DU GERS and W. KOPACZEWSKI (Zeitsch. Chenz. I n d . Kolloide 191 I 9 239-240).- Twenty grams of powdered egg-albumin are treated with 20 C.C. of 15% sodium hydroxide and diluted to a litre. The liquid is then heated to its boiling point filtered from the coagulated albumin again heated to boiling after which a 1% solution of copper sulphate is added drop by drop. The liquid assumes a red colour but this changes to violet and finally to reddish-brown. On dialysing the reddish-brown solution a colloidal solution of copper albuminate is obtained. This solution does not give the reactions for copper unless the complex albuminate is decomposed. It is almost neutral towards litmus can be concentrated on a water-bath and its refractive index at 20' is 1.319. The albuminate is reduced by hydrazine hydrate only after prolonged boiling of the solution.Constitution of Phosphor Bronzes. MARIO LEVI-MALVANO and F. S. ORoFrNo (Gcczzetta 1911 41 ii 297-314).-The study of the system Cu-P has shown the existence of a compound Cu,P (Heyn and Bauer Abstr. 1906 ii 855) whilst in the examination of tbe system Cu-Sn a compound Cu,Su was encountered (Giolitti and Tavanti Abstr. 1905 ii 946). Tho authors have therefore investigated the binary system Cu,P-Cu,Sn and the ternary system Cu-Cu,P-Cu,Sn. The compound Cu3Sn has m. p. 710" (Giolitti and Tavanti found 725") ; its transformation point lies at 630" (Giolitti and Tavanti gave 659"). The diagram of the binary system Cu,P-Cu3Sn is characterisad by the presence of an eutectic corresponding with 650" and (probably) H.M. D. 22% of CU,P. The authors discuss the diagrdm of the ternary system Cu-CUSP-Cu,So as constructed with the aid of those of its three constituent binary systems. The results of the thermal investigation of a number of alloys belonging t o this ternary system are also given and interpreted. The paper is illustrated with a number of photomicrographs. R. V. S. 4-2ii. 52 ABSTRACTS OF CHEMICAL PAPERS. Constitution of Aluminium Brasses. MARIO LEVI-MALVANO and M. MARANTONJO (Gnxxettcc 191 1 41 ii 282-297).-Aluminium brasses contain 58-70% of copper 1-4% of aluminium and zinc. The ternary system Cu-Zn-A1 may be divided into two other systems namely Cu-Cu,Al-Zn and C,u,Al-Al-Zn (compare Gwyer Abstr. 1908 ii 284).The authors have investigated the binary system Cu3Al-Zn and those portions of the ternary system Cu-Zn-Cu3Al which include the aluminium brasses. The fusion and solidificaIion of the alloys was effected in an atmosphere of carbon dioxide. The diagram of the binary system Cu,Al-Zn shows that the temperature falls from the melting point of Cu,M to that of zinc. It exhibits two transformation points and a third lies at 570’. Evidence was obtained of the existence of various solid solutions and corre- sponding with the first two transformation points there are two gaps of miscibility in the solid state which were also demonstrated microscopically. In view of the character of its constituent binary systems the ternary system Cu-Cu3A1-Zn yields various mixed crystals and gaps of miscibility.The ternary alloys in the region examined develop heat at the beginning and a t the end of crystallisation but no heat changes mark the passage over gaps of miscibility. These have therefore been partially explored by microscopic methods. The paper is accompanied by a number of photomicrographs The Solubility of Iron Carbide in y-Iron. N. J. WARK (Metallurgie 1911 8 704-713).-The determination of the form of the curve of solubility of iron carbide in y-iron by thermal methods is uncertain. Better results are obtained by heating steels containing varying quantities of carbon in a salt-bath t o 2100° cooling to a definite temperature maintaining that temperature constant for some time quenching rapidly and examining microscopically.Barium chloride is used for baths above 1000° arid a mixture of sodium and potassium chlopides for lower temperatures. The maximum solubility of carbon in solid y-iron is found to be 1.70%. Equilibrium between the solid solution and the separated carbide (cementite) is rapidly attained