DtFI"l' 2 5-IMINODMYDRO-1 2 3-TRIAZOLE. PART I. 265 XXXIV.-2 5-Iminodihydro-l 2 3-triaxole. Part I . Constitution of Dimroth's 5-Anilinotriaxole. By PAVITRA KUMAR DUTT. DIMROTH (Ber. 1902 35 4041 ; Amalen 1909 364 183) claims to have discovered a peculiar case of isomeric change of the diazo-amine type in connexion with his work on the synthesis of 1 2 3-triazole derivatives. For instance 5-amino-1-phenyl-1 2 3-triazole-4-carboxylic ester (I) when heated in alcoholic solution with sodium ethoxide is transformed into an isomeric substance, 5-anilino-1 2 3-triazole-4-carboxylic ester (11). The mechanism of the change which is reversible is explained by the author in the following manner : The same change is brought about by the action of boiling pyridine or by heating the dry substance alone.The two esters on hydrolysis give the two corresponding carb-oxylic acids. Both however give the same acetyl derivative, corresponding to formula I1 inasmuch as it produces the ester I1 on being hydrolysed with aqueous alkali 266 DUTT 2 ~-IMINODIHYDRO-~ 2 ~-TRIAZOLE. PART I. The two carboxylic acids when heated lose carbon dioxide and produce the same substance m. p. 139"' to which Dimroth ascribes the constitution of 5-anilinotriazole (111). Dimroth also prcpared an isomeride of this compound m. p. 110", in a manner which leaves no doubt as to its constitution (Armalen, 1909 364 210). He represents this substance as 5-amino-l-phenyl-1 2 3-triazole (IV). The latter (IV) when heated alone a t 145" or boiled with pyridine, is converted into the former (111).The mechanism of the change is according to Dimroth tlie same as that of the ester I into 11, and he accounts in a similar way for the conversion of 5-methyl-amino-l-phenyl-1 2 3-triazole (V) into 5-anilino-l-methyl-1 2 3-triazole (VI) by heating with water or pyridine. IT - IT No attempt has been made by Dimroth to elucidate t'hc coil-stitution of what he calls 5-anilinotriazole (111). The formation of a nitroso-compound is not sufficient proof of its constitution, and one of the objects of this investigation was to find evidence for or against this formula. The conversion of Dimroth's 5-methylamino- 1 -phenyltriazole (V) into 5-anilino-1 -methyltriazole (VI) which he regards as a proof of its constitution merely shows that the mechanism of the isomeric change whatever it may be, is the same in all three cases.It appeared to the author that Dimroth's hypothesis is untenable on the ground that the same changes are brought about in a solvent like dry pyridine or by heating the dry substances only. I n this paper a simpler explanation of this change is formulated. Assuming the structure of Dimroth's 5-methylamino- l-phenyl-triazole to be correctly represented by V it can be easily imagined that under the influence of heat or various reagents the hydrogen atom of the methylamino-group migrates from the nitrogen to the adjacent carbon atom of the triazole ring the consequent redistribu-tion of the valencies of the different atoms forming part of the ring being represented in VII.Ring closure by union of the free valencies of the nitrogen atoms would immediately ensue and the four-membered ring (VIII) be produced DUTT 2 ~-IMINODIHYDRO-~ 2 ~-TRIAZOLE. PART I. 267 PhN-Now from its method of formation Dimroth's 6-anilino-1-methyl-triazole may be regarded as having undergone the same isomeric change (IX -+ Y). On comparing the two formuk VIII and X it is easily seen that they represent the same substance. The change of the substance V m. p. 102" into the isomeride (X) of m. p. 172" is thus made clear. Thc constitution of the latter is however not thereby proved for it may be represented by VIII (or X) or by that (VI) of Dimroth's 6-anilino- 1 -methyltriazole the hydrogen atom attached to the tertiary carbon atom (formula VIII or X) having wandered to the negatively substituted nitrogen atom with conse-quent opening of the four-membered ring.I n order to