THE CHEMXSTLiY OF DlBROJIOPROPTLT~IOCARBI3llDE. 1 7 I II.--.Yhe Chemistry of Uibromo~~1.opyltl~iocal.bimide : and the Actioiz o f Bronzi.lze and of Iodine up011 A I1 y It I b i o ~ e a . By AUGUSTUS E. DIXOS, M.D. 13 a short communication, made a few years ago to this Society (Trans., 1892, 61, 545), I showed that bromine unites readily with ally1 thiocarb imi de, thereby prodncing per- dibromopropylthiocar- bimide, CH2Bi-CHBr.CH,.NCS, a dense, almost colourless oil, of high refractive power., volatile i n a current of steam, nearly insoluble in water, miscible with alcohol, possessing thc characteristic odmr, and, generally speaking, the ordinary properties of a liquid thiocar- himide. In its chemical :.elations, however, it manifested certain peculiari- ties; thus, for example, it combined readily, on warming, with aniline, but instead of affording the expected dibromopropylphenyl- thiocarbarnide, CH,Br*CHBr*CH,.NH.CS.NH.CsH,CH, 1 mol.HBr was eliminated, the product being the hjdrobromide of a closed chain basic coniponnd, from mhich the frce base, a solid, was liberated by TOL. LXIX. c18 DIXON : THE CHEMISTRY treatment with caustic alkali. mentioned, the action was believed to occur, ultimately, as follows. For reasons set fort,h in the payer Further, the thiocarbimide interacted spontaneously with alcoholic ammonia, but instead of di bromopropylthiourea, ammonium bromide was obtained, together with a viscid, basic syrup, which refused to crystallisc, and was not then further examined. That the aotistitiients in question should fail to afford dibromo- propylthiourea, was, at the time, a matter of surprise, especially as the latter compound appeared, on reference t o Beilstein’s Handbuch,* to have been already prepared and examined by Maly, who obtained (Zeit.f. Chem., 1867, 42), from bromine and allglthioures, a welt- defined solid substance, C4H,N2SBr = ? CSNzH3*CH2-CHBr*CH2Br. Returning later, however, t o the study of this and similar com- pounds, their properties seemed hardly to accord with tho above con- stitution ; thus, according to Maly, the dibromide, C4H8N,S,Brz, when treated with moist# silver chloride, exchaaged but 1 atom of bromine for chlorine, producing the chlorobromide, C4H8N,S,BrCl ; again, with moist silver oxide, a strongly alkaline hydroxybromide, C4H8N28,Br( OH), was formed, from which, b y the action of hydrochloric acid, the chlorobromide was regenerated.Maly did not, fail to notice the significance of these peculiarities, for he refers expressly to the special fnnction of one bromine atom, which acts like the bromine in ammonium bromide;? in the case of tlre chlorobromide, the chlorine atom plays an analogous part. Moreover, further experiments on the interaction of alnmonia w i t h dibromopropylthiocarbimide suggested the desirability of re- examining the dibromo-derivative ; and this has now been done, with results leading to the conclusion that it and the other allied sub- stances (including a di-iodide) arc not halogenised thionreas, but salts of basic riug-compounds, derived probably from the parent- f Unfortunate!y I had not then access to the original paper.t “ Giebt man dem Thiosinnaminbromur folgende Formel : C H 5Br}N,Br *H in der ein Atom Broin das eine Ammoniak ZIII’U Ammonium rnacht, also innerhalb drs Radicals st,eht, wahrend dss zweite die Stelle einnimmt, wie das Brom im Broniammonium, oder das Chlor im Salmiak, so ist wold zu erwarten, dass sich diese beiden Bromatome rerschieden verhalten werden ” (Zoc. cit.).OF DIBROMOPROPYLTHIOCARBIMIDE. 19 t'ype, CH,< CH*S cHl,N>C*NH2, that is, p-amidopenthiazoline, or, using the nomenclature proposed by me (Trans., 1895: 67, 564), tri- methylene-pn- thiourea. B R O M J N E A N D ALLPLTHIOUREA. The addition Pl-Gduct, ClH8N2S,Br2, was prepared by adding 1 mol. of bromine, dissolved in chloroform, to allylthiourea (1 mol.) in alcohol ; cooling by means of a freezing mixture had little influence on the yield, which amounted, in one case, where it was employed, to 72i per cent.of the theoretical, and i n another, where it was not, to 71 per cent. After removing all the solid matter, a residue was obtained, consisting of a rather dark brown syrup. which, at, the temperature of the water bath, evolved fumes of hydrogen bromide. The solid product deposited slowly ; it was, as described by Malg, soluble in water and alcohol, very easily in the foimer, more sparingly in the latter; in other solvents, it was in- soluble, o r nearly so. Its melting point fell somewhat below that given, namely 146--14i0 ; I observed in two succeshive prcLparations 1:39-140" and 139*5-140-5° (con-.), respectively, apparently with- out decomposition.The aqueous solution had an acid reaction, was not desulphurised by tre;ttrnent with alkaline lead tartrate, and yielded with ammonia- cal nitrate of silver a whitish precipitate (not readily affe-ted by exposure to sunlight), which did not blacken even on boiling the mixture. Presumably, therefore, the cornpound was not a mono- substituted thiourea (Dixon, Trans., 1893, 63, 319). Actiorr qf Caustic Alkali.