946 ORR ROBINSON AND WILLIAMS : LXXX.-The Action of Halogens on Piperonal. By ANNIE MARY BLEAKLY ORR ROBERT ROBINSON and MARGARET MARY WILLIAM s . THE displaement of groups by nitroxyl accompanying the nitra-tion of phenol ethers has frequently been observed but it does not appear to have been recorded that a similar reaction occurs in the chlorination o r bromination of certain of these substances. I n the preparation of bromopiperonal in acetic acid solution the yield of product is not satisfactory and this caused us to suspect that a certain proportion of the formyl group is displaced by the halogen which we found on investigation to be the case. I n a neutral solvent such as carbon disulphide or carbon tetrachloride, the yields of bromopiperonal and of 6-chloropiperonal (I) are excellent and by-products are reduced to a minimum but in acetic acid solution the bromination of piperonal produces a certain amount of 4 5-dibromocatechol methylene ether (this vol., p.913) and the chlorination of the aldehyde yields 4 5-dichloro-catechol methylene ether (11) in addition to chloropiperonal and substances of undetermined nature which are decomposed by aqueous sodium carbonate and probably owe their formation to attack of the methylene ether group by the halogen. Dichloro-catechol methylene ether was also obtained by the action of chlorine on a sodium carbonate solution of piperonylic acid or of 6-chloropiperonylic acid (111) obtained by the oxidation of chloropiperonal. 4-Chloro-5-bromocatechol methylene ether (IV THE ACTION O F HALOGENS ON PIPERONAL.947 was obtained by the action of bromilze on an aqueous sodium carbonate solution of chloropiperonylic acid but not except in traces from chlorine and bromopiperonylic acid under similar conditions. The behaviour of chloropiperonal on nitration resembles that of bromopiperonal and 4-chloro-5-nitrocatechol methylene ether (V) and 4-chloro-5 6-dinitrocatechol methylene ether (VI) were successively obtained. The latter on reduction furnishes a chlorodiamine which was isolated as a phenanthra-phenazine derivative. (IV) Ex P E R I M E N T A L. Brornination of Piperonal in Acetic Acid Solution. Piperonal (100 grams) dissolved in acetic acid (200 c.c.) was gradually treated with bromine (40 c.c.) dissolved in acetic acid (100 c.c.) any rise of temperature being checked by cooling.After remaining overnight the crystals were separated by filtra-tion and found to be pure 6-bromopiperonal (37.8 grams). The filtrate was mixed with water and the solid collected and heated with an aqueous solution of sodium hydrogen sulphite until no further alda,hyde passed into solution. The substance was then dissolved in ether and the ethereal solution extracted with repeated quantities of aqueous sodium hydrogen sulphite. The bromopiperonal was recovered from the hydrogen sulphite extract by the additicn of sodium carbonate and after crystallisation from methyl alcohol weighed 62 grams. The ethereal solution was washed with sodium carbonate and with water dried and evaporated and the crystalline residue weighed 14.8 grams.It was crystallised from alcohol and identified as 4 5-dibromocatechol methylene ether melting a t 86O. The substance from this source persistently crystallised in needles whereas it had formerly been obtained in leaflets but the melting point of a mixture was 8 6 O . The formation of the dibromo-derivative must be due in the first place to that of the monobromecompound in its turn obtained by a direct displacement of the formyl radicle by bromine. Thi 948 ORR ROBINSON AND WILLIAMS : follows from the observation that bromopiperonal is stable to bromine in acetic acid solution. A specimen was found to be unchanged after the attempted reaction had been prolonged during a week. 6-Chloropiperonal (I). A stream of chlorine was passed into a solution of piperonal (25 grams) in acetic acid (50 c.c.) until the product of the reac-tion commenced to crystallise.This required about four hours. Water was added and the solid collected and dissolved in ether. The ethereal solution was washed with aqueous sodium carbonate and then with sodium hydrogen sulphite solution until a test por-tion gave no milkiness on the addition of sodium carbonate. The ethereal solution was dried and evaporated and the residue (3.4 grams) was identified as 4 5-dichlorocatechol methylene ether. The hydrogen sulphite solutions were treated with excess of sodium carbonate and the precipitated crystalline substance was collected and crystallised from aqueous alcohol when 17.1 grams were obtained. The substance was further purified by crystallisation from methyl alcohol from which it separated in long colourless needles melting at 1 1 5 O : 0.1294 gave 0.1019 AgC1.C1-15.3. 