122 DIXON AHD HAWTHORN^ : xIII.- The Action of Acid Chlorides on Thiozc?*ecxs. By AUGU~TUS EDWARD DIXON and JOHN HAWTHORNE. IT is well known t h a t alkylogens can unite with thiourea cnd with certain of i t s derivatives, in particular with those where univalent hydrocarbon radicles replacs one, two, or three hydrogen atoms of the nitrogenised groups in CS(NH,), or NH:C(SH)*NH,. Since, in the products, the alkyl, R, of the alkylogen, RX, is com- bined with the sulphur atom, the change, where a thiourea is concerned,THE ACTION OF ACID CHLORIDES ON THIOUREAS. 123 may most simply be explained by supposing the hydrogen of the group to be substituted by the alkyl group, thus : NH:C(NH,)*SH -+ RX =NH:C(NH2)*SR + HX. Being highly basic, the products retain the acid, KX, which is eliminated by alkali, leaving the free base or pseudothiourea isomeric with the ordinary thiourea containing the same radicle.Where union is effected between an alkylogen and a thiocarbamide, for example, PhNH*CS*NH( C,H,), a molecular change of the latter may be supposed to occur with formation of PhN:C(NH-C,H,)*SH (or its tautomeride), which then interacts as shown above. Probably an additive compound, such as C,H,-NH PhNH>C<zR, is first formed, and a hydrogen atom then withdrawn (as HX) from the NH group. If the sulphur atom is already couldned with a hydrocarbon group, the free base (but not its salts) can unite with the radicle of an alkylogen; in this case the radicle attaches itself to a nitrogen atom (see, for example, Bertram, Ber., 1892, 25, 48).Concerning the behaviour of acylogens with thioureas, our know- ledge is very limited. So long ago as 1875, Claus described (Ber., 1875, 8, 42) a molecular compound CH,N,S,C,H,OCI, obtained by acting on thiourea with acetyl chloride below 40"; the product is said to dissolve unchanged in cold alcohol, but to decompose when the solution is heated, without formation of acetylthiourea. Benzoyl chloride, on the other hand, acts on thiourea to produce benzoyl- thiourea, but only a t 120' (Pike, Be?.., 1873, 6, 755). If, however, acetyl or benzoyl chloride is slowly added t o a pyridine solution of thiocarbanilide, a monoacetgl or a dibenzoyl derivative is formed (Deninger, Ber., 1895, 28, 1332), in which, presumably, the acid group is directly attached to nitrogen.More recently it has been shown (Trans., 1903, 83, 565) that thiourea unites very readily with methyl or ethyl chlorocarbonate to form in each case an additive product, CSN,H,,RO*COCl, the hydro- chloride of a base, NH,*C(NH)*S*CO,R, and, moreover, that from monosubstituted thioureas similar compounds may be obtained ; in so far as these products contain the acyl or oxidised group united with 'the sulphur, they are strictly analogous t o the alkylogen derivatives already mentioned. Furthermore, it has been shown (Dixon, Trans. 1906, 89, 909) that a similar combination occurs when thiourea is brought into contact with phenyl chlorocarbonate, PhO*COCl, there being formed the hydrochloride of a base, NH,*C(NH)*S.C02Ph, isomeric with a non- basic carbophenoxythiourea, PhO*CO*NH* C(NH)*XH, described in the same paper.Certain differences are noticeable amongst these psmduderivatires ; K 2124 DIXON AND HAWTHORNE : for exaniple, when the radicle, R, of the group, CO*OR, united with the sulphur atom, is fatty, the product readily loses carbon dioxide, and the radicle thereupon attaches itself directly to sulphur, forming an alkylpseudobase, NH,*C(NH)*S*R, whereas if this radicle is aromatic a phenol results, thus : NH,*C(NH)*X*CO,Ph + H,O = PhOH + CO, + NH,.C(NH)*SH. I n no case, however, where a chlorocarbonate was united to thiourea or to a monosubstituted thiourea did the radicle CO-OR, or any part of it, become attached to nitrogen. The group CH,*CO is of course more highly electronegative than the group CH,*O*CO and its congeners ; nevertheless, in view of the above facts, a possible explanation of the phenomenon observed by Claus suggests itself, namely, that acetyl chloride may to some extent play the part of an alkylogen, its acetyl group becoming combined with the sulphur so as to yield the hydrochloride of a pseudo- or basic form : NH,*C(NH)*SH c Cl*CO*CH, = NH,*C(NH)*S*CO*CH,,HCl.To throw light on the subject and to investigate the power of com- bination between acplogens and thioureas in general, the present inquiry was commenced. This is still incomplete, but circumstances having now arisen which will preclude us from continuing it jointly, we have the honour to lay before the Society an account of the principal results so far attained. Acetyl Chlor.icle nnd l’hiowea.As Claus gives scarcely any description of his compound (Zoc. cit.) beyond the statement that it is highly unstable, being decomposed a t about 40°, yet dissolves unchanged in warm alcohol, it was necessary to re-examine the substance, When finely powdered thiourea was covered with acetyl chloride, union occurred, with evolution of heat and formation of a bulky, lustrous, deliquescent white powder, apparently crystalline ; this, how- ever, was always more or less impure, as on determiniug the chlorine and sulphur respectively, figures were obtained corresponding to mixtures of additive compound and thiourea, in which the former was present to the extent of from 94 to 96 per cent. The only analytical figure given by Claus (Zoc. cit.) is that for chlorine, namely, 22 1 per cent., whilst CSN,H,,C,H,OCl requires C1= 26-98 per cent.; from this it would appear probable that he w a ~ dealing with a mixture containing 4 per cent. of thiourea and 96 per cent. of additive compound. Attempts to combine thiourea, suspended in benzene, with acetyl chloride proved unsuccessful, but ultimately a pure product wasTEIE ACTION OF ACID CHLORIDES ON TEIIOUREAS. 125 obtained by adopting the following method. To a nearly saturated solution of thiourea in warm acetone, excess of acetyl chloride, diluted with the same solvent, was added gradually ; the precipitate, consist- ing of minute, soft, pearly plates, was collected, washed thoroughly with acetone, and dried, first, by exposure t o warm air and finally in a vacuum desiccator, The yield amounted to more than 90 per cent. of the theoretical.Prepared in this way, the compound was fairly stable ; it was odour- less when dry, and, although deliquescent, was not sufficiently so to preclude its being weighed in an open vessel for analysis. When heated in a narrow tube it melted sharply at 109' with copious effervescence. Chlorine was determined by fusing a weighed quantity with pure caustic soda, and subsequently with nitre ; the product, dissolved in water, was acidified by nitric acid and the mixture treated with excess of N/10 silver nitrate. After most of the nitrous acid had been expelled by boiling, the remainder was destroyed by urea, and the silver in solution determined by Volhardt's method, using N/10 ammonium thiocyanate. I n all the chlorine determinations given later, except where it is otherwise stated, a similar method was adopted : 0.309 required 19.8 C.C.N / l O silver nitrate ; C1= 22.75. C,H70N2C1S requires C1= 22.98 per cent. The substance dissolved very freely in cold water, yielding an acid solution which remained clear on treatment with N/lO caustic alkali, of which two equivalents were required for saturation ; no thiocyanate was present in the neutralised solution : 0.1545 required 19.95 C.C. NjlO NaOH; theory requires 20 C.C. It was considerably less soluble in absolute alcohol, the solution, when warmed with sulphuric acid, developing the odour of ethyl acetate. The aqueous solution yielded with silver nitrate a white precipitate, blackened instantly on the addition of ammonia, and was readily desulphurised by heating with an alkaline solution of lead, with formation of a brilliant mirror of galena.Moreover, the aqueous solution, when evaporated to a small bulk, yielded thiourea, which melted a t 