deuterium exchange of decaborane with deuterium chloride under

Conclusive support for this supposition is af- forded by hydrogen bromide-acetic acid equilibra- tion of 2a-bromo-2/3-methylcholestan-3-one (II). (m.p...
0 downloads 0 Views 238KB Size
COMMUNICATIONS TO THE EDITOR

4998

reasonable analogy to earlier generalization^,^ the bromo ketone was assigned3 the expected 2pbromo-2a-methylcholestan-3-one (I) formulation. We now wish to demonstrate that the situation is considerably more complicated as was uncovered by means of optical rotatory dispersion.‘j According to the “axial haloketone rule,”7s8 a 2P-bromo-2cu-methyl-3-keto steroid in the chair form (IA) should exhibit a positive Cotton effect, contrary to the observed (Fig. 1) negative one of 30

25

20

.

15

-

‘0

c

10

0

e

0

5

0 (L



0

e

” 01

n-5 u)

-25

30

s1

atom was established by dehydrobromination, I yielding A’-2-methylcholesten-3-one3 and I11 giving 2a-methyl-h4-cholesten-3-one. 3,9 The spectral properties of I demonstrate that the bromine atom is axial and its powerful, positive Cotton effect curve (Fig. 1) is fully c o n s i ~ t e n with t ~ ~ ~the chair forniulation IA. Similarly, the spectral and rotatory dispersion (Fig. 1) data show that the bromine atom in I11 is equatorially oriented. Identical isomers (I, 11, 111) were encountered in the 2aniethylandrostan- 17P-ol-3-one series. The above results have several important iinplications: (a) the “axial haloketone rule”7 is applicable to boat as well as chair forms; (b) chair form IA is energetically favored over the boat IIB, but the latter is preferred over its corresponding chair form IIA. That this is due to the presence of the angular methyl group (the chair form IIA having the electrostatically unfavorable equatorial orientation of the bromine as well as a 1,3-diaxial dimethyl interaction, none of which are found in the boat IIB) will be demonstrated in a forthcoming paper reporting the bromination of analogous 19nor steroids and where no boat form is encountered. The fact that the product of the kinetically controlled bromination is 11, while the thermodynamic product is I appears to be contrary to the earlier generalizations6 We intend to comment on this point in another paper, together with additional experimental evidence. (9) J. A. K. Quartey, J . Chcm. SOC., 1710 (1958). (IO) Stanford University, Stanford, California.

-10

-

Vol.

I

I

I

I

300

400

500

600

A

DEPARTMENT OF CHEMISTRY STATE UNIVERSITY CARLD J E R A S S I ~ ~ WAYNE DETROIT,MICHIGAN, AND &-EVILLE FINCH SYNTEX, S. A . ROLFMAULI APDO.POSTAL 2679 MEXICO,D. F. RECEIVED JULY15, 1959

(my 1.

Fig. 1.

the bromo ketone ( m . p 136-13S0).3 The boat form IIB of the 2a-broino-2P-methyl-3-keto isomer, however, satisfies all criteria, the bromine atom occupying an axial orientation and the “axial haloketone rule’ ’ 7 , 8 predicting a negative Cotton effect.

DEUTERIUM EXCHANGE OF DECABORANE WITH DEUTERIUM CHLORIDE UNDER ELECTROPHILIC CONDITIONS

Sir:

