2306
COMMUNICATIONS TO T H E
EDITOR
Vol. 72
molecular formula c24H&. Anal. Calcd. : c, 74.29; H, 9.34. Found: C, 74.53; H, 9.18. The additional carbon atom was shown to be in a methoxyl group by Zeisel determination (calcd. 7.99; found 8.59). When an ethanol solution of potassium hydroxide was used for the reaction a different product, the corresponding ethoxy compound, was obtained, m. p. 143-144.5') [.ID - 30.6' (chloroform). These results clearly prove that the reaction led to the formation of an ether rather than an alcohol. An examination of Marker's analytical data shows that there is equally good agreement for a methoxy derivative. The addition of alcohol to an a,P-unsaturated 20-ketosteroid is entirely analogous to the baseHICKRILL CHEMICAL RESEARCH FOUNDATION W. VON E. DOERING catalyzed addition of alcohols to methyl acrylate LAWRENCE H. KNOX or acrylonitrile to yield @-alkoxy propionic acid KATONAH, NEWYORK derivative^.^ The product formed from 3PRECEIVED APRIL6, 1950 acetoxy- Ab*l6-pregnadien-2O-one upon treatment with methanolic potassium hydroxide is therefore THE ACTION OF METHANOLIC POTASSIUM HY- 3P-hydroxy-l B-methoxy- A5-pregnen-20-one. The DROXIDE ON A A'B-20-KETOSTEROID: ISOLATION reaction is reversible, and an equilibrium mixture OF 3p-ACETOXY- 16-METHOXY-A'-PREGNEN-20- containing 69% of the 16-methoxy compound ONE spectrophotometrically is reached from either Sir : compound a t 23' in 3% methanolic potassium The reaction recently described by Marker' hydroxide in about two hours. for the introduction of a 17-hydroxyl group could (5) Koelsch, THISJOURNAL, 65, 437 (1913). have great significance in the synthesis of adrenal SLOAN-KETTERING INSTITUTE FOR CANCER RESEARCH DAVIDK. FUKUSHIMA cortical hormones. Since no evidence was offered T. F. GALLAGHER for the presence of a 17-hydroxyl group other than NEWYORK,N. Y. RECEIVED MARCH 21, 1950 failure to acetylate the newly introduced oxygen function, and since the reaction postulated was improbable, we have investigated the problem and A NEW ROUTE TO 11-KETOSTEROIDS have arrived a t a completely different conclusion. Sir: 30-Acetoxy- 16-pregnadien-20-one was reWhereas 3-hydroxyl groups usually resist atfluxed with a solution of potassium hydroxide in tack by N-bromosuccinimide in aqueous acetone,' methanol. As evidenced by the ultraviolet methyl 3a-hydroxy-ga,lla-oxidocholanate (I, absorption spectrum, E::*, = 121 a t 2390 A., ap- from the 9,11-ethylene,2*3m. p. 135.4-136.4', proximately half the product no longer contained [ a I z 5 D +22' Chf.4 Anal. C25H4004: c, 74.21, the alp-unsaturated carbonyl group of the initial H, 9.97. Found: C, 74.51; H, 10.09) is oxidized compound. Acetylation followed by chromatog- to the 3-ketone (11, m. p. 129-130°, [ C U ] 4-4' ~~D raphy resulted in the separation of 30-acetoxy- DL4 Anal. C]25H3804: C, 74.52; H, 9.52. A5016-pregnadien-20-oneand a compound, m. p. Found: C, 74.34; H, 9.38. Semicarbazone, m. p. 158.5-159.5'; [ a I z 5-28.5' ~ (chloroform), which 209-210'; Wolff-Kishner reduction to 9 a ) l l a showed no absorption in the ultraviolet above oxidocholanic acid, m. p. 158-159', [ a ] 2 $17' 2~ 2250 A. The possibility that the compound was DL5 Anal. C&H3803: C, 76.96; H, 10.23. a 3P-acetoxy-17-hydroxy- A5-pregnen-20-onewas Found: C, 76.93; H, 10.42). Chromic acid oxireadily eliminated by comparison of the physical dation of I or I1 gives the hemiketal 111, rn. . constants with the known epimers ( 1 7 d m. p. 119-120' (variable), [ a I z 3+102' ~ Chf., X,"', 234-235'; [ a ] D -41.8' [dioxane]; 17@-3m. p. 2.85, 5.78, 5 . 8 5 ~ . Anal. Cj25H3805: C, 71.66; 187-188'; [a]D -61.3' (chloroform)). A D9.16. Found: C, 71.89; H, 9.20. Acetate homosteroid was similarly discarded as a possible H, (BFa-AczO), p. 148.6-149.8', [aIz4D +100.3' interpretation. The most important evidence Chf. Anal. m. C27H4006: 70.38; H, 8.77. against these structures was provided by infrared Found: C, 70.27; H, C,8.76. ether spectrometry4 which revealed that no free hy- (CH30H-HBr), m. p. 124.8-125.5', Methyl [a]'*D f92.3' droxyl group was present in the product. (1) Fieser and Rajagopalan, THIS JOURNAL, 11, 3935, 3938 The elementary analysis conformed to the (1949). (2) Seebeck and Reichstein, Hclv. Chim. Acto, 26, 536 (1943). (1) Marker, THISJOURNAL, 71, 4149 (1949). mp (log E ) a t 235 (4.26)) 327 (3.78), 368 (3.59) and 392 (3.11) and minima a t 273 (3.25)) 363 (3.57) and 387 (3.30). I1 absorbs four molar equivalents of hydrogen in ethanol with platinum oxide giving an oil oxidized with permanganate to pimelic acid, m. p. 99-102' (m. p. 100- 102' in admixture with authentic material, m. p. 101102'). These facts are consistent with the assignment of tropolone (IIa) as the structural hypothesis for I1 but do not completely exclude IIb. This investigation was supported in part by a research grant from the National Institutes of Health, Public Health Service, and in part by a research grant from the American Cancer Society.
