198 FRANKLAND CHALLXNGZR AND NIOgOLL$ : XXILIL-The Prepayati0.n of Moiiomethyladine. By PERCY FARADAY FRANKLAND FREDEEICK CHALLENGER and NOEL ALBERT NICHOLLS. WHEN aniline is submitted to the action of the usual llzethylatillg agents it is difficult to limit the course of the reaction to the intro-duction of a single methyl group. We have investigated the pro-duction of monomethylaniline by three methods with a view t o obtain it in a pure condition. (1) The process described in the German Patent 75854 and with slight modifications in the French Patent 212506 consists in con-densing aniline with formaldehyde and then reducing with zinc dust and concentrated aqueous sodium hydroxide until a test por-tion dissolves t o a clear solution in acetic acid. Any unreduced methyleneaniline CH,:N*C',II, is thereby converted into insolubl THE PREPARATION OF MONOMETHYLAN1L;fNE.199 anhydroformaldehydeaniline (CH,:N*C,H,),. Using 100 grams of aniline we have found the reduction to be completed in twelve to fifteen hours and have obtained about 55 per cent. of the theoretical yield of methylaniline. The Patent Specificatiolns make no mention of the yields obtained. The principal advantage of this method lies in the fact that the product although rich in aniline contains only traces of dimethyl-aniline. The presence of the aniline may be the result of incom-plete condensation with formaldehyde in the first instance or of the hydrolysis of the methyleneaniline by the hot water or of both causes. The primary base can be recovered in the form of its zincichloride which is almost insoluble in water the methylated bases not being affected by this reageat (Monutsh.1888 9 514). The method appears t o work very satisfactorily. Morgan (English Patent 102834) who has also studied the reduc-tion of methyleneaniline suggests that the poor yield of methyl-aniline may be due in addit.ion to hydrolysis to the conversion of some of the methyleneaniline into s-dipheriylmethylenediamine, CH,(NH*C,H,)2 and anhydrof oraaldehydeaniline . We have failed to obtain more than traces of monomethylaniline from the last-named substance (compare however Goldschmidt Chem. Z e i t . , 1904 28 lZZ9) but when s-diphenylmethylenediamine (Eberhardt and Welter Ber. 1894 27 1804; Eibner Annalen 1898 302, 349) is reduced with zinc dust and alkali hydroxide under the con-ditions of the earlier patents aniline and monomethylaniline are produced in equal amounts probably according to the equation CH,(NH*C,H,) + 2H = C,H,*NH2 + C,H,*NH*CH,.We attempted to diminish the hydrolysis by performing the reduction in concentrated alcoholic solution. The yield of methyl-aniline was however only about 46 per cent. No better results were obtained by increasing the quantity of formaldehyde; 1.5 molecular proportions gave rise t o some dimethylaniline whilst with 10 molecular proportions considerable quantities of this base were foxmed.* This appears t'o be due to the interaction of monomethylanihe and formaldehyde giving rise t.0 s-diphenyldimethylmethylene diamine which then undergoes reduction.This reaction would, moreover be analogous to the reduction of s-diphenylmethylene diamine. The condensation productl of formaldehyde and methyl-* Compare the action of formaldehyde on methylaniline in acid solution (Goldschmidt Zoc. cit.) and on methyl-o-toluidine (Braun Ber. 19OS 41 2153), also on ammonium chloride (Werner T. 1917 112 844). Sea also Yinnow,. Ber. 1894,27 3166; Cohn Chem. Zeit. 1900 24 564. K 200 ERANKLAND CHALLENGER AND MUHOLLS : aniline (Braun Ber. 1908 41 2147) was therefore prepared and reduced under the usual conditions with the result that much dimethylaniline was produced. (2) The process described in various works of reference consists in