164 NTERENSTETN THE CONSTJTUTTON OF CATECHTN. PART TTI. XVIII.-The Constitution of Catechin. Part I I I . Synthesis of Acacatechin. By MAXIMILIAN NIERENSTEIN. IN the previous communications (T. 1920 117 971 1151) a provisional formula was suggested for catechin (I). A t the same time it was shown that catechin* tetramethyl ether yields on reduction with metallic sodium and alcohol and subsequent methyl-ation (compare Kostanecki and Lampe Ber. 1909 40 720), 3 4 21 4' 6~-pentamethoxy-au-diphenylpropane (11) which sub-stance may be oxidised to 3 4 2/ 4/ 6l-pentamethoxydiphenyl-aoetic acid (111). It was also shown that the acyl chloride of this acid (111) is converted by the action of diazomethane (compare Clibbens and Nierenstein T. 1915 107 1491) into 3 4 21 4/ 6'-pentamethoxydiphenylmethyl chloromethyl ketone (VI) from which 3 4 21 41 6/-pentamethoxy-au-diphenylpropane (11) is obtained on reduction with metallic sodium and alcohol.The present communication describes the synthesis of 2 4 6 31 4~-pentahydroxy-3-phenylchroman (I). This was carried out according to the following scheme 3 4 21 4' 61-pentameth-oxydiphenylmethyl chloromethyl ketone (VI) was converted into 4 6 31 4~-tetramethoxy-3-phenylchroman-2-one (V) which was reduced; to 2-hydroxy-4 6 31 4~-tetrarnethoxy-3-phenylchroman * The catechin used was obtained from Merck at Darmstadt in 1911 and 1920. After purification both preparations melted at 175-177" which is the melting point given by A. G. Perkin and Yoshitake (T. 1902 81 1183 1173) for catechin from gambier NIERENSTEIN THE CONSTITUTION OF CATECHIN.PART III. 165 (I+). pentahydroxy-3-phenylchroman (I). The latter product yielded on demethylation 2 4 6 3' 4l-0 0 0 MeO/\/><yH MeO/\/\' F€X2 MeO/\OMe Me0 CH t- Me0 CH -+- MeO/ I ICH-CO*CH,CI \./ I I t !!\,CH*OH \/\p / I \. I \ )OBIe '\I L O M e O O M e OMe OMe (IV. 1 (V.1 WI.) 2 4 6 31 4/-pentahydroxy-3-phenylchron1an (I) yields the original 2-hydroxy-4 6 3' 41-tetramethoxy-3-phenyl-chroman (IV). Both the latter product (IV) and 4 6 3/ 4I-tetra-methoxy-3-phenylchroman-2-one (V) give on reduction with metallic sodium and alcohol and subsequent methylation, 3 4 21 41 61-pentamethoxy-aa-diphenylpropane (11) which is in every respect identical with Kostanecki and Larnpe's (Zoc.cit .) methylated reduction product from Merck's catechin. 2 4 6 31 4/-Pentahydroxy-3-phenylchroman (I) is in every respect identical with acacatechin which occurs in Acacia catechu (A. G. Perkin and Yoshitake T. 1902 81 1169; Perkin T., 1905 87 398). The identity of these two substances is evident from the following comparative summary of melting points and from the fact that there is no depression in the melting points of the respective synthetic products when mixed with acacatechin and its penta-acetyl and pentabenzoyl derivatives kindly sent by Professor A. G. Perkin. 2 4 6 3 l 41-Yentahydroxy-3-phenylchroman sinters a t 140°, resolidifies a t about 180° on slowly raising the temperature and melts and decomposes a t 204-205O.OMe On methylation, Q 166 NIERENSTEIN THE COXSTITUTION OE' CATECHIN. PART 111. Acacatechin sinters a t about 1 40° resolidifies on slowly raising the temperature and melts and decomposes a t 204-205O (Perkin and Yoshitake loc. c i t . p. 1169). 2 4 6 3/ 41-Penta-acetoxy-3-phenylchroman sinters a t about 130° and melts and decomposes at 158-16O0. Penta-acetylacacatechin melts a t 158-159O (Perkin loc. c i t ., 2 4 6 3' 4/-Pentabenzoyl-3-phenylchroman melts at 181-183O. Pentabeiizoylacacatechin melts at 181-183O (Perkin and 2-Hydroxy-4 6 3/ 4/-tetramethoxy-3-phenylchroman melts at Acacatechin tetramethyl ether melts a t 152-154O (Perkin Zoc. cit. p. 400). 2-Acetoxy-4 6 31 4/-tetramethoxy-3-phenylchroman melts a t 136-137O.Monoacetylacacatechin tetrainethyl ether melts a t 135-13'7O (Perkin loc. c i t . p. 400). When fused with alkali 2 4 6 3' 4/-pentahydroxy-3-phenyl-chroman (I) gives phloroglucinol and protocatechuic acid. It has, however not been possible t o identify acetic acid which is also formed when acacatechin is fused (Perkin and Yoshitake Zoc. c i t . , p. 1170). This is probably due to the small amount of material (2.3 grams) which was used. p. 399). Yoshitake Zoc. cit. p. 1171). 152-153'. E X P E R I M E N T A L . 