being complete after ten minutes a t constant temperature. The highest temperature a t which the separation of cementite occurs and the lowest temperature at which pure martensite is observed are both determined the mean of the two being taken as the temperature of crystallisation of cementite for that alloy. Equilibrium is also reached from the other side by heating quenched specimens in a salt- bath and again quenching. The two series of observations coincide. The solubility curve thus obtained lies somewhat above that determined by the thermal method.The Removal of Rust from Iron in Reinforced Concrete. EDUARD DONATH (Zeitsch. angew. Chem. 1911 24 2355-2356. Compare Abstr. 1911 ii 897).-Rohland’s view that the active agent in the removal of rust is calcium hydrogen carbonate is not confirmed by experiment. Only that part of the ferric iron which is in combina- tion with ferrous oxide is converted into calcium ferrite. R. V. 8. C. H. D. C. H. D.INORGANIC CHEMISTRY. ii. 53 The Removal of Rust from Iron in Reinforced Concrete. PAUL ROHLAND (Zeitsch. angew. Chem. 191 1 24 2356).-Polemical against Donath (preceding abstract). KICOLA PAPPAU~ (Zeitsch. Chem. Ind. Kolloide 1921 9 233-239).-The coagulation of ferric hydroxides by different electrolytes has been examined.The hydroxide was prepared according to Graham’s dialytic method and the con- centration of the added electrolytes was varied. The observations show that the coagulation is brought about by the anions and that the chlorine bromine iodine and nitrate ions are about equally effective whereas the coagulating power of the hydroxyl ion is much greater. For anions of different valency the efficiency increases with the magnitude of the electrical charge. Non-electrolytes even in concentrated solution hRve no action on colloidal solutions of ferric hydroxide. H. M. D. F. DUCELLIEZ (BUZZ. Xoc. chim. 191 1 [iv] 9 1b17-1023).-8 study of these alloys has been made on the lines adopted by Vigoroux and Bourbon for nickel-zinc alloys (Abstr.191 1 ii l095) and a definite compound CoZn4 has been isolated and characterised (compare Lewkonja Abstr. 1908 ii 853). Alloys of these two metals containing more than 81.6% of zinc are non- magnetic and become more brittle as the quantity of zinc diminishes. When powdered and treated with cold 1% hydrochloric acid the acid being frequently renewed they leave a residue having the composition expressed by the formula CoZn ; this compound may also be isolated by treating the alloys with dilute nitric acid. It has DO 7.43 (calc. 7 ~ 1 1 ) ~ is nonmagnetic oxidises in the air is attacked slightly by boiliug water feebly by 1% hydrochloric acid and more strongly by 1% sulphuric acid. In these acids the residue slowly becomes magnetic due to deposition of cobalt from the solution.The com- pound also becomes magnetic when placed in aqueous solution of cobalt sulphate or chloride due to solution of the zinc but not in a solution of cobalt nitrate. When heated to a red heat it burns in oxygen air or chlorine. Alloys containing less than 81.6% of zinc are all magnetic and are attacked by acids without having any characteristic residue. The results of determinations of the E. M.E developed by a series of the alloys opposed to a pole of zinc in a solu- tion of zinc sulphate are plotted in the original and the curve shows a break when the amount of zinc in the alloy reaches 81% and corresponds with the formation of CoZn,. The analyses were made by dissolving the alloy in nitric acid adding excess of hydrochloric acid evaporating t o dryness re-dissolving in water containing a few drops of hydrochloric acid and precipitating the zinc as sulphide.A little sulphuric acid was added to the filtrate which was then evaporated to dryness and the cobalt weighed as sulphate. T. A. H. Preparation of Stannic Iodide and its Solubility in Certain Organic Solvents. F. ALEX. MCDERMOTT (J. Amer. Cl~em. Xoc. 1911 33 1963-1965).