171-172' and was identified by comparison with pure thiourea (m. p. 171-172°) and by the mixed melting point method. The decomposition by water proceeds, therefore, as shown by the equation : C,H70N2ClS + H20 = HCl + CH,*C02H + NH,*C(NH)*SH. Consistently with this, when the aqueous solution was treated in presence of dilute nitric acid with excess of A7/10 silver nitrate, the precipitate collected, and the residual silver determinediby Volhardt's126 DIXON AND HhWTAOlZWE : method, it was found that two molecules of silver salt were absorbed for each molecule of hydrochloride taken; of these two molecules, one combined with the hydrochloric acid and the other formed the molecular additive compound CSN2H,,AgN03, described by Reynolds (Trans., 1892, 61, 251).Claus’ observations were so far confirmed, that in this decomposition by water no sign mas detected of the formation of acetylthiourea ; in other words, the acetyl group is not combined with a nitrogen atom. Another experiment, made by treating one molecular proportion of the bydrochloride, dissolved in anhydrous alcohol, with an alcoholic solution containing one equivalent of sodium, gave a similar result, sodium chloride being precipitated, and the filtrate, by concentration, yielding crystals of ordinary thiourea. I t does not follow, howover, that the acetyl chloride is held merely by some attraction such as that whereby water of crystallisation is retained in certain compounds, for the fact that acetic acid as well as hydrochloric acid is formed on hydrolysis is equally consistent with the view that partition of the acetyl chloride occurs when it unites with thiourea, the chlorine becoming associated with hydrogen and the acetyl group attaching itself to the sulphur atom t o form NH,*C(NH)*S*CO*CH,. Snch a compound is basic in type, in the same sense as NH,*C(NH)*S*C,HS and its congeners, and since the presence of an acetyl group in place of the electropositive alkyl must greatly weaken the basic character, it is quite to be expected that combined hydrogen chloride, if present, should exert towards alkali the same acidity as if i t mere free.I n presence of water the combined acetyl group also displays full activity, as just shown, but considering that this group is eliminated by dilute alkali, even when united with nitrogen in ccb-disubstituted thiocarbamides, it was to be expected that it would very readily be separated from sulphur, for which element (as may be noticed in the case of thiolacetic acid) its affinity i s but feeble.* Moreover, t h a t the acylogen is not held as acetyl chloride of crystal- lisation seems proba.ble from the behaviour of the compound on heating, for if a t 109’ the acylogen simply passod off, thiourea alone should be left, whereas it will be shown later that this is not the case. Furthermore, that hydrogen chloride is held as such, combined in the molecule, may be inferred from the fact that it is possible, + Aectylthinuwn.p r t s vcry readily witli the ncetyl group. Thus, when 1 mole- cule of the piire substance, clissolvcd i n cold water, was m i x d with excess (2 inole- cules) nf &-/lo alkali, mid the solution allowed to stand for ,z ccrtaiii tiinc before titrating back with N/10 acid, WP, fonnd that, after five iriinntes’ standing, seven- tenths of 1 molecule of alkali had been a1 solbed a n d , after tcii minutes, exactly 1 inolecule.THE ACTION OF ACID CHLORIDES ON THIOUREAS, 12'7 as shown i n the following experiment, to replace it by a different acid. To a concentrated aqueous solution of the acetyl compound, slight excess of a saturated aqueous solution of picric acid was slowly added ; the resultant precipitate of minute, yellow, interlacing needles, when thoroughly washed with water and dried, meltod sharply a t 120'. Since the original compound is somewhat readily dissociated by water, with formation of thiourea (which gives no picrate), and the picrat.e itself does not escape hydrolysis, a poor yield was.obtained. The product was free from chlorine; its aqueous solution, when treated with silver nitrate followed by ammonia, gave a black precipitate, and was readily desulphurised by heating with an alkaline solution of lead. That tho acetyl group still remained i n combination was shown by dissolving a portion i n alcohol, acidifying with sulphuric acid, and warming, when the odour of ethyl acetate became distinct : 0.347 gave 0.2354 BaSO,.0.3292 ,, 56.8 C.C. moisf; nitrogen at 14' and 753 mm. N = 20.2. C,H,O,N,S requires S = 9.22 ; N = 20.17 per cent. S = 9.3. Now, a solution of acetylthioiirea in water yields no picrate with aqueous picric acid, it is somewhat sparingly soluble in cold water, and does not dissolve more readily in cold dilute hydrochloric acid ; when picric acid is added to the solution in the latter, no precipitate is formed unless the hydrochloric acid is sufficiently concentrated, in which case picric acid itself crystallises oub. Accordingly, the above substance is the p i c m t e of an acetyl-9-thio- urea, NH,*C(NH)*S*CO*C€I,, isomeric with the compound melting a t 165O, obtained by Nencki (Bey., 1873, 6, 599) from thiourea and acetic anhydride, and by Doran (Trans., 1905, 87, 341) from acetyl- thiocarbimide and ammonia.Action ofNeat.-A quantity was melted in a test tube immersed in a bath of sulphuric acid, the temperature of which mas kept between l l O o and 115'. The liquid bubbled freely, fumes of hydrogen chloride being evolved together with an odour recalling that of thioacetic acid ; it then became brown, and soon began to solidify. After some twenty minutes the effervescence had almost ceased, when the now solid residue was withdrawn from the bath and twice crystallised from boiling water, being obtained i n small, white needles free from chlorine and melting at 166" (corr.). It gave the usual thiocarbamidic reaction (desulphixrisation) with hot alkalino solution of lead, and when heated with alcohol and dilute sulphuric acid developed the odour of ethyl acetate.A specimen of pure acetylthiourea, attached t o the same thermometer, melted a t the same temperature, and when approximately equal weights of the two substances were mixed, t h e128 DISON AND HAWTHOltNE : melting point of the mixture was found to be unchanged; conse- quently the product was acetylthiouren. From the results of these various observations, we conclude that in the additive compound of acetyl chloride with thiourea the acetyl group is united directly to sulphur, the resultan5 molecule being basic, not alone in type, but also to some extent in character, and that when heat is applied the acetyl group migrates to a nitrogen atom so as to yield ordinary acetylthiourea : NH,*C(NH)*S*COiMe,HCl= HCl + N H: C(NH*COMe)*SH. It may be that this transference of the group named occurs as the primary effect of heat, in which case the rosultant acetylthiourea, being for all purposes non-basic, could not retain the hydrogen chloride previously held by the molecule of more basic configuration and character ; or possibly the hydrogen chloride, being feebly held by so weak a base, is parted reczdily from it by increase of temperature ; if so, the wandering of the acetyl group from sulphur t o nitrogen, for which i t has much more affinity, might occur readily enough. At present it is not easy to decide between these hypotheses, but in view of the observation previously mentioned, that withdrawal of the combined hydrogen chloride by means of a single equivalent of sodium ethoxide fails to produce acetylthiourea, there is at least some ground for believing that in this case the transfer of the acetyl group from sulphur to nitrogen is not conditioned independently of tempera- ture.The mechanism of this additive change being so far explained, we may now describe the results of experiments made with other thio- iireas and acylogens. Acetpl Chloride and PhenyMiourea. On mixing finel y-divided phenylthiourea, suspended in benzene, with excess of acetyl chloride, union occurred immediately without material rise of temperature, the product, the yield of which amounted to 93 per cent. of the theoretical for a molecular additive compound, being apparently crystalline. The same compound, but in a state of higher purity, was obtained by adding considerable excess of acetyl chloride to a tepid, concentrated solution of the thiourea in acetone.On cooling t'his mixture, lustrous, white, flattened prisms were deposited ; these were colourless when dry, and melted, if quickly heated, a t 94" with copious effervescence. The melting point is dependent on the duration of heating, becoming markedly lower if this is prolonged : 0.2305 gave 0.2345 BaSO,. 