4 6 31 4~-Tetran~etl~ozy-3-phe.nylckro?nut~-2-o~~e (V). This substance may be prepared according t o either of the following methods : ( a ) A suspension of 5 grams of 3 4 2/ 4' 6/-peiitamethoxydi-phenylmethyl chloromethyl ketone (VI) in 50 C.C.of water is vigorously heated for thirty-two hours with 20 grams of sodium hydrogen carbonate. The cold solution is filtered and acidified with dilute hydrochloric acid when a bulky white precipitate is formed. The o-hydsoxy-3 4 21 41 6~-pentamethozy-aa-diphen.y1-MeO/)OMe ( (YH* CO c H,* OH Me0 ,,!, ()OM* OMe (VII. NIERENSTEIN THE CONSTITIJTION O F CATECHIN. PART 111. 167 propane-&one (VII) thus formed crystallises from dilute alcohol in small prismatic needles which melt and decompose at 121-122O. It is soluble in the usual organic solvents with the exception of light petroleum. The yield is 71 per cent. of the theoretical (Found * C= 63.8 ; H = 6.3. C,,H,,O requires C=63.8; H=6*4 per cent.). o-Hydroxy-3 4 2' 41 6'-pentainethoxy-aa-diphenylpropane-/3-one (VII) is converted into 4 6 3l 41-tetramethoxy-3-phenyl-chroman-2-one (V) by dissolving 5 grams in 50 C.C.of acetic anhydride and heating for eight hours on a water-bath with 25 C.C. of a solution of 4-32 grams (1 mol.) of acetyl chloride in 100 C.C. of acetic anhydride. After reducing the volume to about 50 c.c., water is added and the precipitate collected and warmed with a dilute solution of sodium hydroxide. The solid remaining is filtered off and freed from alkali. It crystallises from absolute alcohol in small pointed needles melting a t 146-147O. The pro-duct is soluble in the usual organic solvents. The yield is 51 per cent. of the theoretical (Found f- C =66.1; H=5.7. C,,H,,O, requires C = 66.3 ; H = 5.8 per cent.).( b ) To a solution of 5 grams of 3 4 2' 4' 6l-pentamethoxy-diphenylmethyl chloroinethyl ketone (VI) in 75 C.C. of dry benzene is added 0.1 t o 0.3 gram of aluminium chloride and the solution heated on a water-bath for twelve to sixteen hours. The solid left on evaporation of the benzene is purified by dissolving in 50 C.C. of 80 per cent. alcohol filtering and adding 300 C.C. of water to the filtrate. The precipit,ate thus obtained crystallises from absolute alcohol in small pointed needles which melt at 146-147O. The melting point of a mixture with the substance prepared accord-ing to the previous method shows no depression. The average yield of six preparations is 86 per cent. of the theoretical (Found 4 : C=66*2; H=5.6. Calc. C=66*3; H=5-8 per cent.).When condensed with phenylhydrazine in acetic acid solution, the phenylhydrazone C,,H,,O,:N,HPh is obtained. It crystallises from acetic acid in yellow needles which melt and decompose a t 236-239O (Found N = 6.5. C,,H,,O,N requires N = 6.4 per cent.). On reduction with metallic sodium and alcohol and subsequent methylation with diazomethane 2 grams of 4 6 31 4'-tetrameth-oxy-3-phenylchroman-2-one (V) gave 1-7 grams of 3 4 21 4/ 6'-pentamethoxy-aa-diphenylpropane (11) which melted at 83-84O. This melting point was not depressed when the substance was mixed with (1) the same substance obtained from catechin tetramethyl ether and (2) the synthetically prepared product (Zoc. cit. p. 1153). ''- Dried at loo" 168 NIERENSTEIN THE CONSTITUTION OE' CATECHIN.PART III. 2-Hydroxy-4 6 31 4/-tetrametkoxy-3-phenylchro~nan (IV). Three grams of 4 6 3' 4/-tetramethoxy-3-phenylchroman (V) dried a t looo are dissolved in 75 C.C. of acetic anhydride and heated in a boiling-water bath for five hours with 20 grams of zinc dust dried a t looo. The filtered solution is diluted with water and the precipitate collected after twenty-f our hours. The crude product is dissolved in 70 C.C. of alcohol and heated for several hours with 20 C.C. of a 20 per cent. solution of sulphuric acid. The product is again precipitated with water and treated in the cold with a 10 per cent. solution of sodium hydroxide when nearly all the solid dissolves. The filtered alkaline solution is acidified with dilute sulphuric acid and the precipitate crystallised from alcohol with the aid of animal charcoal.The average yield of five prepar-ations is 62 per cent. of the theoretical. This substance melts a t 152-153O which is the melting point (152-154°) given for acacatechin tetramethyl ether (Found * C = 65.8 ; H = 6.6. Calc. : C=65.9; H=6-4 per cent.). It differs however in one respect from Perkin's acacatechin tetramethyl ether in that it does not give the indigo-blue coloration with acetic and nitric acids observed by him for his product ( l o c . c i t . p. 400). Since this difference might be due t o the formation of an isomeric tetramethyl ether during direct methylation of