Asymmetric Weitz-Scheffer epoxidation of α,β-enones by optically active hydroperoxides: Control of enantioselectivity through metal-coordinated or hydrogen-bonded templates

Article abstract of DOI:10.1002/1099-0690(200202)2002:4<630::AID-EJOC630>3.0.CO;2-Y

The enantioselective epoxidation of the α,β-enones 1 with the optically active hydroperoxides (-)-(S)-2, catalyzed by the bases KOH or DBU, affords the enantiomerically enriched oxo epoxides 3 in high yields and with good enantioselectivities (up to 90% e

Full text of DOI:10.1002/1099-0690(200202)2002:4<630::AID-EJOC630>3.0.CO;2-Y

FULL PAPER
Asymmetric Weitz؊Scheffer Epoxidation of α,β-Enones by Optically Active
Hydroperoxides: Control of Enantioselectivity through Metal-Coordinated or
Hydrogen-Bonded Templates
Waldemar Adam,*[a] Paraselli Bheema Rao,[a] Hans-Georg Degen,[a] and
Chantu R. Saha-Möller[a]
Keywords: Asymmetric synthesis / Epoxidation / Chiral peroxides / Template structures / WeitzϪScheffer reaction
α β
S 2
1
Si
αR βS
−
3a
Re
3
ee
ee
αS βR -
3a
1a
Introduction
An alternative strategy is the use of metal catalysts with
optically active ligands.[5] For example, diethylzinc and
molecular oxygen in the presence of (1R,2R)-N-methyl-
pseudoephedrine or polybinaphthol have proven to be an
effective optically active oxidant for the enantioselective
epoxidation of α,β-enones.[22,23] Moreover, tert-butyl hydro-
peroxide (TBHP)[24Ϫ26] may be used as an achiral oxygen
The asymmetric WeitzϪScheffer epoxidation of electron-
poor alkenes was first explored by Wynberg[1Ϫ4] and co-
workers, more than two decades ago. As chiral inductors,
they used quaternary ammonium salts derived from cin-
chona alkaloids; the chemical function of these optically
active agents is to act as phase-transfer catalysts (PTCs).[5]
Over recent years, this approach has enjoyed much atten-
tion. Indeed, continuous improvement of the PTCs[6Ϫ10]
and extensive variation of the oxygen donor (e.g., the most
commonly used H2O2,[6,7] tert-butyl hydroperoxide,[8] and
cumene hydroperoxide,[9] and also hypochlorite[7,10]) have
provided excellent enantioselectivities. Similarly well estab-
lished and no less successful is the polyamino acid based
JuliaϪColonna method.[11Ϫ17] In this epoxidation, the
chiral oxidant is constituted by heterogeneous poly--leuc-
ine in combination with NaOH/H2O2, which is employed
in a triphasic system. Biphasic media with the urea/H2O2
adduct as oxygen source and DBU as base,[18,19] as well as
with a THF-soluble poly--leucine copolymer,[20] have also
been developed, and provide good enantioselectivities. In
addition, optically active bases have been employed most
recently for inducing asymmetry in the WeitzϪScheffer
epoxidation.[21]
source
in
combination
with
optically
active
lanthanideϪbinols as asymmetric inductors; high enanti-
oselectivities have been achieved with this system.
Despite the fact that optically active hydroperoxides are
now conveniently available through enzymatic kinetic res-
olution,[27,28] such chiral inductors have so far received little
if any attention for the preparation of enantiomerically en-
riched epoxides of α,β-enones through the WeitzϪScheffer
reaction. These chiral oxygen sources have previously been
utilized for the electrophilic TiIV-catalyzed asymmetric ox-
idation of sulfides and allylic alcohols,[29Ϫ33] which estab-
lished their potential synthetic value. Recently, two re-
ports[34,35] on the use of optically active hydroperoxides in
enantioselective oxidations have appeared. In a very recent
preliminary[36] report, we demonstrated the efficacy of op-
tically active hydroperoxides for the epoxidation of α,β-en-
ones through the asymmetric WeitzϪScheffer reaction, cat-
alyzed by KOH (Scheme 1). For the enantiomeric enrich-
ment, a template structure held together by Kϩ, in which
steric interactions between the β-substituent of the enone 1
and the substituents at the chirality center of the peroxide
anion 2Ϫ provide the necessary stereochemical differenti-
[a]
Institut für Organische Chemie, Universität Würzburg,
Am Hubland, 97074 Würzburg, Germany
Fax: (internat.) ϩ 49-(0)931/888-4756
E-mail: adam@chemie.uni-wuerzburg.de
Supporting information for this article is available on the
WWW under http://www.eurjoc.com or from the author.
630
WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/0204Ϫ0630 $ 17.50ϩ.50/0 Eur. J. Org. Chem. 2002, 630Ϫ639

Asymmetric WeitzϪScheffer Epoxidation of α,β-Enones by Optically Active Hydroperoxides
FULL PAPER
ation, was proposed to account for the observed preference
for the (αS,βR)-epoxide 3, through Si-face attack.
Results
For the asymmetric WeitzϪScheffer epoxidation with op-
tically active secondary hydroperoxides, the enantiomer-
ically pure (Ͼ 99% ee) (Ϫ)-(S)-1-phenylethyl hydroperoxide
(2a) was selected as oxygen donor for preliminary screening
(Scheme 2). In order to optimize the reaction conditions,
the initial experiments were performed on chalcone 1a as a
model substrate, by varying the base, solvent and the reac-
tion temperature. KOH as the base in CH3CN at Ϫ40 °C
proved to be the best conditions for the asymmetric epoxid-
ation (Table 1).
Scheme 1. Facial selectivity in the WeitzϪScheffer epoxidation with
optically active hydroperoxides 2
It was of synthetic interest and mechanistic importance
to explore the performance of tertiary amines (e.g., 1,8-di-
azabicyclo[5.4.0]undec-7-ene, DBU) as base catalysts in
combination with optically active hydroperoxides as chiral
oxygen sources for the enantioselective epoxidation of α,β-
enones. A preliminary experiment revealed that the epoxide
was formed in a low enantiomeric excess (9% ee) but, more
significantly, that the opposite enantiomer was expressed as
the major product (Scheme 1). This unexpected dichotom-
ous behavior in the two bases Ϫ the inorganic base KOH
delivering the (αS,βR) enantiomer and the organic base the
(αR,βS) congener Ϫ merited a more detailed investigation.
The purpose was to explain the mechanistic circumstances
that underlie the fact that, with the same prochiral α,β-en-
one 1 and the same optically active hydroperoxide 2, oppos-
ite enantiomers of the epoxide 3 are favored by the mere
choice of the type of base. Together with a full report on
our previous preliminary results relating to the KOH-cata-
lyzed asymmetric epoxidation of α,β-enones 1 by the optic-
ally active hydroperoxides 2, we offer here an integrated
mechanistic concept for this enantioselective oxy func-
tionalization in terms of a common template structure for
both base catalysts.
Table 1. Optimization of the reaction conditions in the enantiose-
lective WeitzϪScheffer epoxidation of the enone 1a by (Ϫ)-(S)-1-
phenylethyl hydroperoxide (2a)[a]
Entry
Base
Solvent
T
Conversion Yield[b] ee[c] [d]
[°C] (%)
(%)
(%)
1
2
3
4
5
6
7
8
9
10
11
12
13
NaH
NaH
Na2CO3 CH3CN
Cs2CO3 CH3CN
K2CO3 CH3CN
K2CO3 DMF
K2CO3 neat
K2CO3 CH3CN
K2CO3 CH3CN Ϫ10
KOH
KOH
Et2O
CH3CN
20 Ͼ 99
20 Ͼ 99
96
95
91
98
92
84
88
90
72
96
99
95
94
7
17
18
22
25
20
22
33
35
42
51
45
48
20
95
20 Ͼ 99
20 Ͼ 95
20 Ͼ 95
20 Ͼ 95
0
94
75
CH3CN Ϫ30 Ͼ 99
CH3CN Ϫ40 Ͼ 99
CH3CN Ϫ40 Ͼ 99
CH3CN Ϫ40 Ͼ 99
NaOH
CsOH
[a]
Enone 1a (0.1Ϫ0.5 mmol), hydroperoxide 2a (1.0 equiv.), and
[b]
[c]
base (1.1Ϫ1.2 equiv.).
a Chiracel OD column with 2-propanol/hexane (5:95) as eluent;
error Յ 3% of the stated values. The configuration of the major
Isolated epoxide 3. HPLC analysis on
[d]
enantiomer is (αS,βR), determined by comparison with literature
data.
