ACTION OF ACIDIC OXIDES ON SALTS OF HYDROXY-ACIDS. 1451 XC1V.-The Actioiz of certain Aciclic Oxides on Salts of Hydroxy-acids. 111. By GEORGE G. HENDERSON, D.Sc., M.A., and JOHN M. BARR. COMPOUNDS formed by the interaction of antimonions oxide with potassium hydrogen mucate and potassium hydrogen malate respec- tively were described in a former communication to the Society (Trans., 1895, 67, 1030), but, a,s the analysis of these substances showed that their composition was somewhat complex, it appeared1452 HEXDELSON AND BXRR: THE ACTION O F desirable to prepaye other similar salts in order to confirm the for- mulae assigned to the first. This has now been done, and the action of arsenious oxide on primary mucates and malates has also been examined, with the results detailed below.In addition, the action of antimonious and arsenious oxides on salts of certain hydroxy- benzoic acids has been studied, and the investigation of the be- haviour of those acidic oxides towards salts of certain typical hydroxy-acids has thus been completed. Our experiments were then extended to other acidic oxides, and we selected in the first place molybdenum and tungsten trioxides, and in the second place silicon and titanium dioxides. Compounds of the two former oxides with primary tartrates are described below ; titanium dioxide appears to form compounds which are a t present under investigation ; silicon dioxide, on the other hand, apparently does not react directly with primary tartrates. Antimonious Oxide and Primary Malates. Amnlonircm antimoitiomalate was prepared and purified in the same way as the potassium salt already described (Zoc.cit.), observation of the same precautions being necessary in order to obtain a satisfactory yield, €or in this case also the action of the oxide on the malate proceeded very slowly, and prolonged heating of the solution was essential. The new salt. was obtained in large, colourless crystals, which are easily soluble in water and fairly soluble in dilute alcohol ; it decomposes when heated to about 115O, giving off ammonia, and leaving a brown residue which is insoluble in water. The salt also decomposes when an aqueous solution is boiled for a short time, unless excess of antinionious oxide is present. The addition of caustic alkalis or mineral acids to the solution causes immediate de- composition.Analysis of the salt, dried in air, gave the following figures. Calculated for 2Sb20 (C, H,O;NH,) ,:Sb,O (C4H60,) 4,20H20. Found. Sb .......... 25.53 25-1 7 HzO ......... 12.87 12.55 The water was determined in a vacuum over sulphuric acid. The composition of the potassium salt agrees better with a formula similar t o the one given above than with that previously assigned to it. Found, Sb = 25-69, K = 11.42, H,O == 4-52; calculated foP 2Sb~O(C~H~O~K)~,Sb~O(C~H~0~)~,7H~0. S b = 85-76 ; I( = 11.17 ; H,O = 4.51. sent them as double salts. from them by any of the methods emploged. It will be seen that the formulae assigned to these compounds repre- No simpler compounds could be obtainedCERTAIN ACIDIC OXIDES ON SALTS OF HYDROXY-ACIDS.1453 Antimonious oxide dissolves slowly in a boiling solution of sodium hydrogen malate, and after removal of nnaltered malate by crystalli- sation, the addition of alcohol to the solution causes the precipita- tion of a thick, coloui*less syrup, which, when tested qualitatively, was found to contain antimony in abundance, sodium, and organic matter, and which, no doubt, consisted principally of sodium anti- moniomalate. We did not succeed, however, in obtaining the pure salt in a crystalline condition, and the syrup underwent spontaneous decomposition on standing. Arsenious Oxide and Primasy Malates. A8 formerly stated, although arseuious oxide dissolves freely in a solution of potassium hydrogen ma.late, attempts to obtain a com- pound corresponding to the antimony salt were unsuccessful.Ex- periments with the primary sodium and ammonium salts likewise failed, although, in the case of the latter, the existence of a compound was at least indicated. On boiling the oxide in a solution of the primary ammonium salt for some time, allowing excess of oxidc to crystallise out, and then adding alcohol,. crystals separated which were found to contain 12-13 per centt. of arsenic. They were fairly soluble in cold water, but always decomposed when subjected to purification. A salt corresponding in composition to the antimony compound would contain 17.37 per cent. of arsenic, and the crystals obtained might be a mixture of this with a little unaltered malatle. The preparation of a barium salt was also tried, but without success.A?