KALBF AND ROBINSON A SYNTHESIS OF MYRICETIN ETC. 181 XXVIIL-A Synthesis of Myricetin and of a Galangin Monomethyl Ether Occurring in Galanga Root. By JAN IZALFF and ROBERT ROBINSON. ALLAN and ROBINSON (J. 1924 125 2192) have described an extremely simple method of preparation of 7-hydroxy-3-methoxy-flavone by hydrolysis of the product of the action of benzoic anhydride and sodium benzoate on o-methoxyresacetophenone a t 180-185". On applying the reaction to w-methoxyphloraceto-phenone (Slater and Stephen J. 1920 117 316) we have now obtained an excellent yield of 5 7-dihyciroxy-3-methoxyflavone (I), which is identical with the galangin monomethyl ether isolated from galanga root the rhizome of Alpinia officinarum (Hance) by Testoni (Gazzetla 1900 30 ii 327). The formation of phloro-glucinol and benzoic acid by the aerial oxidation of an alkaline solution of the substance was observed by A.G. Perkin and Allison (J. 1902 81 472) and led to the suggestion that it is galangin 3-monomethyl ether. This conclusion is now confirmed by syn-thesis and the occurrence of the methyl group in this position in the molecule of a naturally occurring flavonol is of interest because there is reason to believe that certain of the anthocyanidins are similarly constituted in respect of this structural detail. The attractive view that the anthocyanin and anthoxanthin pigments are intimately and genetically related derives support from the accumulation of such coincidences. Galangin itself is obtained by demethylation of the methyl ether and this flavonol has been previously synthesised by v.Kostanecki and Tambor (Ber. 1899, 32 2260). 0 0 Myricetin was first isolated by A. G. Perkin and Hummel (J., 1896 69 1287) from the bark of Myrica nagi (Thunb) and as a result of this and subsequent investigations (Perkin J. 1902 81, 204; 1911 99 1721 ; Perkin and Phipps J. 1904 85 62) has been regarded as 3 5 7 3' 4' 5' :-hexahydroxyflavone (I1 with H in place of Me). Myricetin is somewhat widely distributed in nature and it is the flavonol which should yield a delphinidin salt on reduction in acid solution. The synthesis of the colouring matter is a further example of the new and convenient method of prepara-tion of the flavonols. o-Methoxyphloroacetophenone was heate 182 KALFF AND ROBINSON A SYNTIIESIS OF MYRICETIN with trimethylgallic anhydride and sodium trimethylgallate and the product hydrolysed.The 5 7-dihydroxy-3 3' 4' 5'-tetra-methoxyflavone (11) thus obtained yields myricetin on demethyl-ation. Phloroglucinol is converted in this manner into mvyricetin through two isolated intermediate stages only and the synthesis is also available as a preparative method. E x P E R I AT E N T A L. Galangin 3-Monomethyl Ether (I).-A mixture of w-methoxy-phloroacetophenone (5 g . ) sodium benzoate (6 g.) and benzoic anhydride (15 g.) was heated (oil-bath at MOO) during 8 hours, after which alcohol (75 c.c,) was added to the dark red semi-solid mass and potassium hydroxide (8.5 g.) in water (10 c.c.) gradually introduced to the boiling solution. After refluxing for Q hour the greater part of the alcohol was evaporated the residue dissolved in water and the flavonol precipitated as a brown powder by saturating the liquid with carbon dioxide collected washed and dried (6.5 g.).The substance was purified through its diacetyl derivative prepared by the action of boiling acetic anhydride during 2 hours. After two crystallisations from alcohol (charcoal) the substance was obtained in long very pale yellow needles m. p. 175-176" (Found : C = 65.3; H = 4.4. C,,H,,O requires C = 65-2; H = 4.4 yo). Testoni (Gazzeita 1900 30 ii 327) states that the diacetate derived from the galangin monomethyl ether from galanga root melts a t 175-176'. The diacetylgalangin methyl ether was hydrolysed by means of an excess of 10% aqueous potassium hydroxide on the steam-bath.The phenol precipitated by hydrochloric acid may be crystallised from alcohol acetic acid or ethyl acetate in rectan-gular yellow plates m. p. 299" soluble in dilute aqueous potassium hydroxide to an intense yellow solution (sodium salt yellow needles) and