April 5 , 1954
SYNTHESIS O F 4- (hZflS-4'-HYDROXYCYCLOHEXYL)-CYCLOHEXANONE
1733
night. Nothing was eluted by petroleum ether, but 1:l acetic acid and the solution diluted t o saturation a t :he boilpetroleum ether-benzene eluted a solid that on crystalliza- ing point; needles of XV separated, m.p. 148-149 , mixed tion from methanol gave a mat of fine, flat needles (70 mg.), m.p. 149-150". The enol XV gives a strong ferric chloride test but does m.p. 120-121', CYD -74.5' Chf (c 1.33), XEtoH 240 tnp not react with diazomethane. When a petroleum ether (8,360), XCh' 5.74, 5.86, 6.10, 8.0 p . Anal. Calcd. for C&& (442.66): C, 78.85; 13, 10.48. solution is shaken with Claisen's alkali no yellow color develops but a gelatinous enolate separates a t the interface. Found: C, 78.83; H , 10.69. An attempt to condense XV urith ethylene glycol in acetic A comparison sample of A6-cholestene-3~-ol-4-oneacetate acid with boron fluoride etherate led only to recovery of XV. prepared by Miguel A. Romero according to Petrow and A4-Cholestene-4-01-3-one is readily eluted from acid-washed Starling14 melted a t 118-119", CXD-77.4' Chf (P. and S. alumina by 4: 1 petroleum ether-benzene. give m.p. 123-124", CYD -76" Chf, hEtoH 240 m ~ ~ A ~ ) . The acetate28 (XVI), prepared with acetic anhydride in mixture with the above sample was undepressed in m.p. pyridine a t 25', crystallized very slowly from methanolCholestane-3,6-dione Bis-ethylenethioketal (XIV).28-A water, m.p. 100.5-101.5". It was recrystallized by dismixture of 300 mg. of A6-cholestene-40r-ol-3-one acetate, 1 solving it in cold methanol and adding water until a turbid cc. of ethanedithiol and 1 cc. of boron fluoride etherate was emulsion separated. After standing for some time a t 5' stirred in a test-tube, when a stiff orange paste soon resulted. this changes to a paste of fairly well formed crystals, m.p. After 5-10 min. methanol was added to produce a thin paste 1OO-10lo, CXD +95.3' Chf (c 1.035), XEtoH 247 (15,100); of suspended white solid. The collected and dried product XChf 5.71, 5.93, 6.12 p . (290 mg., m.p. about 190') was dissolved in dioxane and AnaZ. Calcd. for C29H4803(442.66): C, 78.68; H , 10.48. the solution diluted with ethanol and then water. A crystal powder separated and was crystallized more satisfactorily Found: C, 78.81; H , 10.51. The pheny$ydrazone of XVz8separated in a crystal crust, from hexane, in which it is moderately soluble; the solution m.p. 188-189 Recrystallization fzom ethanol gave lemon slowly deposited rosettes of fine needles, m.p. 219-220'. Recrystallization from hexane gave silken needles, m.p. yellow micro crystals, m.p. 190-191 Anal. Calcd. for CatHmONg (490.75): C, 80.76; H, 219-220', CYD +30.4' Chf ( c 1.09); no depression in m.p. 10.27. Found: C,80.91; H, 10.20. on admixture with an authentic sample. The quinoxaline derivativez8 of XV was prepared most Anal. Calcd. for C S I H ~(552.73): ~S~ C, 67.36; H , 9.48. successfully by heating 200 mg. of the enol and 200 mg. of Found: C, 67.78; H, 9.57. sublimed, colorless o-phenylenediamine a t 141' for 15 min. A4-Cholestene4-01-3-one (XV).ze-A solution of 455 mg. The solidified melt on crystallization from a solution in diacetate (VI) in 25 cc. of 95% of A~-cholestene-4~~-01-3-one ethanol was treated with 0.5 cc. of 36% hydrochloric acid, oxane diluted with ethanol and then water gave 80 mg. of Recrystallization from cool, the quinoxaline, m .p. 188-190'. refluxed for 3 hr., diluted with 4 cc. of water and Re- dioxane-methanol afforded pale buff-colored plates, m.p. when 313 mg. of needles separated, m.p. 148-149 crystallization. from methanol or methanol-water gave 200-201', XEtoH 239, 321 mp (27,400; 8,870); lit.z m.p. 207-208'. needles, m.p. 150-151", CYD +80.2' Chf (c l.lO), XEtoH Anal. Calcd. for C33H48N2 (472.72): C, 83.84; H, 10.24. 