Initially, the reaction was carried out in Et2O with NaH
as base (Entry 1), under which conditions the reaction was
complete within 2 h at ambient temperature (ca. 20 °C). Al-
though the epoxide product 3a was obtained in excellent
yield, only poor (7% ee) enantioselectivity was observed.
When the reaction was performed in the more polar
CH3CN as solvent, the ee value of the epoxide 3a was
higher, but still only 17% (Entry 2). The weaker bases
Na2CO3, Cs2CO3, and K2CO3 in CH3CN (Entries 3Ϫ5)
gave no substantial improvement in the enantioselectivity
of the epoxidation of 1a, only up to 25% ee for K2CO3.
Other polar solvents such as DMF (Entry 6), yielded the
epoxide with even lower enantioselectivity (20% ee), as did
the reaction in the absence of solvent (22% ee, Entry 7). An
increased ee value of 33% and a degree of conversion of
94% was observed with the K2CO3/CH3CN combination at
0 °C (Entry 8). Further lowering of the temperature to Ϫ10
°C increased the enantioselectivity only to 35% (Entry 9),
with the disadvantage of a substantially reduced reactivity,
that is, conversion of only 75% after 48 h. In order to
shorten the reaction time and to allow even lower temper-
Scheme 2. Base-catalyzed enantioselective WeitzϪScheffer epoxid-
ation of enones 1 by optically active hydroperoxides 2
Eur. J. Org. Chem. 2002, 630Ϫ639
631

W. Adam, P. Bheema Rao, H.-G. Degen, C. R. Saha-Möller
FULL PAPER
atures to be examined, the stronger base KOH was neces- strate 1a (up to 51% ee, Entries 1Ϫ4). With the standard
sary. Thus, the KOH/CH3CN combination afforded the hydroperoxide 2a, the (αS,βR)-epoxide 3b was obtained
(αS,βR)-epoxide 3a in 42% ee at Ϫ30 °C (Entry 10) and in quantitatively in 44% ee (Entry 6). The 4-chloro-substituted
51% ee at Ϫ40 °C (Entry 11), nearly quantitatively at both hydroperoxide 2b gave the epoxides 3a and 3b (Entries 2
temperatures. To lower the temperature still further and 7) with about the same ee values (48 and 40%) as ob-
(CH3CN freezes at Ϫ46 °C), toluene, CH2Cl2, and ethyl served with the unsubstituted hydroperoxide 2a (Entries 1
ether (data not shown in Table 1) were used as solvents; and 6). Use of hydroperoxide 2c (Entries 3 and 8), which
however, not even traces of epoxide product were detected bears a β-naphthyl substituent, resulted in slightly lower en-
between Ϫ40 and Ϫ78 °C under these conditions. Use of antioselectivities (43 and 38% ee). With the sugar-derived
the still stronger base nBuLi in THF at Ϫ78 °C resulted in hydroperoxide 2d (Entries 4 and 9), the lowest ee values (14
a complex mixture of products, presumably due to destruc- and 31% ee) in this series were registered. The (Ϫ)-(S)-1-
tion of the chiral ROOH 2 (data not shown in Table 1). A phenylethyl hydroperoxide (2a) was hence found to be the
change in the alkali metal ion in the hydroxide base resulted oxygen source of choice for this asymmetric WeitzϪScheffer
in somewhat lower ee values of 45% for NaOH (Entry 12) epoxidation of α,β-enones.
and 48% for CsOH (Entry 13). From these results, KOH/
The roles of steric and electronic factors in the enone
CH3CN at Ϫ40 °C are so far the best conditions for the
substrate 1 were subsequently examined, in order to estab-
asymmetric WeitzϪScheffer epoxidation of enone 1a with
lish whether they influenced this asymmetric epoxidation.
With the hydroperoxide 2a under the optimized conditions,
the substituents on the α,β-enone were varied in a system-
atic way (Table 2, Entries 1, 6, 10Ϫ16, and 18). Firstly, the
the optically active hydroperoxide 2a.
Attention was next focused on the choice of the best
chiral hydroperoxide (Table 2).
aromatic ring of the enone carbonyl functionality was sub-
stituted in the para-position (Entries 10Ϫ12). Electron-do-
Table 2. Enantioselective WeitzϪScheffer epoxidation of the enones
1 with the optically active hydroperoxides 2, base-catalyzed by
nating 4-methyl (substrate 1c, Entry 10) and 4-methoxy
groups (substrate 1d, Entry 11), and also the electron-with-
drawing 4-bromo substituent (substrate 1e, Entry 12) did
not significantly (48Ϫ54% ee) influence the enantioselectiv-
ity, compared to enone 1a (Entry 1).
KOH[a]
Entry
Enone
R*OOH
Yield (%)[b]
ee (%)[c] [d]
1
2
3
1a
1a
1a
1a
1a
1b
1b
1b
1b
1c
1d
1e
1f
2a
2b
2c
2d
2a
2a
2b
2c
2d
2a
2a
2a
2a
2a
2a
2a
2a
2a
99
95
87
91
94
99
96
86
92
98
97
95
97
96
98
95
91
90
51
48
43
14
6
Unlike the carbonyl aryl group, its counterpart in the β-
position of the α,β-enone did affect the enantioselectivity.
Thus, a regular trend to higher ee values was displayed by
the electron-donating para-methyl group in substrate 1f
(57% ee, Entry 13), and even more so by the para-methoxy
substituent in derivative 1g (61% ee, Entry 14). In contrast,
the electron-withdrawing 4-nitro substituent in the enone 1h
(42% ee, Entry 15) lowered the enantioselectivity.
4
5[e]
6
44
40
38
31
54
53
48
57
61
42
75
11
90
7
8
9
10
11
12
13
14
15
16
17[e]
18
In order to examine steric effects, the β-alkyl-substituted
enones 1b and 1i were employed as substrates in this asym-
metric WeitzϪScheffer epoxidation, the β-methyl-substi-
tuted enone 1b (Table 2, Entry 6) having already been men-
tioned in the optimization. Its epoxidation yielded the same
(αS,βR)-epoxide 3b enantiomer as substrate 1a (Entry 1),
but with a somewhat lower ee value of 44%. The sterically
more demanding β-tert-butyl-substituted enone 1i
(Entry 16) afforded the (αS,βR)-epoxide 3i with the highest
ee value yet observed (75%) and in 95% yield.
1g
1h
1i
1i
1j
[a] Enone 1 (0.1Ϫ0.5 mmol), hydroperoxide 2 (1.0 equiv.), and KOH
(2Ϫ3 equiv.) in CH3CN at Ϫ40 °C; quantitative consumption of
both the enone 1 and hydroperoxide 2 was observed, except for
Entries 3 and 8, for which the conversions of 1a and 1b were 90%
[b]
[c]
and 92%.
Isolated epoxide 3.
HPLC analysis on a Chiracel
OD column; except Entries 10Ϫ12 and 18 for which a Chiracel
OB-H column was used; error Յ 3% of the stated values.
configuration of the major enantiomer is (αS,βR). In the pres-
The conformation about the single bond between the car-
bonyl functionality and the enone double bond as a struc-
tural variable in the substrate 1 was next investigated. For
this purpose, the conformationally rigid cyclic olefins 1j and
1k were chosen, the derivative 1j possessing fixed s-cis and
[d]
The
[e]
ence of 8-crown-6 ether.
