36 TWISS: THE ACTION OF HYDROGEN PEROXIDEV.-I’he Action of fTydro,qen Peroxide on the SodiumA lk y 1 Thiosti lphates.By DOUGLAS FRANK TWISS.A SHORT time ago it was shown by Fichter and Sjostedt (Ber., 1910,43, 3422) that organic sulphur compounds, such as benzyl sulphideand disulphide, can be subjected to electrolytic oxidation in glacialacetic acid containing hydrogen chloride, with formation ofsulphoxides, sulphones, and sulphonium compounds. I n the hopeof obtaining similar results with organic diselenides, dibenzyldiselenide was submitted t o similar treatment, but the result wasdisappointing ; the diselenide molecule is apparently less stable thanthe disulphide, and the only isolated products were benzyl chlorideand seleniumON THE SODIUM ALKYL THIOSULPHATES.37As the preparation of oxidation products of the diselenides, forexample, the diselenoxides or diselenones, appeared to be of interest,endeavours were made to achieve this end by the application ofhydrogen peroxide, which has proved useful with the analogoussulphur compounds (compare Hinsberg, Ber., 1908,41, 2836 ; 1910,43, 289). Again, however, the results were not those expected, forin acetic acid solution the oxidation gave almost entirely benzylalcohol and selenious acid. Experiments in this direction have alsobeen recently made incidentally by Fromm and Martin (AnmaZen,1913, 401, 185), who merely remark that they did not succeedin converting dibenzyl diselenide into the diselenoxide by means ofhydrogen peroxide. ThO experiments recorded in the present paperwere completed before this publication was made,I n the failure of the two above methods, experiments were triedon the oxidation of sodium benzyl selenosulphate by hydrogenperoxide, and as the results appeared tu be favourable, it prelimin-ary investigation was made with the analogous thiosulphatecompounds.It is found that in acid solution the sodium alkyl thiosulphatesare smoothly oxidised by hydrogen peroxide, with the formationof the corresponding disulphides.The process is very convenientas a method of preparation of the latter class of substance, andappears to possess some slight advantage over the methodspreviously described for the conversion of sodium alkyl thiosulphatesinto disulphides (Price and Twise, T., 1907, 91, 2021; 1908, 93,1395; 1909, 95, 1489).It is interesting that whereas electrolyticreduction necessitates the application of neutral or alkaline solu-tions (T., 1907, 91, 2021), oxidation by hydrogen peroxide requiresan acid solution. I n the presence of excees of alkali, oxidationyields no insoluble substance, but appears to attack the character-istic disulphide grouping, with formation of sulphur oxy-acids andof organic carboxylic acids. The apparently general applicabilityof the method is evidenced by the preparation of disulphides ofthe radicles benzyl, o-nitrobenzyl, pnitrobenzyl, and even allyl.An extension of the method to potassium o-nitrobenzylseleno-sulphate gave the production of di-o-nitrobenzyl diselenide, thusencouraging the hope that it may be useful for the preparation ofdiselenides also.Oxidation of DibenzyZ Diselenide.A cooled suspension of 13 grams of dibenzyl diselenide in 75 C.C.of glacial acetic acid was stirred continuously whilst 20 C.C.ofhydrogen peroxide solution (30 per cent.) were added slowly indrops from a tap funnel. After the introduction of the hydrogel38 TWISS: THE ACTION OF HYDROGEN PEROXIDEperoxide, most of the solvent was removed by distillation underdiminished pressure. The yellow residual oil in the flask partlycrystallised on keeping, and the crystals were found to consist ofselenious acid, which on reduction by a mixture of hydrochloricand sulphurous acids yielded practically the whole of the seleniumoriginally present in the diselenide.The oil from which the selenious acid had crystallised, whendistilled, gave a large fraction (b.p. 203--210°), consisting ofbenzyl alcohol.The only other substance that could be isolated was one whichcrystallised in very small quantity from the acetic acid