COMMUNICATIONS TO THE EDITOR
M a y 2 0 , 1958
and alcohols has been reported to be catalyzed by boron trifluoride-etherate.2 We have found this reagent also to be effective in catalyzing the methylation of alcohols with d i a ~ o m e t h a n e . ~For example, the yield of P-cholestanol methyl ether was essentially unchanged by replacing the fluoroboric acid catalyst with an equimolar amount of boron trifluoride e therate. (2) Lf. S. Newman and P.F. B e a 1 , T ~ r sJOURNAL,^^, 5161 (1950). (3) Since this manuscript was written, similar observations were reported by E. Muller and W, Rundel, Angew. Chem., 7 0 , 105 (1958).
2383
groups developed an intense blue on standing for several months in sealed capsules. Volatile products incompletely characterized were obtained by the distillation of trimethylsilanol and aluminum isopropylate. The use of diethylsilanediol in place of trimethylsilanol with the appropriate isopropylate gave non-volatile polymeric substances which have not been characterized. Acknowledgment.-This work has been supported by the Office of Naval Research.
CONTRIBUTION No, 2332 DEPARTMENT OF CHEMISTRY GATESAXD CRELLIN COLLEGE JOSEPH D. DANFORTH OF CHEMISTRY MARJORIE C. CASERIO GRINNELL LABORATORIES GRINNELL, IOWA OF TECHNOLOGY CALIFORNIA INSTITUTE RECEIVED FEBRUARY 15, 1958 PASADENA, CALIFORNIA J O H N D. ROBERTS OF CHEMISTRY DEPARTMENT OF WISCONSIN hf. NEEMAN UNIVERSITY ~VILLIAMS. JOHNSON MADISON,~VISCONSIN TOTAL SYNTHESIS OF ALDOSTERONE RECEIVED APRIL14, 1958 Sir: CONDENSATION OF TRIMETHYLSILANOL TITANIUM ISOPROPYLATE
WITH
Sir: The preparation of glassy polymeric products containing Si-0-Ti bonds has been reported.' Because of interest in modified polymers, i t seems worthwhile to report a simple synthesis of several monomeric compounds containing silicon and titanium. The synthesis involves mixing trimethylsilanol and titanium isopropylate in the molar ratios desired in the product, usually in a solvent, and distilling to recover substantial yields of the product. A typical preparation for Ti [OSi(CH3)3]4 is given. T o 0.1 mole (28.4 g.) freshly distilled titanium isopropylate in a 125-ml. distilling flask was added 70 ml. of 5.77 molar trimethylsilanol (0.4 mole) in dibutyl ether. The mixture became warm, was allowed to stand for about an hour, then distilled in vacuum to remove ether solvent and recover the reaction product. UYth no special precautions the recovery was 50-70y0 of theoretical. The products obtained in separate preparations using appropriate ratios of silanol to titanium isopropylate are summarized in Table I.
Carboti,/Calcd. % ' \Found Hydrogen, Calcd. Found % ?Z2'D
OC. B.P.iMm.
i
45.85 45.71 9.62 9.48 1.4490 114 13
I
II
m
TABLE I T i (OCaH;j,(OSi(CHd2
Formula
We disclose herewith a preparation of the adrenal hormone aldosterone (Ij by a highly stereoselective total synthesis which is basically different from previous approaches. The dihydroxy-ketone I I ( R 1 = H , R 2 = PH, R 3 = H 2 j 2was converted into the li-furfurylidene derivative (R3 = CHC4H30)m.p. 193-194' which, on treatment with methacrylonitrilc in methanolic methoxide, was transformed into the adduct 11 (R1 = H , R 2 = -CH,CH(CHa)CN, R3 = CKCIH 3 0 ). 3 , 4 Acetylation, followed by ozonolysis, then
'Ti ( O C I H I ) ~ . (OSi(CHa)Jz
41.83 41.65 9.36 9.33 1,4408 120 11
Ti (OSi(CHY)~)~
35.65 35.81 8.98 9.05 1,4300 100 2
The condensation product containing two silanols and two isopropyl groups would be expected to exhibit two arrangements if the titanium is square and planar. No separation was attempted. Each of the titanium products was a water white liquid as prepared, but those containing isopropoxy ( 1 ) K . A. Andrianov. T. N.Janina and E . N. Kh,rustaleva, I z w s l , A k n i l . .Vai(k, S . S . S . R . , nliiei, Khinz X n z i k , 7!JR--X04 ( I R j i i ) ; Chrm. : l b ~ h . , 51, :it87 (1957).
