Unfunctionalized terminal olefins are readily epoxidized,
but more electron-deficient terminal olefins such as butadiene
monooxide (entry 14) or allyl acetate (entry 19) define a
boundary of electron-deficient substrates for this catalytic
system. Allylic alcohols are efficiently epoxidized, although
alcohol oxidation can be a significant side reaction; 2-cyclo-
hexen-1-ol (entry 18) is oxidized to 2,3-epoxycyclohexan-
1-ol (77%), 2-cyclohexenone (19%), and a small amount of
2,3-epoxycyclohexanone (4%). As 2-cyclohexenone is un-
reactive under the oxidizing conditions, the 2,3-epoxycyclo-
hexanone results from a secondary oxidation of the initial
2,3-epoxycyclohexan-1-ol product.
Epoxidation of styryl olefins (entries 9-12) under the
acidic conditions of the reported reaction may be considered
to be unconventional due to the acid sensitivity of the styryl
epoxide products. Yet, most styryl substrates are efficiently
epoxidized. Styrene and cis-â-methyl styrene (entries 9 and
11) show lower selectivities for the epoxide product due to
the formation of phenylacetaldehyde (37%) and propio-
phenone (26%), respectively, during the course of the
reaction. Appropriate control experiments indicate that these
carbonyl products do not result from epoxide opening and
rearrangement, suggesting a secondary oxidation pathway
or oxidant.29
Compared to [MnII((R,R)-mcp)]2+, the active oxidant
created with peracetic acid and 1 is a less potent epoxidizing
agent, providing an opportunity for regiospecific epoxidation
of polyenes.1 The two trans-alkenes of ethyl sorbate are
electronically distinct, yet [MnII((R,R)-mcp)]2+ with peracetic
acid epoxidizes them in a 4:1 product ratio of monoepoxides.
Use of 1 and peracetic acid epoxidizes only the more
electron-rich alkene (entry 15). Indiscriminant epoxidation
of butadiene is also found with [MnII((R,R)-mcp)]2+, while
1 exclusively forms the monoepoxide product, providing a
convenient route to this useful epoxide synthon (entries 13
and 14).30
The nature of the active oxidant has been probed by
intermolecular competition reactions. A competitive reaction
of cyclooctene and vinylcyclohexane shows a 100:1 prefer-
ence for the former at low conversion (∼20%) confirming
the electrophilic nature of the active oxidant. The 5:1
preference for epoxidation of cis-2-heptene over trans-
heptene contrasts with the reactivity of “planar” salen- or
porphyrin-based catalyts that are chemoselective for cis-
olefins. The retention of the original cis sterochemistry in
the epoxide product of cis-2-heptene with 1 (<1% trans-
epoxide) supports a nonradical oxidation process.
tant metal speciation and ligand exchange rates. Reactions
performed with commerical peracetic acid (1% H2SO4) result
in a reaction solution pH of ∼2, and the epoxidation of vinyl
cyclohexane is complete within 5 min with >90% yield
(Table 1). In contrast, strong acid free peracetic acid as the
oxidant creates a reaction solution pH of ∼5, and only ∼20%
epoxidation is observed in 5 min.31 The epoxidation ef-
ficiency of 1 with strong acid free peracetic acid is restored
by adjusting the pH to 2 with either HClO4 or H2SO4,
accentuating the proton sensitivity of this reaction.
Given the complex composition of the peracetic acid
oxidant (H2O, HSO41- (pKa ) 2.0), CH3CO2H (pKa ) 4.7),
CH3CO3H (pKa ) 8.2)) and the sequential pKa values of the
water molecules ligated to [((phen)2(H2O)FeIII)2(µ-O)]4+ (pKa
) 5.0 and 6.5),32 this proton dependence is understandable.
Assuming a tetracationic µ-oxo ferric dimer as the resting
state catalyst of 1, water displacement at one or both ferric
sites by a monodentate or bridging peracetic acid molecule,
respectively, must precede the formation of the active
oxidant. Lower pHs ensure facile water exchange from the
metal, as hydroxide or acetate ligation to the iron will be
minimized. Bisulfate, perchlorate, or nitrate ligation to the
iron site(s) do not appear to be competitive with other ligands
under the reaction conditions. Whether one or both of these
sites is necessary for activation of the peracetic acid is cur-
rently unknown. However, the inactivity of [((phen)2(Cl)FeIII)2-
(µ-O)](Cl)2 with strongly ligated chlorides suggests that labile
coordination sites are required. Chloride inhibition of other
oxidation reactions with iron catalysts has been previously
noted.33
A related µ-oxo ferric dimer, [((mep)FeIII)2(µ-O)(µ-
OAc)]3+ (3), is proposed by Jacobsen as the resting state of
their efficient epoxidation catalyst that uses H2O2 in acetic
acid/MeCN.2 Que has reported dramatically different reactiv-
ity of 3.34 Yet, the conditions used in each report are quite
different; Jacobsen’s include HOAc, while those of Que are
more neutral and do not include acetic acid. In our hands, 3
is an efficient terminal olefin epoxidation catalyst under
acidic conditions. Curiously, [((phen)2FeIII)2(µ-O)(µ-OAc)]3+,20
a complex structurally similar to 3, is an impotent terminal
alkene epoxidation catalyst under comparable acidic condi-
tions. These differences highlight the sensitivity of the
epoxidation reaction not only to the catalyst but to the
reaction conditions as well (i.e., pH).
In this paper we have demonstrated that iron complexes,
easily synthesized from inexpensive and commercially avail-
able compounds, are efficient catalysts for the epoxidation
of alkenes using peracetic acid as the oxidant. The temporal
efficiency of both the in situ catalyst preparation and the
epoxidation reaction itself provide significant advantages over
other epoxidation procedures for terminal olefins. This
system constitutes a second example of a bioinspired non-
Aside from the intermolecular competition studies, the
nature of the active species has been difficult to assess. In
the presence and absence of commercial peracetic acid, UV-
vis, EPR, and magnetic susceptibility measurements of 1
show insignificant changes, suggesting that the predominant
species in solution is a strongly antiferromagnetically coupled
µ-oxo ferric dimer. However, a strong pH dependence on
the efficiency of the epoxidation reaction implicates impor-
(31) For initial formation of epoxide at various pHs, see Supporting
Information.
(32) DubocToia, C.; Menage, S.; Vincent, J. M.; AverbuchPouchot, M.
T.; Fontecave, M. Inorg. Chem. 1997, 31, 6148-6149.
(33) Nam, W.; Lim, M. H.; Oh, S.-Y.; Lee, J. H.; Lee, H. J.; Woo, S.
K.; Kiem, C.; Shin, W. Angew. Chem., Int. Ed. 2000, 39, 3646-3649.
(34) Ryu, J. Y.; Kim, J.; Costas, M.; Chen, K.; Nam, W.; Que, L. Chem.
Commun. 2002, 1288-1289.
(29) Under the standard reaction conditions, styrene oxide does not react.
(30) Butadiene reactions were performed on a 0.025 mol scale and
isolated as solutions in pentane or toluene.
Org. Lett., Vol. 5, No. 14, 2003
2471