-On the addition of strong caustic potash (or even ammonia) to the aqueous solution, a heavy basic oil was. precipitated, containing nitrogen, sulphur, and bromine. The super- natant clear liquid, when acidified and treated with chlorine water, reacted freely for bromine ; inoreover, with silver nitrate, it gave i t yellowish precipitate of silver bromide.Thus a partition of the bromine had occurred, a probable explanation of which, in view of the above phenomena, lay in the abstraction by tbe alkali, from the original substance, of a mol. of hydrogen bromide- C,H,N,SBr, - HBii = CIH,N,SBr; and the results of a roughly quantitative experiment went to support, this idea ; 5.1 grams of the dibrominated compound yielding nearly 3.5 grams of basic oil, whilst, according to the equation, 3.6 grams should have been obtained. The oil is clear, almost colourless, iiiiscible with spirit, soluble in c 220 DIXON : T€IE CHEMISTRY niucli cold water, or moderately at the boiling temperstare ; it is strongly alkaline t o litmus, has a barely Ferceptible basic odonr, and when dissolved in dilute hydrochloric acid, reacts freely with clllo- rine water for bromine.If exposed for some time to air, i t darkens, becomes hard, and insoluble in spirit ; i t dissolres now only slowly in boiling hydrochloric acid, from which i t is precipitated hy alkali as an amorphous, whit3ish powder ; t h e latter, on heating, swells up and carbonises without melting ; it appears also t o be insoluble, save in liquids which decompose it,. The nature cf tliese latter changes mas not investigated. Actioiz of IIych-oLronzic ucid.-Dilnt e hydrobromic acid was added i n slight excess, to LZ qiiantity of the freshly precipitated base; t h e mixture wzs at first turbid, but became clear a s t h e point of neutra- lisation was approached ; t h e solution was then evaporated to a, snialI bnlk on t h e x-xter bath, and left orerniglit ; wliite c r p t d s sepa- rated, freely soluble in water, l m r i v g t h e appeamncc, and exhibiting d l t h e properties, of the compound cbtained hy t h e direct union of lxomiiie with allylthionrea, and me1 ting, after i~eci~ystnllisation, at the same temperature, namely, 139-1-lOG.Action of Hydrochloric ncid.-IIf, then, Maly ’s alIylthiourea di- bromide is t h e hydrobromide of a base, C4HiN,SBr, t h e latter, on treatment with hydrochloric acid, ought t o afford a hydrochloride, C4H7N2SBr,HC1, identical with t h e clilorobromide obtained b y him from tlie dibromide and moist silver chloride. “his is, in fact, the case, as shown by the following experiment. A quantity of‘ t h e basic oil was treated with a slight excess of hydrochloric acid, and the clear solution concentrated by evapora- tion on the water bath ; on cooling, beautiful I-osettes of long, very hard, white prisms were deposited, having t h e properties ascribed to t h e chlcrobromide, and me1 ting, after being prei-iouslv well washed with spirit, a t the temperature recorded by MaIy, namely 129-130”.This compound, like tlie hjdrobromide itself, reacts freely for bromine 011 the addition of chlorine water. From this sjuthesis, i t may be concluded t h a t the hingle bromine atom which the ‘. dibro- mide ” gives u p 011 treatment withmoist silver chloride is t h e one not included in t h e C,H7N2SBr fraction of t h e molecule. It seemed of interest to t r y whether dibromopropylthiocnrbimide would exchange any of its bromine for chlorine under t h e abovc treatment ; an experiment was therefore made by allowing the tbio- carbimide, in aqueous spirit, t o remain (in the dark) in contact wit11 excess of freshly precipitated silver chloride, shaking frequently.After 10 days, during which interval slight desulphurisation bad occurred, t h e mixture \\-as heated for a, few minutes, and then steam distilled j a brilliant oil mine over, i n which the sulphur and nitrogenOF DIBROMOPROPYLTHIOCARBIhI1I)E. 21 were determined, with the result that it proved to be the unchanged dibromopropyl tliiocnrbimide. Actlo77 of Pic& acid.-In order to obtain the base in a form which admitted OF easy purification for analysis, i t was converted into the picrate, either by treating the liquid base in alcoliolic solution with picric acid, o r by adding the latter to an aqueous solution of the hydrobromide, got from bromine and allylthiourea.In either case, a finely d i d ed, canary-yellow precipitate fell, which, on recrystalli- sation froni dilute spirit, was obtained in miii~ite, glitteriiig, pale- yellow prisms, melting a t 187-188" (corr.). If heated rapidly over a gas flame, it melts, begins to effervesce, and then deflagrates, leaving a porous, carbonised residue. The picrate is rather sparingly soluble in boiling water, still less so in cold ; more easily, though by no means freely, in boiling alcohol^ 1 t is oxidised by boiling with nitro-hydrocliloric acid, with formation of sulphuric acid, but yields no sulphur to boiling alkaline, lead, 01- silver solutions.The formula was checked by analysis. 0*2017gave 0.2074 CO,, and 0.0365 H,O. 0.2074 ,, 29.4 C.C. nioist nitrogen at 15" and 759 mm. N = 16.59. C,H,ilr',SBr,C6H2(?\T0.?)3*OH requires C = 28 23 ; I3 = 2.36 ; C = 28.04; H = 3.11. 0.2992 ,, 0.168:3 BnS04. S = d.43. ,-' N = 16.55 ; S = 7-55 per cent. These figures-although that for the hjdrogen is unduly high- sufice t o establish the composition of the base as CIHiN,SBr, and not C4E8X2SBr(OH) ; the latter, Maly's b~o~~~tl~iosinr~anznzol.liumoxyd- hydmt, would requirc C = 27.13 ; H = 2-26 ; N = 15 88 ; S = 7-25. The percentage composition of this compound differs only slightly from that of tho base C,YiN,SBr (even assuming them both to be uncombined), and i t seems not unlikely that the material analysed';' by him was the base in question, possibly containing moisture, or else sliglitly impure.A 111 inowia a nil Dib9-omopropy Ithioca 14 imicle. The chcinistry of the iiiteract,ion between allylthioures and bro- mine laving been hhus far cleared up, an investigation of that between ammonia and dilro~noprop~l thiocarbirnide was now pro- ceeded with. Previous experiment had, as already stated, established the non-formation from t!iesc constituents of