in sulphuric acid to a yellow solution exhibiting green fluor-escence (Found C = 67.2; H = 4.4. CI6Hl2O5 requires C = 67.6; 13 = 4.3%). The properties of this substance agree with those ascribed to natural galangin monomethyl ether by Testoni (Zoc. cit.) and the identity was proved by the fact that the melting point of the synthetical material was not lowered by admixture with the natural product for a specimen of which we are greatly indebted to Professor A.G. Perkin. Demethylation by means of boiling hydriodic acid (d 1.'7) mixed with a quarter of its weight of acetic anhydride during 40 minutes resulted in the formation of galangin m. p. 214-215" after crystallisation from alcohol. Myricetin 3 3' 4' 5I-Tetramethyl Ether (II).-Pyridine (52 g.) was gradually added to a solution of trimethylgalloyl chloride (35 g.) (Perkin and Weizmann J. 1906 89 1655) in ether (250 c.c. AND OF A GALANGIN MONOMETHYL ETHER ETC. 183 and after 2 hours ice and then water was gradually introduced, and the trimethyl gallic anhydride which is sparingly soluble in ether collected washed with dilute aqueous sodium carbonate solution and dried (yield 75%). This product melted a t 159" (corr.) and was employed without further purification (Fischer and Freudenberg Ber.1913 46 1129 give 160-161" [corr.] as the m. p. of the pure substance). A mixture of trimethylgallic anhydride (23 g.)? sodium trimethylgallate (10 g.) and o-methoxyphloro-acetophenone (4.5 g.) was heatcd a t 175" for 3 hours. The mixture was a t first completely fluid but as the reaction proceeded it became a dark red paste. As in the previous example the flavonol was isolated by treatment of the crude product with an alcoholic solution of potassium hydroxide (8 g.) and precipitation of the phenolic material by means of carbon dioxide after removal of the alcohol and solution of the residue in water. The crude material (8.2 g.) was acetylated by boiling with an excess of acetic anhydride and the derivative crystallised from alcohol.3 3' 4' 5'-Tetramethylmyricetin diacetate was obtained in long pale yellow needles m. p. 159" (corr.) (Found C = 60.3; H = 4.9. C23H2a0, requires C = 60.2; H = 4.9%). On hydro-lysis by means of boiling concentrated hydrochloric acid this derivative yields myricetin tetramethyl ether which crystallises from alcohol in thin glistening pale yellow plates m. p. 276-277-5" (Found C = 60.6 ; H = 4.8. CI9Hl8O8 requires C = 60.9 ; H = 4.8%). This sparingly soluble substance is almost devoid of mordant dyeing properties. Myricetin.-The tetramethyl ether was demethylated by treat-ment for 2 hours with a boiling mixture of colourless hydriodic acid (d 1.7) (4 parts) and acetic anhydride (1 part). The red crystalline myricetin hydriodide was collected decomposed by hot water, and the yellow precipitate isolated.The substance crystallised from aqueous alcohol in bright yellow needles and showed all the highly characteristic reactions of myricetin as described by Perkin and his collaborators (Eoc. cit.). The remarkable behaviour with alkalis and with mineral acids the reactions with lead acetate and ferric chloride and the dyeing properties were examined. The melting point was about 360° but as stated by Perkin the deter-mination is difficult on account of blackening. For this reason the substance was converted in the usual manner into the hexa-acetate, which crystallised from alcohol in long silky colourless needles, In. p. 214-215" * (corr.) (Found C = 56.8; H = 3.7. Calc. for C2,H2,0,, C = 56.8; H = 3.9%).uncorrected. * The melting point 211-212' given in the literature is presumabl 184 MORGAN AND YABSLEY RESEARCHES ON We are greatly indebted to Professor A. G. Perkin for a specimen of myricetin hexa-acetate. The derivative prepared from natural myricetin when mixed with the synthetical material did not depress its melting point. Direct comparison further confirmed the identity of the specimens. We desire t o thank the Ramsay Memorial Fellowship Trust for a Fellowship (Netherlands) which has enabled one of us to take part in this investigation. THE UNIVERSITY MANCHESTER. [Received November 20th 13241 