790 LAPWORTH AND WYKES: LXVI.-Tiie Pungent Pyinciples of Ginger. Part Synthetic Prepcwatz’ons of Zingeq-one Methyl- II. zingerone and Some Related Acids. By ARTHUR LAPWORTH and FREDERICK HENRY WYKES. IN Part I. evidence was adduced that the phenolic ketone, (‘ zingerone,” obtained from gingerol the pungent oleo-resin of ginger was 4-hydroxy-3-methoxyphenylethyl methyl ketone (I) or a position isomeride and that ‘( methylzingerone,” the correspond-ing ketone from “methylgingerol,” is its methyl ether (11) 3 4-dimethoxyphenylethyl methyl ketone. CH2* OH,* CO-CH CH,*CH,*CO*CH, /\ /\ (1.1 (11.1 The authors have prepared these and also several important acids‘ that were required for purposes of comparison by synthetic methods which in certain cases represent the first direct syntheses of the compounds in question and in other cases are simpler or give better yields than previous methods.The synthesis of methylzingerone presented no difficulty as 3 4-dimethoxystyryl methyl ketone (11) from veratraldehyde and acetone is easily niade in good yield and is readily reduced by sodium amalgam giving 3 4-dimethoxyphenylethyl methyl ketone, which is identical with methylzingerone (11). CH:CH*CO*CH ciio C H:CH* CO-CH, /\ /\ /\ I ‘ ! ,!OMe ( / 0 1 \ l e OH \/oale OH OMe (111.) (IV. 1 (V.) A similar process for preparing 4-hydroxy-3-methoxyphenylethyl methyl ketone by condensing vanillin (IV) with acetone and reducing the resulting 4-hydroxy-3-methoxystyryl methyl ketone (V) gave very poor yields a t each stage but the final product wa THE PUNGENT PRINUIPLES O F GINGER.PART 11. 791 identical with zingerone. On the other hand good yields were quickly obtained by the following process. Vanillin (V) gives an excellent yield of ethyl vanillylideneaceto-acetate (VI) by Knoevenagel and Albert’s method (Ber. 1904, 37 4476); this unsaturated ester was easily reduced by means of sodium amalgam and the product with excess of alkali was con-verted into an acid doubtless vanillylacetoacetic acid (VII) which when heated lost the elements cf carbon dioxide and was con-verted into zingerone (V). CH :CAc:CO,Et CH,*CHAc*CO,H CH:CAc, /\ 1 , /\ /’\ <)OM0 ( ) 0 Mi3 \/io&le OH OH OH (VI. 1 (VII.) (VIII.) Synthetic zingerone was obtained in crystalline form and had the characteristic sweet odour and pungent taste of the ketone from the decomposition of gingerol.The solid when used to infect samples of ketone obtained from the drug caused these to solidify f o r the first time. The properties of the synthetic ketone were identical in every respect with those of the compound from ginger. An attempt to imitate this synthesis through vanillylideue-acetylacetone (VIII) was not successful and the reduction pro-ducts of the latter compound were not of the type expected. The authors’ work included some simple direct syntheses of hydroferulic acid (XI) and of hydrocaffeic acid (XIII). These have previously been obtained by the reduction of synthetic ferulic and caffeic acids (Tiemann and Nagai Ber. 1878 11 650 672) prepared by the Perkin “ cinnamic acid synthesis.” As the author&’ methods although obvious enough are new if not in principle and are very easily carried out the steps may be indicated.Vanillin (V) was condensed with diethyl malonate or ethyl cyanoacetate yielding diethyl vanillylidenemalonate and ethyl v anill ylidenec y ano ace t a t e (IX) respectively. These were readily reduced a t the double bond and on subsequent hydrolysis with excess of alkali were converted into acids doubtless vanillylmaloaic acid (X) in both instances. The product when heated yielded hydroferulic acid (XI). CH C( CN) *CO,Eb CH2*CH(C0,H) CH,*CH,* CO,H /\ /\ /\ (S.) I ‘ \,/OM‘ OH (XT.) I 792 LAPWORTH AND WYKES A precisely analogous process applied to protocatechualdehyde instead of vanillin gave hydrocaffeic acid (XIII) through the inter-mediate condensation