dibromopropylthiourea ; but the facts which Lad meanwhile come t o light pointed to such a iiegative result as at leasf probable, for the union of the compounds in question might be expected t o result in the production of the hydro- * No analysis is givcn in liis paper.22 DIXON : THE CHEMrSTRY bromide of the brominnted base, and since this salt is decomposed by ammonia, as shown above, the prodiicts to be anticipated are those which were actually obtained, namely, a basic oil, together with ammonium bromide, C,H,Br,*NCS + 2NH3 = NH4Rr + C4H,N,SBr.Dibromopropylthiocarbimide was dissolved in a considerable excess of strong alcoholic ammonia ; interaction commenced in a few seconds, with evolution of so much heat that the mixture began to boil, and, on cooling, crystals of ammonium bromide were deposited. When evaporated a5 F t r as possible on the water-bath, the product con- sisted of a brownish oil, mixed with solid matter ; this was treated with excess of dilute hy(lrobromic acid, again evaporated, and ex- tracted with absolute alcohol, which left most O F the ammoniuni hvomide undissolved, and by repeating the evaporation and extraction, x little more ammonium bromide wag removed.The clear solution as then mixed with a few drops more of dilute hydrobromic acid, arid allowed to evaporate spontaiieously; next day a crop of solid matter had formed, entangled in a tenacious mother liquor ; the latter was drained off, not without difficulty, by aid of the pump, and the residue recrystallised from water, and well washed with spirit. The product was colourless, freely soluble in water, soluble in alcohol, had nn Acid reaction, gave, on the atdtlition of caustic potash, a basic oil, afforded bromine when treated with chlorine water, and in short,, resembled precisely the hydrobromide obtained from allylt,hiourea ; its melting point was 139-140' (corr.), that is, identical with that observed for the said hydrobromide.For further identification, the thiocarbimide and ammonia were combined as before, the precipitated ammonium bromide was dis- solved by adding water, and the oil by dilute hydrochloric acid ; the clear solution, when mixed with caustic alkali, deposited a basic oil, which w-as dissolved in spirit and a little hydrochloric acid, and treated with picric acid. It gave a somewhat viscid, yellow prccipi- tate, which, after recrystal lisa,tion, had the appearance and general properties of the picrate already descx-ibed, melting at nearly the same temperature, namely, 183-184" (corr.). Summarising the principal facts elicited from the above experi- ments, it may be statcd tliat- 1.Bromine unites directly with allylthiorirea to form R compound, C,H,N,SBr, ; this compound is the hydrobromide of a strong base, C,H,N,SBr, which is liberated from it as an oil on the addition of caustic alkali. 2. The free base combines with hydrobroniic, hydrochloric, and picric acids, yielding C4H,N2SR r,HB r, CIHiN2SB r, HCI, and C4H7N,SBr,CsH,( NO,),*OH,OF DIBROMOPROPTLTHIOCARBIMIDE. 28 respectively. The first of these is identical with the addition product of alljlthionrea and bromine. 3. The same base, 'together with ammonium hromide, is formed by t8he action of alcoholic ammonia on dibromopropylthiocarbitnide. Constitution of f he Base, C,H,N,SBr.-The mechanism by which the above base is produced niay be explained as follows.First, having regard t o the very favourable conditions realised for its formakion, the initial stage consists, probably, in the production of (an unstable) dibromopropylthiourea ; and since the mono-substituted thioureas, a t all events when undergoing chemical change, and parti- cularly under the influence of substances containing halogen, assume generally the '' labile " form, the base may be supposed t o originate from the decomposition of either C31T,Brz*N:C( SH)*NH,, or C3H,Brz*NH*C (SH):NH. Secondly, the tenacity with which the base retains its sulphur, not only on treatment with alkaline lead solution, but even aft-er boiling with the much more potent ammoniacnl silver nitrate, points strongly to the existence of a sulphur atom, held in closed chain between two other polyvalent atoms.Such a linkage would necessarily entail the withdrawal of the SH hydrogen (to form hydro- gen bromide), and the action would then occur in one of the two following ways, probably the former. CH,* S 1. CHBr<CH CHzBr H,s,>C*NH, = HBr + CHBr<CH2.N>C*NH,; 2. CHBr<CH,---NH>C:NH CH,Br HS = HBr + CHBr<C,H,"ZG:>C:NH; 2-- wit.h, of course, the alternative possibility that the bromine in the @position might be removed instead, for example, CBzBr*FHBr HS CH,Bi.*$lH S CHZ - N CH,.A >C*NH, = HBr + ,>C*NHp Moreover, these structures accord much better with the highly basic cliaracter of the substance than those where the SH is supposed to persist, for instance, CHB~<:&~YC.SH. Other evidence of a more direct kind goes to confirm the views above stated.By combining methylthiocarbimide with hromethyl- amine, Gabriel obtained (Ber., 1889, 22, 1148), not methyl. brom- ethylthiocarbimide, CH3.NH.CS.NH.C~z.CH,Br, but the hydro- bromide of a bme, " n-metByl-ethylene-+-thiourea," * In tbe same paper, he conjectures (p. 1145) that Maly's dibromide,24 DiXON : THE CHEMISTRY Hirsch, in like manner, starting from /3-~romopropylamine, pre- pared (Bey., 1890, 23, 971) with methyl and allyltliioc,zl.bimie~, l3Me.S the " propylene " derivatives, CH,--N >C.NIIMc, and >C*NEIAlI, flHDle*S CII, --N respectively ; the structu1.e of these compounds was established by their behaviour on oxidation, the first-named, for example, yielding methylamine, caybonic anhydride, and taurine.YH2* S