heating aniline hydrochloride with methyl alcohol in an auto-clave to 180-200°.s The best result obtained by us in a series of thirteen experiments was a 55 per cent.yield of monomethylaniline. Experiments in sealed tubes confirmed these results. We have also investigated a variation of the above method in which aniline hydrochloride and methyl alcohol are heated together in the presence of glacial acetic acid (Ber. 1897 30 3072). The most favourable result was a yield of 56 per cent. of methylaniline. (3) The demethylation of dimethylaniline by heating with aniline hydrochloride a t above 1 8 0 O . In the most favourable experinients, a yield of 55 per cent. of methylaniline was obtained. Aniline was also heated’to a high temperature with dimethylaniline hydro-chloride with similar results but it was found necessary to heat f o r a much longer period.This was probably due to the more basic nature od dimethylaniline. A condition of equilibrium between the three bases and hydrochloric acid would appear t o be established. At the time these experimenh on demethylation were performed (1916) we were unable to find any record of similar work on this subject apart from the well-known phenomenon of the transfer a t above 300° of alkyl groups from nitrogen to the ring. According ta Schultz (“ Chemie des Steinkohlentheers,” 1900, 3rd ed. I p. 98) monoethylaniline may be prepared by heating aniline hydrochloride with diethylaniline. No yields or references to the original literature are given but the reactJon is obviously not quantitative since it is stated that the hydrochloride of diethyl-aniline remains in solution.EXPERIMENTAL. Prepmatiom of Monornethyla,n&ne accwding t o D.R.-P. 75854. Materials Used.-One hundred grams of aniline 80.5 grams of formaldehyde (37 per cent.) 30 grams of methyl alcohol 25 grams of sodium hydroxide solution (D 1-38) 150 grams of zinc dust, 1 litre of water and 225 grams of sodium hydroxide solution. monomethylaniline obtained in this way are very contradictory. See Wahl-Atack “Organic Dyestuffs,” p. 70 ; Cdn ‘* Intermediate Products,” p. 61 ; Lunge ‘‘ Chem. Tech. Untersuchungsmethoden,” Vol. III. p. 761 ; Friedkinder, “ Fortschritte der Teerfarbenfabrikation,” 1877-1887 p. 6. * The published statements concerning the yields o THE PREPARATION OF BZON0METRYLANIL;INE. 201 The first four ingredients were mixed in a widemouthed bottle fitted with a stirrer and avreflux condenser the zinc dustl and water then added and the temperature raised to about 90° The remainder of the sodium hydroxide solutio'n was gradually intro-duced and the stirring continued a t this temperature until after about twelve to fifteen hours the methyleneaniline had disappeared.The bases were then distilled in a current of steam extracted with ether and a portion convert,ed into the nitrosoamine. Yield of bases 102 grams. Theory=115 grams. Thirty grams of the mixture gave 23 grams of dry phenylmethyl-nitsosoamine whence t.he total yield of monomethylaniline is 54.5 per cent. of the theoretical. P r e p r a tioia of Monom e t h y la nili n e according to Fr enc 11 Pat e n t 212506. I n these experiments the quantities of material and method of procedure were as described above with the exception that the whole of the sodium hydroxide solution was added a t once.In one case where particularly efficient stirring was employed, the methyleneaniline had disappeared in six and a-half hours. I n this experiment the mixed bases contained 65 per cent, of mono-methylaniline (by the nitrosoamine metho'd of analysis) correspond-ing with a4yield of 51.5 per cent. Other experiments with the same quantities and under similar conditions gave yields of 41.5 and 53.5 per cent. of the theoretical quantity of