A51
B
(2) Hegner and Reichstein, Hrlu. Chim. A c f a , 42, 828 (1941). (3) Shoppee and Prius, ibid., 26, 201 (1943). (4) We wish to express our appreciation to Dr. Konrad Dobriner of this Institute, who determined and interpreted the infrared spectra for us.
(3) Mattox, Turner, Engel, McKenzie, McGuckin and Kendall, J . B i d . Chcm., 162, 569 (1946). (4) Chf. chloroform; Di. dioxane; An. acetone. (5) Alther and Reichstein, Rclu. Chim. Acta, 16, 492 (1943), give [U]D +18.8' An. for the methyl ester.
-
-
-
May, 1950
COMMUNICATIONS TO THE
EDITOR
2307
method of preparation from other sources is being sought. CONVERSE MEMORIAL LABORATORY LOUISF.FIESER HARVARD UNIVERSITY HANSHEYMANN CAMBRIDGE 38, MASS. SRINIVASA RAJAGOPALAN RECEIVED APRIL15, 1950 HO COMPOSITION OF SYNTHETIC LIQUID FUELS. I. PRODUCT DISTRIBUTION AND ANALYSIS OF Cs-Cs PARAFFIN ISOMERS FROM COBALT CATALYST
I
Sir : We wish to call attention to the following diagram which was unfortunately omitted from our recent paper entitled "Composition of Synthetic Liquid Fuels. I. Product Distribution and Analysis of CrCS Paraffin Isomers from Cobalt Catalyst," THISJOURNAL, 72, 1212 (1950).
Chf., Xgyx 5.77, 5 . 8 5 ~ . Anal. C~BH~OOS:. C, 20 72.19; H, 9.32; OCH,, 14.34. Found: C, 72.21; H, 9.05; OCH3, 13.99. Thioethyl derivative d (VI, RSH-HBr-HOAc), m. p. 67.2-68.5') [ a ] 2 2 D d 16 .*m +loo' Chf. Anal. C27H4204S: C, 70.08; H, ,al 9.15; S, 6.93. Found: C, 70.24; H, 9.32; S, 12 6.76. Desulfuration of VI with Raney nickel and hydrolysis afforded 3a,9a-oxido-ll-ketocholanic w0 acidj6 m. p. 174.9-175.6') [ a I z 6 D $88.1' Chf.; Wolff-Kishner reno depression in mixed m. i 8 duction of I11 gave Ag(llfcholenic acid, m. p. t 139.4-140.8', [ C U ] ~ +40.2' ~D Chf., methyl ester,' y 4 0 m . p. 67.4-68.2') [ a l Z 2f42.8' ~ Chf.; Clemmensen reduction gave 11-ketocholanic acid, 0 methyl ester,7 m. p. 89.4-90.5', [ a l Z 0 D 4-48.1' An.4 In acetic acid over platinum the hemiketal absorbed two moles of hydrogen and afforded two 0.8 triol esters (V), isolated as the monoacetates. Methyl 3a-acetoxy-9 a,11/3-dihydroxycholanatej m. p. 166-167', +55.1' Chf., 2.86, 0.6 5.8811. Anal. C27H4406: C, 69.79; H, 9.54. i Ql Found: C, 70.22; H, 9.81. The 3p-epimer) m. p. 0.4 171.2-172.4', [aIz1D +33.6' Chf., 2.86, 5 . 8 5 ~ . Found: C, 69.99; H, 9.75. Chromic acid oxidation of the triol monoacetates with 0.2 chromic acid gave the 11-ketones: 3a, m. p. 149150.4') [ a ] 2 2$83.4' ~ Chf. Anal. C27H4206: C, 70.10; H, 9.15. Found: C, 70.25; H, 9.16; 0 5 10 15 20 25 3/3, m. p. 134-135.5' and 143.6-144.6O, [ a ] " ~ Length of hydrocarbon chain, n. +58.2' Chf., XEhJX2.87, 5.89-5.94~. ClemmenFig. 1.-Part a is the distribution of hydrocarbon prodsen reduction of methyl 3a-acetouy-9a-hydroxy11-ketocholanate, followed by esterification and ucts from atmospheric pressure tests of cobalt catalysts acetylation gave methyl 3a-acetoxy-1l-keto- at about 190OC. Part b is a plot of Herington's term 0. cholanate18 m. p. 129.4-131.8', [ a l Z 2$59.4 ~ * against n, where 8. is the probability that a hydrocarbon radical on the surface will appear as product of chain j0, not depressed by an authentic sample. The 9,ll-oxide I is a t present available only length n rather than grow to higher molecular weight. from methyl Ag(ll)-lithocholenate,obtainable in These graphs are necessary for a complete undermoderate yield from desoxycholic acid, but a way standing of the discussion presented in that paper. is now open for utilization of such an intermediate CENTRAL EXPERIMENT STATION for the synthesis of cortisone and a satisfactory
2
xeix
(6) Turner, Mattox, Engel, McKenzie and Kendall, J.Bioi. Chcm., 166, 345 (1946). (7) Reich and Reichstein, Hels. Chim. Acta, 46, 562 (1943). (8) Lardon and Reichstein, ibid., 40, 586 (1943).
BUREAUOF MINES R. A. FRIEDEL R. B. ANDERSON PITTSBURGH, PENNSYLVANIA RECEIVED APRIL10, 1950' (1) Manuscript originally received January 12, 1949.