product (XII).CH:C(CN)*CO,Et CH ,*CH,*CO,H /\ /\ !,)OH OH (XII.) OH (XIII. ) It is worthy of remark that all the unsaturated phenolic ketones and esters (I V VIII IX and XII) give solutions in alkali which in the thinnest layers exhibit an intense yellow colour that dis-appears on reduction a t the double bond. The effect is perhaps most pronounced with (XII) which exhibits with boric acid a reaction very like that associated with curcumin the colouring matter of turmeric. The constitution which Kostanecki suggests for curcumin is (IH:CH*CO* CH,-CO*CH CH /\ in which case it is obviously very closely relat-ed to the similar compounds dealt with in the present paper.The question of the groupings essential t o the pungency of gingerol zingerone and similar compounds is one which one of us is hopeful of reserving for a short time. So far it would appear certain that the presence of the free phenolic hydroxyl group is essential and also not. improbably the ketonic carbonyl suitably disposed in a saturated chain att*ached t o the phenolic residue. E x P E R I M E N T A L . I . S p t h e s i s of Zingerone (4-Eydroxy-3-methoxyp~~enylethyl Methyl Ketone). Preparation of Ethylcarborntovanillin, OM0 -CO,Et*O/ \CHO \-/ As vanillin and its acetyl and benzoyl derivatives did not readily condense with acetone the ethylcarbonato-derivative which is new, was prepared by adding one molecular proportion of ethyl chloro-formate to vanillin dissolved in the requisite quantity of AT-sodiur THE PUNGENT PRINCIPLES OP GINGER.PART 11. $93 hydroxide. The derivative separated in the cold in small white needles and after an hour was collected and crystallised from hot alcohol. Found C = 63.3 ; H = 5.9. CI1Hl2O4 requires C = 63.5 ; H = 5.8 per cent. The substance crystallises from alcohol in slender needles and melts a t 65O. When it was dissolved in excess of acetone and the mixture treated with a little dilute sodium hydroxide an intense yellow colour was pro-duced and apparently a small quantity of the desired condensa-tion product was formed as the neutral product of the reaction, when reduced with sodium amalgam gave an oily mixture which had a pronounced pungent taste. The authors have not yet had opportunities to follow up these observations.It is slowly hydrolysed by cold dilute alkali. R eductisni of Vanillylidene Derivatives of A cet ylacet one a nd E t h y l Acetoacetate. Formatioit of Zingerone. Vanillylideneacetylacetone which was prepared by Knoevenagel and Albert's method (Ber. 1904 37 4480) forms an intensely yellow solution in sodium hydroxide; this colour is discharged by shaking the solutioii with sodium amalgam and on saturating the resulting liquid with carbon dioxide an oil doubtless vanillyl-acetylacetone HO*C,H3(0Me)*CH2*CH(CO*CII,), is deposited and ultimately tends to crystallise; as the oil when heated with acid or alkaline hydrolytic agents gave no product with the characters of vanillylacetone its further investigation was not undertaken.Ethyl vanillylideneacetoacetate which was also obtained in nearly theoretical yield by Knoevenagel and Albert's method (luc cit. p. 4476) crystallised in pale yellow needles melting a t 112*5-113*5° and as Knoevenagel and Albert give the melting point as 120-121° the author? analysed their product. (Found, C = 63.3 ; H = 6.0. The ester was dissolved in 10 per cent. aqueous sodium hydroxide and the intensely yellow solution shaken violently with washed, fluid sodium amalgam the whole being kept very cool. When the solution no longer displayed a yellow colour in thin layers it was separated from mercury mixed with 35 per cent. of its weight of solid sodium hydroxide and heated for about eight hours on the water-bath when it was coslsd saturated with carbon dioxide, and extracted with ether to remove unhydrolysed ethyl vanillyl-acetoacetate OMe*C,H3(OH)*CEf,~CHAc*C0,Et an oily