yHz.S02* OH >C.NHMe + 2H20 + 3 0 = NH,Me + CO, + c H,." C H,.NH, These and other similar experiments by Gabriel and his pupils, have shown the easy mobility of a halogen atom, under the circum- stances mentioned ; whence, by analogy, the general struct,in*e of t h e basic dibromopropylthiocarbimide derivative may be inferred, save as regards the position of its bromine. I t has been pointed out, how- ever, by P a d (Bey., 1891, 24, 4'253) that bromirie is more readily e1imin:ited from the than from the /3-position (see also Andreascb, .Mo.nntsh., 1889, 5, 33) ; t h o constitution of the base is, therefore, presumably CHBr<cH~.N>C.NIIZ,~ CH *S that is, p-anlido-y-bmmopenthi- nzoline. Regarded a s a dihydric thiourea derivative, and marking the points of substitution in the carbon ring, a, /3, pi, reckoning from the nitrogen attachment ( n ) of the latter, its structure niay also be systematically expressed, accoding to the nomenclature referred to above, by the name /3-bromo trimethylene-\1/?~-tliiourea.It was hoped that by reduction of the base with nascent hydrogen, its bromine might undergo replacemciit, and the position of the latter be dircctly ascertained by the production of either triniethylenti- pi= propylene-pseudot hiourca, t h u s cH3'7 '- '>C.NH, ; CH,*PU" >C-NH, + 2H = HBr + CH,Br*$: H - S C H,*N but the methods tried, namely, treatment with sodium amalgam, or with tin and hydrochloric acid, led only t o disruption of the molccule. C,T€,N~SEr,, slioiild be regarded as the Ii~drobrornide of a siinilarly constituted base j I only noticed this statement after practically all tlie work in this paper hod been completed.* Probably capable of interac?ing also, uiicler suitable conditions, in tlie desinotropic form, CEB~<:$.;;>C:XII.OF DIBROMOPROPTLTHIOCARBIMIDE. 25 The readiness with which tho brominnted thiocnrbimide parts with an atom of i t s halogen, thereby forming the balo‘id salt of a base (or fme hydrogen bromide) will be seen from what follows, t o be a s marked with primary and secondary amines (and even with alcohols) as in the case of ammonia ; i t seems, in fact, as though the broininsted thioureas,* mono-, di-, or tri-ecbstituted, are all alike incapable of existence under ordinary conditions.The question arises, therefore, whether, haviiig regard t o the apparent similrtrity of tlie cases t o those above-mentioned, the “ cEilor~l1yltbiouren ’’ and “ bromalljlthiourca,” briefly recorded by Henry (Be?.., 1872, 5, ISS), as resillting froin the action of ammonia on a-chlor- and bt.orn-tlllylthiocsrbimides, respec- tively, are not also halo’id salts, namely, tlie hydrochloride a n d CH,:C -- S hydrobromide of a basic ring compound, (:l12.K>C*NH2. I hope, later on, to re-examine t h c x substn~~ccs. T O D I N E A S D d L L P L T H I O U H E B . By combining the above materials in alcobolic solution, Maly ob- tained (Zeit. j’. Chcm., 1869, 258) almost colourless crystals, soluble i n water and alcohol, l q i n n i n g to ixelt, m-ith decomposition, a t 90°, and consisting of CJ38N2S,T2.Like the corresponding bromo-deri- vative, this diiodide, on treatment with silver chloride, exchanged half its lialogen for clilorine, yielding thereby t h e chloriodide, In repeating t h e former experiment, I added tlic calculated quantity of solid iodine, to allylthiouwn, in alcoholic solntion. Heat was crolved (this was checked by external cooling with water) and the iodine quickly dissolved, yielding a clear, brown so:u tioii ; on conceii- trating, and allowing i t to cool, i t solidified t o a mass of haid. brownish crystals, which when washed with benzene and spirit, became nearly white, and melted a t 136-5-133*5° (corr.) to a reddish liquid. There is so marked a divei3gence between this melting point, and t h a t one given by Maly, t h a t i t was considered ndrisable to check the formula by analjsis ; figui-es, however, were obtnined, agreeing with those required for the diiodo-addition product.C,H,N,S,CII. 0.2036 gave 0-255 AgI, and 0.1262 BsSO,; I = 68-73; S = 8 52. C,H,K2SI, requires I = 68.59 ; and S = 8.66 per cent. The cornpound is readily soluble in water and alcchol, insoluble i n benzene; its aqueous soiution is sharplg acid to litmus, gires t h e reaction f o r iodine, after the addition of chlorine water, and is not * This does not apply to compounds in which the bromine is included in an aromatic ring. A U C ~ as bro~iophen~ltliioui~ea.26 DISON : THE CHEMISTRY desnlphurised, either by ammoniacnl silver, or by alkaline lead salts, even on boiling. Action of Caustic Alkali.-Ten grams of the di-iodide were dissolved in a little water, and the solution mixed with a, slight excess of strong caustic potash a dense, nearly colourless, very tenacious oil at once sank to the bottom.Aftei- a short time, the supernatant liquor was poured off and examioed ; it was found to contain much alkali iodide. The residual oil (washed with water) becnme pasty a t -8', hut did not solidify ; it was very sparingly soluble in water or dilute alkali, insoluble in ether, miscible with a little spirit, readily soluble in hydrochloric acid, strongly alkaline to tcst-paper, and reacted freely for iodine when treated with chlorine water. I n this case, therefore, a s in that of the corresponding bromo-compound, part of the halogen is removed by treatment with cold alkali, and part Left in R basic residue.In order to get the latter in a form fit for analysis, i t was dissolved in dilute hydrochloric acid, and picric acid added to the clear solution. The bright yellow, amorphous precipitate thus prodiiced was collected, well washed with water, and dissolved in boiling alcohol ; on cooling, the solution deposited spherical aggre- gates of orange-yellow, minute prisms, darkening a t 172', and melting, somewhat indistinctly, at 176-177'. If heated more strongly, it turns reddish-black, swells up, and effervesces, evolving violet fumes of iudine. I t is practically insoluble in boiling water ; the alcoholic solution, if mixed m7ith starch a i d chlorine water, and then diluted, gives no blue coloration.S = 6.94 ; I = 26.62. 