Epoxidations of the aryl-substituted enone 1a 1k s-trans geometries. The optimized reaction conditions af-
(Entries 1Ϫ4) and the alkyl-substituted enone 1b forded an ee value of 90% in the epoxidation of s-cis enone
(Entries 6Ϫ9) were carried out with the optically active hy- 1j (Table 2, Entry 18), the highest enantioselectivity ob-
droproxides 2aϪd under the optimized reaction conditions served so far. A contrary effect was observed with the s-
shown in Table 1. The ee values for the epoxidation of the trans enone 1k (not shown in Table 2). At Ϫ40 °C, no trace
β-methyl-substituted enone 1b were somewhat lower (up to of epoxide was detected even after 24 h; however, when the
44% ee, Entries 6Ϫ9) than for the β-phenyl-substituted sub- temperature was raised to Ϫ20 °C, ca. 20% of the enone
632
Eur. J. Org. Chem. 2002, 630Ϫ639

Asymmetric WeitzϪScheffer Epoxidation of α,β-Enones by Optically Active Hydroperoxides
FULL PAPER
had been converted into the epoxide after 48 h, but no op-
tical activity was observed.
The role of the potassium ion in this asymmetric
WeitzϪScheffer epoxidation was tested with the aid of the
potassium-chelating 18-crown-6 ether. In the case of enone
1a, the ee value diminished dramatically from 51% (Table 2,
Entry 1) to 6% ee (Entry 5) with the chelating agent present
under otherwise standard conditions. The 75% ee found in
the epoxidation of substrate 1i (Entry 16) dropped to
11% ee (Entry 17) when the 18-crown-6 ether was added.
Alternatively, the use of the organic base DBU (4a) instead
of KOH in acetonitrile (Table 3, Entry 1) also gave a much
decreased enantioselectivity of only 9% ee with the enone
1a, with the opposite enantiomer (αR,βS)-3a in fact being
produced.
In view of this drastic diminution in the enantioselectiv-
ity, it was important to examine the amine-mediated reac-
tion with the enone 1a in more detail. The epoxide 3a was
also obtained with a very small ee of 4% in favor of the
(αR,βS) enantiomer (Entry 2) in (protic) methanol and with
DBU (4a) as base. In the nonpolar toluene (Entry 3), the
reaction proceeded with 90% conversion to yield 98% of the
(αR,βS)-epoxide 3a in 40% ee with DBU (4a).
The comparatively small tetramethylguanidine (4b) pro-
vided a lower (9% ee) enantioselectivity (Entries 3 and 4)
than DBU (4a). In the case of the base 4c, in which the
seven-membered ring of DBU is replaced by methyl groups,
an enantioselectivity (38% ee) similar to that obtained with
DBU (4a) was observed (Entries 3 and 5). With the nonco-
ordinating Schwesinger base 4d,[37,38] the reaction pro-
ceeded with a very low ee (4%) to afford the (αS,βR)-epox-
ide 3a; that is, the opposite enantiomer to that formed with
DBU (Entries 3 and 6). A similar effect was observed when
LiCl was added to the DBU-mediated epoxidation, from
which the (αS,βR)-epoxide 3a was obtained with the low ee
value of 6% (Entry 7).
Since DBU (4a) in toluene had proven to be the amine/
solvent combination of choice in the asymmetric epoxid-
With weaker organic bases (pKa Ն 11) such as triethyla- ation with optically active hydroperoxides, the ROOH
mine, diisopropylamine, pyridine, DMAP, and Tröger’s structure was varied for the enones 1a (Entries 3 and 8Ϫ10)
base, no reaction was observed even after 72 h at ca. 20 °C and 1b (Entries 11Ϫ14) as substrates. In general, the hydro-
(data not shown in Table 3). The stronger bases 4bϪd peroxide 2a was slightly less enantioselective than 2b and
yielded the desired epoxide 3a with varying enantioselectivi- 2c. For enone 1a, the 4-chlorophenyl-substituted hydroper-
ties (Entries 4Ϫ6).
oxide 2b afforded the (αR,βS)-epoxide with a marginally
Table 3. Enantioselective WeitzϪScheffer epoxidation of enones 1 with the optically active hydroperoxides 2, base-catalyzed by the
amines 4[a]
Entry
Enone
R*OOH
Conversion (%)
Yield (%)[b]
ee (%)[c]
Configuration[d]
1[e]
2[f]
3
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1b
1b
1b
1c
1f
2a
2a
2a
2a
2a
2a
2a
2b
2c
2d
2a
2b
2c
2d
2a
2a
2a
2a
2a
2a
2b
88
93
90
94
85
Ͼ 95
Ͼ 99
93
78
75
81
78
76
70
85
88
84
42
95
95
98
96
94
93
99
96
86
92
98
96
97
91
93
97
94
95
93
86
88
9
4
40
9
38
4
6
44
46
43
25
34
39
31
41
37
36
7
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αS,βR)
(αS,βR)
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αR,βS)
(αS,βR)
(αS,βR)
(αR,βS)
(αR,βS)
4[g]
5[h]
6[i]
7[j]
8
9
10
11
12
13
14
15
16
17
18
19
20
21
1g
1i
1j
35
51
44
13
54
72
1l
1l
[a]
Enone 1 (0.1Ϫ0.5 mmol), hydroperoxide 2 (1.0 equiv.), DBU (1.1Ϫ1.2 equiv.), except for Entries 4Ϫ6, in toluene, except for Entries 1
[b]
[c]
and 2, at ca. 20 °C.
propanol/hexane (5:95) as eluent; error Յ 3% of the stated values.
used as solvent. [f] MeOH was used as solvent. [g] Base 4b was used. [h] Base 4c was used. [i] Base 4d was used. [j] LiCl was used as additive.
Isolated epoxide 3 based on conversion of the enone 1.
HPLC analysis on a Chiracel OD column with 2-
[d]
[e]
Configuration of the major enantiomer is given. CH3CN was
Eur. J. Org. Chem. 2002, 630Ϫ639
633

W. Adam, P. Bheema Rao, H.-G. Degen, C. R. Saha-Möller
FULL PAPER
higher ee value (44%) than observed for 2a (Entries 3 and of α,β-enones 1 with the optically active hydroperoxides 2
8). Similar enantioselectivities were obtained for the β- (Scheme 2) decisively depends on the nature of the catalyz-
naphthyl-substituted hydroperoxide 2c (46% ee, Entry 9) ing base. Two types of base exist: On the one hand there
and the sugar-derived 2d (43% ee, Entry 10). For the β- are the alkali metal hydroxides (Tables 1 and 2), the inor-
methyl enone 1b, a more pronounced difference in the ee ganic KOH the most effective, and on the other hand the
values was noted. Thus, while hydroperoxide 2a displayed organic tertiary amine bases such as DBU (Table 3). Even
only 25% ee (Entry 11) for the (αR,βS)-epoxide, the ee value though KOH and DBU induce opposite senses in the en-
increased to 34% with the para-chloro derivative 2b antioselectivity with the same optically active (Ϫ)-(S)-hy-
(Entry 12). The β-naphthyl-substituted hydroperoxide 2c droperoxide 2, a common mechanistic explanation applies
further improved the enantioselectivity to 39% ee for the enantiofacial differentiation (Figure 1), as attested
(Entry 13), while the sugar-derived hydroperoxide 2d af- to by the similar behavior of the two types of bases
forded the epoxide 3b in only 31% ee (Entry 14). Although (Tables 1Ϫ3).
the hydroperoxide 2c gave the highest enantioselectivity, the
degrees of conversion were higher with the simple phenyl-
substituted hydroperoxide: for ROOH 2a, they were 90%
and 81% (Entries 3 and 11), while with ROOH 2c they were
only 78 and 76% (Entries 9 and 13) for the enones 1a and
1b, respectively. Hence, the more reactive hydroperoxide 2a
was used to assess the influence of the enone substitution
pattern on the enantioselectivity.