(1) (a) J. Schmidlin, G. Anner, J . R. Billeter and A . !T'eit.triit, Experieiztia, 11, 365 (1953), el s e q . to J . S. Schmidlin, Q. Anner, J . ;: Billeter, K . Heusler, H. Uebermaiser, P. Wieland and A . Wett.;+rl.i Helv. Chim. A c t a , 40, 2291 (1957): (b) A . Lardon, 0. Schindler .:n I T. Reichstein, { b i d . , 40, W i f i (1937) ( 2 ) W. S. Johnson, R. Pappo and W. 1'. Johns, T H I SJ O V R X A I . , 78, 6339 (1956). (3) This product was a mixture of Cnn epimers t h a t could be employed in the succeeding steps without separation because t h e a s y n metry of Czo was ultimately eliminated. I n the preliminary study t!;e mixture was separated a t the diketone stage I I I a ( R ' = R2 = C O C H B , R y = Ac) and each epimer examined separately in the succeeding rewtions. ( 4 ) C / . W. S. Johnson, I). 0 . RIartin, R. Pill)pn. S. D . Darling and 11. A . Clement. Proceeriitrgy. .iR (l9;7),
1'5S6
COMMUNICA4TIONSTO
THE
EDITOR
Vol.
so
saponification, gave on acidification the lactone- zation of the lip-epimer.'" Treatment of VIb (IC' dicarboxylic acid IIIa (R' = R 2 = C 0 2 H , R 3 = = H , R 2 = COCHzOBc) with potassium carbonate H ) which was transformed, through reaction of in aqueous methanol gave the C1; epimeric mixthe diacid chloride acetate I I I a (R'= R2= COCl, ture la from which dl-aldosterone readily was isoR 3 = Xc) with dimethylketene aceta1,j into the lated. The product was identified with authentic diketo-lactone IIIa (R' = R 2 = COCH,, R2 = materiallo by paper chromatographic behavior, in=IC) [isomer a! m.p. 1'72-173"; isomer b? m.p. frared spectroscopy and physiological activity. '?,' 145-146 "1. Rearrangement with trifluoroperace(12) Satisfactory analytical d a t a have been obtained for t h e n ? \ v tic acid6 afforded the triacetate I I I a (R' = R 2 = substances reported above. OAc, R3 = =IC) [isomer a, m.p. 1.10-141";isomer (13) W e are grateful t o the Sterling-Winthrop Research I n s t i t u t e . b, m.p. 142-143OI3 which, on mild saponification the Upjohn Company, t h e IVisconsin Alumni Research Foundation followed by Srbromoacetamide oxidation and re- and the National Science Foundation for assistance. acetylation, gave the 3-ketodiacetate [isomer a, L ~ I L L I A M S. JOHNSOS OF CHEMISTRY JOSEPH C. COLLISS 1n.p. 135-136'; isomer b, m.p. 14i-14S.5°].3 DEPARTMENT RAPHAEL PAPPO UXIVERSITV OF L~ISCONSIN Bromination and dehydrobromination' afforded MADISON, L~~ISCOKSIX MORDECAI B. RCBIS the unsaturated ketone I I I b (R' = R 2 = OAic) APRIL 18, 1938 RECEIVED [isomer a, m.p. 183-184"; isomer b, m.p. 15413.5°],3 which was converted? into the ketal IIIc (R1= R 2 = OXc) [isomer a, m.p. 155-156"; isoT H E COMPLEX FORMED FROM COBALT mer b, m.p. 175-1i1i"].3 Saponification gave the HYDROCARBONYL AND BUTADIENE diol I I I c (R'= R2 = OH), [isomer a, m.p. 1%Sir: 186"; isomer b, m.p. 203-207"3 which was transformed, by selective reaction with 2, .?