obtain evidence for one or other of these formulz! (VI and VIII) and also in support of the scheme just formulated, two different series of compounds have been made under exactly similar conditions. In one series 5-hydroxy-1-phenyl-1 2 3-triazole-4-carboxylic ester (Dimroth Ber. 1902 35,4040) prepared by condensing phenylazoimide with malonic ester was converted, by successive treat'ment with phosphorus pentachloride and with p-toluidine into an ester the free acid of which by loss of carbon dioxide readily gave a substance m. p. 171-172" which would, according to the above hypothesis have the constitutlion The substance does not give an acetyl or an N-nitroso-derivative and therefore its constitution must be represented by XI.Similarly in the second series of experiments starting from p-tolylazoimide and malonic ester and going through the same course of reactions but substituting aniline for p-toluidine the free acid obtained yielded by loss of carbon dioxide the identical substance (XI) m. p. 171-172". The constitution given to the substance and the hypothesis formulated above may therefore be regarded as established although the evidence is mainly indirect. Dimroth's 5-anilinotriazole should therefore be regarded as 2 5-imino-I-phenyldihydro- 1 2 3-triazole, PLN<;;>NH tH (XIII.1 This formula accounts for all its properties as yet examined. For example i t gives a colourless mononitroso-compound whic 268 DUTT 2 5-IMINODIHYDRO-1 2 3-TRIAZOLE.PART I. dissolves in alkalis forming a colourless solution. On keeping, especially in sunlight and on recrystallisation from alcohol it changes to a greenish-yellow compound. The latter gives a red solution in alkalis from which the yellow nitroso-compound is precipitated on the addition of acids. These changes may be represented as follows : (XIV.) PhN<-N>N*NO 8 -+ PhN<&T>N*NO 8 or CH CH CH Colourless. PhN<i:>N*N( OH)*ONa 8 CH Colourless salt. (XV.) PhN<N>N*NOH IH PhN<t>N*NONa N I I CH C ---Coloured Coloured salt. The nitroso-compound when heated with water or sodium bicarbonate solution gives the original phenyliminodihydrotriazole (XIII) and when heated with aniline forms the same compound together with aminoazobenzene.The substance XI11 gives a monomethyl (m. p. 198-199') and a monoacetyl (m. p. 141-1416") derivative as would be expected from its structure. If Dimroth's formula were correct a disub-stituted product should be formed in each case. It should however, be mentioned that the monomethyl derivative described in this paper differs in its melting point from Dimroth's 5-anilino-1-methyl-triazole (m. p. 172") and it is hoped that further work will reveal the constitution of the former. Treated with an oxidising agent it readily decomposes giving off nitrogen and phenylcarbylamine. Its alcoholic solution gradually decomposes at the ordinary temperature and smells strongly of carbylamine, and the same odour is developed even by the carefully purified, odourless substance after keeping in a dry bottle for some time.It seems as though the internal strain of the four-membered ring is sufficient to make the whole molecule unstable. Dimroth's acid ester (11) should according to the evidence put forward in this paper be regarded as having the constitution : Lastly the substance XI11 is not very stable. -N -~ (xvr.) PhN<CH>NH ?? C-CO,R It is obvious that the transformation of the ester I into XVI and vice versa by the action of heat or various reagents can also be accounted for by the same scheme as given above (p. 266). The iminodihydrotriazole derivatives having two differen DUTT 2 5-IMINODIHYDRO-1 2 3-TRIAZOLE.PART I. 269 radicles in the four-membered ring contain an asymmetric carbon atom. Experiments on the resolution of these compounds are in progress. EXPERIMENTAL. PhenyZax0imide.