0.2010 in 16.06 C,H gave At=0*347. 6-Chloro-3 4-methyle?zedioxystyryl methyl ketone (6-chEoro-piperonylideneacetone) C,,H,03C1 is obtained by the addition of dilute aqueous potassium hydroxide to a solution of the aldehyde in three times its weight of acetone. After an hour water was added and the nearly colourless precipitate collected and crystal-lised from alcohol. The pale yellow needles melted sharply a t 1 5 8 O and dissolved in sulphuric acid to a reddish-yellow solution, and the substance was recovered unchanged on the addition of water. I n chloroform solution bromine was rapidly absorbed, but the dibromide was oily.Bi-6-chloromethylenedioxystyryl ketone C,9H,20,C1, is obtained by the condensation of the aldehyde with the above described sub-stance or by heating the chloropiperonal in alcoholic solution with acetone and concentrated potassium hydroxide. It is very sparingly soluble and its formation is a matter of seconds. The substance was washed with hot alcohol and crystallised from acetic acid. The felted citron-yellow needles melted and decomposed a t 265O and dissolved in sulphuric acid to an intense blue solution which became purple and then yellm on the addition of water. 6-ChZoro~*peronyZic acid C,H,O,Cl was produced on oxidising M.W.=180. C,H,03C1 requires C1= 19.0 per cent. M.W. = 184 THE ACTION OF HALOGENS ON PIPERONAL 949 the aldehyde in benzene solution by stirring with a 2 per cent.solu-tion of potassium permanganate. The reaction proceeds slowly and when the benzene gave no residue on evaporation the excess of permanganate was destroyed and the acid obtained by acidification of the filtered aqueous solution. It was precipitated as a caseous solid which wm collected and crystallised from acetic acid. The colourless needles melted a t 202O and the substance is very sparingly soluble in water. It gives a pale yellow solution in sulphuric acid which becomes olive-green on gently heating, whereas piperonylic acid under these conditions gives an intense red solution which may be due to a condensation in the ortho-position. 4 5-Dichlorocntechol Methylene Ether (11). This substance may be obtained as already mentioned by the chlorination of piperonal and in this case i t seems that it may be obtained by the further action of chlorine on chloropiperonal, although the reaction is not smooth.It is also obtained by pass-ing chlorine into a solution of piperonylic acid which is always kept alkaline by the addition of aqueous sodium carbonate. Finally the same substance was prepared by chlorinating catechol methylene ether in acetic acid solution. It crystallises from alcohol in slender colourless needles melting at 8 2 O : 0.1218 gave 0.1866 AgC1. c1=37*5. C,H,O,Cl requires C1= 37.2 per cent. The substance is sparingly soluble in sulphuric acid and the solution is colourless but becomes reddish-yellow on the addition of a trace of nitric acid.The reaction is therefore not so charac-teristic as that of the corresponding dibromo-derivative which gives a rhodamine-red solution under these conditions. The con-stitution of the derivative follows from its production from chloro-piperonylic acid which must have the orientation of chloro-piperonal and therefore of chloronitrocatechol methylene ether, which as shown below must be a 4 5-derivative. 4-Chloro-5- bromocat echo1 M e t hyleii e Ether (IV). This substance could not be obtained by t,he chlorination of bromopiperonal in acetic acid solution and only in traces as the result of application of a large excess of reagent by the action of chlorine on an alkaline solution of bromopiperonylic acid. It was obtained in small amount by the bromination of chlore piperonal and comparatively readily by the action of bromine on an alkaline solution of chloropiperonylic acid.It was found tha 950 ORR ROBINSON AND WILLIAMS : the best conditions were to add the bromine water and aqueous sodium carbonate alternately to a dilute solution of the sodium salt of the acid and the completion of the reaction was judged by the amount of the precipitate formed. This was collected and crystallised from