com-pound which represents the first stage in the reduction process.CI4Hl6O5 requires C =.63*5 ; H = 6.1 per cent.) 11 794 LAPWORTH AND WYKES: The aqueous residue was next acidified and extracted with ether, the latter then being dried and evaporated. The oily extract was belated in a vacuum when a t first carbon dioxide was evolved, and the residual material distilled a t 175-210°/15 mm. I n order to separate the products the distillate was dissolved in aqueous sodium hydroxide which was then saturated with carbon dioxide and extracted with ether (“ phenolic extract ”), the aqueous residue being subsequently acidified with hydrochloric acid and again extracted with ether (“ acidic extract ”).The “phenolic extract” when dried and evaporated left a brown oil. This was dissolved in N-sodium hydroxide solution and treated in the cold with ethyl chloroformate when the bulk of the material in solution was converted into an oil which solidified on scratching the vessel with a glass rod. The solid was dissolved in ether thO solution dried and allowed to evaporate spontaneously. Large tabular crystals were deposited which on recrystallisation from light petroleum formed colourless hex-agonal plates melting a t 47.50. Found C = 62.7 ; H = 6.7. CI4H,,O requires C = 63.1 ; H = 6.8 per cent. The substance was identical in all respects with the ethyl-carbonate-derivative of the ketone (“ zingerone ”) obtained from gingerol.I n order to obtain the free phenolic ketone the foregoing com-pound was heated on the water-bath with dilute aqueous sodium hydroxide until a homogeneous liquid resulted. Excess of hydro-chloric acid was then added and the cooled product extracted with ether. After drying and evaporating the ethereal extract left a residue which was distilled in a vacuum when the dis-tillate set to a crystalline mass. The solid material obtained as above was purified by dissolv-ing it in dry ether adding enough petroleum t o cause a turbidity, allowing the latter t o settle and then infecting the clear solution with a trace of solid 4-hydroxy-3-methoxyphenylethyl methyl ketone. On spontaneous evaporation the liquid deposited lustrous, flat colourless crystals which had the odour of the above ketone and melted a t 36-37O.Found C = 67.7 ; H = 7.2. C,,H,,O requires C = 68.0 ; H,= 7.3 per cent. 4-Hydroxy-3-methoxyphenylethyl methyl ketone obtained in this way had an extremely pungent taste and was in every resEect identical in properties with “ zingerone.” As has already been mentioned the ‘* zingerone ” from the natural ~ource was no THE PUNGENT PRINCIPLES OF GINGER. PART 11. 795 obtained in crystalline condition until infected with a trace of the synthetic ketone which had solidified spontaneously after dis-tillation in a vacuum. 4-Hydroxy-3-methoxyphenylethyl methyl ketone is not the only product which is formed by reduction of ethyl vanillylideneaceto-acetate and subsequent hydrolysis.The " acidic extract " (com-pare p. 794) gave on evaporation a small quantity of an acid which was moderately soluble in cold water and readily so in hot; this formed colourless leaflets melting a t 133-134O and was a t first believed t o be hydrocaffeic acid which also crystallises in leaflets and melts a t 137-139O. The substance obtained as above, however unlike hydrocaff eic acid gives no coloration with ferric chloride and when it is mixed with hydrocaffeic acid the mixture melts a t 127-132O. A titration with alkali using phenol-phthalein as indicator gave an equivalent' for thi4 acid of about 240; the quantity of this material obtained in the pure state was, however too small to permit of further investigation. Oxidation and Reduction of 3 4-Dimethoxystyryl Methyl Retone.Formation of Dimethylcaffeic Acid and of Methylzingerone. 