0.2305 required 20.2 C.C. N/10 AgNO,. 8 = 14.0. C,H,lON,CIS requires S = 13.88 ; C1= 15.40 per cent. C1= 15.7.THE ACTTON OF ACID CHLORlDES ON THIOUREAS. 129 When treated with sulphuric acid the solid additive cornpound effervesced, evolving fumes of hydrogen chloride. It was freely soluble in cold water, yielding an acid solution, from which, if not too dilute, prismatic crystah of plienylthiouren separated ; in the aqueons mother liquor both hydrochloric and acetic acids were present, but no thiocyanic acid.It is plain t h a t in the additive compound the acetyl group is not directly associated with llitrogen ; otherwise scetylphenylthiocarb- amide (either aa- or ab-) must be produced by hydrolysis on contact with water. The possibility of acetyl chloride being held in some sort of mechanical combination was negatived, just as in the case of the corresponding thiourea derivative, by the observation t h a t the hydro- chloric acid may be eliminated and a picrate of the " base '' may be obtained. Owing to the ready dissociation of the original compound by water, its aqueous solution must lie combined quickly with the picric acid ; otherwise little or no picrate is formed ; moreover, the picrate itself, although tolerably stable when once obtained, is dissociated by much water if this is present when combination takes place.The picrute was obtained in minute, lemon-yellow needles, becoming highly electrical on friction ; they were free from chlorine, almost insoluble in cold water (but dissociated by boiling with it.), and melted to a deep bromine-coloured liquid a t 187-1 8S0, with previous darkening. The substance was decomposed by warming with caustic potash, and hence was desulphurised when heated with an alkaline solution of lead : 0,4113 absorbed 19% C.C. N/lO barium chloride. S = 7.7. C,5H,,0,N,S requires S = 7-56 per cent. Ordinary substituted thioureas and thiocarbamides (that is, those i n which the substituting radicles are attached t o nitrogen) do not yield picrates readily, if at all.Thus, when ccb-acetylphenylthiocarb- amide, dissolved in acetone, was treated with an aqueous solution of picric acid, brilliant plates were deposited ; these, however, became pearly white on washing, and proved on examination t o be nothing more than the unaltered thiocarbamide, precipitated by t h e water used as solvent. Neither was it found possible t o combine picric acid with an-acetylphenylthiocarbamide, AcPhN*CS*NH,, dissolved in alcohol, acetone, or water. In a further experiment, made by leading a large excess of dry hydrogen chloride through a nearly saturated solution of the cca-compound in cold acetone, no hydrochloride was pre- cipitated, nor, after evaporating t h e solution t o a pasty consistence and removing some oily product (having an odour of mercaptan) by washing slightly with alcohol, did the solid residue contain any chlorine ; it consisted, in fact, of the original thiocarbamide, nearly pure.130 DlSON AND IIAWTHORKE : Phenylthiourea, in water or alcohol, gave no picrate ; moreover, when dry hydrogen chloride was led through its solution in acetone, no solid was produced ; instead, hydrogen sulphide escaped, and the residual liquid had a strong odour of mercaptan.So far as may be judged from these experiments, i t seems a justifiable conclusion that mono- and di-substituted thioureas or thiocaybamides are almost devoid of basic characters, but that a molecule having the configuration NH:C(SH)*NH, becomes basic when an organic radicle is substituted for the SH hydrogen, and does so independently of whether the sub- stituting radicle is itself electropositive or electronegative, this cbaracter affecting merely the strength of the resultant base.For the various reasons set forth above, we infer that the additive product of acetyl chloride and phenylthiouren is tt definite chemical compound, namely, the hydrochloride of an acetylated phenylthiourea, in which the acetyl group is joined to the rest of the molecule through the sulphur atom ; otherwise, iminoacet~ylthiolplienylcarbamic acid, or, according t o the nomenclature suggested by one of us (Trans., 1895, 6'7, 5 64), acetyl- q-v-phenylthiourea, C6H,NH*C(NH) * S*CO*CH,.This represents a typical basic form 01- peeudothiouren, analogous to the known derivatives, having distinctly positive radicles attached to the sulphur at'om. AS a rule, members of the latter class are hydro- lysed more o r less readily by alkali, that is, as soon as the combined acid is withdrawn, but in such cases the sulphur atom passes off in combination with the alkyl group as meroaptan. Action of HecLt.-A quantity of the hydrochloride contained in a test- tube immersed in a bath of sulphuric acid mas heated slightly above its melting point until the effervescence (due principally to the escape of hydrogen chloride) ceased; the liquid, which had an odour of thio- acetic acid, now gradually solidified, and the residue, on crystallisation from boiling water, separated in glistening leaves melting at 170-171° and consisting of ab-acetylphenylthiocarbamide. Heat, therefore, just as in the case of the corresponding thiourea derivative, brings about a movement of the acetyl group from the sulphur to one of the nitrogen atoms, whilst the molecule changes in configuration from the imino- tbiocarbamic to the thiocarbamidic form.Action of AZ?caZi.-Attempts to neutralise the dilute aqueous solu- tion with standard alkali failed to give concordant results owing to the difficulty of attaining a definite end point ; it was noticed, however, that if the alkali was run in quickly, before phenylthiourea had time to separate from tho aqueous solution, and tho now turbid mixture mas cleared by warming, the liquid, as it cooled, deposited first brilliant plates and then phenylthiourea in needles or prisms.When tho solid was added directly to a slight excess (about 2 molecules) of N/10 alkali and the mixture warmed uptil it became clear, the solution, ODTHE ACTION OF ACID CHLORIDES ON THIOUREAS, 131 cooling, deposited only the brilliant spangles ; these were devoid of bitter taste and consisted of ab-acetylpbenylthiocarbamide (m. p. This transfer of the acetyl group to nitrogen seems to take place only with the ready formed hydrochloride, for when phenylthiourea was crystallised from solutions containing sodium acetate and chloride, or acetic and hydrochloric acids, or from the latter mixture after neutral- isation by alkali, no sign could be detected of the production of nb-ace tylphenyl thiocarbamid e.Moreover, when phenyl thiourea, dis- solved in weak caustic alkali (2 molecules), was treated with excess of acetyl chloride, the solution, on cooling, deposited nothing but un- changed phenylthiourea. The symmetrical or ab-thiocarbsmide, then, is formed on removal of the combined hydrogen chloride, whether this be effected by heat or by the action of caustic alkali in excess. That a transfer of the acyl radicle from sulphur to nitrogen should take place under the influence of heat is not surprising, a number of cases having now been observed of the movement of a n acid group from one nitrogen atom to another within the thiourea molecule. Thus, for