-The method of preparing aromatic azoimides previously described (T. 1921 119 2088) has been modified SO that fairly large quantities can be made in a short time and a t a reasonable expense. Twenty-eight grams of aniline dissolved in 90 C.C. of strong hydrochloric acid and 90 C.C. of water are diazo-tised with 21 grams of sodium nitrite dissolved in 60 C.C. of water. The diazo-solution is added to a solution of 51 grams of p-toluene-sulphonamide and 45 grams of sodium hydroxide in 700 C.C. of water a t the ordinary temperature. After some time the mixture is steam-distilled and the azoimide purified in the ordinary way.Yield 26-27 grams. p-ToZyZaxoimide.-The same procedure is followed in this case. Forty grams of p-toluidine dissolved in 120 C.C. of strong hydro-chloric acid and 90 C.C. of water are diazotised with 28 grams of sodium nitrite in 60 C.C. of water. The diazo-solution is added to 51 grams of p-toluenesulphonamide and 60 grams of sodium hydroxide dissolved in 700 C.C. of water. Ethyl 5-Amino-1-phenyl-1 2 3-triaxole-4-carboxylate (Formula I).-A cold solution of 4.6 grams of sodium in 100 C.C. of anhydrous ethyl alcohol is mixed with 22 grams of phenylazoimide and 22 grams of ethyl cyanoacetate are gradually added with frequent shaking and with external cooling. After leaving over-night the mixture is diluted with about 600 C.C.of water. The precipitate (yield 29-30 grams; m. p. 125-126") crystallises from benzene in needles and from alcohol in plates the melting point remaining unchanged (Found N = 24.26. Calc. N = 24.13 per cent.). The filtrate and washings when acidified with concentrated hydrochloric acid give a crystalline precipitate (prisms) of the free acid which melts a t about 140" with decomposition. Yield 5-6 grams. The crude acid is crystallised from alcohol with charcoal treatment and then melts a t 141-142" with decomposition (Found C = 53.01 ; H = 4-20; N = 27-65. Calc. C = 52-94 ; H = 3-92 ; N = 27.45 per cent.). The same carboxylic acid is obtained by hydrolysing the above ester with aqueous alcoholic sodium hydroxide. The acetyl derivative of the ester obtained by boiling itswith an equal weight of acetic anhydride for half an hour separates from spirit in needles m.p. 103-104" (Found N = 20.51. Calc., N = 20.44. Dimroth gives m. p. 90" and N = 20.11 per cent.). Ethyl 2 5-Imino-1-phenyldihydro-1 2 3-triaxole-4-carboxylate (XVI).-Thirty grams of the ester I and 70 C.C. of dry pyridine Yield 31-32 grams. VOL. CXXTTT. 270 DUTT 2 5-IMINODIHYDRO-1 2 3-TRUZOLE. PART I. are heated together under reflux at 150-160" for five to six hours. The mixture is then poured into 500 C.C. of water cooled and acidified with hydrochloric acid. The precipitate which resembles cotton wool in texture and appears under the microscope as a mass of long hairy crystals is collected dissolved in spirit and the solution treated with a large volume of aqueous sodium carbonate (10 per cent.).A little unchanged insoluble ester having been removed the filtrate is acidified and the imino-ester (yield 28-29 grams ; m. p. 124-126") recrystallised from dilute alcohol ; needles, m. p. 129-130" (Found N = 24.21 per cent.). The acetyl derivative of this ester obtained as in the previous instance melts a t 103-104" alone or mixed with the acetyl derivative of ester I. The meth,yl derivative of the ester is prepared by boiling 11.6 grams (1 mol.) of the substance in a solution of 4.6 grams (2 atoms) of sodium in 100 C.C. of anhydrous alcohol together with 30 grams (more than 2 mols.) of methyl iodide until neutral. The alcohol having been distilled off and water added to the residue the mixture is made slightly alkaline and cooled and the solid (yield 8 grams) recrystallised from benzene ; needles m.p. 98-99" (Found N = 23.16. Cl2HI4O2N4 requires N = 22.76 per cent.). The substance is therefore a monomethyl derivative. Its constitution has not yet been proved. Unlike the parent ester it does not give phenyl-carbylamine when treated with sodium hypobromite. When hydrolysed witlh alcoholic sodium hydroxide it gives the free acid, and the latter on recrystallisation from spirit melts at 223" with decomposition (Found N = 25.64. CloH,,02N requires N = 25.68 per cent.). Conversion of the Imino-ester (XVI) into the Amino-ester (I).