Tetrahedron Letters N0143, pp, 3829-3837, 1965. Pergamon Press Ltd. Printed in Great Britain.

CONPORllATIORAL EFFECTS IN 'IHE OPENING OF

CIS/TRANS 1,4-DIALKYL SUBSTITUTED 1,2-

CYCLQHEXENE EPOXIDES WI'RI ACETIC ACID

John C. Leffingwelll and E. Earl Royals

Denartment of Chemistry

Emory University, Atlanta, Georgia (Received 23 August 1965)

In our continuing studies on the reactions of

teroene oxides2 we wish to report the following cases

of stereospecific epoxide ooenings in the E-menthane

series.

Q 0 9 I II III

The peracid epoxidation of limonene (I), menthene-1

(TI) and menthene-3 (III) leads, in each case, to

aporoximately a 1:l mixture of the corresoonding cis -

and trans epoxides3 as wonld be exuected from the

recent findings of Rickborn and Lwo 3d . Althouflh

the cis/trans epoxide mixtures are senarable by dis- --

tillation only with difficulty, they may be separated

conveniently on a 15 ft. 15:: Carbowax 2OM on chromo-

sorb W GLPC column 4 . In each of the three oases cited,

reaction of the cis epoxide (IVa,b,c) with a sodium -

3829

3830 No.43

(b) R1= CH3, ~92' CH(CH3)2

(c) Rl= CH(CH3)2, RZ= CH3

acetate buffered solution of acetic acid leads exclu-

sively to a single voduct containing trans tertiary

acetoxy and secondary hgdroxyl grouns and trans 1,4-

alkyl groun.s (VII a,b,c), as would be expected by an

extension OP the First-Plattner rule 5 to r?onocxolic

system 3a , assuming that R2 "fixes" the confbrmstion of

the cgciohexane ring system. Similarly, Feactioc of the

n?-*

Y-

.

VI

*,

?I+ '1111

Oh

VII

-9 *,&7- b)H

SX

trans epoxidss (Ya,b,c) leads predominantly6 tc apoxjde

opening resulttng from cleavage of the "equatorial"-C-O

No.43 3831

1inkaEe in the transition state (VIII) to afford the

diaxial products (IXa,b,c) with secondary acetoxy and

tertiary hgdroxyl pouos. The woof of structure in

each case follows from ayrolysis of the acetate to

yield !mown products 2,3a . For examule, (k)-cis-3- -

menthene epoxide (IVc) affords a hydroxyacetate (VIIc)

which pyrolyzes at 390' to a m!.xture of (+)-trans-3-

menthene-5-01 (X)7a and (k)-menthone-isomenthone (XI) in

a product ratio, 7:3. The (?)-trnns-3-menthene epoxide

affords a crystalline (+)-hydroxyscefste (TXc), m.p.

80.5-89.5°, which gyrnlgzed at 450' to (t)-trans-2-

3832 No. 43

menthene-4-017b (XII). In each ease, the infrared

spectra of the pyrolysis products were superposible on

those of authentic samules 7a,b . Further structure proof

af (X) came from hydrogenation (PtO2 - ethyl acetate)

to a mixture of 88% (?)-isomenthol (XIII) and 12% (f)- -

neomenthol (XIV). - Hydrogenation of (XII) afforded trans-

E-menthan-2-01 (XV),

The high degree of stereosoeoificity in these

epoxide ooenings leads us to pronose the following:

(1) The aporeoiable amount of abnormal,

"diequatorial" , onening observed with

nuoleouhilio reagents on monooyolio ois- -

1,4-dialkyl cyolohexene epoxides (vide

infra) does not necessarily arise from

reaction of the less‘ stable oontormational

isomer' (as is observed in conformationally

mobile oyolohexene epoxide systems') but

rather, perhaps, from the lnate difficulty

of nucleophilio displacements at tertiary

centerslO.

(2) The 4-alkyl substituent determines the oon-

formation in whioh'the ois- and trans-6 -

1,4-alkyl substituted cyolohaxene-1,2-

epoxides react with acidio reagents (i.e.,

?!CAc) via a normal, "diaxial", opening:

(Furst-Plattner Rule5).

No. 43 3833

Thus the lithium aluminum hydride (LAFI) reduction

of (+)-cis-1,2-limonene epoxide (IVa)3a~lla affords a -

product ratio of 157: cis-,8-ternineol (XVI) via "effective

diequatorial opening" and 65% (i-)-neodihydrooarveol (XVII) -

via "normal diaxial opening". With other nucleoohilic

rearents, i.e., dimethylamine, the oercentage of

"diequatorial" openinn is even greater (30-40s) llb

.

There are two nays in which s-IC:FL diequatorial ooeninrs

may occur; (a) through diaxial ooenin? of the less stable

epoxide conformer with subsequent inversion (orocess

"A") or (b) via diaxial oneninp; of a "twist" transition

state (nrocess rrD"). Inasmuch as there should be a

considerable steric (shielding) effect exhibited by the

"axial" 4-al'xyl vrouo on the incoming nucleonhile in

process “A”, in addition to the already enerretically

un?avorahle enoxide conforma:ion, we tend to Tavor

3834 R-o.43

process “IY’.