3 4-Dimethoxystyryl methyl ketone was prepared by condensing veratraldehyde with acetone (compare Francesconi and Cusmano, Gazzetta 1908 38 ii 70 et sep.) and purifying it by recrystallisa-tion from light petroleum. C,,H,,O, requires C=69*9; H=6*8 per cent.) The compound crystallises from carbon tetrachloride in micro-scopic leaflets. When it is warmed with concentrated hydro-chloric acid it gives a deep red coloration doubtless due to the intermediate formation of veratraldehyde which gives a similar reaction. 3 4-Dimethoxystyryl methyl ketone is readily oxidised when shaken with aqueous sodium hypobromite being converted into dimethylcaffeic acid C,H,(OMe),*CH:CH-CO,H which was isolated in small flat needles (from water) rnelting a t 180-181O.Reduction of 3 ddimethoxystyryl methyl ketone with the aid of Paal and Skita's or Willstatter's methods did not proceed in a very satisfactory manner but when an alcoholic solution of the compound was shaken with liquid sodium amalgam the bright yellow colour rapidly lost its intensity and when this process was carried out in presence of excess of potassium hydrogen carbonate, good yields of the desired reduction product were obtained. It was isolated by diluting the aqueousalcoholic solution with wat'er, extracting with ether and shaking the ethereal extract with (Found C = 70.0 ; H = 6.8. I I* 796 LAPWORTH AND WYKES: freshly prepared sodium hydrogen sulphite solution.The solid hydrogen sulphite compound was collected washed with ether, dried and then decomposed by warming it with excess of aqueous sodium carbonate. By extracting the resulting liquid with ether, 3 4-dimethoxyphenylethyl methyl ketone was obtained in quantity corresponding with about 80 per cent. of that theoretically possible. It was purified by crystallisation from methyl alcohol. Found C = 69.4 ; H = 7.8. C,,H,,O requires C = 69.2 ; H = 7.7 per cent. The substance formed colourless odourless needles melting at 55-56O and its identity with “ methylzingerone,” obtained by methylating “ zingerone ” or by the decomposition of ‘‘ methyl-gingerol,” was established by the usual methods. Its colour reactions with hydrochloric acid and the properties of its oxirne (long white needles m.p. 920) were indistinguishable from those of the corresponding compounds obtained from ‘‘ gingerol.” 111. Synthesis of Hydroferulic Acid. E t h y 1 Vanill ylidenec yanoac e t at e, HO( )CH:C(C N)-C0,Et . Vanillin and ethyl cyanoacetate were mixed in molecdar pro-portions and heated on the water-bath until a homogeneous liquid was obtained when a few drops of piperidine were added and the heating was continued until a test portion solidified com-pletely. Alcohol (twice the weight of vanillin present) was added, and the whole allowed to cool. The crystals obtained were washed with dilute hydrochloric acid dried and crystallised from alcohol. With material obtained from mother liquors the total yield of condensation product approached that theoretically possible.OMe .~ ___ Found C = 62*2 ; H,= 5.3. Ethyl vanillylidenecyanoacetate separates from alcohol in yellow needles melting a t 107O. It dissolves in aqueous sodium hydroxide giving a solution which shows an intensely yellow colour even in thin layers; it is reprecipitated unchanged from this solution on the additioa of acids. C,H,,O,N requires C = 63.1 ; H = 5.3 per cent. Formation of Hydroferulic Acid from Ethyl Vanillylideize-cyanoacetate. The reduction of ethyl vanillylidenecyanoacetate was accom-plished by means of sodium amalgam in precisely the same manne THE PUNGENT PRINCIPLES OE GINGER. PART 11. 