0*%094 gave 0.1032 AgI and 0.1057 BaS04. The interaction may be thus represented- C ~ H , ~ z S I , ~ ~ H ~ ( ~ ~ ? ) 3 * ~ H requires S = 6-80; I = 26.92 per cent. CSN2H3*CH2*CH:CH2 + 21 = C H I < ~ ~ ~ : ~ > C * N H , , H I , the iodisecl base, accordingly, being p-arnido-y-iodopent biRzolin e, other- wise /%iodotrimethylene-+n-t,hiourea. When the base is tlreated with ammoniacal silver nitrate, a white precipitate forms, which is scarcely coloured, even by expostire to direct sudight. With iodine monobromide interaction occurred, but no bromiodopropyltliio- cmbimide could be obtained. Iodine appears not to combine with allylthiocarbimide. ACTION O F O R G A N I C BASES ON D J B R O M O P R ~ O P Y L T H I O - CARBINITDE. Numerous experiments were made in this direction, using primary and secocdary bases, and phenylhydrazine.Interaction occurredO F DIRROMOPROPPLTHIOCARBIMIDE. 27 ceadily in every instance on warming. but the bases, when liberated irom the corresponding hydrobromides by caustic alkali, were, very frequently, uninvi ting viscid masses, sometimes having the consist- ence of birdlime or pitch, and resisting all the attempts made to get, them into suitable condition for analysis. The cases where deiinite results were attained me recorded below. Dibromopi.r~ylthiocal.bimide and Payatoluidine. These, in molecular proportion, mere separately dissolved in nearly absolute alcohol, the solutions mixed, and the alcohoi driven off on the water bath. The residue, a thick, apparent.ly uncrystallisa\~le, acid syrup, consisting of the hydrobromide, was dissolved in dilute hydrochloric acid, and the clear, yellow solution treated with excess of caustic soda.The yellowish, heavy oil which mas precipitated, i n ft short, time changed to a whitish golid ; the alkaline liqiior poured off from this contained sodium bromide. The solid was twice re- ci-ptallised froiii alcohol, and t,hns obtained in rosettes of pointed, white prisms melting at 124-125' (corr.). hiialysjs gave the following result. 0.2097 afforded 0.1743 BaSO+ S = 11.42. CllHI3N2SBr rcquires S = 11-24 per wnt. The interaction may be thus formulated- CH,Rr.CHBr*CH,-NCS + C,H7.NH2 = CHBr<~~$'>C*NH.C,H: + HBr, the product being, accordingly, p-paratolylamido-(y-bromopcnthiazo- line, or, using the " thiourea " nomenclature, P-bromotrimethylene- ~/n,-v-paratolylthiourea." Tlie substanco is very faintly aikaline t'o test-paper, freely soluble in warm alcohol, much less so in cold, insoluble in water. It is not affected by boiling w i t h alkaline lead solution, and gives, with am- moniacal silver nitrate, a white, amorphous precipitate, which is not blackened by heating the mixture.* Possibly, since thiocarbimide and a monosubstituted basc are here interact- ing, the intermediate formation of a fliiocarlamide, S C < ~ ~ ; ~ ' C H B r ' C H ~ B r , might be anticipated, with subsequent, break-doxn into SC < ~ ~ ~ CHBr and HBr ; but, in the first place, even if a thiocarbamide mwre initially produced, the deformation of its molecule into the " labile " (thiourea) configuration, under Llie iiifluence of the halogen, is to be expected ; and, secondly, a compound having the structure indicated, would almost certaidy give up its sulphur without difficulty t,o rsmmoniad silver nitrate.28 DIXON : THE CI-IElIISTRY After numerous nnsuccessfnl attempts to isolate this CoinpouiicI, the following method was found t o answer.Orthotoluidine and the thiocnrbirnide, in molecular proportion, were dissolved in spirit,, and the mixture slowly evaporated on the water bath, so as to expel most of the solvent. The residue, rz thick, jellow s p u p , was ti-ansferred to :t flask, and distilled with steam until the distillate ceased to be turbid ; the residue in the ff ask, a clear, colourless solntion, mixed with ti-accs of a dark oil, which refused to come over, was separated from the latter by filti.ntion through paper, and the filtrate concen- trated on the water bath.On cooling, the hydrobromide sank as a clear, pale brown oil, wliicli slioxTred no sig~is of solidifying cven after some days' standing. It was then dissolved in a, sufficiency of warm water containing a tracc of spirit (about 30 volumes were required) and caustic potash added ; tlie base was thereapm precipitated as :I paste, which in a short time bncame tongti. I n this condition i t was very freely soluble in boiling sphit, from which. on cooling, it was deposited in well-formed crystals ; after a couple of recr-ystnllisatioiis from the same Eolvent, it formed beautiful, vitreous, apparently rhombic, plates, melting at 13 l.