278 mp (13,000), XHexane 275 mp (14,570), XChf 2.92, 5.98, Found: C, 83.52; H , 10.10. 6.10 p . Oxidation of XV.z8-Oxidation of 0.5 g. of A4-cholesteneAnal. Calcd. for C27H1402(400.62): C, 80.94; H , 11.07. 4-013-one with hydrogen peroxide in ethanol-sodium hyFound: C, 81.18; H, 11.06. droxide and crystallization from acetic acid according to Similar acid hydrolysis of 221 mg. of As-cholestene-4a-01- Butenandt26 gave spars of dihydro-Diels acid, m.p. 257%one ethylene ketal acetate ( I b ) afforded 131 mg. of XV, 259'; diester (diazomethane, two crystallizations from m.p. 146-148'. The enol XV was also obtained by adding methanol), m.p. 123-124", no depression on admixture with 0.5 cc. of Claisen's alkali to a suspension of 150 mg. of A6- a comparison sample34 made by esterifying Diels acid with cholestene-4~~-01-3-one acetate (VI) in 5 cc. of methanol a t diazomethane and hydrogenating the diester in acetic acid 25". In a few minutes the substance dissolved to a slightly in the presence of platinum catalyst and a trace of perchloric yellowish solution and then an enolate separated. This acid. was collected, suspended in methanol and acidified with
.
.
Ip .
(33) See V. A . Petrow, 0. Rosenheim and W. W. Starling (footnote), J . Chcm. Soc., 135 (1943).
[CONTRIBUTION FROM
THE
(34) Experiment by Dr. Wei-Yuan Huang.
CAMBRIDGE, MASS.
LABORATORY OF ORGANIC CHEMISTRY OF
THE
UNIVERSITY OF WISCONSIN ]
Steroid Analogs Lacking Ring C. 111. Synthesis of 4- (trans4'-Hydroxycyclohexyl) -cyclohexanone BY A. L. WILDS,CLIFFORD H. SHUNK2b AND CARL H.
H0FFMANzbq3
RECEIVED SEPTEMBER 17, 1953 Reduction of 4,4'-dihydroxybiphenyl (111) has given the truns,truns-, &,trans- and a small amount of &,cis-bicyclohexyl4,4'-diols (IV), separated as the dibenzoates. Oxidation of the pure isomers or the mixture gave the diketone VI which could be partially reduced to the truns-keto alcohol I. Better methods for synthesizing this compound involved partial oxidation of Iruns,truns-diol IV directly or as the monobenzoate.
I n extending our synthesis of analogs of the steroidal hormones lacking ring C' to the analogs (11) of testosterone, i t was necessary to develop a synthesis of the bicyclic keto alcohol I as the key (1) Preliminary report, A. L. Wilds, C. H. Shunk and C. H. Hoffman, THIS JOURNAL, 11,3266 (1949). Paper 11, A. L. Wilds and C. H. Shunk, ibid., 71, 2388 (1950). (2) (a) Wisconsin Alumni Research Foundation Research Assistant, 1941-1942; ( b ) Merck and Co., Inc., Rahway, N. J. (3) Wisconsin Alumni Research Foundation Research Assistant, 1946-1947; Ciba Pharmaceutical Products Fellow, 1948-1949.
intermediate.
The present paper describes the An accompanying paper reports work leading to the cis isomer and the evidence on which is based the assignment of configuration of these geometrical isomers and related bicyclohexyl derivatives. Although 4,4'-dihydroxybiphenyl(II1), the starting material for our synthesis, has been prepared for many years by tetrazotization and hydrolysis of benzidine, the procedures available have not
trans isomer of this structure.
A . L.
1734
\~'ILDS,
CLIFFORD H . Srrvsr;
AND
c.1~1, H . HOFFMAX
1'01. 76
in 4370 yield, the remainder consisting of diol and some dibenzoate readily hydrolyzed for reuse. Oxidation of the monobenzoate gave the bcnzoatcl of the fmn\-kcto alcohol (VII) in 94yo yield.