monomeljhyl-aniline. Reduction in the Presence of Excess of Formaldehyde. I n two experiments on 100 grams of aniline in which 0.5 mole-cular proportion of formaldehyde was used in excess the yields of monomethylaniline were 46.5 and 55 per cent#.I n the first case, about 17 grams of the hydrochloride of p-nitrosodimethylaniline were obtained on nitrosificabion. With an excess of 9 molecular proportions of formaldehyde two experiments according to the German Patent gave scarcely any monomethylaniline ; the reactions proceeded very slowly and in one case much dimethylaniline was produced. Reduction in Concentruted Alcoholie Solution. One hundred grams of aniline 25 grams of formaldehyde (36 per cent.) 840 grams olf alcohol 88 grams of solid sodium hydr 202 FRANKLAND CHALLENGER AND NICHOLLS : oxide 25 grams of aqueous sodium hydroxide (D 1-38) and 150 grams of zinc dust' were vigorously stirred together. The reaction was complete in seven and a-half hours a t about 60° and the yield of methylaniline was 46.5 per cent.of the theoretical. Reduction o f s-niphen?/Zmeth?/leitediamz'ne. Thirty grams of s-diphenylmethylenediamine 125 grams of sodium hydroxide solution (D 1-38) 75 grams of zinc dust 500 C.C. of water and 25 grams of alcohol were mixed and vigorously stirred for ten and a-half hours a t 70-90°. When a portion dissolved to a clear solution in dilute acetic acid (in which s-diphenylmethylene-diamine is but sparingly soluble) the products were distJlled in a current of steam. Twenty-two grams of mixed bases were obtained which gave 13 grams of phenylmethylnitrosoamine corresponding with a yield of 10.2 grams of monomethylaniline. From the diazonium chloride solution 10 grams of phenol mere isolated corresponding with 10 grams of aniline.Rduction o f the Condensation Product of Formaldehyde arid Mon onze t hy Zaniline. Fifty grams of monomethylaniline 19 grams of formaldehyde (36 per cent.) and 5-10 C.C. of aqueous so'dium hydroxide were mixkd well shaken and allowed t o remain overnight. The condensation product was separated by extraction with ether and reduced with a mixture of 150 grams of zinc dust. 1000 C.C. of water 60 grams of methyl alcohol and excess of sodium hydr-oxide solution a t about< 80°. After one and a-half days the bases (44 grams) were removed by steam distillation. Nitrosification showed the product to contain 35 grams of monomethylaniline, whilst 9 grains of ~~-nitrosodimethylanili~ie hydrochloride were obtained.In these experiments the methylaniline was determined as the nitrosoamine and the aniline as phenol after decomposition of the diazonium salt. Dimethylaniline was separated and weighed as p-nitrosodimethylaniline hydrochloride but as some of this always remained in solution the figures for the tertiary base are low THE PREPARATION OF MONOMETHYLANILINE. 203 I n experiment 8 aniline (140 grams) and sulphuric acid (16 grams) were used instemad of aniline hydrochloride. Experiments 9-13 were made in sealed tubes and in Nos. 11 12 and 13 acetic acid (12 grams) was added. Percentqge yield Aniline Time w-Experi- hydro- Methyl Tempera- in Methyl- Dimethyl-ment. chloride. alcohol. ture- hours. Aniline. aniline. aniline. 