-dimethylAlthough Prichard has reported the preparation benzenesulfonyl chloride in pyridine, into the mono- of a complex betm-een butadiene and dicobalt octaester I I I c (R'= OH, R2 = OS02C6H3hIe2).Oxi- carbonyl under Oxo conditions, l only an analysis dation u-ith Sarett's reagent converted the Czo hy- for cobalt and an analysis by the Orsat technique of droxyl to ketone; thus both epimers yielded the the sulfuric acid decomposition products were given. same product3 which, with potassium t-butoxide, \Ye have now been able to prepare and study underwent cyclizationg to the keto-lactone IVc, m.p. the stereochemical configuration of the complex 194-196". This substance mas partially isomerized formed between butadiene and cobalt hydrocarby base into the l i p epimer T'c, imp. 210-214' (re- bony1 in the absence of Oxo conditions. ported,'" 2013-208"). The latter substance and the Potassium cobaltcarbonylate was made to react corresponding ketone T%, n1.p. 215-219" (re- with glacial acetic acid and liquid butadiene in a ported,la 218-1320') were shown, by infrared and high pressure vessel. After twelve hours, a redmixed m.p. comparisons, to be identical with brown liquid, distilling at 33-35 o at less than 1 inm. authentic specimens. l o Vc has been converted pressure, was obtained. into aldosterone.la Analysis showed it to have the empirical coniAlthough our main objective was thus realized position CO(CO)~C~H:.I n benzene it gave a we hoped to obviate certain difficulties attending molecular weight of 195>compared to a calculated the last stages of the synthesis due t o the $-oriented molecular weight of 198. The compound is diaC1; side-chain'" by operating in the 170 series. magnetic and contains no acidic hydrogen. Tilth two mole-equivalent. of lithium aluminuni The infrared and ultraviolet absorption charachydride IVc was selectively reduced to the lactol teristics indicate the disappearance of the conjuT'Ic (R1 = H , R 2 = CHOHCH,). n1.p. 190-194", gated diolefin structure and the appearance of a which was hydrolyzed to I r I b (R1= H , R2 = structure producing absorption characteristics simCHOHCH3), m.p. 11'7-133 ', then treated with ilar t o those of cis monoolefins. The sharp peak methanol and acid to yield the lactol ether VIb a t 703 cm. which in the cobalt hydrocarbonyl has (R: = CHd, R2 = CHOHCH:j), m.p. 168-169.3". been a ~ s i g n e d ' .to~ the hydrogen vibrations also has Oxidation with Sarett's reagent afforded the ke- disappeared. tone VIb (R' = CH3, R2 = COCH3), m.p. I(KHowever, carbon monoxide could react with l(jSo, which was converted'l into the 21-acetoxy butadiene to give cyclopentanone, which might compound YIb (R'= CH:I, R' = COCH~OAC), have some aromatic character and give a complex m.p. 138-140". Hydrolysis of the lactol ether similar to cyclopentadiene bis-carbonyl cobalt; with TO:;; acetic acid gave dl-lTa-aldosterone-21- however, the infrared data do not seem to be in line acetate, T-Ib (R'= H, R 2 = COCHZO-IC),m.p. with such a structure. 1 ~ X j - 1 T O o . The infrared spectrum was identical Experimental details and a reinterpretation of with that"' of material obtained by partial isomeri- the mechanism of the oxo rcactionJ,5fi will be rc( 5 ) S. RI. hlcElvain and G. I ) . ( i ) Cf. I