~tiinoiiiot~s Ozide and Prirnavy Mucates. Ananzoniztrn antimowiomucnte, like the potassium salt (Eoc. cit.), was obtained by boiling antimonious oxide for several days with a soln- tion of the primary mucate. On concentrating the solution, a powder separated, which, after recrystallisation fi-om hot water and drying in the air, gave the following figures on analysis. Calculated for F O U I ~ . 2SbO (NH,) CsHsOs, (NH.JC6H,Os,’7LT20. Sb .......... 22.18 22-26 H,O ......... 11.50 11-70 From these figures the compound would appear to be a double salt, but whether this be so 01- not, it behaves exactly like the potassium compound when it is reci.gstallised several times, namely, its com- position changes, and the ,pimple salt is finally obtained. This is seen by the following analysis of the salt prepared by recrystallising the original substaiice until the percentage of antimony remained constant.1454 HENDERSON AND BARR: THE ACTION OF Calculated for Found.SbO(NH4)C4H,08,3 H,O. Sb.. ...... 28.64 28.82 EItO ...... 12-24 12-98 The water was determined in a vacuum over sulphuric acid. The salt is thus obtained in the form of a white, finely crystalline powder, only sparingly soluble in cold, but fairly in hot water. 111 its other properties it closely resembles the potassium salt. In preparing sodiicnz nntimoniomucate, SbONaC4H808,3H,0, a similar process was employed, but there was this difference, that the crude product, when recrystallised twice, had a composition agreeing with that calculated for the simple salt, there being no appearance of the formation of ail intermediate substance, Found.Calculated. Sb 28.13 28.50 H,O ........... 12-77 12.82 The sodium salt is also a crystalline powder, rather more solable in water than the potassium and ammonium salts, but very like them in other respects, On the addition of barium acetate to a solut'ion of the sodium salt, a white precipitate forms after a short time. The powder, washed with water and dried in air, xas found to contain 24 per cent. of antimony, but decomposed when it was heated w i t h water with the object of purifying it by 1-ecrjstallieation. A barium salt corre- sponding to the sodium salt woiiltl contain 25 per cent. of antimony. ............ Arsenious Ozide and P&n ary Mucates. As already stated, the arseniomucate of potassium, owing to its instability, could not be obtained in a state of purity.The same results followed when the preparation of sodium and ammonium salts was tried, for though the oxide dissolved freely in solutions of both salts, and gave crystalline products containing considerable quantities of arsenic, the instability of these was such t,hat no definite com- pounds could be prepared free from adrnixtum with unaltered rnucates . Antim oiiious and A?.seruioibs Oxides u i d JIami?elntes. I n experiments on the action of these oxides on alkali lactatep, thick sjrups were obtained containing some antimony or arsenic respectively, but they could not be further purified, and wlieii a solu- tion of barium lactate was used the results were practically the same.Hence, instead of lactates, we took the alkali salts of mandelic (phenyl-lactic) acid, which crystallise well, in the hope of obtaiiiingCERTAIN ACIDIC OXIDES ON SALTS OF HYDROXY-ACIDS. 1455 crystalline products, but it was found that antimonious oxide was almost insoluble in solutions of these salts, and that, although arsenious oxide dissolved to some extent, apparently it formed no compounds with the mandelates. Antirnonious and Arsenious Orides and Salicylates and Gallates. Having found that antimonious and arsenious oxides react with the salts O F various alcohol acids to form definite compounds, we proceeded to examine the behavioiir of these oxides towards bydroxy- benzoic acids, selecting in the first instance salicylic and gallic acids.The oxides were added to boiling solutions of alkali salicylates and gallstes respectively, but, even after prolonged boiling, no reaction took place. Antimonious oxide was practically insoluble in the solutions, and arsenious oxide merely dissolved to the usual extent in the water of the solutions without forming any compounds with the salts present. In view of these results, experiments with hydroxy- beiizoic acids were not further prosecuted. Mooiybderctiw Tj-ioxide and Primary Tartrutes. To a boiling, aqueous solution of sodium hydrogen tartrate (2 mols.), molybdenum trioxide (1 mol.) was added in small quanti- ties until all was dissolved ; the solution was then filtered, cgncen- trated on the water bath, cooled, and niixed with alcohol, when a white, crystalline powder was precipitated.The crystals were puri- fied by dissolving them in water and reprecipitating with alcohol, repeating the process several times, and finally drying in air. Found. MOO,( NaC,H,O,),, 3H20. Calculated for Mo.. ...... 