-Ten grams of the ester and 160 C.C. of anhydrous alcohol are heated under reflux while a current of dry hydrogen chloride is passed through for six to seven hours.The alcohol is now distilled off the residue treated with a large volume of aqueous sodium carbonate (10 per cent.) and the precipitate after being washed and dried (yield 2.5 grams) recrystallised from benzene. The product melts a t 125-126" (Found N = 24.61 per cent.) and is identical with the ester I. From the filtrate the imino-ester is obtained on the addition of hydrochloric acid. Dimroth (Annalen, 1909,,364 205) by heating the substance in alcoholic solution at 1SO" for three hours obtained a 23 per cent. conversion into the amino-ester (I) but he did not isolate the product. Hydrolysis of the Acetyl Derivative of the Ester.-Ten grams of the acetyl compound dissolved in alcohol are heated in a current of dry hydrogen chloride as in the previous experiment.After tw DUTT 2 5-IMINODIHYDRO-I 2 Q-TRIAZOLE. PART I. 271 hours' heating the mixture is treated with sodium carbonate solution. A little insoluble matter (0.7 gram) is collected and recrystallised from alcohol. It melts at 124-126" (Found N = 24.26 per cent.) and is identical with the ester I. The sodium carbonate solution on the addition of hydrochloric acid gives a considerable quantity of the ester XVI. The acetyl compound is therefore derived from the latter. 2 5-Imino-l-phenyldihydro-1 2 3-triaxole (Formula XIH).-Dimroth (Ber. 1902 35 4041) prepared this compound by heating the acid corresponding with the ester I and obtained a yield of 73 per cent. of the crude substance. A better method is as follows : 20.4 grams of the free acid and 50 C.C.of dry pyridine are heated together under reflux at 120-130". After half an hour the evolu-tion of carbon dioxide ceases and after two hours the liquid is poured into 500 C.C. of cold water acidified with hydrochloric acid, and the solid collected; m. p. 138-139" yield 14.5 grams (90 per cent .). After recrystallisation from alcoholic solution with charcoal treatment it is obtained in colourless rhombic crystals m. p. 140" (Found N = 35-15. Calc. N = 35.0 per cent.). The picrate obtained by mixing alcoholic solutions of the sub-stance and of picric acid forms needles m. p. 163" with decomposi-tion (Found N = 25.53. C,H8N4,C6H30,N3 requires N = 25.19 per cent.). The acetyl derivative is obtained by warming the substance with three times the quantity of acetic anhydride on the water-bath until the mixture just begins to darken.Prolonged heating results in decomposition producing a black sticky mass. The mixture is poured into water made alkaline with sodium carbonate and the precipitate crystallised from spirit ; needles m. p. 141-141.5" (Dimroth gives m. p. 137-138') (Found N = 28.14. Calc., N = 27.72 per cent.). On hydrolysis with aqueous sodium hydroxide the original substance (XIII) is obtained. The methyl derivative of the substance (XIII) is obtained in the same way as in the case of methylation of the ester (XVI). It crystallises from absolute alcohol in thick plates and melts at 198-199" (Found N = 32-32. C,H,,N requires N = 32.18 per cent.).Unlike the parent substance i t is insoluble in alkali and does not give carbylamine when treated with alkaline oxidising agents. Nitroso-derivative of the Base X I I I .