D. Shabashov, M.P. Doyle / Tetrahedron 69 (2013) 10009e10013
10011
impact on diastereoselectivity (Scheme 1).14 The same is true with
the use of rhodium acetate, but the structure of the aldehyde
employed as reductant, which has not been previously explored,
does have a significant influence on diastereoselectivity. Results
were subjected to the same conditions used for oxidation
of simple alkenes in Table 1 [1.0 mol % Rh2(OAc)4, iPrCHO
(3.0 equiv) in acetone under oxygen]. Oxidation of neryl acetate 7
provided the mono-epoxide 10 as major product in 69% isolated
yield (Table 2, entry 1) and in a 2.7-molar excess over its di-
epoxide, which was obtained in 26% isolated yield. The cis-ge-
ometry of the reactant neryl acetate double bond was not pre-
served in the di-epoxide product. Epoxidation of geranyl acetate
8 produced a separable mixture of mono- and di-epoxide prod-
ucts. The mono-epoxide 11 was obtained in 59% isolated yield,
and the di-epoxide was isolated in 22% yield (Table 2, entry 3).
Interestingly, decreasing the amount of aldehyde from 3.0 to
1.5 equiv provided only mono-epoxidation products for neryl
and geranyl acetate 10 and 11, which were obtained in 65 and
74% yield, respectively (Table 2, entries 2 and 4). Excellent se-
lectivity was also attained in differentiating terminal and in-
ternal double bonds in linalyl acetate 7dthe internal epoxide 12
was isolated as the sole epoxidation product in 82% yield (Table
2, entry 5).
Oxidation of farnesyl acetate 13 using 4.0 equiv of iso-
butyraldehyde and 1.0 mol % of rhodium acetate catalyst gave far-
nesyl acetate di-epoxide 16 as the major product in 42% yield and
a mixture of mono-epoxides 15 and 14 in 9% combined yield
(Scheme 3). Selective formation of the di-epoxide of farnesyl ace-
tate has been reported using mCPBA,17c peracetic acid,17d or
DMDO,17a,b but the Mukaiyama epoxidation of trienes has not been
previously reported.
Scheme 1. Diastereoselectivity in Mukaiyama epoxidation of cholesteryl benzoate
compared with peracid epoxidation.14
from the epoxidation of cholesteryl acetate 4 using a variety of al-
dehydes show that increasing the steric size of the alkyl substituent
increases selectivity for formation of the 5,6
up to a dr of approximately 5:1. The low product yield obtained
from the use of -methylcyclohexylcarboxaldehyde is the result of
b-isomer 6 (Scheme 2)
Mechanistic studies are uniform in suggesting a radical mech-
anism for the Mukaiyama epoxidation reaction.18 Mizuno18b and
Valentine18c have proposed acyl peroxy radicals while Kaneda19 is
favoring peracids as oxygen transfer species. In 2000, Beak has
reported the free radical internal epoxidation of 2,2-dimethyl-5-
phenylpent-4-enal using oxygen and catalytic amount of TBHP as
initiator.20 His studies clearly suggested acyl peroxy radicals as
oxygen transfer species. Later, Feiters reported mechanistic studies
evaluating the role of metal catalyst and ratio of aldehyde to alkene
in the epoxidation of alkenes18a and concluded that by increasing
the aldehyde to alkene ratio oxidation proceeds through an in-
ternally formed peroxyacid. This conclusion is, however, in dis-
agreement with Mukaiyama’s original discoveries and our present
studies. Epoxidation of cholesterol derivatives clearly shows dif-
ferences in the epoxide isomer ratio when peracid or Mukaiyama
epoxidation is used (Scheme 2). To resolve this mechanistic dis-
agreement we applied the conditions of Table 1 to the oxidation of
2,2-dimethyl-5-phenylpent-4-enal. If the outcome of the reaction
is not dependent on the initiation step, as we suspected, we should
obtain same products as in Beak’s report. Indeed, the products
obtained are the same (Scheme 4), and this result points to acyl
peroxy radical rather than peracid participation in the epoxidation
of alkenes.
When the epoxidation of cholesteryl acetate (Scheme 2) was
performed using 2-methyl-2-phenylpropanal as the aldehyde re-
ductant, no epoxidation products (5 and/or 6) were observed over
the same reaction time. Instead, reactant cholesteryl acetate 4 was
isolated, and both cumyl alcohol (major) and cumyl hydroperoxide
(minor) were isolated and identified. This outcome is consistent
with the steric influence of the aldehyde in the delivery of the
epoxide oxygen and with the rapid loss of carbon monoxide from
the acyl radical formed by hydrogen abstraction from 2-methyl-2-
phenylpropanal.
a
a slow turnover, and unreacted starting material remained after the
reaction time of 3 h.
Scheme 2. Effect of aldehyde on diastereoselectivity in the Mukaiyama epoxidation of
cholesteryl acetate.
Epoxidation selectivity for reactions with terpenes has been
investigated by several groups.15e17 Moderate selectivity for mono-
and di-epoxide products (20 and 58%, respectively) of neryl acetate
has been reported using a catalytic system containing perox-
otungstophosphate and hydrogen peroxide as the terminal oxi-
dant.15 In another study selectivity in epoxidation of geranyl acetate
utilizing in situ prepared dioxirane from OxoneÒ and tri-
fluoroacetophenone was greatly affected by temperature, where at
Based on these observations the mechanism for epoxidation is
proposed to occur in the manner described in Scheme 5. Reaction is
propagated by acyl radical formation that then reacts with molec-
ular oxygen to form the acyl peroxy radical. Oxygen atom transfer
from the acyl peroxy radical to the alkene occurs in a two-step
process that includes addition of the acyl peroxy radical and
0
ꢀC the mono epoxide was obtained in 76% yield, while at room
temperature oxidation resulted in a 1:1 mixture of mono and
diepoxides.16
To assess selectivity for epoxidation under Mukaiyama con-
ditions with rhodium acetate as the catalyst selected terpenes