instance, Wheeler has shown (Amer. Chetu. J., 1902, 27, 270) that aa-acetylphenylthiocarbamicle, AcPhN*CS*NH, (compare Hugershoff, Ber., 1899, 32, 3649), is changed by fusion into the nb-compound, AcNH*CS*NHPh, and Johnson and Jamieson (ibid., 1906, 35, 297), that various acyl-$-thioureas, for example, Ez,N*C(SAle):NH, undergo a like change, one acyl group moving from its original attachment in the amino-position and becoming united a t the imino-group.I n the case of our acylphenyl derivative, heat appears to determine that change whereby the most stable configuration is produced. On the other hand, that che presence of dilute caustic alkali should lead under such mild conditions to the same ultimate result was both unexpected and puzzling. Water, as previously stated, removes the acetyl group from the hydrochloride, thereby producing phenylthiouren ; and since phenylthiourea, in contact with alkali, resists acetylation by acetyl chloride, the final change, however accomplished, could scarcely be accounted €or by initial hydrolysis of the additive compound into its original constituents.Presumably, therefore, the formation of n6-acetylphenylthiocarbamide must be explained either by some change occurring in the $-base itself, when liberated from its hydrochloride, or else through some influence exerted on the former by the alkali. This could obviously be tested by removing the combined hydro- chloric acid under conditions such as to preclude the resultant organic product from exposure to tho action of free alkali, and noting if the product was still the same. The first experiment in this direction mas conducted by adding 170-1 7 1").132 DIXON AND HAWTHORNE : gradually to a solution of the hydrochloride (1 molecule) in nearly anhydrous alcohol, one equivalent of sodium previously dissolved in a separate portion of the same solvent, any material rise of temperature being prevented.The precipitate of sodium chloride was separated, and the clear, strongly acid filtrate allowed t o evaporate spontaneously ; t h i s liquid gave no reaction for thiocyanic acid, thereby differing from solutions which had been treated with alkali to neutrality or in excess. On concentration, white crystals were deposited, free from chlorine, somewhat sparingly soluble in water, and containing both thiocarb- amidic sulphur and the acetyl group. Our hope that dissociation of the original compound would be avoided by the use of strong (99 per cent.) alcohol instead of water was not realised, for the product melted very indistinctly at 129-132" and had a bitter taste, which proved t o be due to its containing a very appreciable quantity of phenylthiourea.By means of cold chloroform, in which it is almost insoluble, the latter mas separated ; the chloroform was then evaporated, and the residue, when crystallised from dilute alcohol, obtained in long, pointed prisms, melting, if rapidly heated, at 139O. The alcoholic solution darkened gradually when mixed with neutral silver nitrate or at once on treat- ment with the nmmoniacal nitrate ; desulphurisation occurred readily on boiling with an alkaline solution of lead. The presence of an ncetyl group was proved by warming the substance with alcohol and sulphiiric acid, when the odour of ethyl acetate became distinct.I n the second experiment, the hydrochloride, dissolved as before in strong alcohol, was treated with pure, dry calcium carbonate; when the effervescence had ceased, the unattacked carbonate was removed by filtration, and on evaporating the filtrate to a small bulk a t the atmo- spheric temperature, precisely the same results were obtained as when the hydrochloric acid was eliminated by means of sodium ethoxide, the purified end product resembling in every respect t h e Substance melting a t 139' previously described. A sulphur determination gave the figures required for the free " base," PhNH*C(NH)*S*CO*CH,, or for its isomeride, ab-acetyl- phenyl t hiocarbamide, PhNH CS-NH *CO C H, : 0.194 absorbed 20.1 C.C.X / l O barium chloride. S= 16.6. C9H,,0N,S requires S = 16.50 per cent. This substance, however, could not be ah-acetylphenylthiocarbamide, which crystallises in brilliant leaves melting a t 170-171", neither, on account of its comparatively high melting point, could it well be the '' base '' formulated above, since a compound having t.he structure of the latter might be expected to melt at about 50". h t as the phenyl group is undoubtedly attached to nitrogen, t h e only remaining isomeride proper to this series is a thiourea or thiocarbamide havingTHE ACTION OF ACID CHLORIDES ON TIIIOUREAS. 133 both the phenyl group and the acetyl group attached to the same nitro- gen atom, that is, assuming the compound to be a thiocarbamide, AcPhN.CS*NH,.That the substance in question had this composion was made certain by the following observations : (1) when heated a t or slightly above its melting point it presently resolidified, being converted by the fusion into the isomeric ab-acetplphenylthiocarbamide ; (2) when dissolved in weak aqueous caustic alkali it yielded the last-named symmetrical compound; (3) when treated with strong alkali the acidified mixture gave an intense reaction for thiocyanic acid. These are the properties of the substance obtained by Hugershoff (Zoc. cit.) by dissolving phenyl- thiourea in acetic anhydride at 80’ ; to this he incorrectly assigned the symmetrical or ab-formula, an error subsequently corrected by Wheeler (Zoc. c i t .) , who placed beyond doubt the fact of its being an aa-deriv- ative. The chemical identity of our product with that of Hugershoff was further established by the observation that a specimen of his com- pound, prepared according to his directions, melted, within a degree, at the same temperature as ours, and when the two were mixed in equal proportions the melting point underwent no perceptible change. These experiments show that caustic alkali determines by its presence a change in the product initially formed by the removal of the coni- bined hydrochloric acid from our hydrochloride, since if this removal is effected with a limited quantity of alkali, or in the absence of any alkaline substance, the product is not identical, but isomeric with that obtained in presence of excess of alkali, and the former product, when isolated and then brought into contact with free alkali of a certain strength, is changed into the latter [see (2), above].(i) Phenylthiourea is not acetylated, in presence of alkali, by treat- ment with acetyl chloride, but (ii) Acetyl chloride unites spontaneously with phenylthiourea to form the hydrochloride of a feeble ‘( base,” acetyl-$-v-phenylthiourea, PhNH*C(NH)*SAc. (iii) This base, when liberated in alcoholic solution, undergoes isomeric change, the acetyl group migrating, a t the ordinary tempera- ture, to the phenylated nitrogen atom to form PhAcN*CS*NH,. (iv) The last product, if heated or if brought into contact with dilute alkali, undergoes f urtiher isomeric change, the phenyl and acetyl groups now becoming att,ached to different nitrogen atoms, and thus yielding AcNH- CS N HPh.