-Ten grams of the freshly purified base dissolved in 400 C.C. of dilute hydrochloric acid (10 per cent.) are treated a t 5" with 5 grams of sodium nitrite dissolved in 100 C.C. of water. The colourless precipitate is washed wit'h ice-cold dilute hydrochloric acid and with ice-water and dried in a Its constitution has not yet been proved. L 272 PUTT 2 5-IMINODIHYDRO-I 2 3-TRJAZOLE. PART I. vacuum desiccator in the dark. It melts with decomposition a t 117-118" (Dimroth gives m. p. 115-116") (Found N = 37.6. Calc. N = 37.03 per cent.). It dissolves in dilute sodium hydroxide solution in the cold forming a colourless solution which when warmed gradually turns red and decomposes.On prolonged exposure to sunlight the colourless solid turns greenish-yellow and then gives a red solution in alkali. When recrystallised from cold alcohol the coloured crystals are obtained. Both substances give Liebermann's nitroso-reaction. The nitroso-compound is insoluble in sodium bicarbonate solution in the cold and when heated gradually dissolves with partial de-composition (smell of carbylamine). The precipitate obtained on acidifying the clear solution melts at 139-.140" after recrystal-lisation from alcohol and is identical with the parent substance (XIII) from which the nitroso-compound is derived. The same change occurs when a suspension of the nitroso-compound in water is heated.Action of Aniline on the Nitroso-compound ( X I V or XV).-Five grams of the nitroso-compound are heated with 10 C.C. of aniline on the water-bath for about two hours the aniline is distilled off in steam and the residue extracted with hot water and filtered. Crystals appear when the filtrate is cooled. These are filtered off and the clear filtrate used again for extracting the above residue. The total crop of crystals is then thoroughly dried and exhaustively extracted in a Soxhlet apparatus with light petroleum. Thc extract deposits a reddish-brown crystalline substance which melts at 124-126" and is aminoazobenzene (Found N = 21-51 per cent.). The insoluble residue remaining in the Soxhlet is dissolved in alcohol decolorised by boiling with animal charcoal, and crystallised.Colourless crystals are obtained which melt a t 139-140" alone or mixed with the substance XI11 (Found: N = 35.2 per cent.). Ethyl 5-Hydroxy-1-phenyl-1 2 3-tria~ole-4-carboxylate.-A cold solution of 9-5 grams of sodium in 125 C.C. of anhydrous alcohol is mixed with 50 grams of phenylazoimide and 70 grams of et'hyl rnalonate are added gradually with shaking and cooling. The mixture having been heated under reflux for about two hours the alcohol and unchanged azoimide are distilled off in steam and the hot aqueous solution of the residue filtered. On acidifying the filtrate with hydrochloric acid a semi-solid mass separates which solidifies on cooling. It is filtered washed and dried. Yield 70-75 grams.On recrystallising from spirit the ester separates in yellow prisms m. p. 72-73' (Found N = 18.34. requires N = 18-02 per cent.). Yield 11 grams DUTT 2 6-IMINODIHYDRO-1 3 3-TRIAZOLE. PART I. 273 Ethyl ~-H~drozy-l-p-tolyl-l 2 3-triaxole-4-carboxylate is made in the same way as the preceding ester. Using 9 grams of sodium, 125 C.C. of anhydrous alcohol 45 grams of p-tolylazoimide and 56 grams of malonic ester 65-70 grams of the crude ester are obtained. It crystallises from spirit in dull yellow prisms m. p. 98-99" (Found N = 17-13. CI2H1,O3N3 requires N = 17-0 per cent.). Ethyl 5-Chloro- 1 -phenyl- 1 2 3-triaxole-4-carboxylate, NPh<"Y CCl:CH*CO,Et -Thirty-four grams of the hydroxy-ester and 40 grams of phos-phorus pentachloride are heated together on a gently boiling water-bath under reflux for about an hour and a half.It is well to add the phosphorus pentachloride in several portions by momentarily disconnecting the flask as otherwise the reaction is likely to become too vigorous. The phosphoryl chloride is then distilled off in a vacuum from a water-bath and the residue dis-solved in ether. The ethereal solution is washed several times with water the ether distilled off and the residue crystallised from methyl alcohol; colourless needles m. p. 80-81" (Found : N = 17.09. The same substance is obtained by the action of nitrous acid on the ester I dissolved in concentrated hydrochloric acid (Dimroth Annalen 1909,364,212). Ethyl 5-Chloro-1-p-tolyl-1 2 3-triazole-4-carboxylate.