/l/OH
b
/-\A,
Experimental6,
' . /
O / V
4,4'-Dihydroxybiphenyl (111).-The procedure of HirschB I I1 has been improved as follows.g A solution of 200 g. of technical benzidine (Eastman Kodak given material of high purity in satisfactory yield. C o . ) in a hot mixture of 210 ml. of concentrated hydroAn improved procedure was developed in the chloric acid and 2150 ml. of water was cooled t o 25", an course of our work leading to the biphenol (111) additional 235 ml. of hydrochloric acid added, and the suspension cooled and stirred below 10". A solution of 151 g. in 60-65% yields from technical benzidine. Hydrogenation of the biphenol with Raney of sodium nitrite in 450 ml. of water was added slowly, 20 minutes after complete addition and 1 g. of urea nickel proceeded smoothly to the perhydro deriva- stirred added. The diazonium solution was added slowly t o a hot tive IV and by crystallization readily afforded one (85-90") solution of 215 i d . of sulfuric acid in 4300 ml. of geometrical isomer of this diol (map. 216', trans, water, heated 10 minutes longer, cooled, filtered and dried. trans). Further crystallization gave material melt- This solid was added'o t o 400 ml. of acetic anhydride, the was heated for 3 hours a t reflux, cooled and filtered, ing a t 182.5-183.501 which, however, proved not to mixture giving 295 g. of crude diacetate; this material was distilled be a single isomer but a molecular compound of the a t 0.14.5 mm. from a two-bulb flask" and crystallized from 216' isomer with a second isomer (cis,trans) melting benzene, giving 159 g. of nearly colorless 4,4'-diacetoxya t 178'. The most satisfactory method of separa- biphenyl, m.p. 161-162.5', and 12.6 g., m.p. 160-162'. Distillation and recrystallization of the material from the tion was found to be fractional crystallization of the acetic anhydride filtrate gave an additional 40-50 g., m.p. dibenzoates, which did not form molecular coin- 1X-16Oo, bringing the total yield of the 4,4'-diacetate t o pounds, and with which the solubility differences 6 2 4 5 % . The purest sample melted a t 163-164" (reported among the isomers were accentuated. Saponifica- 1fi2.5-163.5°).'2 From the filtrates was obtai2ed 3-4% of (reported the 2,4'-diacetoxybiphenyl, m .p. 99.5-101 tion of the pure dibenzoates readily led to the pure 91.5-96.E1"~), arising from the corresponding iqomer of bendiols. In this way the hydrogenation mixture zidine as an impurity. gave the trans,trans-diol in 37-43YG yield together T o a suspension of 7.5 g. of the diacetate in 200 ml. of alwith 12-23Y0 of the &-,trans-diol. Only a trace cohol was added over a five-minute period 22 g. of sodium hydroxide in 50 ml. of water. After heating the suspension (1-3y0)of the cis,&-diol (m.p. 197') was formed. for 30 minutes, 750 ml. of water was added followed by Oxidation of the pure diols or the stereoisomeric sufficient 1:1 hydrochloric acid t o make the solution strongly mixture with chromium trioxide readilv gave the acid. The mixture was diaested on the steam-bath for 1 diketone VI in 62% over-all yield from'I11, This hour, cooled and filtered. T h e washed and dried ( a t 80') amounted to 33-51 g. (97-99%), m.p. 279-282", method is superior to that reported by Schaefgen material and generally was suitable for hydrogenation without further and Flory4*6 involving dehydrogenation of 11' over purification. In other runs the 4,4'-dihydroxybiphenyl was copper-chromium oxide (17% yield from 111). recrystallized from alcohol as shiny white leaflets (95-99% yield), with crops melting from 279.5-281.5" o = p ) - - a - o to 282-284 ' (nitrogen-filled, sealed capillary). H O - ~ ) - - ~ - O H The monobenzoate of 4,4'-dihydroxybiI11 1-1 phenyl mas prepared by shaking a solution of I 1.86 g. of the dihydroxy compound in 15 ml. CrOa/ of water, ml. of dioxane and 10 ml. of 1.I 3 N sodium hydroxide with 1.25 g. (0.9 equivalent) of benzoyl chloride for 5 minutes a t 5'. H Dilution gave 2.36 g. (90%) of monobenzoH)>od7a
