Dual mechanisms of acid-catalyzed rearrangement of enol ester epoxides: Enantioselective formation of α-acyloxy ketones

Full text of DOI:10.1021/ja990124f

4080
J. Am. Chem. Soc. 1999, 121, 4080-4081
Dual Mechanisms of Acid-Catalyzed Rearrangement
of Enol Ester Epoxides: Enantioselective Formation
of r-Acyloxy Ketones
Scheme 1
Yuanming Zhu, Karl J. Manske, and Yian Shi*
Department of Chemistry, Colorado State UniVersity
Fort Collins, Colorado 80523
with protic acids such as p-TsOH led to a facile rearrangement
with retention of configuration (Table 1, entry 1). The reaction
was complete within 10 min as monitored by TLC, and the
product showed only a 3% decrease in ee compared to the starting
epoxide. In addition to catalysis by protic acids, we found that
this rearrangement could also be catalyzed by Lewis acids.
Interestingly, the ee of the product varied dramatically with the
Lewis acids. For example, 85% ee was obtained for the product
ReceiVed January 13, 1999
Enol ester epoxides can rearrange to R-acyloxy ketones or
aldehydes under thermal or acidic conditions.1 Investigations of
,2
these rearrangements have largely been carried out on ster-
oids,2
a,b,d,g,h,k,l,n
and the mechanistic conclusions have been based
on stereochemical analysis of the diastereomeric products. Rear-
rangements under thermal conditions proceed via intramolecular
migration of the acyloxy group with inversion of configuration
when Sn(OTf)
ee was obtained when La(OTf)
2
was employed (Table 1, entry 2), but only 15%
was used (Table 1, entry 4).
3
Strikingly, it was found that the S isomer (inVerted product)
actually became the major product with certain Lewis acids (Table
2a,g
at the carbon to which the acyloxy group migrates. In contrast,
most acid-catalyzed rearrangements occur with retention of
1
, entries 6-12). High enantioselectivities (80-91% ee) were
configuration.2
b,h,k,l
While the proposed mechanisms have provided
6
obtained with YbCl
3 3 3
, ErCl , AlMe , and silica gel (Table 1,
reasonable explanations for the observed stereochemistry in these
entries 6, 8, 9, and 12).
3
cases, some uncertainty remains. Recently, we reported the
The results presented in Table 1 indicate that there are two
distinct pathways involved in the acid-catalyzed rearrangement
of enol ester epoxides, leading to two different enantiomers.
Although a full understanding of the factors controlling this
competition has not been attained, the acidity of the catalyst seems
to play an important role. For example, when Yb(OTf) was used
3
as the catalyst, the R enantiomer of the rearranged product was
obtained in 66% ee (Table 1, entry 5). On the other hand, when
a notably weaker Lewis acid YbCl was used, the S enantiomer
3
enantioselective epoxidation of enol esters using a fructose-derived
4
ketone as catalyst. The availability of enantiomerically enriched
enol ester epoxides prompted us to study the factors involved in
the rearrangements of these epoxides under acidic conditions with
the aim of developing a route to enantiomerically enriched
5
R-acyloxy ketones. Herein we wish to report our preliminary
studies in this area.
Our studies started with (R,R)-1-benzoyloxy-1,2-epoxycyclo-
hexane (1) as a test substrate (Scheme 1). Treating epoxide 1
was obtained in 82% ee (Table 1, entry 6). Pathways a and b
outlined in Scheme 2 provide plausible mechanisms for the results.
In pathway a, the complexation of a strong acid to the epoxide
(
1) For leading reviews on enol ester epoxide rearrangements, see: (a)
McDonald, R. N. Mech. Mol. Migr. 1971, 3, 67-107. (b) Riehl, J.-J.; Casara,
P.; Fourgerousse, A. C. R. Acad. Sci. Paris, Ser. C 1974, 279, 79-82.
oxygen of 3 leads to cleavage of the C -O bond to form
1
(
2) For examples of acid-catalyzed and thermal rearrangements of enol ester
epoxides, see: (a) Soloway, A. H.; Considine, W. J.; Fukushima, D. K.;
Gallagher, T. F. J. Am. Chem. Soc. 1954, 76, 2941-2943. (b) Leeds, N. S.;
Fukushima, D. K.; Gallagher, T. F. J. Am. Chem. Soc. 1954, 76, 2943-2948.
carbocation intermediate 5. Subsequent acyl migration with
retention of configuration gives acyloxy ketone 6. In pathway b,
the complexation of a weak acid to 3 weakens both epoxide bonds,
facilitating acyloxy migration with inversion of configuration
(7).7
The discovery of the two different rearrangement pathways
prompted us to investigate more substrates to test the generality.