*S-lS5*5O (corr.).0.2039 gave 0.1672 BitSO,. CI,H,,N2SRr requires S = 11.24 per cent. The 01-thotolyl base is feebly alkaline to litmus ; when pure, i t is somewhat sparingly soluble in boiling alcohol, much less so in cold, and issoluble in water. I t s alcoholic solntion, when treated with alkaline lead tartrate, is not desulphurised ; even tlie whitish prccipi- tate, which falls on the addition of ammoniacal silver nitrate, may be boiled without the least sign of darkening. S = 11.27. By evaporating on Lhe water bath mixed alcoholic solutions of di- bromopropylthiocnibimide and p-naplithylamine, tlie hydrobromide was obtained as a tenacious, brownish syrup, difficultly soluble i n cold water. It was dissolved in dilute hydrochloric acid, the solu- tion mixed with excess of caustic alkali, and the lmxipitsted free base, a thick, browii oil, washed with water, arid set aside ; after a few days it solidified, and by repeated recrgstaliisat,ion from spirit, nsing animal cl~arcoal, brilliant, nearly white, prism3 wei-e obtained, softening at 189", and melting between 190 slid 191" (con-.).0,1992 gave 0.1430 BaS04. S = 9-86. C14HlJK2SBr requires S = 9.93 per cent.OF DIBROMOPROPYLTIIIOChRCIM1DE. 29 The base is neutral t o litmus ; slightly soluble in boiling water, easily in boiling alcohol, rather sparinqly in cold ; it is not visibly affected by boiling with alkaline solution oE lead. Ammonincal nitrate of silver precipitates a white, ainorplious substance, not rcadily coloured by exposnre to white light, and not blackened on boiling ; this precipitate is insoluble iii diluto nitric acid.but is decom- posed by the concenbrated acid, with formation oE silver bromide. Operating as iii the precerliilq case, an extremcly viscid brown hydrobromide was obtained, practically insoluble in \I ater ; the base set free from the hydrochloric solution of tEc hydrobromide fornied .z tenacious lmstt', eventually acquiring a pitch-like consistence, i t WAS iris01 uble in water, rniacibl2 with ~ a r ~ i spirit, but (lid n o t crystalliscb from the latter. I n a second espe~irnent, the base was steam-distilled in order t o remove any uncliaiiged thiooarbimide cir nnphthyl,zrnine, but die product, even aEter long keeping, liad solidified to o n l j :L trifling extent.For the pyrpose oE analysis, thc picrate T V ~ S prepared, ancl purified by recrjstallisntion from boiling acetic acid. The rnyrtle- greec prisms which sepnratcd on cooling 7,vere dissolved in hot hydro- chloric acid ; OR pnrlly nentralisinq thc solution, the pic:*ate was thrown down as a bright ydlow powder. S = 5-80. 0.2008 gave O.OS46 BaSOI. C,,H1,N,SBr,C,H2( NO,),*O€€ requires S = 5.82 per ccnt. The base, in alcoholic solution, is not desulphurised by boilinq with alkaline lead iartratc ; on the addition of ammoniacal nitrate of silver a wliite precipitate fulls, not blackened by boiling, but changivg rapidly in sunlight, first to rose colour, and then to deep purple. Dilute nitric acid, added to an alcoholic solution of the base, affords :L rich, but evanescent, purple - bl u e coloration. Interaction occurred spontaneously on adding the thiocarbimide to (:tlcoholic) met,hylanilinc, with evolution of so much heat that the mixture began to boil.The blackish-brown syrup thus obtained ivc2.i treated as described for the corresponding ortliotolyl compoun1,i. After steam-distillation, the filtered residue was of a, pale, clxi-e t colour ; when concentrated and allowed to cool, it solidified t o a slate- coloured, crystalline mass, becoming white 011 mashing with spirit ; by recrystallisation from hot water, small wliite prisms were ob-30 DIXON : THE CHE?clISTRY tained melting at 183-184" (corr.). The hydrobromide is \Terj freely soluble in alcohol aid in hot water, much more sparingly in cold ; the solution has a marked tendency to remain supetwtturatcd. its aqueous solution reacts sharply acid, is not desulphurised by boil- i n g with alkaline lead, or ammoniacal silver salts, and affords bra- mine when treated with chlorine water.0.2001 gave 0.128 RaSO,. S = 8-80. C,,H,,N,SBr,HBr requires S = 8.75 per cent. By mixing the aqueous solution with caustic potash, the base I V ~ S thrown down as a heavy, brownish oil, which did not solidify in R freezing mixture ; i t is insoluble in water, soluble in alcohol, etliei-, benzene, arid hydrochloric acid ; the latter solution gives no reaction for bromine after treatment with chlorine water. Alcoholic silver nitrate precipitates a nearly white silver derivative ; if this be covered with nitric acid, no immediate change takes place, but after a short time sudden and violent actiou occurs with evolution of nitrous fumes, silver bromide being left. CH 9s p- Pipe;.idy 2- y - bromope t I t hiuzoliue , C H B I*< cH-.N> C*N C,H I ,, . From the thiocarbimide, in alcohol, and piperidine : here, as with the other secondary base (nietliylauiline), 1-igorous action spontane- ously occurred, the mixture boiling freely. On cooling, beautiful, anemone-like, crystalline tufts appeared, and the whole inass pre- sently solidified ; the product,, after two rewystnllisatious from boil- ing spirit, formed brilliant, colourl(w, vitreous prisms, freely soluble in hot, but only sparingly in cold, alcohol, and melting at 189-190° (COW.), with previous sintering a t 1 8 8 O . It dissolves also 'in cold water, yielding a solutioii which is ncutml to litmus ; this solution, when treated with chloroform and chlorine water, gives the reaction for bromine.0.2588 gave 0.1792 BaS04. S == 9.51. CSHl,N,SBr,HBr requires S = 9.31 per cent. The aqueous solution is not desulphurised by boiling with alkaline lead tartrate, but becomes clear yellow ; with ammoniacal silver nitrate in excess, a bulky, white precipitate falls ; the lat.ter is very sensitive to light, a rich purple colour developing after a few seconds' exposure in the sun. By adding dilute caustic potash, the free base mas a t once pre- cipitated as a clear, almost colourless, strongly alkaline syrup, iusoluble in water and alkali, soluble in alcohol, ether, and acids ; it contains halogen, as shown by the copper oxide test, but does iiok react for it on treatment with chlorine water.OF DIBROMOPROPYLTHIOCARBiMIDE.31 The alkalinit,y of the piperidyl base is so great as to allow of titra- tion by standard acid ; but on account of its practical insolubility in water, this is best performed indirectly. 