-This sub-stance is made in the same way as the preceding.It crystallises from absolute alcohol in colourless prisms m. p. 61-62' (Found : N = 16.07. Using 35 grams of the hydroxy-ester and 45 grams of phosphorus penta-chloride about 12 grams of the pure chloro-derivative are usually obtained and a little more can be recovered from the mother-liquor. Action of p-Toluidine on Ethyl 5-Chloro- 1 -phenyltriaxole-4-carboxylate.-Forty grams of p-toluidine and 25 grams of the ester are heated together on a water-bath for five to six hours. The excess of p-toluidine is then distilled off with steam the residue dissolved in ether the ethereal solution washed several times with water and the ether distilled off. The residue which consists of ethyl 2 5-p-tolylimino-1-phenyldihydro-1 2 3-triazole-4-carb-oxylate is hydrolysed with alcoholic potassium hydroxide.The aqueous alkaline solution of the free acid is saturated with carbon dioxide decolorised by treatment with animal charcoal and filtered. On adding hydrochloric acid to the cooled filtrate a colourless precipitate of the free acid is obtained m. p. 84-47" with decomposition ; yield about 25 grams. Its alcoholic solution, when warmed on the water-bath gradually evolves carbon dioxide, Calc. N = 16.7 per cent.). C,,H,,O,N,Cl requires N = 15-81 per cent.) 274 DUTT 2 &mINODIHYDRO-1 2 3-TRIAZOLE. PART I. forming the substance XI which is however best obtained by heating the free acid in a paraffin-bath at 150-160". When no more carbon dioxide is given off the residual mass is dissolved in hot alcohol decolorised and allowed to crystallise.It forms elongated plates or needles which are colourless with a silvery lustre and melts at 171-172" (Found C = 71.6; H = 6.0; N = 22.6. Cl,Hl,N4 requires C = 72.0; H = 6.6; N = 22.4 per cent.). 2 5-p-Tolylimino-1-phenyldihydro-l 2 3-triaxole is a neutral substance insoluble in hydrochloric acid or sodium hydroxide solution. It is soluble in glacial acetic acid from which it crystal-lises well and does not react with acetic anhydride even on prolonged boiling. Action o f Nitrous Acid on Xil.-1*3 Grams of the substance, dissolved in 25 C.C. of glacial acetic acid are mixed with 10 C.C. of dilute hydrochloric acid (10 per cent.) and treated with 2 C.C. of 25 per cent. solution of sodium nitrite.The nitrous acid acts very sluggishly and the mixture gradually turns yellow. After eight hours the clear solution is poured into 400 C.C. of cold water. The yellowish-orange gummy mass that separates is filtered washed with water (yield 1.2 grams) dissolved in spirit and water added in small quantities when an emulsion is formed from which orange prisms are obtained. These melt indefinitely a t 135-140". The percentage of nitrogen in various crops of crystals varies from 23.2 to 23.5. The nitrogen content for a mononitroso-compound should be 25.09 per cent. As the substance does not give Liebermann's nitroso-reaction it cannot be the N-nitroso-derivative. The nitrous acid must have reacted in a way which is a t present inexplicable. Action of Aniline on Ethyl 5-Chloro- 1 -p-tolyltriaxole-4-carboxylate. -Forty grams of aniline and 26.6 grams of the ester are heated together on a water-bath for five to six hours and the resulting mass is treated as described above (p. 273). The crude free acid usually amounts to about 25 grams and melts with decomposition at 76-81". When heated in a paraffin-bath as before it loses carbon dioxide and the product on recrystallisation from alcohol melts at 171-172" alone or mixed with the substance XI. Both have the same crystalline form and behave exactly alike with various reagents. I n conclusion the author wishes to thank Professor J. B. Cohen, F.R.S. for criticism and encouragement and the Research Fund Committee of the Chemical Society for a grant which has partly defrayed the expenses of this investigation. THE ORGANIC CAEMICAL LABORATORIES, THE UNIVERSITY LEEDS. [Received October 26th 1922.