17.76 18-32 H,O ....... 10.52 10.30 The wat,er was determined in a vacuum over sulpliuric acid. Sodium molybditartrate was t'hus obtained in %he form of a white, Crystalline power, easily soluble in cold water, but only sparingly in dilute alcohol. When exposed to light, or when heated to about go", it decomposes, turning brown. If an aqueous soiution is boiled for some time, it turns blue, pointing to decomposition of the salt and reduction of the rnolybdic acid, and the addition of alkalis or mineral acids also causes decomposition. Molybdenum trioxide was also found to dissolve in a, boiling, aqueous solution of potassium hjdrogen tartrate, the same propor- tions being taken as in the preparation of the sodium salt, and, on the addition of alcohol to the solution, a gelatinous precipitate was formed, which solidified on standing in contact with the mother1456 HENDERSON AND RARR: THE ACTION OF liquor. The crystalline powder obtained in this \my was very soluble in water, but only sparingly in dilute alcohol, aud, like the sodium salt, i t decomposed on exposure to light or heat ; no doubt, it was the corresponding potassium salt, but,, owing to its instability, whicli rendered purification difficult, i t was not analysed.When ammonium hydrogen tartrate was used, the molybdenum trioxide was again found to dissolve, and, when alcohol was added to the solution, R gelatinous precipitate was thrown down ; but, this compound could not be obtained in R crystalline state.Twugsten T&Xide nicd l'&nar$ Ta!rtrates. Sodium fwzgstitartrate was prepared by adcling tungsten trioxide (1 mol.) to a boiling ayueoiis solution of sodium hydrogen tartrate (4 mols.) in small quantities until all was dissolved. I n this case it was necessary to use excess of the tartrate, for otherwise some decomposition took place, and the solubion turned blue when boiled for any length of time. When dissolution o€ the oxide was com- pleted, the solution was conccntrated on the water babli, the excess of tartrate removed by crystdlisntion, and alcohol added t:, the liquid ; a colourless syrnp was then precipitated, which solidified after standing for some time i n contact with tbe mother liquor.The crystals were purified by dissolving them in water and reprecipitating with alcohol, the operation being repeated unt'il the composition was constant. The salt was dried in the air and anslysed. Calculated for Found. W02 (NaC4H40G) r,5H20. W. .. .... .. 28-89 28.39 HZO . . . . . . 13.79 13-88 The water was determined at 100". The salt crystallises in white plates, which are easily soluble in water, hiit only sparingly in dilute alcohol. It begins to decompose when heated t o about loo", and the addition of alkalis or mineral acids to the solution also causes decomposition. Potassiam tungstiturtrate, W02( KC,H,O6),,4~H2O, was prepared and purified in a similar manner. Like the sodium salt, it is pre- cipitated by alcohol as a colourless syrup, which solidifies on standing. It is then obtained as a white, crystalline powder, fairly soluble in hot, but only very sparingly in cold water. It is decomposed by heat, and by the addition of alkalis or mineral acids to its solution, or when its aqueous solution is boiled for any length of time. Analpis of the air-dried salt gave the following figures. Found, W = 28.03, H,O = 12.85; calculated, W = 27-42, R,O = 12.07. The same process was used in prepai+ing the ammonium salt, and crystals were obtained, oE wbich the composition approximatedCERTAIN ACIDIC OXIDES ON SALTS OF HYDROXY-ACIDS. 1457 to the formula W02(NH~CiH~06)2,2H20, but the mlt was not prepared free from all impurity, as it readily decomposes when heated with water. The harizcm sa2t is precipitated on adding bariam acetate to a solu- tion of the sodium salt; it is a white powder, insoluble in water. Analysis of the washed and air-dried precipitate gave the following figures. Calculated for Found. WO2(C4H4Ob=)2Ba. w ............ 27.34 28.35 Ba 21.61 21-11 ............ Chemical Laboratory, Clasgozo und West of Scotland Technical College.