Of particular interest was the possibility of generating either
enantiomer of an acyloxy ketone in high ee from a single
enantiomerically enriched enol ester epoxide. As shown in Table
2, p-TsOH proved to be an effective catalyst for rearrangement
via pathway a for a variety of epoxides, giving products with
retention of configuration in high stereospecificity. In most cases,
(c) Gardner, P. D. J. Am. Chem. Soc. 1956, 78, 3421-3424. (d) Johnson, W.
S.; Gastambide, B.; Pappo, R. J. Am. Chem. Soc. 1957, 79, 1991-1994. (e)
Shine, H. J.; Hunt, G. E. J. Am. Chem. Soc. 1958, 80, 2434-2435. (f) House,
H. O.; Thompson, H. W. J. Org. Chem. 1961, 26, 3729-3734. (g) Williamson,
K. L.; Johnson, W. S. J. Org. Chem. 1961, 26, 4563-4569. (h) Nambara, T.;
Fishman, J. J. Org. Chem. 1962, 27, 2131-2135. (i) Draper, A. L.; Heilman,
W. J.; Schaeffer, W. E.; Shine, H. J.; Shoolery, J. N. J. Org. Chem. 1962, 27,
,8
2
1
1
1
1
5
727-2729. (j) Riehl, J.-J.; Lehn, J.-M.; Hemmert, F. Bull. Soc. Chim. Fr.
963, 224-227. (k) Rhone, J. R.; Huffman, M. N. Tetrahedron Lett. 1965,
395-1398. (l) Williamson, K. L.; Coburn, J. I.; Herr, M. F. J. Org. Chem.
967, 32, 3934-3937 (m) McDonald, R. N.; Tabor, T. E. J. Am. Chem. Soc.
967, 89, 6573-6578. (n) Smith, S. C.; Heathcock, C. H. J. Org. Chem. 1992,
7, 6379-6380.
(3) For example, it was observed that treating an enol ester epoxide of
epiandrosterone acetate with silicic acid at 50 °C for 17 h led to the formation
of a mixture R and â epimeric 16-acetoxy ketones with a ratio of 1.8:1 in a
(6) For examples of silica gel catalyzed rearrangements, see refs 2a, b,
and d. Both inversion and retention of configuration were reported.
(7) While the acidity of the catalyst is an important factor affecting the
competition of the two pathways, the size and the coordination number of the
Lewis acid could also be important. Full elucidation of these factors awaits
further studies.
16.7% isolated yield (see ref 2d). The R isomer resulted from a rearrangement
with retention of configuration, and the â isomer resulted from a rearrangement
with inversion. On the basis of the observation that the R and â isomers did
not undergo isomerization under the reaction conditions, it was suggested that
these two isomers came from two competing reactions. However, it is not
clear whether these two reactions were induced by acid alone or both acid
and heat since the reaction was carried out at 50 °C.
(8) To test whether the rearrangements proceed intermolecularly or
intramolecularly, crossover experiments were carried out using a mixture of
1-acetoxy-1,2-epoxycyclohexane and 1-benzoyloxy-1,2-epoxycycloheptane as
(
4) Zhu, Y.; Tu, Y.; Yu, H.; Shi, Y. Tetrahedron Lett. 1998, 39, 7819-
822.
5) For leading references on nonenzymatic synthesis of enantiomerically
7
3 3
substrates under acidic conditions (p-TsOH, YbCl , AlMe , silica gel). In the
(
case of p-TsOH, small amounts of crossover products (2-benzoyloxycyclo-
1
enriched hydroxyketones and their derivatives, see: (a) Enders, D.; Bhushan,
V. Tetrahedron Lett. 1988, 29, 2437-2440. (b) Davis, F. A.; Sheppard, A.
C.; Chen, B.-C.; Haque, M. S. J. Am. Chem. Soc. 1990, 112, 6679-6690. (c)
Davis, F. A.; Chen, B.-C. Chem. ReV. 1992, 92, 919-934. (d) Hashiyama,
T.; Morikawa, K.; Sharpless, K. B. J. Org. Chem. 1992, 57, 5067-5068. (e)
Reddy, D. R.; Thornton, E. R. J. Chem. Soc., Chem. Commun. 1992, 172-
hexanone and 2-acetoxycycloheptanone) were detected by GC and H NMR.