As in the KOH-mediated epoxidation, the aryl substitu-
ents on the enone were varied first. Electron-donating sub-
stituents did not significantly change the enantioselectivity.
For example, the para-methyl-substituted aryl group, both droperoxides, where Tϩ (either the Kϩ or R3NHϩ ion) represents
Figure 1. Template structures TS-A [(Si) face] and TS-B [(Re) face]
for the base-catalyzed asymmetric oxygen transfer in the
WeitzϪScheffer epoxidation of α,β-enones by optically active hy-
the templating agent; to facilitate the discussion, the (Si) face and
at the carbonyl functionality (substrate 1c) as well as at the
(Re) face descriptors for substrate 1a have been adopted for all
β-position of the enone (substrate 1f), resulted in the corres-
ponding epoxides with enantioselectivities close to those
obtained for the phenyl derivative 1a (Entry 3); the ee
others; actually the priorities differ only for substrate 1b
The important feature pertinent for enantiocontrol in the
values thus were 41% for the substrate 1c (Entry 15) and structures TS-A and TS-B is the template-type aggregation
37% ee for substrate 1f (Entry 16), the corresponding of the enone substrate 1 and the activated peroxide ion 2Ϫ
(αR,βS)-epoxide enantiomer being observed in each case. by the templating agent Tϩ, either the potassium ion (Kϩ)
Substrate 1g, with a para-methoxy-substituted aryl group for KOH, or the ammonium ion (R3NHϩ) for the DBU.
at the β-position, also afforded a similar ee value of 36% The tightness of aggregation in the template dictates the
(Entry 17). The large tert-butyl substituent in the β-position efficacy of stereodifferentiation through the steric interac-
of enone 1i resulted in a very low enantioselectivity of 7% tions between the three components, namely the prochiral
ee (Entry 18), with the (αS,βR) enantiomer slightly favored. enone substrate 1, the chiral hydroperoxide 2, and the tem-
Significant for this substrate is the fact that the opposite plating cations Kϩ or DBUHϩ. In particular, the aryl group
sense of enantioselectivity was obtained, since all other en- is the largest substituent on the stereogenic center of the
ones yielded the (αR,βS)-epoxide under these conditions optically active hydroperoxide anions 2Ϫ and thus points
(Entries 3, 11, 15Ϫ17). Enone 1j also bears a β-tert-butyl away from the enone substrate 1, exposing the smaller
group, but has a fixed s-cis conformation. Like the acyclic methyl group and the hydrogen atom for steric interaction
analog 1i, the enone 1j was epoxidized with a low ee value with the enone substrate and the templating agent (Tϩ).
of only 13% ee (Entry 19) in favor of the (αS,βR) enanti- Provided that the templating agent does not display any
omer. The poor enantioselectivity was matched by the slug- competitive steric demands (as is the case for DBU), of the
gish reactivity of the s-trans-fixed enone 1k, which was con- two template structures TS-A and TS-B, the latter is of
verted to an extent of only 30% after 125 h at 20 °C (in 5% higher energy. TS-B is therefore disfavored, due to the more
ee; data not shown in Table 3).
pronounced steric repulsion between the methyl group of
The terminally disubstituted enone 1l yielded the the chiral hydroperoxide ion and the β-substituent of the
(αR,βS)-epoxide 3l with an ee value of 54% for the hydro- prochiral enone substrate. To substantiate this mechanistic
peroxide 2a (Entry 20) and 72% ee for 2b (Entry 21). These supposition, we shall recapitulate the salient experimental
represent the highest enantioselectivities for this series of features of this asymmetric WeitzϪScheffer epoxidation im-
DBU-mediated asymmetric WeitzϪScheffer epoxidations.
plicit in Tables 1Ϫ3 (Scheme 2).
As expected from the template structures TS-A and TS-
B (Figure 1), the enantioselectivity for both types of bases
depends on the substitution in the enone substrate 1. For
example, the steric size of the β-substituent significantly in-
Discussion
The results in Tables 1Ϫ3 show that the extent and sense fluences the extent of asymmetric induction: a range of
of asymmetric induction in the WeitzϪScheffer epoxidation about 31% in the ee values results from the variation of the
634
Eur. J. Org. Chem. 2002, 630Ϫ639

Asymmetric WeitzϪScheffer Epoxidation of α,β-Enones by Optically Active Hydroperoxides
FULL PAPER
β-R substituent in the case of KOH (Table 2, Entries 1, 6 face attack Ϫ is clearly less serious than that with the hydro-
and 16), and even as much as 47% in the case of DBU peroxide methyl group in TS-1i(Re) of the (Re)-face attack.
(Table 3, Entries 3, 11 and 18) for the enones 1a (R ϭ Ph), Thus, in the KOH-catalyzed reaction, the sterically less
1b (R ϭ Me), and 1i (R ϭ tBu). In contrast, electronic hindered (Si)-face attack is favored and provides the
factors have a negligible influence on the enantioselectivity; (αS,βR)-epoxides 3, as confirmed experimentally (Table 2,
this was tested by varying the β-aryl substituents (KOH: Entry 16).
Table 2, Entries 13Ϫ15; DBU: Table 3, Entries 3, 16 and
17).
Furthermore, the fact that the s-cis conformation of the
enone functionality is essential for effective enantiofacial
discrimination for both the DBU- and the KOH-mediated
reactions is indicative of the template structures in Figure 1.
Thus, the s-cis-fixed enone 1j (KOH: Table 2, Entry 18;
DBU: Table 3, Entry 19) gives rise to a higher enantioselec-
tivity than the corresponding open-chain substrate 1i
(KOH: Table 2, Entry 16; DBU: Table 3, Entry 18), whereas
the s-trans-fixed substrate 1k is scarcely (and, moreover, un-
selectively) converted (data not shown in Tables 2 and 3).
In order to achieve better enantiofacial differentiation,
the catalyzing base must be available for coordination with
the reactants, as depicted in structure TS-A and TS-B. If,
Scheme 3. Potassium-coordinated template for the enantioselective
for example, the potassium ion is sequestered by the 18-
crown-6 ether (Table 2, Entry 5), or the noncoordinating
Schwesinger base 4d is used (Table 3, Entry 6), or LiCl is
added to compete with DBU for coordination (Table 3,
Entry 7), the enantioselectivity is significantly lower than
epoxidation of the tert-butyl-substituted enone 1i with the optically
active hydroperoxide 2a, catalyzed by KOH
To relieve the entropic constraint of the required s-cis
under the normal reaction conditions (Table 2, Entry 1, conformation in the template (Scheme 3), the enone 1j, the
Table 3, Entry 3). Thus, the nature of the base (KOH or rigid geometry of which obligates this conformation, was
DBU) and its coordinating ability are crucial in the control chosen as substrate. This indeed resulted in the highest en-
of the enantioselectivity.
antioselectivity yet observed with this oxidation system,
All these experimental data form the basis for the pro- 90% ee in favor of the (αS,βR)-epoxide 3j. This fact nicely
posed template structures TS-A and TS-B (Figure 1), which corroborates the validity of the proposed template effect in
govern the efficacy of the asymmetric WeitzϪScheffer epox- Figure 1. In contrast, for the conformationally fixed s-trans
idation with optically active hydroperoxides. In the (Si)-face enone 1k, the enantiodifferentiating coordination of the en-
(TS-A) and the (Re)-face (TS-B) attacks, the (Ϫ)-(S)-perox- one substrate and the optically active hydroperoxide anion
ide ion 2Ϫ approaches the s-cis conformationally aligned by the templating Kϩ ion in the proposed way (Scheme 3)
enone, directed by the base-derived templating agent (Tϩ is is difficult, since the carbonyl oxygen atom and the β-car-
either Kϩ or R3NHϩ). The simultaneous coordination of bon atom of the enone substrate are too far apart for effect-
the enone substrate and the peroxide anion induces the de- ive coordination with the Kϩ ion and the peroxide anion.
cisive steric interactions between the three aggregated spe- Consequently, the substrate 1k is epoxidized unselectively
cies, which determines the extent and sense of the asymmet- (ca. 0% ee). Moreover, the s-trans-fixed substrate is compar-
ric oxygen transfer. The role of this template in the enanti- atively unreactive, which indicates that the s-cis conforma-
ofacial differentiation is now illustrated for the individual tion of the enone in the template is not only a prerequisite
cases.
for effective enantiofacial discrimination, but also beneficial
When KOH is used as base in this asymmetric for the reactivity.