797 as was used for the reduction of ethyl vanillylideneacetoacetate (p. 793). The reduction product [mainly no doubt a mixture of vanillylcyanoacetic acid HO*C,H,(OMe)*CH,*CH (CN) *CO,H , and its ethyl ester] was heated with excess of potassium hydr-oxide until the ester present had been completely hydrolysed when excess of acid was added and the liquid extracted with ether.The latter was evaporated and the residue heated in a vacuum to expel carbon dioxide; the hydroferulic acid left was purified by converting it into its lead salt which is very sparingly soluble, in water and decomposing the latter in aqueous suspension with hydrogen sulphide. Hydroferulic acid obtained in the above way crystallises from hot water in stout white needles melting ah 89-90°. It dis-solves readily in hot water less readily in cold and only spar-ingly in concentrated hydrochloric acid.Its aqueous solution gives no coloration with ferric chloride. (Found C = 60.6 ; H=6*2. C,,H120 requires C=60*6; H=6*1 per cent.) The acid agreed very closely in properties with those assigned by Tiemann and Nagai (Ber. 1878 11 650) to the acid obtained by the reduction of ferulic acid. Diethyl vanillylidenemalonate HO*C,H,(OMe)*@H:C(CO,Et),, was prepared from vanillin and ethyl malonate by Knoevenagel and Albert’s method (Zoc. cit. p. 4481). It had the properties ascribed to i t by these authors. On reduction with sodium amalgam and subsequent treatment in the manner described in the case of ethyl vanillylidenecyane acetate i t yields hydroferulic acid. Iv. Synthesis of Hydrocaffeic A c i d . Hydrocaffeic acid was prepared by Tiemann and Nagai (Ber., The 1878 11 672) by the reduction of synthetic caffeic acid.following is a Con,densation simple alternative synt,hesis. of Protocatechualdehyde with Ethyl Cyanoacetate. Formation of Ethyl a-CyanocafJeat e, OH HO’ \CH:C(CN)*CO,Et. \-/ This condensation was effected by means very similar to those used in similar condensations in previous sections but it was found desirable to dilute the mixture of aldehyde and ester with a little absolute alcohol. The product which was contaminate 798 THE PUNGENT PRINCIPLES OF GINGER. PART 11. with a brown impurity was purified by extraction with benzene and recrystallisation therefrom. Found C= 61.6 ; H =4*9. CI2H,,0,N requires C = 61.8 ; H = 4.7 per cent. Ethyl a-cyanocaffeate is a yellow microcrystalline solid which melte somewhat indefinitely a t 162-166O and dissolves in alkali to give an intensely orange solution.A test-paper made by dipping paper in an alcoholic solution of this ester behaves towards boric acid in much the same way as does turmeric paper, that is if moistened with boric acid solution and then heated it turns brown but the temperature required is somewhat higher than with turmeric. Formation of Hyd.rocaffeic Acid from Eth.yZ a-Cyanocafleate. A solution of ethyl a-cyanocaffeate in alkali was reduced with sodium amalgam until colourless. The resulting liquid which very quickly turned brown on exposure to air was made strongly alkaline by the addition of solid sodium hydroxide boiled for two days under a reflux condenser cooled acidified and extracted with ether. The ethereal extract gave an oil which soon deposited crystals; these were not isolated but the whole was heated t o expel carbon dioxide then dissolved in water treated with animal charcoal and allowed to crystallise. The crystals of hydrocaffeic acid obtained in this way were hexagonal leaflets melting a t 138-139O were moderately soluble in cold water readily so in hot and their aqueous solution gave a green colour with ferric chloride changing t o a rich purple on the addition of ammonia. These properties correspond closely with those assigned to hydro-caffeic acid by Tiemann and Nagai (Zoc. cit.). The authors have also found that hydrocaffeic acid is readily obtained from hydroferulic acid by heating i t with dilute hydro-chloric acid (about 5 per cent.) a t 200° for six hours in a closed tube. Acknowlsdgmente are due to Mrs. L. Kletz Pearson who kindly carried out some preliminary experiments on the preparation of vanillin and verat.raldehyde derivatives. ORGANIC CHEMICAL LABORATORIES, THE UNIVERSITY MANCHESTER. [Received April 2 7th 19 1 7.