(v) The hydrochloride (ii) changes by melting, with loss of hydrogen134 DXXON AND BAWTHORNE : chloride, into AcNH*CS*NHPh’; by solution in cold water it yields phenylthiourea. This succession of movements of the acetyl group is exhaustive ; the acetyl, combined at first with sulphur, can be driven to the phenyl- amino-group, and thence, by an easy transition, to the remaining nitrogen atom; in other words, it may occupy in succession, and in a given existing plienyl thiourea molecule, every place where, according t o our present notions, i t could conceivably be attached. No less than six distinct forms of acetylphenylthiourea may be formulated, namely, AcPhN*CS*NH,, PhNHICS-NHAc, NHPh.C(NAc)*SH, NHPh*C(NH).SAc, NHAc’C(NH)-SPh, and AcPhN*C(NH)*SH ; these are all essentially different, and do not include mere tautomeric variants (for example, PhNH6C(NH)*8Ac +X NH,*C(NPh)hSAc), and.there doea not a t present seem to be any valid reason for supposing that any one of them is incapable of existence. Nevertheless, in view of the free mobility of both hydrogen and acetyl in this very “ plastic ” molecule, it will doubtless be no easy matter to prepnre and to keep the three forms which still remain unknown. Of these, one contains the aryl group combined with the sulphur atom; it may be noted in passing that all attempts hitherto made to fix the aryl group in this +-form have been unsuccessful (see Trans., 1906, 89, 909).I n reference to Hugershoff’s observation that acc-thiocarbamides containing one acyl and one aryl group yield thiocyanic acid on treat- ment with strong alkali, AcPhN*CS*NH, + KOH = AcNHPh -I- IiSCN + H,O, a number of experiments wore made in order t o learn if this property is peculiar to members of that class. The compounds examined included (i) monosubstituted thioareas, both acvl and alkyl ; (ii) di- substituted thiocarbamides, ua- and ab- of the alkyl or aryl series, or of mixed varieties, and ccb-derivatives of the acyl-alkyl or acyl-aryl class ; (iii) trisubstituted thioureczs derived from acyl, aryl, and alkyl thiocarbimides by combination with secoiidsry bases of various sorts. Without giving a detailed list of a11 the substances employed, it may suffice to say that riot one of them afforded the slightest reaction for thiocyariic acid when treated with alkali followed by hydrochloric acid and ferric chloride.On the otlier hand, our acetyl-$-v-phenylthiourea hydrochloride reacted readily for it, owing to partial conversion into Hugershoff’s compound, and the same is true of the various additive compounds from acylogens and monosubstituted thioureas, which we describe in the following pages. This test, therefore, appears to be a characteristic one for compounds of the class named. It is stated above that acc-acetylphenylthiocarbamide melted a t 1 3 9 O , the temperature recorded by Hugershoff (Eoc. cit.) for this compound. Nevertheless, we had a t first much difficulty in reconciling the meltingTHE ACTION OF ACID CHLORIDES ON THIOUKEAS.135 point of our product with that given by him, or indeed, in arriving at any really definite melting point. This was ultimately found to be due to the slowness with which we conducted the heating, for on working rapidly, the substance melted a t 139". At our own slow rate, not only did the substance melt at temperatures varying in different determinations from about 133O to 137", but also, when a specimen of Hugershoff's compound (prepared from acetic anhydride according to his directions) was heated a t the same time as ours, it showed a like behaviour. Moreover, the substance, if heated for some time a t 129") gradually softened, but did not liquefy, and on raising the temperature, no further change could be observed until between 150" and 160' or even a trifle higher, when i t melted to a clear liquid.It appeared doubtful, therefore, whether the compound really possessed a true melting point, for, since a t a temperature many degrees below that of liquefaction, considerable change may occur within not many minutes (into the ab-compound), it was t o be expected that at about 139O this change mould be very rapid, Such, in fact, is the case, for if the compound, heated quickly, was removed the instant it liquefied and cooled a t once, the now solid material, when put back into the appardtus, even at 145", no longer melted, thus showing that the process of conversion had gone far. The solidified product also, when tested with alkali, gave but a trifling reaction for thiocyanic acid.Now although the rate of change is very rapid indeed a t the liquefy- ing point, it is considerably retarded at temperatures not far removed from this, and hence it seemed probable that the liquefaction was con- ditioned, not by the melting of the na-compound, with subsequent change to the ab-form, but through the production of a mixture of both in proportions continuously varying, so that a t some particular moment the most fusible mixture would result ; in which case, if the tempera- ture was high enough, it must melt. Two narrow tubes, as nearly equal in all respects as possible, were charged to the same depth with two fine powders, one consisting of the cba-corn- pound, the other of a substitnee melting at 141-142'. The bath being kept steadily a t 143", both tubes were immersed simultaneously and attached to the same thermometer. I n ten seconds, the substance of higher melting point liquefied suddenly; after a total interval of forty-five seconds, the nu-compound, which meanwhile had scarcely changed in appearance, also suddenly melted.Ten seconds, therefore, were required for the establishment inside the tubes of a temperature not less than 141-142", which is above the maximum '' melting point " of the un-derivative j presumably the remaining thir ty-five seconds were occupied, not in melting it, but in effecting such a relative amount The following experiment peems to confirm this view.136 DIXON AND EIAWTI1OIINE : of conversion as to produce a mixture fusible a t the temperature already attained within the tube.Acetyl Chloride and o-Tolylthiourea. When to a saturated solution of o-tolylthiourea in cold acetone rather more than the calculated quantity of ncetyl chloride was added, and the mixture cooled, a crystalline, white solid was soon deposited, melting a t 96" with effervescence ; the same product was obtained, with evolution of heat, by mixing the constituents in presence of benzene, the latter method giving 95 per cent. of the theoretical yield for a molecular additive compound : 0.2445 required 9.8 C.C. N/lO silver nitrate. C1=14.2. C,oH,,0N2S,HCl requires CI = 14.50 per cent. The pure substance dissolved readily in cold water, yielding an acid solution, from which in a short time white crystals of o-tolylthiourea began to separate ; in the solution both hydrochloric and acetic acids were present, but no thiocyanic acid.When heated slightly beyond its melting point i t effervesced freely, evolving fumes of hydrogen chloride ; the liquid now resolidified, and the solid, when recrystallised from boiling alcohol, in which i t was rather sparingly soluble, formed brilliant prisms. The cold alcoholic solution gave immediately, with aqueous silver nitrate, a black pre- cipitate, and had an odour of ethyl acetate when warmed with . sulphuric acid. The melting point of this new product, 1 S2- 1 8 3 O , was less than a degree below that of a specimen of pure ab-acetyl-o-tolyl- thiocarbamide, and when equal weights of the two were mixed, the melting point of the mixture was still 182-183". I n this case, therefore, as in that of the phenylic homologue, mith- drawal of the combined hydrochloric acid by heat is associated with movement of the acetyl group from sulphur to nitrogen, acetjl-$-o-tolyl- thiourea hydrochloride changing to ab-acetyl-o-tolylthiocarbamide, NHPh*C(NH)*SAc,HCl= HC1+ NHPh*CS.NHAc.The solution of the hydrochloride, i f mixed without delay with aqueous picric acid, gave a yellow picrate in minute needles. By treating the hydrochloride, dissolved in absolute alcohol, with one equivalent of sodium ethoxide, separating the precipitated sodium chloride, and treating the crystalline residue left by evaporation of the alcohol, as described in the corresponding experiment with the phenyl derivative, white prisms were obtained melting a t 139.5". They were free froin chlorine, gave the usual desulphurisation reactions with lead and silver salts, and when treated with strong alkali yielded a pasty mass, redcting abundantly for tliiocyitnic acid. TI& product wdsTHE ACTION OF ACID CHLORIDES ox THIOUREAS.137 obviously .Hugershoff's aa-wetyl-o-tolylthiocarbamide, melting point 140' : 0.208 gave 0.3318 BaSO,. S= 15.3. C,H,,ON,S requires S = 15-38 per cent. It may