WeitzϪScheffer epoxidation, the increasing size of the β-
The vital importance of the templating potassium ion for
substituent [that is, the order methyl (Table 2, Entry 6), enantiofacial differentiation, as presumed in Scheme 3, is
phenyl (Entry 1), and tert-butyl (Entry 16)] raises the en- unequivocally demonstrated when 18-crown-6 is employed
antioselectivity from 44 to 75% ee in favor of the (αS,βR)- as a sequestering agent. The drop in enantioselectivity, both
epoxide enantiomer. This shows the fundamental import- for enone 1a (Table 2, Entries 1 and 5) and for enone 1i
ance of the steric interaction between the β-R enone sub- (Table 2, Entries 16 and 17), observed on addition of the
stituent and the peroxide ion 2Ϫ. In Scheme 3, this is illus- potassium-complexing crown ether, substantiates the pro-
trated for the tert-butyl-substituted enone 1i, hitherto the posed mechanism in Scheme 3. Thus, the potassium ion co-
most selective derivative and thus the best suited to illus- ordinates the prochiral enone 1i and the chiral peroxide an-
trate the steric interactions that control the enantioselectiv- ion 2aϪ to afford the two diastereomorphic transition struc-
ity. In the transition structures TS-1i(Si) and TS-1i(Re), the tures TS-1i(Si) and TS-1i(Re), of which the former is pre-
steric repulsion between the tert-butyl group and the hydro- ferred and the epoxide (αS,βR)-3i prevails (Scheme 3;
gen atom of the hydroperoxide in TS-1i(Si) Ϫ i.e., the (Si)- Table 2, Entry 16).
Eur. J. Org. Chem. 2002, 630Ϫ639
635

W. Adam, P. Bheema Rao, H.-G. Degen, C. R. Saha-Möller
FULL PAPER
The use of the amine base DBU (4a) in CH3CN at 20 °C in the KOH-mediated case (Table 2) imply that the hydro-
(Table 3, Entry 1) in place of KOH (Table 2, Entry 1), gen-bonded template (Scheme 4) for the DBUHϩ ion is a
shows two additional revealing aspects of the asymmetric looser aggregate than that produced by the oxygen atom
WeitzϪScheffer reaction with optically active hydroperox- ligating Kϩ ion (Scheme 3), and the steric interactions for
ides 2: On one hand, the extent of the enantioselectivity is the enone/R*OOH/DBU system are thus weaker.
much lower; on the other hand, while KOH selects the
Be that as it may, the opposite senses of enantioselectivity
(αS,βR)-configured epoxide 3a, DBU selects the opposite found for the DBU- and the KOH-mediated asymmetric
(αR,βS) enantiomer. Clearly, this indicates a change in the WeitzϪScheffer epoxidations still need to be accounted for.
oxygen-transfer process, since none of the factors that have In this context, the size of the organic base (Table 3,
been addressed so far would explain the preference for the Entries 3Ϫ5) appears to be important. The substantially
(Re)-face attack observed with DBU. Thus, DBU and KOH lower (9% ee) enantioselectivity observed with the smaller
differ significantly in their enantiofacial control; that is, tetramethylguanidine (4b) in Table 3 (Entry 4) and the sim-
DBU facilitates the (Re)-face attack [(αR,βS)-epoxide] and ilar ee values (ca. 40%) for the larger bases 4c (Entry 5)
KOH the (Si)-face attack [(αS,βR)-epoxide]. This dicho- and DBU (Entry 3) indicate that the large ammonium ions
tomy demands a more detailed scrutiny of the interactions derived from the amine bases 4a and 4c favor the (Re)-face
between the substrate, the ammonium ion, and the peroxide attack of the peroxide anion on the enone substrate in the
anion in the template structures.
hydrogen-bonded template 1b(Si)϶ (Scheme 4). Indeed, the
The DBU results in Table 3 suggest that a nonpolar, ap- template structures reveal that the protonated amine inter-
rotic solvent such as toluene (44% ee, Entry 3) promotes acts sterically with the substituents at the chirality center of
(αR,βS) selectivity, compared to that seen with the polar the peroxide anion in competition with the steric repulsions
acetonitrile (9% ee, Entry 1), while the protic methanol of the β-substituent of the enone. Hence, the steric effects
(4% ee, Entry 2) reduces it even further. The addition of of the DBUHϩ ion counteract those of the β-enone sub-
LiCl to the DBU-mediated reaction in toluene counteracts stituent, the net stereochemical outcome being a balance
the effect of the amine base on the enantioselectivity between these two opposing trends. In particular, domin-
(Table 3, Entry 3), and provides the opposite (αS,βR)-epox- ance of the steric interaction by the large DBU favors (Re)-
ide enantiomer in only 6% ee (Table 3, Entry 7). A similar face attack and affords an excess of the (αR,βS)-epoxide,
low enantioselectivity (4% ee) of opposite configuration while dominance by the β-substituent on the enone double
(αS,βR) is observed for the noncoordinating (no hydrogen bond, as in the case of the β-tert-butyl-substituted enone
bonding) Schwesinger base 4d (Table 3, Entry 6). These ex- 1i, favors the (Si)-face attack and the (αS,βR) enantiomer
periments implicate hydrogen bonding as the mode of the prevails. Thus, the observed (αR,βS) selectivity for the ma-
aggregation, and we therefore propose the protonated jority of the substrates in Table 3 indicates that the steric
amine (DBUHϩ) as the hydrogen bond donor. Hydrogen interactions between the DBUHϩ ion and the small hydro-
bonding is known to be obstructed by polar (CH3CN) or gen atom at the stereogenic center of the peroxide anion are
protic solvents (MeOH),[39] and the Schwesinger base 4d is minimized in the template structure 1(Re)϶. In comparison,
incapable of hydrogen bonding, facts that explain the poor for the higher energy 1(Si)϶ alternative, the more obstruct-
enantioselectivity. Consequently, in the template structures ive methyl substituent at the chirality center of the peroxide
of Figure 1, the protonated amine DBUHϩ acts as the tem- anion interacts with the DBUHϩ ion. This case is illus-
plating agent and coordinates the peroxide anion and the trated in Scheme 4 for the enone 1b with a β-methyl sub-
enone substrate in a template (Scheme 4), analogously to stituent, for which the (αR,βS)-epoxide 3b is favored.
the Kϩ ion in the KOH-mediated reaction (Scheme 3).
