Organic Letters
Letter
suggested that the highest s value could be obtained under the
following reaction conditions: 0.1 mmol substrates and 10 mol
% catalyst A in mesitylene (1.5 mL) at 25 °C for 47 h.
Scheme 3. Gram-Scale Preparation and Synthetic
Applications
a
With the optimized reaction conditions in hand (Table 1,
entry 16), we investigated the substrate scope of the
established kinetic resolution of peroxides through an
enantioselective γ-elimination reaction (Scheme 2). The
peroxides were obtained from the corresponding olefins,
aldehydes, and t-BuOOH or cumyl hydroperoxide under the
catalysis of n-Bu4NBr; the detailed procedures are shown in the
SI. Substrates with electron-donating R1 groups at the phenyl
ring allowed the enantioselective γ-elimination to give chiral
peroxides (S)-1b−f with high s values and with up to 94% ee;
the corresponding epoxides (2R,3S)-2b−f were also obtained
with good enantioselectivities. para-Chloride-substituted sub-
strate racemic 1g was tolerated in our reaction system, and
chiral peroxide (S)-1g was recovered in 44% yield with 90% ee.
Epoxide (2R,3S)-2g was formed with a moderate ee value.
Similarly, para-bromo-substituted substrate 1h also afforded a
good selectivity with an s value of 31. By substituting the
phenyl ring with naphthalene, 2-furan, or 2-thiophene,
peroxides 1i−k were also efficiently resolved with high
enantioselectivities (s value up to 42). After the investigation
of the R1 group of peroxides, we determined the substrate
scope with different R2 groups. Substrates with electron-
donating R2 groups resulted in chiral peroxides (S)-1l (methyl)
and (S)-1m (tert-butyl), respectively, with 98% ee and 93% ee,
and epoxides (2R,3S)-2l and (2R,3S)-2m were obtained with
good enantioselectivities as well. Using halides such as fluoride,
chloride, and bromide as R2 groups located in the para position
of phenyl ring, corresponding substrates were suitable for the
resolution to give enantioenriched peroxides (S)-1n−p (up to
97% ee) and epoxides (2R,3S)-2n−p (up to 84% ee), and
selectivity factors as high as 37 were obtained. The absolute
configurations of (S)-1p and (2R,3S)-2p were determined by
X-ray crystallography, and others were analogously assigned.
When R2 was an ortho-chloro or meta-bromo group on the
phenyl ring, chiral peroxides (S)-1q and (S)-1r were recovered
after an enantioselective γ-elimination process with promising
results (96% ee and up to 42% yield), and epoxides (2R,3S)-
2q-r were obtained with up to 81% ee. After substituting the
tert-butyl peroxide group with a dimethyl phenyl group, the
corresponding substrate could also be resolved to give chiral
peroxide (S)-1s in 41% yield with 87% ee. Nevertheless, the
reactions were not successful with α-ester peroxide and alkyl
ketone peroxides under standard reaction conditions.
a
For reaction conditions, see the SI.
both products were obtained in good yield with high ee.
Treating epoxide (2R,3S)-2p with SnCl4 afforded a chlorinated
open-chain product 6 instead of a cyclized adduct, as
reported.20 The absolute configuration of product 6 was
assigned through X-ray crystallography.
In an attempt to gain some insights into the mechanism of
this enantioselective γ-elimination transformation, we meas-
ured the independent initial rates of parallel reactions with
substrates ( )-1p and α-proton-deuterated substrate ( )-1p-
D under the standard reaction conditions (Scheme 4). The
primary intermolecular kinetic isotopic effect (KIE)21 was
calculated to be 2.88, indicating that the deprotonation of the
α-proton of the carbonyl group was probably involved in the
rate-determining step.
a
Scheme 4. Kinetic Isotopic Effect Experiment
After the substrate scope of the reaction was investigated, we
planned to further demonstrate the synthetic potential of our
reaction. First, a gram-scale reaction was carried out to prepare
(S)-1p and (2R,3S)-2p under the optimal reaction conditions,
and the chemical yield and enantioselectivity showed almost
no variation (Scheme 3). Next, a few functional trans-
formations of chiral substrates and products were conducted.
As expected, the enantioenriched peroxide (S)-1p could be
converted into the corresponding peroxide 3 in 80% yield with
good stereoselectivity (93% ee and 4:1 dr) by sodium
borohydride. Upon the treatment of (S)-1p with 2,4-
dinitrophenyl hydrazine in methanol at 40 °C for 6 h, product
4a was formed as the major product in 73% yield with 93% ee
value. Moreover, the utility of the epoxide from this γ-
elimination-based kinetic resolution was also explored. Epoxide
(2R,3S)-2p was sufficiently transformed into alcohol 5 and
ester 7 through a reductive and oxidative process, respectively;
a
For reaction conditions, see the SI.
D
Org. Lett. XXXX, XXX, XXX−XXX