here be noted that the hydrochloride, both on heating and on treatment with sodium ethoxide, had an odour of thioacetic acid, owing probably to partial decomposition of the I' base " when liberated : U,H,*NH*C(NH)*S*(:OIe = C,H,*N:C:NH + COMe*SH. Acetyl Chloride ccncl p-Tolylthiourecc. p-Tolylthiourea is so sparingly soluble in acetone that the precipita- tion method is not well suited for preparing the additive compound except on a small scale ; when obtained in this way it formed lanceolate prisms.It was prepared in larger quantity by mixing together the finely-powdered thiourea and acetyl chloride, whereupon sufficient heat was evolved to evaporate R portion of the latter ; the mixture was then ground up in a mortar, the solid collected a t the pump, washed with light petroleum, and dried in a vacuum, The melting point was 102-103' with much effervwcence, and the yield amounted to 80 per cent. of the theoretical : 0.2445 required 9.8 C.C. NI10 silver nitrate. C1= 14.2. C,,H,,ON2S,HC1 requires C1= 14-50 per cent. If not contaminated with unchanged p-tolylthiourea (which may be extracted by repeatedly shaking the powder with acetone) the hydro- chloride dissolved readily in water, the solution quickly becoming turbid from the separation of p-tolylthiourea ; no thiocyanic acid was contained in the liquor. The thiourea melted at 181'; Staats (Ber., 1880, 13, 136) gives 1S2'.When the hydrochloride was treated with cold aqueous caustic alkali the mixtur; reacted readily for thiocyanic acid, thereby showing the formation of the ua-acetyl-pt,olylthiocar bamide. Acelyl Chloride uncl ab- Diphen ylthiocarbamicle. According to Deninger (Bey., 1895, 28, 1322), thiocarbanilide cannot be acotylated by the Schotten-Baumann method ; similarly, we found (see p. 131) that phenylthiourea is not acetylated by acetyl chloride in presence of caustic alkali. When, however, diphenylthiocarbamide, suspended in benzene, was mixed with excess of acetyl Chloride, the solid gradually changed to a clear, yellow oil, which crystallised on standing.The product, when powdered and washed thoroughly with benzene, was a white powder, fuming in moist air, and having an odour of hydrochloric acid; i b VOL. XCI. L138 DIXON AND HAWTHORNE : was apparently insoluble in water, and began to decompose, with effervescence, at about 106" : 0.3065 absorbed 9.7 C.C. N/lO silver nitrate. The yield amounted to only 50 per cent. of that calculated for R molecular additive compound. Dilute alkali withdrew all the com- bined acid, leaving thiocarbanilide. The compound of diphenylthiocarbamide with acetyl chloride is dis- tinctly less stable than that of monophenylthiourea, since the former evolves visible fumes when exposed to moist air, whilst the latter does not, Acetyl Chloride and di-o-170lyZthiocccrbamide. On mixing these together, using excess of acetyl chloride, a yellow liquid was formed; this, when treated with light petroleum, gave a paste which presently solidified.The product, when broken up and dried, had little odour, and melted sharply, with copious effervescence, at 135-136O : C= 11.24. C15H150N,CIS requires C1= 11.58 per cent. 0.3345 required 9.7 C.C. NjlO silver nitrate. 0,3345 gave 0.262 BaSO,. C1= 10.3. S = 10.8. C17Hl,0N,C1S requires C1= 10.6 ; S = 10.45 per cent, Benxoyl Chloride and Thiourea. These substances combined at once in presence of benzene to form a white powder melting a t about 116' ; the yield was poor, amounting to only 54 per cent.of the theoretical : 0.2165 required 9.7 C.C. N/10 silver nitrate. C1= 15.9. C,H,ON,S,HCl requires C1= 16.4 per cent. When the combined acid was removed by adding calcium carbonate t o a solution of the hydrochloride in 99 per cent. alcohol, no benzoyl- thiourea was found in the filtrate, but ordinary thiourea instead ; this behaviour is similar to that observed in the case of the corresponding acetyl compound (see p. 126) when treated with a limited amount of sodium ethoxide. Reference has already been made to Pike's observation that thio- urea and benzoyl chloride i f heated to 120' yield benzoylthiourea; an attempt was therefore made to ascertain whether the latter substance would be produced by heating the above additive compound, A quantity was melted in a test-tube, immersed in a sulphuric acid bath : hydrogen chloride and a little hydrogen sulphide were evolved, and soon the mass solidified; the temperatiire was now raised to 126' to complete the action, and after some five minutes the tube was removed and cooled, The product, nearly insoluble in cold water,THE ACTION OF ACID CHLORIDES ON THIOITREAS.139 was boiled with a large quantity of this solvent and the solution filtered from a trace of pasty solid. The filtrate crystallised imme- diately, giving small, vitreous prisms which dissolved readily in cold alkali; this solution, when nixed with a lead salt and heated, was desulphurised with formar,ion of a speculum of lead sulphide. The solid had an intensely bitter taste; it was easily soluble in hot alcohol, somewhat sparingly so in cold, and the solution, when warmed with sulphuric acid, had an odour of ethyl benzoate. It crystallised from boiling water in needles melting a t 169-170°, and hence con- sisted of benzoylthiourea, which melts according to Pike (Zoc.cit.) at There can be no doubt as to the position of the acyl group in benzoylthiourea, since lsliquel has shown (Ann, Chim. Phys., 1877, [v], 11, 313) that it is produced from benzoylthiocarbirnide and ammonia. By heating tho additive compound, therefore, it loses hydrogen chloride, and the benzoyl group thereupon transfers itself from the sulphur to one of the nitrogen atoms. Benzoyl chloride gave with phenylthiourea a tenacious paste ; with o-tolylthiourea it yielded a solid additive product which was not examined in detail, 169 -1 70'.Benxoyl Chloride and p- ToZyZtAiourea. By direct union of these constituents a soft, white powder, decom- posing with effervescence at 137--138O, was obtained in nearly quantitative yield. The substance melted in boiling water, and the filtrate, on cooling, deposited long needles, which mere, apparently, benzoic acid. It was desulphurised by alkaline solution of lead, and when warmed with alcohol and sulphuric acid gave the odour of ethyl benzoate : 0.613 required 20.1 C.C. , ! l o silver nitrate. C1= 11.6. C,,H,,ON,S,HCl requires C1= 11 -5 per cent. On warming the substance with caustic alkali and treating the resulting mixture with hydrocliloric acid, followed by ferric chloride, a blood-red coloration appeared ; as this reaction points in the case of acetyl derivatives of monosubstituted thioureas to the presence of an aa-disubstitution compound, it seemed probable that the thiocyanic acid yielded by the benzoyl derivative had a like origin.Hugershoff does not appear to have inquired whether the benzoyl radicle is similar to the acetyl group as regards the power of forming labile thiocarb- amides ; me therefore conducted the following experiment to learn whether, from our supposed bsnzoyl-$-v-tolylthiourea hydrochloride, C,H?*NH*C(NH)*S*COPh,HCL, an acc-derivative could be produced, PhCO*N( C,H,)*CS*NH, [or perhaps PhCO*N( C,H,) C( NH) *SH], con- L 2140 DIXON AND HAWTHORNE : vertible in turn into the known nb-benzoyl-p-tolylthiocarbamide, PhCO*NH*CS*NH*C,H,. From a quantity of the freshly-prepared hydrochloride, disscjlved in cold anhydrous alcohol, the combined acid was withdrawn by excess of calcium carbonate, the filtered liquor mas then evaporated, and the solid residue purified by chloroform i n the manner previously described for the corresponding acetylphenyl derivative.The viscid solid, left by evaporation of the chloroform, was treated with pure ether, which separated it into (1) a residue, giving no thiocyanic reaction with alkali, and melting after one recrystallisation from alcohol a t 167-15S0, and after a second a t 158-159P (uncorr.), and (2) a filtrate. The latter on evaporation left a solid, reacting with alkali for thiocyanic acid. On crystnllising this from alcohcl, needles melting a t 157-158' were obtained, resembling the preceding and giving no thiocyanic reaction; moreover, the liquor from these now reacted but faintly with alkali for thiocyanic acid. The substance melting a t 158-159" proved to be a thiocarbamide containing the benzoyl group, and since it gave no thiocyanic acid, was presumably not the acb- but the ah-compound, namely, benzoyl-p- tolglthiocarbamide.This melting point is materially lower than that recorded by Miqnel (Zoc. it.)