Another illustrative case is substrate 1i, with the large
Evidently, the significantly lower ee values for the DBU- tert-butyl substituent (Scheme 4), epoxidation of which re-
catalyzed epoxidation (Table 3) compared to those observed sults predominantly in the (αS,βR) enantiomer of the epox-
Scheme 4. Hydrogen-bonded template for the enantioselective epoxidation of the enones 1b and 1i with the optically active hydroperoxide
2a, base-catalyzed by DBU
636
Eur. J. Org. Chem. 2002, 630Ϫ639

Asymmetric WeitzϪScheffer Epoxidation of α,β-Enones by Optically Active Hydroperoxides
FULL PAPER
ide 3i. Thus, the large tert-butyl substituent in this case ov- Conclusion
ercompensates for the effect of the DBU base. The larger
In summary, a template effect has been proposed in order
to interpret the extent and sense of the enantioselectivity in
the base-catalyzed, asymmetric WeitzϪScheffer epoxidation
of α,β-enone substrates with optically active hydroperox-
ides. In the template structure, made up of the enone sub-
strate, hydroperoxide, and base catalyst, the enantiofacial
differentiation is governed by the steric interactions be-
tween the aggregated species. When KOH is used as base,
the (αS,βR) enantiomer of the epoxide 3 enantiomer is
formed in up to 90% ee, due to the less severe steric repul-
sions between the β-substituent of the enone and the sub-
stituent at the chirality center of the peroxide anion during
the (Si)-face attack. In the DBU-mediated epoxidation,
competition between the steric interactions of the optically
active peroxide anion with the enone β-substituent and the
DBUHϩ (usually dominant) favor the (Re)-face attack,
which results in the (αR,βS) enantiomer of the epoxide 3 in
up to 72% ee. The asymmetric WeitzϪScheffer epoxidation
of the α,β-enones 1 with the optically active hydroperoxides
2, catalyzed by KOH or DBU, affords both enantiomers of
the epoxides 3 with ee values that range up to good, which
should be valuable for asymmetric organic synthesis.
steric repulsion occurs between the β-tert-butyl group and
the methyl substituent of the stereogenic center in the per-
oxide anion during (Re)-face attack, represented by the
template structure 1i(Re)϶. Thus, the (Si)-face approach is
favored, since the smaller hydrogen atom is encountered by
the enone tert-butyl group in the alternative template struc-
ture 1i(Si)϶. The competing steric effects between DBU and
the enone β-substituent not only explain the opposite senses
of the enantioselectivities for the various substrates in
Table 3, but also justify the lower ee values of the DBU-
mediated asymmetric WeitzϪScheffer epoxidation.
As in the KOH-mediated epoxidations, the validity of the
s-cis conformation depicted in the template structures is
corroborated by the conformationally fixed substrates 1j
and 1k. The s-cis-configured enone 1j facilitates the forma-
tion of the enantiodifferentiating template and thus results
in a higher enantioselectivity (Table 3, Entry 19) than seen
with the corresponding conformationally flexible enone 1i
(Entry 18). In contrast, the s-trans-fixed enone 1k is again
less reactive and unselective (5% ee) in its epoxidation under
the DBU-mediated reaction conditions (data not listed in
Table 3).
In the epoxidation of the terminally disubstituted enone
1l, the steric effect of the β-cis substituent in the enone and
the steric demand of the DBU combine to yield the highest
(up to 72% ee) enantioselectivities so far achieved with the
DBU base (Table 3, Entries 20 and 21). Analysis of the
steric congestion in the template structures 1l(Re) and 1l(Si)
clarifies this significant improvement in the extent of asym-
metric induction.
Experimental Section
General Remarks: 1H and 13C NMR spectra were recorded with
Bruker AC 200 (1H: 200 MHz, 13C: 50 MHz) or Bruker AC 250
(1H: 250 MHz, 13C: 63 MHz) spectrometers, IR spectra were meas-
ured with a PerkinϪElmer 1600 FT-IR spectrophotometer. The
HPLC analyses were carried out on chiral columns (Daicel
CHIRALCEL OD and OB-H) with a Kontron instrument, fur-
nished with a UVIKON 720 spectrophotometer and an IBZ Masse-
technik (Hannover, Germany) CHIRALYSER 1.6. The optical ro-
tations were determined with a PerkinϪElmer 241 MC polarimeter.
Solvents and commercially available chemicals were purified by
standard procedures. Enone 1a is commercially available (Riedel-
de Hae¨n). Enones 1bϪl were prepared according to literature pro-
cedures (1b,[8] 1c,[40] 1d,[41] 1e,[42] 1f,[43] 1g,[44] 1h,[45] 1i,[46] 1k,[47]
1l[48]).
(E)-2-(2Ј,2Ј-Dimethylpropylidene)-1,2,3,4-tetrahydronaphthalene-1-
one (1j): NaOH solution (10%, 1.20 mL) was added to a mixture
of tetralone (1.46 g, 10.0 mmol) and pivalaldehyde (1.20 g,
11.0 mmol) in 6.5 mL of ethanol (77% in water) and the mixture
was stirred for 48 h at 25 °C. The ethanol was removed in a rotary
evaporator (25 °C, 20 Torr) and the residue was diluted with 10 mL
of water and extracted with ether (3 ϫ 15 mL). The combined or-
ganic phases were washed with water (5 mL) and brine (5 mL) and
dried with MgSO4, and subsequent silica gel chromatography
(Et2O/petroleum ether, 1:9) yielded 1.21 g (56%) of a colorless oil.
1H NMR (200 MHz. CDCl3): δ ϭ 1.25 (s, 9 H, CH3), 2.95 (s, 4 H,
CH2), 6.97 (s, 1 H, CH), 7.21Ϫ7.36 (m, 2 H, aromatic), 7.42Ϫ7.50
(m, 1 H, aromatic), 8.06Ϫ8.11 (m, 1 H, aromatic). 13C NMR
(50 MHz. CDCl3): δ ϭ 26.3, 27.8, 30.2, 32.6, 127.1, 128.4, 128.6,
133.3, 133.8, 134.9, 144.1, 149.8, 189.3. C15H18O (214.14): calcd. C
84.07, H 8.47; found C 83.68, H 8.56.
The (Si)-face attack brings the hydrogen atom at the ste-
reogenic center of the hydroperoxide ion into close proxim-
ity to the phenyl group on the enone double bond, while
the methyl group of the peroxide anion is positioned prox-
imate to the DBUHϩ ion and also to the methyl group of
the enone double bond in the DBUHϩ-coordinated tem-
plate 1l(Si)϶. In contrast, for the (Re)-face attack, the
methyl group of the peroxide anion is positioned proximate
to the phenyl group on the enone double bond and the hy-
drogen atom toward the DBUHϩ ion and also toward the
methyl group of the enone double bond. The latter steric
repulsion of the hydrogen atom of the peroxide anion with
DBUHϩ and the enone methyl group is evidently less ser-
ious and the (αR,βS) enantiomer of the epoxide 3l is thus
preferentially generated by a substantial margin (up to
72% ee).
Enzymatic Kinetic Resolution of 1-(2-Naphthyl)ethyl Hydroperoxide
(2c):[49] Horseradish peroxidase (1.00 mg) was added to a stirred
Eur. J. Org. Chem. 2002, 630Ϫ639
637

W. Adam, P. Bheema Rao, H.-G. Degen, C. R. Saha-Möller
FULL PAPER
mixture of the racemic hydroperoxide 2c (47.0 mg, 0.250 mmol) CH3), 2.38Ϫ2.62 (m, 2 H, CH2), 2.97 (s, 1 H, CH), 3.12Ϫ3.19 (m,
and guaiacol (12.0 mg, 0.100 mmol) in a 0.1  phosphate buffer 2 H, CH2), 7.26Ϫ7.38 (m, 2 H, aromatic), 7.48Ϫ7.56 (m, 1 H, aro-
(pH ϭ 6, 2.5 mL). The reaction was monitored by HPLC, which matic), 8.03Ϫ8.08 (m, 1 H, aromatic). 13C NMR (50 MHz. CDCl3):
showed 99% ee of the hydroperoxide after 1 h. The reaction mixture δ ϭ 26.2, 27.9, 28.4, 32.3, 63.8, 72.4, 126.7, 127.7, 128.6, 132.6,
was filtered through Celite and extracted with ether (3 ϫ 15 mL), 133.9, 143.5, 194.7. [α]2D0 ϭ ϩ 68.4 (c ϭ 1.0, CHCl3, for 100% ee).
and the combined organic layers were dried (MgSO4), concentrated
(20 °C, 30 Torr), and purified by silica gel chromatography (petro-
leum ether/Et2O, 8:1) to yield 17.9 mg (76%) of the (Ϫ)-(S)-hydro-
peroxide 2c (Ͼ 99% ee) and 17.8 mg (82%) of the corresponding
C15H18O2 (230.13): calcd.: C 78.23, H 7.88; found C 78.04, H 7.93.