0.3828 gram, dissolved in 20 C.C. N/10 HCl, required for neuti-ali- sation, 5.5 C.C. N/10 NaOH, corresponding to 14.5 C.C. acid used up to form the hydrochloride. Theory for C9HljN2SBr,HC1 requires 14.58 C.C. ACTION OF ALCOHOLS ON T H E THIOCARBIMIDE. BibToni qwopylt h.ioca&mide and E th y lic A1 cohol. p - E't hoxy -91- b ~ o ~ n o - This reaction was carried out in the expectation o€ obtaining the penthiazoline. snlphuretted urethane-- Eight grams of thiocarbimide, together with excess of anhydrous ethylic alcohol, were sealed up, and heated for about an hour, at a temperature slightly over 100".On opening the tube, some gas escaped, smelling of mercaptan, and blackening lead paper ; the contents of the tube thereupon partly crystallised. The mother liquor, from which the crystals had sepa,rated, was sharpiy acid to litmus, and contained hydrobromic acid. After two recrystallisations from hot alcohol, the solid was deposited in small, white prisms, melting a t 96-97' (corr.) ; they were dried in a, vacuum over sul- phuric acid and analysed, with the following results. 0.2922 gave 0.3072 BaS04. S = 14.45. 0.225 ,, 0.1893 A@. Br = 35.79. CousequentJy, the thiourethane, which would require 10.5 of S and 52.4 of Br, had not been produced, but, as in the corresponding experiments with amines, the addition product, if formed a t all, is unstable, and decomposes under the conditions of experiment, with elimination of the elements of hydrogen bromide, and production of a closed-chain compound.Theory for C6H,,NSOBr. Experiment. S.. ...... 14-30 14-45 Br ...... Y5.67 35.79 The substance is moderately soluble in hot water, much less so in cold (the solutioii has a distinctly acid reaction to litmus), freely i n boiling alcohol, but only sparingly at the ordinary temperature ; it dis- solves also in hydrochloric acid, from which it again separates on the32 DIXON : THE CHEMISTRY addition of dilute caustic alkali. It is not affected by boiling with alkaline lead tartmt e, and gives no colour reaction with ferric chloride.Neutral, or ammoniacal, silver nitrate throws down a ~ello~7~isli-wliite precipitate, which is decomposed by strong nitric acid, in. the cold, with formation of silver bromide ; tliis silver componnd is remark- ably sensitire to actinic light, rapidly changing in ordinary (diffused) claylight, through mauve and violet, to deep purple. On boiling, it i s not desulphurised, but, on tlie contrary, whitens somewhat, thereby losing, to a great extent, its sensitiveness to liglit. A very ~ o u g h coniparstiv-e experiment 011 the latt el* property was made, by adcling silver nitrate t o separate solutions of ammonium chloIide and tlie substance, and exposing the rcsnltant precipitates (Nos. 1 and 2 respectively) to dull cinylight : In one minute, No.2 began t o colour, whilst Nc!. 1 was yet not visibly affected; after seven minutes, a faint mauve coloration was perceptible in No. 1, No. 2, meanwhile, had acquired a strong purple coloui-. Marked annlogy appcars to subsist between the action of dibi-omo- propylthiocarbimide on the nitrogen buses, 011 the one lmnd, and ethylic alcohol on the other. In both cases, one-half of the thio- carbimidic bromine is elit~~inated in the forrr1 of hydrogen bromide ; the failure of the dcohol clerivative to desnlphurise uiider the influencc of lead, and especially of silver salts, points equally to the inclusion of sulp’lrui* as an integral member of an organic r i n g ; and on the principle already mentioned, regarcling the comparatively easy withdrawal of the end bromine atom, it is fairly sai‘e t o conclude that with respect to ring-closing also, the alcohol process runs a like course as in the case of the nitrogen bases (including ammonia).L4ccordingly, the interaction may be, at least provisionally, repre- sented as follows. the former product being the ethylic salt of P-bromotrimethylenc- I/ ?z.tliiocmbamic acid or, mom shortly, p-ethoxy-y-bromopent,hi- a I, ol i lie. This T V ~ S obtained from tlie thiocarbimide and pure inethylic alcohol by heating in a sealed tube for something over an hour a t 1 1 0 - 1 1 5 O ; there was slight pressure on opening it, and tl lit’tle lzjdrogen sulphide escaped, together with a fuming acid gas (HBr). The clear liquid product was evaporated to an oil, tvhich, on rubbingwith a glass rod; solidified ; the solid was dissolred in hot, dilute spirit, and, on cool- ing, separated as n nearly white, crystalline mass, melting, withoutOF D1BROMOPROPYLTHIOCARBIMIDE. 33 decomposition, at 95-96'.i n g with those required for the methoxypenthiazoline. 0.2011 gave 0.180 AgBr. 0.2005 ,, 0.2266 BaS04. S = 15.53. The methoxy-derivative is easily soluble in hot water and alcohol, rather sparingly in the cold. It is neither desulphurised by boiling with alkaline lead tartrate nor by ammoniacal silver nitrate; the latter precipitates a yellowish silver compound, whitening somewhat on standing in the dark, and, like the corresponding ethoxy-deriva- tive, very sensitive to even dull, white light. Figures were obtained on analysis agree- Br = 38.09.C5H,NSOBr requires Br = 38.05 ; S = 15.26 per cent. Dibromopropylthiocarbimide and excess of normal propylic alcohol were heated together as in the preceding experiment; on cooling, t,he originally turbid mixture* had given place to a crystalline solid pZus a clear liquid. There was no pressure on opening the tube, but the contents smelt of hydrogen sulphide, evolved a little acid vapour, and gave a strong acid reaction, due to hydrobrombic acid. After three recrystallisations from spirit, the solid was obtained in white, pyramidal crystals, melting, without decomposition, at 96-97' (corr.) . 