methyl alcohol from which i t separated in flat, satiny needles melting a t 78O. 0.1305 gave 0.1829 AgCl plus AgBr whereas this amount of a substance C,H,O,ClBr requires 0-1838 of the mixed saIts. The reaction of this substance in sulphuric acid on the addition of a trace of nitric acid was intermediate between that exhibited by the dichloro- and dibromo-derivatives.The reddish-brown solution appeared carmine in thin layers. A mixture of about ten parts of the dichlorocatechol methylene ether with one part of the dibromo-compound gave a solution almost identical in appearance with that from the chlorobromo-derivative. Application o f the Canizizzaro Reaction t o Piperonal. The formation of piperonylic acid and of homopiperonyl alcohol by the action of sodium hydroxide on piperonal does not appear to have been described and as we required piperonylic acid and the alcohol was being employed in another investigation we made some experiments on the conditions of the reaction and adopted the following procedure. A solution of sodium hydroxide (200 grams) in water (200 c.c.) was cooled to 40° and added to piperonal (100 grams) contained in a bottle capable of withstanding changes of temperature and the mixture vigorously shaken.Alcohol (50 c.c.) was then added and caused an almost immediate reac-tion and rapid rise of temperature. The bottle was vigorously shaken and the pressure released from time t o time. Soon the temperature fell and the mixture was then allowed to remain overnight. It was then mixed with sufficient water to dissolve the sodium piperonylate and extracted with benzene. The aqueous solution gave 53 grams of piperonylic acid on acidification, and from the benzene 1.8 grams of piperonal were recovered by washing with sodium hydrogen sulphite solution and after drying the extract and removing the solvent 32.5 grams of homopiperonyl alcohol remained in a pure condition.It readily crystallised and could be employed in most experiments in this condition. The substance may be recrystallised from light petroleum and occurs in slender colourless needles melting a t 58O. Fittig and Remsen (Annulen 1871 159 138) give the melting point as 51° but their product was purified by distillation THE ACTION OF HALOGENS O N PIPERONAL. 951 4-Chlorod-nitro ca t echoi! Me thylen.e Ether (V). 6-Chloropiperonal was dissolved in nitric acid (D 1*4) and after half an hour the mixture was added to water and the solid col-lected and crystallised from alcohol in which the substance is rather readily soluble. The nearly colourless needles melted a t 70° and became yebw by the action of light: 0.1231 gave 0.0867 AgC1.C1=17.2. C,H,O,NCl requires C1= 17.4 per cent.. This substance was also obtained from 5-nitro-4-aminocatechol methylene ether by diazotisation of the latter in hydrochloric acid and tr'eatment with copper powder. This connects the sub-stances described in the present cominunication with the series of 4 5-disubstituted catechol derivatives. On reduction the sub-stance yields a crystalline amino-derivative which can be diazo-tised and contains chlorine. 4-C'hl oro-5 6-ditzit.rocatech 01 Methyl e m Ether (VI) . The foregoing substance was dissdved in nitric acid (D 1.52), and a rather vigorous reaction ensued which had t o be moderated by cooling. On the addition of water a precipitate was obtained which was collected and crystallised from alcohol.The sparingly soluble yellow needles melted a t 138-141O : 0.1162 gave 0-0670 AgC1. C1=14-1. This substance becomes orange-yellow on exposure to light. C,H,0,N2Cl requires C1= 14.2 per cent'. 4-Chlor0-l 2-methylenedioxyphenant hrapheiurzin e, The dinitro-derivative just described was reduced in alcoholic aqueous hydrochloric acid solution by the addition of zinc dust, and the filtered solution mixed with excess of sodium acetate was heat'ed during two or three minutes with a solution of phenan-thraquir,one in aqueous sodium hydrogen sulphite. The brownish-yellow precipitate was collected and dried and crystal-lised from xylene. Glistening ochre-orange needles separated which melted and decomposed at 298-300° 952 BROWN AND ROBINSON VERATRICSULPHINIDE. 0.1079 gave 0.0422 AgC1. C1=9*6. C,,H,,O,N,Cl requires C1= 9.8 per cent. The substance dissolves in sulphuric acid to a purple solution, and in neutral solvents to yellow solutions which exhibit intense green fluorescence. UNIVERSITY ox SYDNEY. [Received September 4th 191 7 .