^ namely, 165'. On preparing Miquel's compound, however, by his own method (from benzoylthiocarbimide and p-toluidine), and recrystallising from alcohol until the melting point became constant, a substance was obtained identical in appear- ance and properties with that described above, and melting sharply at the same temperature, 158-1 59' (uncorr.).Bliquel's figure, there- fore, is somewhat too high. So far as may be judged from the single set of experiments detailed above, it would seem that the hydrochloride of benzoyl-$-p-tolylthio- urea can yield, by the elimination of its combined hydrochloric acid, uu-benzoyl-p-tolylthiocarbamide, but that the latter, being rather unstable, is resolved by successive recrystallisations from hot alcohol into the stable or tcb-isomeride, PhCO*NH-CS*NH*C,H,. Benxogll Chloride and Thiocarbani title. When brought together in presence of benzene these substances did not appear t o unite, but on mixing them together directly there was vigorous combination, with considerable evolution of heat. After being washed with benzene, Followed by light petroleum, the product was slightly yellow, and decomposed a t 10S-109°.Yield, 90 per cent. of the theoretical. This substance was distinctly unstable, the loss of material being perceptible during the process of weighing out for analysisTHE ACTION OF ACID CHI,ORIDE;S ON THIOUREA~. 141 0.3784 required 9.5 C.C. N/10 silver nitrate. C,,H,,ON,ClS requires Cl = 9.6 per cent. I n addition t o the experiments described above, the following substances were examined as t o their power of combination when mixed together : C1=8*9. Acetyl chloride and ab-acetylphenylthiocarbamide. Acetyl chloride and ab-benzoyl-o-tolylthiocarbamide. Ethyl chlorocarbona te and tcb-benzoyl-o-tolylthiocarbatnide. Ace ty 1 chloride and carboxy-o- t oly 1 thiourea, C,H7*C0,*NH* C(NH) *SH.Ethyl chlorocarbonate and earboxy-o-tolylthiourea, C7H7*C0,*NH*C(NH)*SH. o-Tolylchlorocarbonate and carboxy-o-tolylthiourea, C,H,* CO,*NH* C(NH)*SH. No action seemed to occur ; in each case tho thiocarbamide employed was recovered and its melting point verified. The presence of an electronegative group in a thiourea appears, therefore, to paralyse, or at least greatly to hinder, its power of combining with acylogens. I n all the preceding cases of combination the radicle R*CO* of the acid chloride employed was highly electronegative in character. With the chloride R*O*COCl, Lhe radicle of which is much more electro- positive, the products as a rule were comparatively stable. Thus, ethyl and methyl chlorocarbonates, when united with thiourea, gave compounds which did not appear to be dissociated by cold water to any material extent, whilst the compounds of methyl chlorocarbonate with phenyl-, o-tolyl-, and p-tolyl-thiourea respectively, were not dis- sociated at all; in fact, when treated with caustic alkali, their solu- tions yielded the corresponding bases, for example, PhNH* C( NH) * S.CO* OMe (Dixon, Trans., 1903, 83, 550).The hydrochloride obtained from thiourea and phenyl chlorocnrbonate (Dixon, Trans., 1906, 89, 909) is a well-marked salt; the kiase, however, could not be isolated by means of alkali, probably because of the feebly positive character of the radicle PhOaCO. Now, if it be true that the general stability of these combinations, and in particular their power of resisting the hydrolysing action of water, depends mainly on the nature of the acyl radicle united with the sulphur atom; it should he possible, by employing acylogens con- taining groups less strongly negative than acetyl and its congeners, t o obtain hydrochlorides proportionately more stable in the sense named than the above.We have tested this conjecture experi- mentally, anci so far as may be judged from the following results it appears on the whole to be substantiated. Of acylogens suitable for the purpose, no abundant choice was avail-142 DIXON AND HAWTHORNE : able ; we selected for investigation the chlorides of certain substituted carbamic acids, R,N*CO*OH, since the group R,N*CO* is distinctly less negative than the acetyl group. Diphen~lcarbccmic Chlode and Thiourea.liolecular proportions were thoroughly mixed and heated in an oil- bath until liquefaction commenced and a trace of effervescence set in ; a t this stage the temperature of the bath was 134'. When cool, the solid residue was powdered, boiled with acetone (which dissolves it to a very appreciable extent), filtered, and washed with more cold acetone; the yield of nearly white powder reached only 55 per cent. of the amount calculated from the equation but a further considerable quantity separated from the acetone solution, The hydrochloride was moderately easily soluble in warm water ; it crystallised from this solvent in short, vitreous prisms having a some- what greasy lustre and melting, with copious effervescence, at 182- 183' (uncorr.) : Ph,N COCl + NH,*C( NH) S H = NH, C( NH) S C 0 NPh,,HCl , 0.3075 required 9.9 C.C.NjlO silver nitrate ; C1= 11.4. C,,HI,ON,CIS requires C1= 11.54 per cent. Water, therefore, did not destroy the salt, The aqueous solution was neutral to litmus; when treated in the cold with rather less than one equivalent of normal alkali, it gave a crystalline white precipitate, which consisted, not of the expected free base, but of diphenylamine; cold alkaline solution of lead gave with the liquor a black precipitate, thus showing that the disruption of the molecule had gone far. When treated with dilute nitric acid or with a solution of potassium nitrate, the aqueous solution of the hydrochloride yielded a crystalline, white precipitate ; the latter, by recrystallisation from boiling water, was obtained in clear, colourless, ffattened, oblique prisms free from chlorine, darkening a t 170°, and frothing a t 176-177".That this substance was a nitrate was shown both by the ferrous sulphate test and by the fact that on warming it with dilute alkali and then treating the mixture with sulphuric acid, a splendid indigo-blue coloration appeared ; this is explained by the liberation by the alkali of diphenyl- amine, which in presence of a nitrate and sulphuric acid gives the well- known blue diphenylamine reaction : 0.334 required 20.5 C.C. N/lO barinm chloride ; 8 = 9.8. C,,H,,ON,S,HNO, requires S = 9.58 per cent. Aqueous picric acid, when mixed with a cold saturated solution ofTHE ACTION OF ACID CHLORIDES ON TMIOUREAS. 143 the nitrate, gave the corresponding picrate in minute, yellow needles rather sparingly soluble in boiling water. Pl~en?lZmethyZctarbanaic Chloride and Thiourea.Combination occurred on the water-bath, and after treatment with acetone, as described in the preceding case, the yield of white solid reached 63 per cent. of tho theoretical; the product melted with effervescence a t about 1 7 5 O , and when dissolved in water gave a neutral solution. Chlorine was determined by acidifying the aqueous solution with nitric acid, filtering, and treating the filtrate according to Volhardt's method : 0.2445 required 9.65 C.C. N/lO silver nitrate ; C1= 14.0. C,H,,ON,ClS requires C1= 14.47 per cent. On adding dilute nitric acid (or potassium nitrate) to the aqueous solution the nitrate was precipitated, which crystallised from boiling water in white, lustrous needles, becoming yellow at 155O, and melt- ing with effervescence to a black liquid at 162".The solution was desulphurised by boiling with ammoniacal silver nitrate, and gave the usual reactions for nitric acid : 0.272 required 20.45 C.C. -@/lo barium chloride ; S = 12.0. The picrate crystallised from water in small, yellow needles melting C,H1lON,S,HNO, requires S = 11.76 per cent. with effervescence a t 174-175". Pheny~et~y~car6~6~n~c Chloride and Thiourea. Prepared by warming the constituents on the water-bath and treating as previously described; the nearly white product melted at 160° with effervescence : 0.2595 absorbed 9.9 C.C. ,V/10 silver nitrate ; C1= 13.9. 