0.2023 gave 0.2019 BaS04. pPropoxy- y- bromopenthiazoline is moderately soluble in hot water, easily in hot spirit, rather spariugly in cold. Like its congeners, it is not desulphurised, either by boiling with alkaline lead solution, or by neutral or ammoniacal silver nitrate. The latter reagent throws down a yellowish, amorphous derivative, changing, on boiling, to white; insoluble in dilute nitric acid, but decomposed by warming wit,h the concentrated acid, with formation of silver- bromide : this silver compound also (unless boiled) is sensitive to light in a high degree, and changes rapidly to mauve even in very dull daylight.I n a rude comparative experiment, similar to that with the ethoxy- compound, silver chloride was barely perceptibly altered in 10 minutes, the other silver derivative being then purple; in 15 niinutes, when a faint bluish coloration was beginning to show with the former, the latter had assumed a deep purple colour.S = 13.70. C,H,,NSOBr requires S = 13.47 per cent. The thiocarbimide dissolves tolerably freely in cold methylic alcohol, much less so in even anhydroua ethylic alcohol ; in propyiic alcohol it is ratner sparingly soluble. VOL. LXIX. D34 DIXON : THE CHEMISTRY A D D E N D U ~ ~ .-Imidazolines. The relations of the azoline compounds with which this paper is concerned are indicated by the typical formuke- Thiazoline. Penthiazoline. Imidazoline. Amongst thiazoline derivatives of the above type (apart from sub- stitukion in the carbon chain), p-hydrogen is alone replaceable ; the imidazolines, on the other hand, present two points where such replacement can occur, namely, CH(p) and NH(n). By the intro- duction into the latter, in either of these positions, oE sulphuretted groups, forms can result, in which the sulphur is contained without the ring, and which are isomeric both with one another and with the thiazolines.Thus, for example, taking a methyl derivative as start- ing point, and aseuming the substitution by a Rulphur radicle to occur solely in the ,mu-position, one thio and two imidic forms are possible, pMethylamidothitlzo1ine. p-Thiomel,hylimidazoline. n-Methylthioimidazoline. with two further possibilities if the n-hydrogen is in like manner replaced, or five isomers in all. The thio-compounds described below are imidazolines of the last-named, or n-, class, and were prepared - - CH2*NH from the imidic base p-methyiimidazoline," 1 >C*CH,. CH,--N p. U e t h y limidazo ly lphen y 1 thiourea, CGHO* N C ( SH) ON < ~ ~ 2 ~ ~ ~ ~ > Phenylt(hiocarbirnide and p-methylimidazoline were separately dis- solved in nearly anhydrous alcohol, and the soIutions mixed ; inter- action commenced at once, with evolution of heat, and white crystal- line matter quickly began to Repapate, the yield amounting to about $5 per cent.of that theoretically obtainable. On treating the product with boiling spirit, nearly t,he whole dissolved, leaving a trifling quantity of white, apparently amorphous, residue ; the filtrate from the latter, when cool, deposited thick, lustrous, white prisms, which, after again recrystallising, melted at 17.3-174' (corr.) to a yellowish c1ea.r liquid. Analysis for sulphur afforded the following result. * The material employed was a sample of " Lysidine," kindly supplied to me by Messre. Jeyes ; it is the 50 per cent. aqueous solution of the above base.35 0 f.’ DlBROMOPROPTLTHIOCARBIMIDE. 0,2359 gave 0.2184 BaS04. Cl,H,2X,S reqnires S = 14.62 per cent. The compound is, accordingly, an addition-product of base and thiocarbimide ; its formula is given above, and it may be named either p-methylimidazolylphenylthiouren, 013 12-pheuyl thiouramidu-p-methyl- imidazoline. It is slowly, and somewhat sparingly, soluble in hot water or alcohol, nearly insoluble in the cold, very sparingly in benzene, easily in cold, concentrated hydrochloric acid. Warm alkali dissolves it to a moderate extent, and the solution, when boiled with alkaline lead tartrate, is readily desulphurised ; the alcoholic solution, when mixed with animonixcal silver nitrate, blackens instantly. Tertiary thioureas, so far as my experience goes, are either not desulphurised at all by alkaline lead tartrate, or else only with great S = 14-33. difficulty, and after prolonged boiling ; this md the to be exceptional cases. following seem p-lllethy li?nidazolyl-orthotolylthiozi,.ea, CH3.CsH,*N:C( SH)*N<CH,.G3H,>N. C(CH )- This cornpound was prepared from the base and orthotolylthio- carbimide, as described for the phenylic homologue; the yield of crude product amounted to 92 per cent. of the theoretical. By crystallising from spirit alone, the product was slightly brownish, but, after further treatment with spirit and animal charcoal, it was obtained in fine, white crystals, melting at 159-159.5’ (corr.). The formula was checked by a sulphur determination. 0.2018 gave 0.202 BaS04. S = 13-76. C,,H1,,N,S requires S = 13.74 per cent. The compound is moderately soluble in water, freely so in boiling alcohol ; it resembles generally the preceding compound in properties, the potash-solu tion, moreover, is desulphurised moderately readily by boiling with alkaline lead tartrate. Ethyl- and aliyl-thiocarbimides also combined with the base on warming ; the products, in both cases, were sticky oils, which did not show any sign of solidification after long standing. Chemical Department, Queeia’s College, Cork.