1 110 32 180" 24 - 48.0 -2 110 32 180 24 27.0 51.5 -3 110 32 180 21 25.5 53.0 16.5 4 220 64 180 '' 36-0 52.0 7.0 5 220 64 190 gi 34.0 55.0 7.0 180 44 26.0 53.0 10.0 180 3 - 45-0 20.0 7 220 96 8 140 84 190 41 32.0 40-5 16.0 10 22 7.0 175 6 13.0 54.0 13.0 11 22 6.4 170 5 - 56-0 -12 22 6.4 225 3 - 41.0 -13 22 6.4 220 3 - 47.5 -above 6 220 64 9 22 6.4 180 32 19.0 58.0 11.0 MethyJatim of Aniline under otlher Conditions, (1) Twenty-six grams of aniline hydrochloride 7 grams of methyl alcohol and 25 C.C.of hydrochloric acid were heated in a sealed tube for six hours a t 200-210°. Nitrosoamine=13.5 grams; yield of methylaniline = 49-5 per cent. (2) Thirty-six grams of aniline zincichloride and 6.4 grams of methyl alcohol were heated in a sealed tube for seven hours a t 1 8 0 O .Nitrosoamine = 7.5 grams ; yield of methylaniline = 28 per cent. (3) Twenty-two grams of aniline hydrochloride 6.4 grams of methyl alcohol and 15 grams of anhydrous calcium chloride were heated a t 160° for three hours. Nitrosoamine=lO grams; yield of methylaniline= 43 per cent. (4 and 5). Twenty-eight grams of aniline (in the presence of 0.2 gram of iodine) were heated in one case with 10 grams in another with 6.5 grams of methyl alcohol f o r nine hours a t about 220° (Knoll and Co. D.R.-P. 250236). I n both cases nitroso-amine= 15 grams. Yield of methylaniline=42 per cent. The Dem e t hylation of Dime t h y luniline. First Series.-Interaction between aniline hydrochloride (13.0 grams) and dimethylaniline (1 2.0 grams) in molecular proportions in sealed tubes 204 THE PRDPARATIOW OF MQNOMETHYLANILLNE.Time Nitr?so- Percentage Experiment. Temperature. in hours. amne. rnonornethyEk2 1 180' 6 12.5 46 2 200 6 16.0 65 3 230-236 5! 15.0 55 Second S e ries.-Inter action be tween cl j met hylaniline hydro-chloride (31.5 grams) and aniline (18.6 grams) in molecular pro-portions in sealed tubes.* Time Nitroso- Percentage yield of Experiment. Temperature. in hours. amine. monomethylaniline 1 180" 3 4 14.5 2 180 6 11.5 21.0 31 200 13 28.0 51.0 1 In Experiment 3 about 5 grams of a white solid separated on diluting the contents of the tube. This melted indefinitely at 144" and after crystellisa-tion from light petroleum indefinitely a t about 160'. It was only superficially examined and appeared to be a tertiary halogen-free base possibly containing methyl groups in the benzene nucleus.When 12 grams of dimethylaniline and 9.5 grams of aniline (molecular proportions) were heated for three hours a t 220° prac-tically no monomethylaniline was produced. Demethylation was found to occur when the two hydrochlorides were heated for three hours atl 180O; thus 13 grams of aniline hydrochloride and 15.8 grams of dimethylaniline hydrochlorid'e (molecular proportions) gave 5 -5 grams of the nitrosoamine corre-sponding with a 20 per cent. yield of monomethylaniline. The Separation of dilailine a d Monomet17??llai1ilinf using Zin\c Chlo.ride. A mixture of aniline and monomethylaniline was treated with an aqueous solution of anhydrous zinc chloride.The precipitate was collected and thoroughly washed with light petro,leum. After evaporation of the solvent the residue of crude methylaniline was weighed and converted into the nitroso-derivative which was removed from the mixture by extracting t7hree times with ether, dried and weighed. The acid liquid which remained after the removal of t h s nitroso-amine and contained traces of benzenediazonium chloride was heated saturated with salt the phenol extracted with ether and finally weighed. The aqueous filtrate from the zincichloride pre-cipitate contained practically no aniline or methylaniline hydro-chlorides. * In Experiment 1 half these quantities were employed EQUILIBRIA IN THE FLEPUCTION OF OXXDEE BY CARBON. 205 The accuracy of this method was checked by regeneration of the aniline from a given weight of the zincichloride Andytical Results.-l’alcem Aniline 20 grams methylaniline 20 grams fused zinc chloride 22 grams water 50 C.C. Obtained .- Zincichloride 37 grams whence aniline= 19.2 grams. Phenylmkthylnitrosoamine 23.7 grams whence methylaniline = 18.7 grams. There was also obtained 0.5 gram of bases from the aqueous filtrate from the zincichloride. Phenol 0.25 gram whence aniline=0*25 gram. THE UNIVERSITY, BIRMINGHAM. [Received February 6th 1919.