the CH bond (TSCH); and (3) in both cases, the steric and/or
electronic communications between the substrate and the
dioxirane are mainly coming from the more substituted end.
On the basis of this analysis, we hypothesized that the ketone
catalyst that induces high enantioselectivity in the epoxidation
of terminal alkenes should also induce high enantioselectivity
in CH bond oxidation.
Shi and co-workers have reported that oxazolidinone
ketones 2 and 3 are very good catalysts for the asymmetric
epoxidation of terminal alkenes such as styrenes.9 On the
basis of the recent theoretic work on the origin of the
enantioselectivity in this asymmetric epoxidation8d and our
above analogy, we reasoned that these catalysts should also
be good for CH oxidation of benzylic alcohols because in
both cases there is a phenyl group to interact with dioxirane
to direct the substrate approach.8d,9 By using a modified
procedure, we synthesized the known ketones 2 and 3 and
the new derivatives 4 and 5 and applied them for the
asymmetric CH oxidation of some benzylic Vic-diols. The
results are collected in Table 1.
To our pleasure, ketone 2 indeed yields a much improved
enantioselectivity in the asymmetrization of meso-hydroben-
zoin, and an ee value of 70% was obtained with 1 equiv of
the catalyst (Table 1, entry 1). For comparison, catalyst 1
yields only 45% ee of the product of this substrate.5 Ketone
3 is an even better catalyst for this oxidation, as a high ee
value of 87% was obtained and only 50 mol % of catalyst
loading was necessary (entry 2). This is the first example of
dioxirane-mediated asymmetric CH oxidation using a cata-
lytic amount of the ketone catalyst. The remote substituent
on the oxazolidinone ring was found to have a subtle
influence on the enantioselectivity of the reaction: with the
size of R reduced from t-Bu to Et or Me, the enantioselec-
tivity dropped slightly from 87% to 80% (entries 3 and 4).
On the basis of our preliminary screening, catalysts 3-5 are
comparable in reactivity, whereas catalyst 3 always yields a
slightly higher enantioselectivity than the other two.
Further study with catalyst 3 reveals that very good ee
values may be obtained for the asymmetrization of var-
ious meso-4,4′-disubstituted hydrobenzoins (g76% ee,
entries 5-9). However, the dependence of the enantioselec-
tivity on the electronic nature of the para substituents that
has been observed for catalyst 15 did not happen in the case
of catalyst 3.
Figure 1. Ketone catalysts utilized for CH oxidation.
enantioselective CH oxidation protocol for the oxidation of
Vic-diols.
Although ketone 1 is an excellent catalyst for the asym-
metric epoxidation of trans- and trisubstituted alkenes,6 the
asymmetric induction is considerably lower in CH bond
oxidation.5 The reason for this is probably due to totally
different steric requirements for these two oxidations.
Through the comparison of the transition state (TS) of CH
oxidation7 with those of epoxidation of different alkene
substrates, we found that these distinct TSs3,8 achieve the
closest resemblance of each other in the cases of CH
oxidation and epoxidation of the terminal alkene (Figure 2):
Figure 2. Transition states for the epoxidation of terminal alkene
(TSE) and the CH oxidation (TSCH).
(1) Both TSs are asynchronous spiro; (2) in the terminal
alkene cases (TSE), the terminal CH2 group is small and not
differentiated in space (regarding the left and right sides of
the forming oxirane plane), as is the hydrogen atom end of
(6) (a) Tu, Y.; Wang, Z.-X.; Shi, Y. J. Am. Chem. Soc. 1996, 118, 9806-
9807. (b) Wang, Z.-X.; Tu, Y.; Frohn, M.; Shi, Y. J. Org. Chem. 1997, 62,
2328-2329. (c) Wang, Z.-X.; Tu, Y.; Frohn, M.; Zhang, J.-R.; Shi, Y. J.
Am. Chem. Soc. 1997, 119, 11224-11235. (d) Frohn, M.; Dalkiewicz, M.;
Tu, Y.; Wang, Z.-X.; Shi, Y. J. Org. Chem. 1998, 63, 2948-2953. (e)
Cao, G.-A.; Wang, Z.-X.; Tu, Y.; Shi, Y. Tetrahedron Lett. 1998, 39, 4425-
4428. (f) Wang, Z.-X.; Shi, Y. J. Org. Chem. 1998, 63, 3099-3104. (g)
Wang, Z.-X.; Cao, G.-A.; Shi, Y. J. Org. Chem. 1999, 64, 7646-7650. (h)
Frohn, M.; Zhou, X.; Zhang, J.-R.; Tang, Y.; Shi, Y. J. Am. Chem. Soc.
1999, 121, 7718-7719. (i) Lorenz, J. C.; Frohn, M.; Zhou, X.; Zhang, J.-
R.; Tang, Y.; Burke, C.; Shi, Y. J. Org. Chem. 2005, 70, 2904-2911.
(7) As one reviewer pointed out, the most recent theoretic work by Sarzi-
Amade´ and co-workers described a perpendicular radicaloid transition state
for this oxidation; however, as the authors conceded, such a mechanism
cannot explain the high selectivity data obtained experimentally (ref 3d).
(8) (a) Houk, K. N.; Liu, J.; DeMello, N. C.; Condroski, K. R. J. Am.
Chem. Soc. 1997, 119, 10147-10152. (b) Crehuet, R.; Anglada, J. M.;
Cremer, D.; Bofill, J. M. J. Phys. Chem. A 2002, 106, 3917-3929. (c)
Bach, R. D.; Dmitrenko, O.; Adam, W.; Schambony, S. J. Am. Chem. Soc.
2003, 125, 924-934. (d) Singleton, D. A.; Wang, Z. J. Am. Chem. Soc.
2005, 127, 6679-6685.
The kinetic resolution of racemic hydrobenzoins was also
studied with catalyst 3. Again, improved enantioselectivity
was observed as compared with catalyst 1. For example, with
3 as catalyst, an ee value of 87% was obtained for the product
of rac-hydrobenzoin, whereas the reported result with catalyst
1 was only 65% ee.5 In most cases, the racemic substrates
yield better enantioselectivities of the products than their
meso counterparts (entries 10-15). For example, the fluoro-
substituted racemic diol generates an ee value of 90% for
the product (entry 12), whereas the corresponding meso diol
(9) (a) Tian, H.; She, X.; Shu, L.; Yu, H.; Shi, Y. J. Am. Chem. Soc.
2000, 122, 11551-1152. (b) Tian, H.; She, X.; Xu, J.; Shi, Y. Org. Lett.
2001, 3, 1929-1931. (c) Tian, H.; She, X.; Yu, H.; Shu, L.; Shi, Y. J. Org.
Chem. 2002, 67, 2435-2446. (d) Shu, L.; Wang, P.; Gan, Y.; Shi, Y. Org.
Lett. 2003, 5, 293-296.
3014
Org. Lett., Vol. 8, No. 14, 2006