0,2595 ,, 9.9 C.C. .V/lO ,, C1= 13.9.0,2595 ,, 19.S C.C. iV/lO barium chloride; S= 12.2. C,,H,,ON,CIS requires C1= 14.07 ; S = 12.33 per cent. The nitrate, precipitated as before, crystallised from boiling water in vitreous prisms changing in appearance at 145", and melting with frothing and green coloration at 154' : 0.2595 required 18.7 C.C. N / l O barium chloride ; S = 11.5. C,oHI,ON,S,HNO, requires S 2= 11.18 per cent. A crystalline piwate melting at 160° was obtained from the mother liquor of the nitrate ; phenylethylcarbamylthiourea (from the corre- sponding thiocarbimide and ammonia) when treated in alcoholic solution with picric acid- gives no; picrate.144 DIXON AND HAWTHORNE : Phenylbenxylcarbamic Chloride and Thiourea. The hydrochloride, being somewhat impure, was dissolved in water and precipitated as nitvate, which crystallised from boiling water in long, stout needles changing colour at 145' and effervescing a t 154' : 0.348 gave 0.235 BaYO, ; S = 9.3.C,,H,,ON,S,HNO, requires S = 9.19 per cent. A crystalline picrafe was obtained melting a t 161"; on analysis : 0.2185 gave 30.8 C.C. moist nitrogen a t 17Oand 773 mm. N = 16.64. C21Hl,0,N,S requires N = 16.34 per cent, Benxpl Chlorocarbonnte and Phenylthiou.?*ea. Reference has already been made in an earlier part of this p p e r (p. 123) to the fact that alkyl chlorocarbonate derivatives of certain thioureas, when treated with alkali, do not behave in the same way as the corresponding compounds obtained from aryl chlorocarbonates. In so far as benzyl chlorocarbonate, although containiiig an arornatic group, is allied rather to the former class, it seemed probable that its compound with phenylthiourea would tend to pass readily into Benzyl- +-phenylt hiourea, NHPh*C( NH)-S C,H,, and an experiment was made in order to learn if this would be the case, As no special interest attached to the production of the additive compound of the two substances named above, the preparation was made under the influence of heat, which it was expected would decompose the additive compound as fast as formed, with elimination of carbon dioxide, but not of hydrogen chloride ; in effect, this proved to be the case.When phenylthiourea was heated on the water-bath with a slight excess of benzylchlorocarbonate in presence of benzene, carbon dioxide escaped with effervescence, and an oil was formed, which presently solidified ; the solid, after being washed, first with benzene and then with light petroleum, amounted to 97 per cent.of the yield calculated from the equation CSN,H,Ph + PhCH,.COCl= CO, + PhNH*C(NH)*S*CH,Ph,HCl. 3-52 grams of the product, dissolved in 500 C.C. of water, were treated with excess of normal alkali, the solid precipitate was then separated, and the filtrate neutralised with normal sulphuric acid, using phenolphthalein as indicator ; 10.5 C.C. of alkali were absorbed by the combined hydrochloric acid, instead of 10.9, as required by the above formula. The residual white powder, having an odour of benzyl mercaptan, crystallised from light petroleum in brilliant, pearly leaves, meiting at SO', which were insoluble in water, easily soluble in alcohol or inTHE ACTION OF ACID CHLORIDES ON THIOUREAS.145 hydrochloric acid ; this solution gave a crystalline picrute and a white mercuricldoride. The alcoholic solution was not affected by silver nitrate, but the mixture, when treated with ammonia and warmed, gave a yellow, flocculent precipitate, and the solution in alcoholic potash, when heated with a lead salt, was not blackened, but yielded instead a bright yellow precipitate. The substance was obviously Werner’s +-base, obtained by him from phenylthiourea and benzyl chloride (Trans., 1890, 57, 295); for this he gives the melting point 81-S2O, whilst our compound, once recrystallised, melted at SOo. With thiourea and benzyl chlorocarbonate similar results were obtained, and eventually Werner’s thiourea base, (Zoc.cit.) was isolated. N H,* C ( N H) S * C H,P h , Sunm c m ~ cbnd Conclusion. I n the following summary of the principal observations described or referred t o above, the statements have occasionally been put in somewhat general terms, although the number of cases tested may have been few. (i) Acetyl chloride or berizoyl chloride combines in molecular pro- portion with thiourea to form t h e hydrochloride of a “ b a s e ” or +-thiourea, in which the acyl group is joined to the rest of the molecule through the sulphur atom, as shown by the typical formula NH,*C(N H)*S*CO*CH,. By treatment of the hydrochloride with water, or by treatment of its alcoholic solution with one equivalent of sodium ethoxide, or with excebs of calcium carbonate, thiourea is regenerated ; if, however, the compound is melted, it loses hydrogen chloride only, the acyl group migrating t o the nitrogen atom so as t o produce acetyl- or benzoyl-thiourea. The hydrochloric acid may be displaced by picric acid, with formation of a sparingly soluble picrate of the base. (ii) Acetyl chloride or benzoyl chloride unites similarly with aryl monwubstituted thioureas, the products being quickly dissociated by water, as in the preceding cases ; it is possible, nevertheless, to obtain a picrate from the hydrochloride. On treating the hydrochloride in alcoholic solution with excess of calcium carbonate or with one equivalent of sodium ethoxide, the combined hydrogen chloride is eliminated, b u t (although the odour of thioscetic acid becomes per- ceptible) not with formation of the corresponding $-base, for example, ArNH*C*(NH)*S*COCH, ; instead, the my1 radicle migrates t o the nitrogen atom combined the aryl group, thus producing the isomeric ccu-disubstituted thiocarbamide, for example, PhAcN*CS-NH,. Under the action of heat or of dilute alkali, or, it may be, even by recrgstsl-146 SMITH AND ORTON : TRANSFORMATIONS O F lisation, the latter again changes, owing to further acyl migration, into an ab-disubstituted thiocarbamide. When melted, the above additive compounds lose hydrogen chloride, thereby changing directly inbo ah-disubstitutecl thiocsrbamides, AcNH*CS*NHAr. (iii) Acetyl chloride or benzoyl chloride unites additively with aryl- disubstituted thiocarbamides ; the products have not yet been examined in sufficient detail to justify a definite statement as t o how the acyl radicle is attached. (iv) Disubstituted cmbamic chlorides unite with thiourea, forming haloid salts of basic forms, XYN*CO*S.C(NH)*NH, ; the nitrates and picrates of such bases are sparingly soluble in water. Caustic alkali destroys the hydrochlorides without liberating a corresponding base, the group XYN*CO* undergoing ready hydrolysis into secondary amine and carbon dioxide. (v) If benzyl chlorocarbonate, PhCH,*O*COCI, is warmed with thiourea or with phenylthiourea, carbon dioxide escapes, and a com- pound is formed such as NH,*C(NH)*S*CH,Ph, in which the benzyl group is attached t o the sulphur atom. Aliphatic chlorocarbonates behave similarly. (vi) Organic acyl chlorides do not appear to be capable of uniting with a thiourea or thiocarbamide containing a distinctly acid radicle. The organic group of such a chloricle, however, sometimes expels from an acid-substituted thiocarbamide, and replaces, a radicle more highly electronegative than itself. CHEMICAL DEPARTMENT, QUEEN’S COLLEGE, CORK.