7,7a-Dihydro-7,7-dimethylnaphth[2,3-b]oxiren-2(1aH)-one (3k): The
epoxidation of the enone 1k according to the general Procedures
A and B resulted in only 30% conversion and very low (5%) ee
values. For this reason, the racemic epoxide was prepared as de-
scribed below.[54] A solution of DMD in acetone (0.89 , 4.90 mL,
440 mmol) at 0 °C was added to a magnetically stirred solution of
4,4-dimethyl-4H-naphthalene-1-one (44.0 mg, 440 mmol) in dichlo-
romethane (3 mL). After this had stirred for 12 h at ca. 20 °C,
another batch of the DMD solution (440 mmol) was added and
stirred until complete conversion of the enone, as determined by
TLC. The reaction mixture was dried (MgSO4), and the solvent was
evaporated (20 °C, 20 mbar). Silica gel chromatography (petroleum
ether/ether, 95:5) afforded 44.4 mg (87%) of the racemic epoxide 3k
as white needles, m.p. 68Ϫ69 °C. 1H NMR (250 MHz, CDCl3): δ ϭ
1.29 (s, 3 H, CH3), 1.69 (s, 3 H, CH3), 3.56 (d, J ϭ 4.3 Hz, 1 H,
CH), 3.72 (d, J ϭ 4.3 Hz, 1 H, CH), 7.32Ϫ7.40 (m, 2 H, aromatic),
7.55Ϫ7.62 (m, 1 H, aromatic), 7.88Ϫ7.92 (m, 1 H, aromatic).
13C NMR (63 MHz, CDCl3): δ ϭ 25.2, 30.1, 36.0, 55.4, 62.5, 126.0,
127.0 127.5, 128.8, 134.2, 147.1, 194.7. C12H12O2 (188.08): calcd.:
C 76.57, H 6.43; found C 76.11, H 6.85.
1
alcohol (96% ee). H NMR (200 MHz, CDCl3): δ ϭ 1.56 (d, J ϭ
6.6 Hz, 3 H, CH3), 5.26 (q, J ϭ 6.6 Hz, 1 H, CH), 7.48Ϫ7.54 (m,
3 H, aromatic), 7.83Ϫ7.91 (m, 4 H, aromatic). 13C NMR (50 MHz,
CDCl3): δ ϭ 14.9, 78.8, 118.8, 120.7, 121.0, 121.1, 122.6, 122.8,
123.5, 128.1, 128.1, 133.6. [α]2D0 ϭ Ϫ91.5 (c ϭ 1.0, CHCl3, Ͼ
99% ee). HPLC: (Ϫ) 16.64, (ϩ) 22.28 min (OD column; iPrOH/
hexane, 9:1; flow 0.9 mL/min).
Epoxides 3
Representative Procedure A for the Epoxidation of Enone 1a by the
Optically Active Hydroperoxides 2a with KOH as a Base: A pre-
cooled (Ϫ40 °C) solution of enone 1a (104 mg, 0.500 mmol)
and
the
optically
active
hydroperoxide
(Ϫ)-(S)-2a
(69.0 mg, 0.500 mmol) in dry CH3CN (3 mL) was added slowly to a
suspension of powdered KOH (56.0 mg, 1.00 mmol) in dry CH3CN
(2 mL) at Ϫ40 °C under nitrogen. The reaction mixture was stirred
for 20Ϫ30 min at Ϫ40 °C, decanted into ice/water (10 mL), and
extracted with ether (3 ϫ 10 mL). The combined organic extracts
were dried (MgSO4), and the solvent was evaporated (20 °C,
30 Torr). Purification by silica gel flash chromatography, with a
petroleum ether/ethyl ether (15:1) mixture as eluent, afforded
110 mg (99%) of epoxy ketone (αS,βR)-3a (51% ee) as colorless
needles (m.p. 62Ϫ63°C). The spectroscopic data were in accordance
with those reported.[50] The optically active epoxides 3bϪj were
prepared from the corresponding enones 1bϪj according to the
above procedure, and the results are given in Table 2.
Acknowledgments
We are grateful to the DFG (SFB-347 ‘‘Selektive Reaktionen
Metall-aktivierter Moleküle) and the Fonds der Chemischen Indus-
trie (doctoral fellowship 1998Ϫ2000 for H.-G. D.) for generous fin-
ancial support. P. B. R. thanks the Alexander-von-Humboldt
Foundation for a postdoctoral fellowship (1999Ϫ2000). We also
thank Roche Diagnostics GmbH for the generous gift of enzymes.
Representative Procedure B for the Epoxidation of the Enone 1a by
the Optically Active Hydroperoxide 2a with DBU as a Base: DBU
(91.0 mg, 0.600 mmol) was added under nitrogen to a solution of
enone 1a (104 mg, 0.500 mmol) and the optically active hydroper-
oxide (Ϫ)-(S)-2a (69.0 mg, 0.500 mmol) in dry toluene (4 mL). The
reaction mixture was stirred magnetically for 24 h at ca. 20°C,
poured into water (8 mL), and extracted with ether (3 ϫ 8 mL).
The combined organic layers were dried (MgSO4) and the solvent
was evaporated (20 °C, 30 Torr). Purification by silica gel flash
chromatography, with a petroleum ether/ethyl ether (15:1) mixture
as eluent, afforded 98.0 mg (88%) of the known epoxy ketone
(αS,βR)-3a (40% ee). The spectroscopic data were in accordance
with those reported.[50] The enones 1b, 1c, 1f, 1g, 1i, 1j, and 1l were
also epoxidized according to this procedure and the results are
given in Table 3. The epoxides 3aϪi and 3l are known; their spec-
troscopic data were identical to those described in the literature.
The configuration of the major enantiomer was in each case deter-
mined by comparison with the literature data (3a,[50] 3b,[51] 3c,[6]
3d,[14] 3e,[8] 3f,[52] 3g,[8] 3h,[12] 3i,[6] 3j,[36] 3k,[9] 3l[53]). The configura-
tion of 3j was tentatively assigned by comparison of the HPLC
retention times and the optical rotation of the reported isopropyl
derivative.[22] Because of the very low enantiomeric excess (5% ee)
observed for 3k, the absolute configuration was not determined.
[1]
R. Helder, J. C. Hummelen, R. W. P. M. Laane, J. S. Wiering,
H. Wynberg, Tetrahedron Lett. 1976, 17, 1831Ϫ1834.
H. Wynberg, B. Greijdanus, J. Chem. Soc., Chem. Commun.
[2]
1978, 427Ϫ428.
[3]
H. Wynberg, B. Marsman, J. Org. Chem. 1980, 45, 158Ϫ161.
[4]
H. Plium, H. Wynberg, J. Org. Chem. 1980, 45, 2498Ϫ2502.
[5]
M. J. Porter, J. Skidmore, Chem. Commun. 2000, 1215Ϫ1225.
[6]
S. Arai, H. Tsuge, T. Shioiri, Tetrahedron Lett. 1998, 39,
7563Ϫ7566.
[7]
B. Lygo, P. G. Wainwright, Tetrahedron 1999, 55, 6289Ϫ6300.
G. MacDonald, L. Alcaraz, N. J. Lewis, R. J. K. Taylor, Tetra-
hedron Lett. 1998, 39, 5433Ϫ5436.
W. Adam, P. Bheema Rao, H.-G. Degen, C. R. Saha-Möller,
[8]
[9]
Tetrahedron: Asymmetry 2001, 12, 121Ϫ125.
[10]
E. J. Corey, F.-Y. Zhang, Org. Lett. 1999, 1, 1287Ϫ1290.
[11]
´
S. Julia, J. Masana, J. C. Vega, Angew. Chem. Int. Ed. Engl.
1980, 19, 929Ϫ931.