I n contrast to the recently reported base catalyzed deuterium exchange reactions of decaborane1a2 with such compounds as deuterium oxide and chloride, we wish to report the results of an electrophilic exchange study. Four successive exchanges of a 10 mmole sample of decaborane with 130 mmole portions of deuterium chloride in the presence of 10 mmole of aluminum chloride and in carbon disulfide solvent (15 ml.) 0 H 0’ ir By H introduced six atoms of deuterium per molecule of IA 11.4 IIR decaborane. Each exchange was carried out for Conclusive support for this supposition is af- 90 hours a t room temperature. The HC1-DCl forded by hydrogen bromide-acetic acid equilibra- gas mixtures were analyzed a t equilibrium by an tion of 2a-bronio-2(?-methylcholestan-3-one (11) infrared method. ,Sfter the third exchange re( m p . 136-138’) which yielded 2P-bromo-Za- action the hydrogen chloride content of the gas mixture was negligible and the recovered decamethylcholestan-3-one (I) ( m p . 120-122’, borane (9.5 mmole) analyzed’ properly for 308 m p , X ~ ~ 5.S5 ~ 1 3p) and 2a-methyl-4a-bromocholestan-3-one (111) 140-141°, Xz;c13 5.77 B I o H D L ~ . I n the infrared this material exhibited B-H p, X ~ 285 ~ m p )~ . The ~ position ~ e of the bromine bridge, B-H terminal and B-D terminal stretching (9) E. J. Corey, THISJOURNAL, 7 6 , 176 (1954). bands. No 7 . 3 0 ~ B-D bridge was observed. (($1 See C. Djerassi, Bull. SOC.Chim. France, 741 (1957).

e:;,

(7) C. Djerassi and W. Klyne, THIS JOURNAL, 79, 150ti (1057). ( 8 ) C. Djerassi, J. Osiecki, R. Riniker and B. Riniker, ibid., 80, 1216 (1958).

(1) hI. 1;. Hawthorne arid J. J. hliller, THIS J O U R N A L , 80, 794 (lY58). (2) I . Shdpiro, hI. Lustig and R. E. Willidms, sbtd., 81, 838 (1959).

COMMUNICATfONS TO THE EDITOR

Sept. 20, 1959

4999

TABLE I I n the B l1 nuclear magnetic resonance spectrum the doublet attributed2 to the 2 and 4 borons was NUCLEAR MACSETICRESONANCE SPECTRA O F SOLANESOL' collapsed to a singlet as was the doublet attributed Assignment Band rh to the 5, 7, 8 and 10 positions of decaborane. / 1 4.45 -CHz-CH=C The latter doublet normally appears as the high \ field portion of a triplet. The proton spectrum of 2 5.10 HO-CH2-CH= this material was composed of the bridge hydrogen 5.21 peak plus the four peaks assigned2 to the com1 1 posite 1, 3, 6 and 9 positions. The slight separa3 6.51 =C--CH2--CH,-C= tion of each of these latter peaks into two equivalent 1 4 6.71 =C-CHI peaks was suggested by the curve shape. The a Concentration, 14% in carbon tetrachloride. 7 = Bll and H1 spectra substantiated each other nicely yo/40 3.50 where yo is the observed band position in and this result corroborates the spectral assignments C.P.S. relative to benzene as external standard. See G. V. of Shapiro and co-workers.2 D. Tiers, J. Phys. Chem., 62, 1151 (1958). Electrophilic iodination of decaborane a t the 2 TABLE I1 and 4 positions was demonstrated r e ~ e n t l y . ~ AREASOF BANDS1 AXD 2 FOR SOLANESOL (TABLE Recently another diiododecaborane was isolated4 RELATIVE I ) AND RELATED COMPOUNDS~ and shown to be the 2,5-isomer. This previous h70. of work coupled with that reported here substantiates deterAi/Az the fact that decaborane is sensitive to electromrna- CalcuFound/ Compound tions lated Found calcd. philic attack a t the 2, 4, 5, 7, 8 and 10 positions. 6 5.0 5.2 1.04 The less saturated 1, 3, 6 and 9 borons2 apparently Coenzyme Qla (III)3 2,3-Dirnethoxy-5-methylare inert to such attack.

+

(3) R . Schaeffer, THISJ O U R N A L , 79, 2726 (1957). (4) M . Hillman, Abstracts of the 135th Meeting of the American Chemical Society, April, 1959, p. 44-M.

6-farnesylbenzoquinone

( W 4 4 1.5 1 . 5 5 1.04 Solanesol (11) 5 4.5b 4.77 1.06 Band 2 in I11 and I V is assigned to the methylene group REDSTONE ARSENALRESEARCH DIVISION Calculated for structure 11. OF ROHM & HAASCOMPANY JOHN A. DUPONT attached to the quinone ring. The corresponding value for I is 5.0. HUNTSVILLE, ALABAMA M. FREDERICK HAWTHORNE RECEIVED JULY 20, 1959 Calcd. for C62H751302: c , 56.11; H, 6.79; I, 34.21.