distillateobtained in the first process after the oxidation; this formed needlesmelting at 55O.Oxidation in acetone solution gave still lecjs satisfactory results,but a minute quantity of the substance melting at 5 5 O was againobtained, but i t was again insufficient for further examination.Oxidution of the AIkyl I'hioszdphute Compounds.The oxidation was effected in aqueous solution, and as the oxida-tion generates acid, all that is required is the addition of hydrogenperoxide solution :2C,H,*CH2*S20,Na + H20, = (C,H,*CH,)S, + ZNaHSO,.The addition of a little acid, however, greatly accelerates the action,so that in a few hours at the ordinary temperature more than90 per cent.of the theoretical yield of almost pure disulphide isobtained.Taking sodium benzyl thiosulphate as a typical example, 3 gramsof this substance were dissolved in 40 C.C. of water, 4 C.C. of dilutesulphuric acid were added, and then 4 C.C. of 30 per cent. hydrogenperoxide solution. The solid which deposited melted a t 71'5Owithout recrystallisation. The oxidation can be accelerated if neces-sary by warming without diminution of the yield. The additionof a little ferrous sulphate, although it catalyses the reaction,renders the product less pure.Similar results were obtained with sodium o-nitrobenzyl thio-sulphate and with sodium p-nitrobenzyl thiosulphate, the produceddisulphides again being almost pure without further treatment.I n experiments in which sodium o-nitrobenzyl thiosulphate wastreated with hydrogen peroxide in aqueous solution containing suffi-cient' sodium carbonate or sodium hydroxide t o maintain an alkalinereaction to litmus, no disulphide or any other solid separated,although the thiosulphate compound was completely changed.Acidification of the solution by hydrochloric acid gave a precipitatewhich was identified as o-nitrobenzoic acidON THE SODIUM ATJKYL THIOSULPHATES.39Sodium a-llyl thiosulphate was not isolated, but a solution preparedin the usual manner by heating together an alcoholic solution ofallyl bromide with an aqueous solution of the equivalent quantityof thiosulphate, was first deprived of its alcohol by partial distilla-tion under diminished pressure; the cold residual solution was thentreated with hydrogen peroxide, when diallyl disulphide slowlyseparated as an undistillable, yellow oil.(Found, S =42.3. C,H,,S,requires S -43.5 per cent.*)An attempt to prepare the disulphide by the action of iodine ona hot solution of sodium allyl thiosulphate (compare Price andTwiss, T., 1909, 95, 1489) did not yield a satisfactory result.Potassium o-nitrobenzyl selenosulphate (1.5 grams), dissolved inapproximately 100 C.C. of water and treated with a little acetic acidand 3 C.C.of 30 per cent. hydrogen peroxide solution, gave a thickdeposit of di-o-nitrobenzyl diselenide, the action being complete inone-half to one hour. When once crystallised from alcohol thesubstance was pure and of the correct melting point.With potassium pnitrobenzyl selenosulphate under similar condi-tions the reaction appeared to proceed in an analogous manner, butthe product was a substance melting at 11B0, the melting point ofthe corresponding diselenide prepared by the action of iodine being107'5O. The cause of this discrepancy is not yetl clear.The application of hydrogen peroxide in this direction for thepreparation of disulphides and diselenides is being further extended,especially as it appears to be particularly well suited for the produc-tion of such compounds where other methods are known to give riseto by-productJs of unpleasant d o u r .It is also hoped to submit the chemical action to a physico-chemical examination.The author wishes to express his appreciation of the friendlyinterest of Dr. T. Slater Price in the above investigation.CHEMISTRY DEPARTMENT,MUNICIPAL TECHNICAL SCHOOL,BIRMINGHAM.* The sulphur was estimated by the electrolytic process of Gasyarini (CTamtta,1907, 37, 426), but considerable care was necessary in the introduction of the nitricacid on account of the vigour of the reaction on mixing