[12]
[13]
S. Julia, J. Guixer, J. Masana, J. Rocas, S. Colonna, R. Annuzi-
ata, H. Molinari, J. Chem. Soc., Perkin Trans. 1 1982,
1317Ϫ1324.
W. Kroutil, M. E. Lasterra-Sanchez, S. J. Maddrell, P. Mayon,
P. Morgan, S. M. Roberts, S. R. Thornton, C. J. Todd, M.
Tüter, J. Chem. Soc., Perkin Trans. 1 1996, 2837Ϫ2844.
S. Itsuno, M. Sakakura, K. Ito, J. Org. Chem. 1990, 55,
6047Ϫ6049.
(2S,3ЈR)-3Ј-tert-Butyl-1,2,3,4-tetrahydrospiro[naphthalene-2,2Ј-
oxirane]-1-one (3j): Epoxide 3j was prepared according to the above
general epoxidation Procedure A in 90% yield as a white powder;
m.p. 62Ϫ63 °C. 1H NMR (200 MHz. CDCl3): δ ϭ 1.10 (s, 9 H,
[14]
[15]
L. Pu, Tetrahedron: Asymmetry 1998, 9, 1457Ϫ1477.
638
Eur. J. Org. Chem. 2002, 630Ϫ639

Asymmetric WeitzϪScheffer Epoxidation of α,β-Enones by Optically Active Hydroperoxides
FULL PAPER
[16]
[17]
[18]
[33]
S. Ebrahim, M. Wills, Tetrahedron: Asymmetry 1997, 8,
3163Ϫ3173.
M. J. Porter, S. M. Roberts, J. Skidmore, Bioorg. Med. Chem.
P. W. Shum, R. J. Saxton, J. G. Zajacek, Asymmetric Epoxid-
ation of Prochiral Alkenes Using a Titanium-Containing Zeolite
and a Chiral Hydroperoxide, US Patent 5663384, 1997; Chem.
Abstr. 1997, 127, 262597.
C. L. Dwyer, C. D. Gill, O. Ichihara, R. J. K. Taylor, Synlett
2000, 704Ϫ706.
M. Aoki, D. Seebach, Helv. Chim. Acta 2001, 84, 187Ϫ207.
W. Adam, P. Bheema Rao, H.-G. Degen, C. R. Saha-Möller, J.
Am. Chem. Soc. 2000, 122, 5654Ϫ5655.
R. Schwesinger, J. Willaredt, H. Schlemper, M. Keller, D.
Schmitt, H. Fritz, Chem. Ber. 1994, 127, 2435Ϫ2454.
O. Legrand, Synlett 1999, 752.
W. Adam, T. Wirth, Acc. Chem. Res. 1999, 32, 703Ϫ710.
S. von Kostanecki, G. Rosback, Ber. Deutsch. Chem. Ges. 1896,
29, 2245Ϫ2247.
E. Bergman, H. A. Wolff, J. Am. Chem. Soc. 1932, 54,
1644Ϫ1647.
P. Dilthy, J. Prakt. Chem. 1921, 101, 177Ϫ212.
V. Hanzlik, A. Bianchi, Ber. Deutsch. Chem. Ges. 1899, 32,
2282Ϫ2287.
W. B. Black, R. E. Lutz, J. Am. Chem. Soc. 1955, 77,
5134Ϫ5140.
W. B. Black, R. E. Lutz, J. Am. Chem. Soc. 1953, 75,
5990Ϫ5997.
K. B. Wiberg, B. I. Rowland, J. Am. Chem. Soc. 1955, 77,
1159Ϫ1163.
E. C. Friedrich, D. B. Taggart, M. A. Saleh, J. Org. Chem.
1977, 42, 1437Ϫ1443.
W. Wayne, H. Adkins, Org. Synth., Coll. Vol. 1955, 3, 367Ϫ369.
Davies, A. G.; R. Field, J. Chem. Soc. 1958, 4637Ϫ4643.
B. Marsman, H. Wynberg, J. Org. Chem. 1979, 44, 2312Ϫ2314.
O. Meth-Cohn, R. M. Horak, G. Fouche, J. Chem. Soc., Perkin
Trans. 1 1994, 1517Ϫ1527.
C. L. Elston, R. F. W. Jackson, S. J. F. MacDonald, P. J. Mur-
ray, Angew. Chem. Int. Ed. Engl. 1997, 36, 410Ϫ412.
C. A. Kingsbury, D. L. Durham, R. Hutton, J. Org. Chem.
1978, 43, 4696Ϫ4700.
1999, 7, 2145Ϫ2156.
[34]
B. M. Adger, J. V. Barkley, S. Bergeron, M. W. Cappi, B. E.
Flowerdew, M. P. Jackson, R. McCague, T. J. Nugent, S. M.
Roberts, J. Chem. Soc., Perkin Trans. 1 1997, 3501Ϫ3507.
J. V. Allen, K.-H. Drauz, R. W. Flood, S. M. Roberts, J. Skid-
more, Tetrahedron Lett. 1999, 40, 5417Ϫ5420.
M. W. Cappi, W.-P. Chen, R. W. Flood, Y.-W. Liao, S. M. Rob-
erts, J. Skidmore, J. A. Smith, N. M. Williamson, J. Chem. Soc.,
Chem. Commun. 1998, 1159Ϫ1160.
T. Genski, G. MacDonald, X. Wei, N. Lewis, R. J. K. Taylor,
Synlett 1999, 795Ϫ797.
D. Enders, J. Zhu, L. Kramps, Liebigs Ann./Recueil 1997,
1101Ϫ1113.
H.-B. Yu, X.-F. Zheng, Z.-M. Lin, Q.-S. Hu, W.-S. Huang, L.
Pu, J. Org. Chem. 1999, 64, 8149Ϫ8155.
[35]
[36]
[19]
[20]
[37]
[38]
[39]
[40]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[41]
[42]
[43]
T. Nemoto, T. Ohshima, K. Yamaguchi, M. Shibasaki, J. Am.
Chem. Soc. 2001, 123, 2725Ϫ2732.
[44]
[45]
[46]
[47]
S. Watanabe, Y. Kobayshi, T. Arai, H. Sasai, M. Bougauchi,
M. Shibasaki, Tetrahedron Lett. 1998, 39, 7353Ϫ7356.
S. Watanabe, T. Arai, H. Sasai, M. Bougauchi, M. Shibasaki,
J. Org. Chem. 1998, 63, 8090Ϫ8091.
W. Adam, U. Hoch, M. Lazarus, C. R. Saha-Möller, P. Schre-
ier, J. Am. Chem. Soc. 1995, 117, 11898Ϫ11901.
W. Adam, R. T. Fell, U. Hoch, C. R. Saha-Möller, P. Schreier,
Tetrahedron: Asymmetry 1995, 6, 1047Ϫ1050.
W. Adam, M. N. Korb, K. J. Roschmann, C. R. Saha-Möller,
J. Org. Chem. 1998, 63, 3423Ϫ3428.
W. Adam, M. N. Korb, Tetrahedron: Asymmetry 1997, 8,
1131Ϫ1142.
H.-J. Hamann, E. Höft, D. Mostowicz, A. Mishnev, Z.
Urbanczyk-Lipkowska, M. Chmielewski, Tetrahedron 1997,
53, 185Ϫ192.
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[32]
H.-J. Hamann, E. Höft, J. Liebscher, ‘‘Preparation of Optically
Active Hydroperoxides and their Use for Stereoselective Oxy-
gen Transfer’’, in Peroxide Chemistry: Mechanistic and Prepar-
ative Aspects of Oxygen Transfer (Ed.: W. Adam), Wiley-VCH,
Weinheim, 2000, pp. 381Ϫ405.
W. Adam, L. Hadjiarapoglou, A. Smerz, Chem. Ber. 1991,
124, 227Ϫ232.
Received September 12, 2001
[O01436]
Eur. J. Org. Chem. 2002, 630Ϫ639
639