Calcd. for C67H&02: C, 57.97; H , 7.08; I, 32.24. Found: C, 56.00, 55.81; HI 6.57, 6.79: I, 34.04, 34.52. The oxidation of solanesol with manganese dioxide yielded the corresponding aldehyde which Sir: The isolation and determination of the structure was isolated as its 2,4-dinitrophenylhydrazone, of solanesol (I), a polyisoprenoid alcohol from to- m.p. 74-75'] E:? = 360 a t 379 mp in ethanol. 75.70: ~ : H, 9.47; bacco, has been described.' Since a synthesis2 Anal. Calcd. for C K ~ H ~ B NC,~ O N, 6.92. Calcd. for C66H84N404: C, 76.67; H, CHI 9.65; N, 6.39. Found: C, 75.86; HI 9.14; N, 7.19. These analytical data support structure 11. IIO(CH2CH=A---CH2),H Similarly, farnesol yielded a triiodobenzoate, I, n = 10 m.p. 46-47', E m 0 1 = 31,000 a t 234 mp in i-octane. 11, n = 9 Anal. Calcd. for CZ2H27I302:C, 37.52; H, 3.86. 0 Found: C, 37.77; H, 3.87. Farnesaldehyde yielded a 2,4-dinitrophenylhydra~one,~ m.p. 105106', Em",= 30,000 a t 380 mp in ethanol. Anal. Calcd. for C Z ~ H ~ ~ NC,~ 62.98; O ~ : H, 7.05; N, 13.99. Found: C, 62.87; H I 6.88; N, 14.50. A 111, n = 10 calculation of the molecular weight of solanesol I\', n = 3 from these Emol values for farnesol derivatives and values of the corresponding solanesol deutilizing solanesol and 2,3-dimethoxy-5-methylhy-the droquinone did not give coenzyme QIO(111),we re- rivatives leads to values of 658 and 655 which supinvestigated the structure of solanesol. We wish port structure I1 (mol. wt. = 631) rather than I to present new data which confirm the basic struc- (mol. wt. = 699). The synthesis of "Q-254" from solanesol for conture but support I1 rather than I. The nuclear magnetic resonance spectrum of firmation of structure6 and the synthesis of coensolanesol (Table I) provides confirmation of the zyme QBand a vitamin K analog is reported in an isoprenoid structure, and relative area measure- accompanying communication. We wish to thank Dr. R. L. Rowland and Dr. M. ments of appropriate bands (Table 11) support Senkus of the R. J. Reynolds Tobacco Co., Winstructure 11. The reaction of solanesol with 3,4,5-triiodoben(3) D. E. Wolf, C. H. Hoffman, N. R. Trenner, B. H. Arison, C. H . zoyl chloride yielded solanesyl triiodobenzoate, m.p. Shunk, B. 0. Linn, 3. F. McPherson and K. Folkers, ibid., 80, 4752 53-54.5', E:Z = 271 a t 234 mp in i-octane. Anal. (1958). COENZYME Q.

XI.

THE STRUCTURE OF SOLANESOL

(1) R. L. Rowland, P. H. Latimer and J. A. Giles, THIS JOURNAL, 7 8 , 4680 (1956). (2) C. H. Shunk, R. E. Erickson, E. I,. Wong and K. Folkers, ibid., 81, 5000 (1959).

(4) '2. H. Shunk, B. 0. Linn, E. L. Wong, P. E. Wittreich, F. M. Robinson and K. Folkers, ibid., 80, 4753 (1958). (51 Y.R. Naves, Heiw. Chim. Acta, 32, 1798 (1949). ( 6 ) N. R. Trenner, B. H. Arison, R. E. Erickson, C. H. Shunk, D. E. Wolf and K. Folkers, THIS JOURNAL, 81, 2026 (1959).