KALBF AND ROBINSON A SYNTHESIS OF MYRICETIN ETC. 181 XXVIIL-A Synthesis of Myricetin and of a Galangin Monomethyl Ether Occurring in Galanga Root. By JAN IZALFF and ROBERT ROBINSON.ALLAN and ROBINSON (J. 1924 125 2192) have described an extremely simple method of preparation of 7-hydroxy-3-methoxy-flavone by hydrolysis of the product of the action of benzoic anhydride and sodium benzoate on o-methoxyresacetophenone a t 180-185". On applying the reaction to w-methoxyphloraceto-phenone (Slater and Stephen J. 1920 117 316) we have now obtained an excellent yield of 5 7-dihyciroxy-3-methoxyflavone (I), which is identical with the galangin monomethyl ether isolated from galanga root the rhizome of Alpinia officinarum (Hance) by Testoni (Gazzetla 1900 30 ii 327). The formation of phloro-glucinol and benzoic acid by the aerial oxidation of an alkaline solution of the substance was observed by A. G. Perkin and Allison (J.1902 81 472) and led to the suggestion that it is galangin 3-monomethyl ether. This conclusion is now confirmed by syn-thesis and the occurrence of the methyl group in this position in the molecule of a naturally occurring flavonol is of interest because there is reason to believe that certain of the anthocyanidins are similarly constituted in respect of this structural detail. The attractive view that the anthocyanin and anthoxanthin pigments are intimately and genetically related derives support from the accumulation of such coincidences. Galangin itself is obtained by demethylation of the methyl ether and this flavonol has been previously synthesised by v. Kostanecki and Tambor (Ber. 1899, 32 2260). 0 0 Myricetin was first isolated by A. G. Perkin and Hummel (J., 1896 69 1287) from the bark of Myrica nagi (Thunb) and as a result of this and subsequent investigations (Perkin J.1902 81, 204; 1911 99 1721 ; Perkin and Phipps J. 1904 85 62) has been regarded as 3 5 7 3' 4' 5' :-hexahydroxyflavone (I1 with H in place of Me). Myricetin is somewhat widely distributed in nature and it is the flavonol which should yield a delphinidin salt on reduction in acid solution. The synthesis of the colouring matter is a further example of the new and convenient method of prepara-tion of the flavonols. o-Methoxyphloroacetophenone was heate 182 KALFF AND ROBINSON A SYNTIIESIS OF MYRICETIN with trimethylgallic anhydride and sodium trimethylgallate and the product hydrolysed. The 5 7-dihydroxy-3 3' 4' 5'-tetra-methoxyflavone (11) thus obtained yields myricetin on demethyl-ation.Phloroglucinol is converted in this manner into mvyricetin through two isolated intermediate stages only and the synthesis is also available as a preparative method. E x P E R I AT E N T A L. Galangin 3-Monomethyl Ether (I).-A mixture of w-methoxy-phloroacetophenone (5 g . ) sodium benzoate (6 g.) and benzoic anhydride (15 g.) was heated (oil-bath at MOO) during 8 hours, after which alcohol (75 c.c,) was added to the dark red semi-solid mass and potassium hydroxide (8.5 g.) in water (10 c.c.) gradually introduced to the boiling solution. After refluxing for Q hour the greater part of the alcohol was evaporated the residue dissolved in water and the flavonol precipitated as a brown powder by saturating the liquid with carbon dioxide collected washed and dried (6.5 g.).The substance was purified through its diacetyl derivative prepared by the action of boiling acetic anhydride during 2 hours. After two crystallisations from alcohol (charcoal) the substance was obtained in long very pale yellow needles m. p. 175-176" (Found : C = 65.3; H = 4.4. C,,H,,O requires C = 65-2; H = 4.4 yo). Testoni (Gazzeita 1900 30 ii 327) states that the diacetate derived from the galangin monomethyl ether from galanga root melts a t 175-176'. The diacetylgalangin methyl ether was hydrolysed by means of an excess of 10% aqueous potassium hydroxide on the steam-bath. The phenol precipitated by hydrochloric acid may be crystallised from alcohol acetic acid or ethyl acetate in rectan-gular yellow plates m.p. 299" soluble in dilute aqueous potassium hydroxide to an intense yellow solution (sodium salt yellow needles) and