In the opening of the cis-dialkyl epoxide (VIo) with -

acetic acid, althou;?h Rl is isopronyl and Rz is methyl,

only a sin-rle oroduc t is obtained (governed by diaxial

opening with R2 equatorial). Therefore, even though the

1,4-dialkyl substituted cyclohexene epoxlde has been

nostulatedL2 to assume a “nearly chair” conformation in

the reaction transition state, the 1-alkyl substituent

nlays litt‘te or no part in determining the conformational

preference of the transition state 13

. It is orobable,

therefore, that the transition states for the reactions

of the cis/‘trans-1,4-dialkyl oyclohexene sooxides as

illustrated above (VI and VIII)

since to develop a true “chair”

bond must effecti.vely be broken

are a bit exaggerated,

conformation the C-O

(or at the least very

stretched), Transition statas such as (VId) and (VIIId)

might be better representations. In such transition

states, as the protonated oxirane starts to assume the

status of an axial hydroxyl there is still effective

overlap between the p-orbi.tal of oxygen and the vacant

p-orbital of the “incipient carbonium ion”. Since Rl

in such a transition state for the ois enoxide (VI) -

can be neit?er axial nor equatorial it should not have

any effect on conformational nreferenoa (as indeed is’-;*

case). Sue;2 overlap would be less favorable for a

No.43

VId VIIId

diequatorial oDeninE (which Is not observed for the cis

epoxides of type IV). The small oercentage (ca. 105) of

an “effective diequatorial opening ” observed for the

trans eDoxides (Va and Vb) may occur by either a orocess

of type “A” or “7S” (vide suora). Althou,?h the results --

oresented do not distinguish between these two mechanisms

it is acain feasible that process “5”

much as the normal diaxial ogeninr of

of (V) would be enhanced b:i formation

tertiary Dositive carbon center.

In any event, tl,,, 2terePspecific

may flunction inas-

a It twist" conformer

of an incipient

cpening of 4-

alkyl substituted cyclohexene epoxides, couoled with the

convenient nyrolysis of the resultant hydroxyacetates

affords an excellent method for ti‘.~: preparation of

specific allylic alcohols.

Acknowledgements. me authors wish to thank Dr. Gunther

Ohloff and Dr. J. Cresham for valuable comments on

va?lous oarts of this work and especially Dr. Eruce

Rickborn who contributed insight into the discussion of

the nossible transition states.

3836 No.43

REFERENCES

(1) l Present address, R. J. Reynolds Tobacco Co., Research Center, Winston-Salem, North Carolina.

(2). E. E. Royals and J. C. Leffingwell, J. Am. Chem. Sc+., a, 2067(1964); 3. Osg. B., &-2Om964); n press.

(3). (a) E. E. Royals and J. C. Leffingwell, in press; (b) A. B. Booth, U. S. Pat. 2,945,068 (1960); (0) W,' P. Newhall, J. Org. w., a, 185(1964); (d) B. Rickborn and S.-Y Lwo, 2. a. Chem., a,

2212.(1965).

(4). J. C. Leffingwell, unuublished results; Ref. 30, footnote 5.

(5). A. Furst and P. A. Plattner, Abstracts of Papers, 12th International Congress of Pure and Applied Chemistry, New York, 1951, p. 409; see also, D. H. R. Barton, 2. Chem. &., 1953, 1027.

(6). In t'?e case of (Va,b) ea. 10% of the abnormal product arising from an effective "diequatorial" opening was observed; see Ref. 3a.

(7). (a) :I. Malcolm and J. Read, 2. Chem. Sot., 1939, 1037; (b) G. Ohloff and G. Uhde,lv.him. s., 48, lO(1965).

--

(B). S. J. Anggal, Chem. and Ind., 1954, 1230; Quart. Revs., &&,212(?& s;ee3al;;lp. Cookson, Chem.. and Ind., -f 9 .

(9). For exsmoles see Ref. 38,~; W. F. Newhall, 2. Or . Chem., & 1637(1959); a, 185(1964); H. Kuozy?& andA. Zabza, Roczniki a., & 733(1963).

(10). E. L. Eliel, Stereochemistry of Carbon Comaounds, McGraw Hill Co., New York,. 19=,-O.

(11). (a) f:. KUCZ~ISK~ and K. Biatkowski, Roczniki Chem., 2,. 299, 311(1959)+ (b) H. Kuozynski and mbza, Bull. Acad, Polon. Sci., Ser. soi. them., m?).

9, 551(1961); Roczniki 3., 9, 773

(12). R. E. Parker and N. S. Isaacs, Chem. e., $& 742(1959).

No.43 3837

(13). The term l-alkyl and 4-alkyl substituent refers to the systematic naming of cyclohexane systems In general. Therefore, in each of the cases cita here, the alkyl substltuent substituted on the epoxide bearing carbon is the 1-alkyl substituent. The fact that the isopropyl group in the ,3-menthene epoxides is on the B-carbon atom according to nomenclature in the p-menthane systems thus should not confuse the reader In the interpretation of the general effect.