-8 method is described for the preparation of C. H. D. Coagulation of Ferric Hydroxide. Cobalt-Zinc Alloys.ii. 54 ABSTRACTS OF CHEMICAL PAPERS. stannic iodide by heating tin and iodine with carbon tetrachloride. The salt is much more soluble in the solvent at the b. p. than at the ordinary temperature and therefore separates on cooling. Stannic bromide can be prepared by adding bromine dissolved in carbon tetra- chloride in small quantities a t a time through the open end of the condenser to which the flask containing the tin and carbon tetra- chloride is attached.Sufficient heat is generated spontaneously to effect the reaction. In carbon tetrachloride a t 22.4" 8.35 grams per 100 C.C. of solution or 5.25 grams per 100 grams of solution ; at 50*0° 20.47 grams per 100 C.C. or 12-50 grams per 100 grams. I n chloroform at 28*0° 12.32 grams per 100 C.C. or 8-21 grams per 100 grams. I n benzene a t 2 0 * 2 O 12.02 grams per 100 C.C. or 12-65 grams per 100 grams The following solubilities of stannic iodide are recorded. E. G . Bismuthides and Inter-metallic Compounds. ALEXANDER C. VOURNASOS (Bey. 1911 44 3266-3271).-To prepare for example sodium bismuthide the author proceeds as follows.One hundred grams of sodium free from crust are added to an excess of pure anhydrous paraffin (b. p. 375-400") the mixture cdrefully heated on the sand-bath until the sodium melts and then the temperature raised to 300-310'. Sticks of pure bismuth (30 grams) are then gradually immersed in the paraffin ; as the bismuth melts the drops fall on to the sodiurri and combine with it crystals of the bismuthide immediately separating. The hest of reaction is so great that the paraffin is heated to the boiling point. The crystals are taken out with a perforated spoon washed with benzene or light petroleum and if necessary treated with amyl alcohol or liquid ammonia t o remove any free alkali metal ; this latter process is seldom necessary.Other cornpouods are prepared similarly. Sodium bismuthide Na,Bi forms small dark grey crystalline lamina? m. p. approx. 776". POtaSBiUm bisrnutlde K,Bi forms small shirring crystals. These bismuthides are readily oxidised on exposure to the air and are decomposed by moisture The compounds Ka,Pb (m. p. 405") and Na,Sn (m. p. 477") were also obtained. When potassium bismuthide is heated at 380-400' in a current of dry 11) drogen gas is absorbed producing a pyrophoric substance. On fuvther heating to 700° potassium vaporises away and a grey attioi ptious powder is left containing hydrogen and bismuth approxi- iiial elg in the proportions corresponding with the compound BiH,. This substance is being further investigated. ANDREW GUHDON FRENCH (Chem. News 1911 104 283).-The new element to which the name canadiurn is given was discovered by the author in the dyke rocks in the Nelson district of British Columbia running in quantities from a few penny-weights up to three ounces to the ton. It occurs pure in semi-crystalline grains and in short rods about half a millimetre in length by one-tenth of a millimetre in thickness. It is also found in metallic particles in the form of scales in platinum- bearing ores. These particles which have a bluish-white colour con- T. S. P. A New Element Probably of the Platinum Group.MlN ERALOGICAL CHEMISTRY. ii. 55 tain the metal alloyed with a volatile substance which may be osmium as it is dispelled by the blowpipe flame leaving a brilliant bead of cttnadium which is white. It is not platinum ruthenium palladium or osmium. It is much softer than these and is quite easily melted by the blowpipe. The new metal does not become tarnished by lengthened exposure to moisture and it is not oxidised by continued heating in the blow- pipe oxidising flame. It is soluble in nit& and Fydrochloric acids and in aqua regb without residue and its solution in nitric acid yields no precipitate with sodium chloride solution. It is not blackened by lengthened exposure to moist hydrogen sulphide or to alkali sulphides nor is it blackened by iodine; its nitrate solution is not precipitated by potassium iodide. Its melting point is somewhat lower than that of fine gold and silver and very much lower than that of palladium. It is electro-negative t o silver in dilute acid solution. T. S. P.