acacatechin 2 grams of 2 4 6 31 4l-pentahydroxy-3-phenylchroman (I) were methylated with methyl sulphate according to Perkin's method.The tetramethyl ether obtained in this way melted a t 152-15307 and gave Perkin's indigo-blue coloration. Several mixed melting points of these two tetra-methyl ethers gave somewhat doubtful depressions of 1.5-Z0 but i t was found that both yielded on acetylation identical mono-acetyl derivatives (m. p. 136-137O) which showed not the slightest depression when their mixed melting point was taken (Found t : C=64.7; H=6*4. Calc. C=65.0; H=6*2 per cent.). It is 'difficult to see why the two tetramethyl ethers should show this difference unless Perkin's colour reaction is due t o a trace of another substance. Ferric chloride produces a violet coloration if added to a suspension of either 2-hydroxy-4 6 3' 4/-tetramethoxy-3-phenyl-chroman (IV) or 4 6 3' 4~-tetramethoxy-3-phenylchroman-2-one (V) in concentrated sulphuric acid.This is due to the 3-phenyl-chroman nucleus (compare Greenwood and Nierenstein T. 1920, 117 1594) and not to the coumaran nucleus as assumed by Kostanecki and Lampe (Ber. 1906 39 4007). On reduction with . * Dried at 110". t Dried at 100" NIERENSTEIN THE CONSTITUTION OF CATECHIN. PART 111. 169 metallic sodium ansd alcohol and subsequent methylation with diazomethane 3 grams of 2-hydroxy-4 6 3/ 4'-tetramethoxy-3-phenylchroman (IV) gave 1.9 grams of 3 4 21 4' 6'-pentameth-oxy-act-diphenylpropane (II) which melted a t 83-84O. This melting point was not depressed by admixture with other prepar-ations of this substance. 2 4 6 3' 4~-Pentahydroxy-3-phenylchroman (I), Four grams of 2-hydroxy-4 6 3' 4/-tetramethoxy-3-phenyl-chroman (IV) dissolved in 70 C.C.of glacial acetic acid are heated a t 130° in a sealed tube for six hours with 4 grams of acetyl chloride.* After opening the tube the content is a t first heated for several &ours with 50 C.C. of a 10 per cent. solution of sulphuric acid hydrogen being passed through continuously. The solution is then concentrated to about 10 C.C. in an atmosphere of hydrogen, diluted with 170 C.C. of water neutralised with solid barium carbonate filtered and extracted several times with ethyl acetate. The residue left on evaporation of the ethyl acetate dried over anhydrous sodium sulphate is purified according t o Perkin and Yoshitake's method (Zoc. c i t . p. 1163) by crystallising i t at first from 25 per cent.alcohol and subsequently dissolving in boiling ethyl acetate and benzene from which mixture it separates with the aid of animal charcoal in wart-like clusters consisting of small, pointed needles. The average yield of four preparations is 78' per cent. of the theoretical. 2 4 6 31 4'-Pentahydroxy-3-phenylchroman (I) is in every respect identical with acacatechin (Found j- C= 61.9. 62.0 ; H =4*9 6.0. Calc. C= 62.1 ; H=4-8 per cent.). It crystallises, like acacatechin from distilled water with 3H,O (Found + H,O = 15.8. Calc. H,O= 15.7 per cent.) and gives all the colour reac-tions of catechin including the phloroglucinol test with pine-wood and hydrochloric aci'd (compare Perkin and Yoshitake loc. c i t ., p.1172 ; Perkin loc. cit. p. 405). Both the penta-acetyl derivative (Found $ C = 60.1 ; H = 5.2. Calc. C = 60.0 ; H =4.8.per cent.) and the pentabenzoyl derivative (Found T C=73.9; H=4*6. Calc. C=74.1; H=4*2 per cent.) were prepared according to Perkin's method (Zoc. cit. p. 399) and * All attempts to demethylate both 2-hydroxy-4 6 3' 4'-tetramethoxy-3-phenylchroman (IV) and catechin tetramethyl ether (m. p. 144-146") from Merck's catechin with hydriodic acid have resulted in the production of amorphous substances. Acetyl chloride also gave good results in the caw of catechin tetramethyl ether. Thus 5 grams of this substanoe gave 3.8 grams of catechin which melted correctly at 175-177". t Dried at 160". $ Dried at 100" 1.70 FTNDLAY AND THOMAS TNFLUENCE OF COLLOIDS Perkin and Yoshitake’s method (Zoc. cifi. p. 1171) respectively. They proved in every respect identical with the corresponding derivatives of acacatechin. The author’s thanks are due to Professor A. G. Perkin for the specimens of acacatechin penta-acetylacacatechin and penta-benzoylacacatechin used in this investigation. He also thanks the Colston Society of the University of Bristol for a grant which has covered the expenses of this research. BIOCHEnf ICAL LAB ORATORY, CHEMICAL DEPARTMENT, UNIVERSITY OF BRTSTOL. [Received Dmmbcr 3014 1920.