in sulphuric acid to a yellow solution exhibiting green fluor-escence (Found C = 67.2; H = 4.4. CI6Hl2O5 requires C = 67.6; 13 = 4.3%). The properties of this substance agree with those ascribed to natural galangin monomethyl ether by Testoni (Zoc. cit.) and the identity was proved by the fact that the melting point of the synthetical material was not lowered by admixture with the natural product for a specimen of which we are greatly indebted to Professor A. G. Perkin. Demethylation by means of boiling hydriodic acid (d 1.'7) mixed with a quarter of its weight of acetic anhydride during 40 minutes resulted in the formation of galangin m.p. 214-215" after crystallisation from alcohol. Myricetin 3 3' 4' 5I-Tetramethyl Ether (II).-Pyridine (52 g.) was gradually added to a solution of trimethylgalloyl chloride (35 g.) (Perkin and Weizmann J. 1906 89 1655) in ether (250 c.c. AND OF A GALANGIN MONOMETHYL ETHER ETC. 183 and after 2 hours ice and then water was gradually introduced, and the trimethyl gallic anhydride which is sparingly soluble in ether collected washed with dilute aqueous sodium carbonate solution and dried (yield 75%). This product melted a t 159" (corr.) and was employed without further purification (Fischer and Freudenberg Ber. 1913 46 1129 give 160-161" [corr.] as the m. p. of the pure substance). A mixture of trimethylgallic anhydride (23 g.)? sodium trimethylgallate (10 g.) and o-methoxyphloro-acetophenone (4.5 g.) was heatcd a t 175" for 3 hours.The mixture was a t first completely fluid but as the reaction proceeded it became a dark red paste. As in the previous example the flavonol was isolated by treatment of the crude product with an alcoholic solution of potassium hydroxide (8 g.) and precipitation of the phenolic material by means of carbon dioxide after removal of the alcohol and solution of the residue in water. The crude material (8.2 g.) was acetylated by boiling with an excess of acetic anhydride and the derivative crystallised from alcohol. 3 3' 4' 5'-Tetramethylmyricetin diacetate was obtained in long pale yellow needles m. p. 159" (corr.) (Found C = 60.3; H = 4.9. C23H2a0, requires C = 60.2; H = 4.9%).On hydro-lysis by means of boiling concentrated hydrochloric acid this derivative yields myricetin tetramethyl ether which crystallises from alcohol in thin glistening pale yellow plates m. p. 276-277-5" (Found C = 60.6 ; H = 4.8. CI9Hl8O8 requires C = 60.9 ; H = 4.8%). This sparingly soluble substance is almost devoid of mordant dyeing properties. Myricetin.-The tetramethyl ether was demethylated by treat-ment for 2 hours with a boiling mixture of colourless hydriodic acid (d 1.7) (4 parts) and acetic anhydride (1 part). The red crystalline myricetin hydriodide was collected decomposed by hot water, and the yellow precipitate isolated. The substance crystallised from aqueous alcohol in bright yellow needles and showed all the highly characteristic reactions of myricetin as described by Perkin and his collaborators (Eoc.cit.). The remarkable behaviour with alkalis and with mineral acids the reactions with lead acetate and ferric chloride and the dyeing properties were examined. The melting point was about 360° but as stated by Perkin the deter-mination is difficult on account of blackening. For this reason the substance was converted in the usual manner into the hexa-acetate, which crystallised from alcohol in long silky colourless needles, In. p. 214-215" * (corr.) (Found C = 56.8; H = 3.7. Calc. for C2,H2,0,, C = 56.8; H = 3.9%). uncorrected. * The melting point 211-212' given in the literature is presumabl 184 MORGAN AND YABSLEY RESEARCHES ON We are greatly indebted to Professor A. G. Perkin for a specimen of myricetin hexa-acetate. The derivative prepared from natural myricetin when mixed with the synthetical material did not depress its melting point. Direct comparison further confirmed the identity of the specimens. We desire t o thank the Ramsay Memorial Fellowship Trust for a Fellowship (Netherlands) which has enabled one of us to take part in this investigation. THE UNIVERSITY MANCHESTER. [Received November 20th 13241