The amounts of these products were found to be concentration dependent (ca
6.8% at substrate concentration of 0.64 M and ca. 1.7% at substrate
3 3
concentration of 0.018 M). In the cases of YbCl , AlMe , and Silica gel, the
crossover products were found to be minimal (less than 0.5% at substrate
concentration of 0.64 M and less than 0.2% at substrate concentration of 0.018
M) (for details see Supporting Information). All of these results suggest that
the acid-catalyzed rearrangements proceed predominantly in an intramolecular
fashion, particularly for pathway b. Further experiments with enantiomerically
enriched enol ester epoxides showed that the enantioselectivities of the
rearranged products were not affected by the substrate concentrations.
1
73. (f) D’Accolti, L.; Detomaso, A.; Fusco, C.; Rosa, A.; Curci, R. J. Org.
Chem. 1993, 58, 3600-3601. (g) Chang, S.; Heid, R. M.; Jacobsen, E. N.
Tetrahedron Lett. 1994, 35, 669-672. (h) Fududa, T.; Katsuki, T. Tetrahedron
Lett. 1996, 37, 4389-4392. (i) Adam, W.; Fell, R. T.; Stegmann, V. R.; Saha-
Moller, C. R. J. Am. Chem. Soc. 1998, 120, 708-714.
1
0.1021/ja990124f CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/08/1999

Communications to the Editor
J. Am. Chem. Soc., Vol. 121, No. 16, 1999 4081
Table 2. Examples of p-TsOH-, Silica gel-, YbCl -, or
Table 1. Effects of Different Acid Catalysts on the Rearrangement
of (R,R)-1-Benzoyloxy-1,2-epoxycyclohexane (1)a
3
a
3
AlMe -Catalyzed Rearrangements of Enol Ester Epoxides
entry
acid
t (min)
ee% (1)b
ee% (2)b
yield(%)c
1
2
3
4
5
6
7
8
9
p-TsOH
Sn(OTf)
10
10
1
40
5
90
10
90
5
17
10
720
93
93
92
93
92
93
93
90
91
91
91
92
90 (R)
85 (R)
26 (R)
15 (R)
66 (R)
82 (S)
12 (S)
80 (S)
87 (S)
67 (S)
30 (S)
91 (S)
89
84
74
88
67
76
48
73
85
54
41
83
2
AlCl
La(OTf)
3
3
Yb(OTf)
YbCl
ZnBr
ErCl
AlMe
AlEt Cl
AlEtCl
Silica gel
3
3
2
3
3
10
11
12
2
2
a
All reactions were carried out in nitromethane under anhydrous
conditions at room temperature using 10 mol % acid catalyst except
entry 12 where 5-10 times (by weight) silica gel (Davisil 35-60 mesh,
pH 7.0) was used. Epoxide 1 was freshly made and stored at -20 °C
b
prior to use to avoid decomposition. The enantiomeric excess was
determined by HPLC (Chiracel OD). The absolute configuration of 2
was determined by comparing HPLC chromatograms with the authentic
sample prepared from commercially available (R,R)-1,2-trans-cyclo-
c
hexanediol. Isolated yield.
a
All reactions were carried out at room temperature with 10 mol %
2
NO ,
Scheme 2
p-TsOH (dried by azeotropic removal of its hydrate) in dry CH
3
or 5-10 times (by weight) silica gel (Davisil 35-60 mesh, pH 7.0) in
CH
CH
3
NO
NO
2
, or 10 mol % YbCl
3
in CH
2
Cl
2
, or 10 mol % AlMe
3
in
b
c
3
2
unless otherwise noted. 100 mol % AlMe
3
was used. 20
mol % AlMe
HPLC (Chiralcel OD) except entries 3 and 6 where enantioselectivity
was used. d Enantioselectivity was determined by chiral
3
1
3
was determined by H NMR shift analysis with Eu(hfc) . For determin-
e
ing absolute configuration of these epoxides see ref 4. Enantioselec-
tivity was determined by chiral HPLC (Chiralcel OD) except entry 3
1
where enantioselectivity was determined by H NMR shift analysis with
3
Eu(hfc) . The values in parentheses are the ee’s after recrystallization.
f
The absolute configurations were determined by comparing HPLC
chromatograms (entries 1 and 2) or 1H NMR shift analysis using
the resulting R-acyloxy ketones were crystalline, and the enan-
tiomeric excess could be further enhanced by recrystallization.
To test the generality of the rearrangement via pathway b, silica
3
Eu(hfc) (entry 3) with the authentic sample prepared from com-
g
mercially available (R,R)-1,2-trans-cyclohexanediol. The R-acyloxy
ketones were hydrolyzed to R-hydroxy ketones, and the absolute
configurations were determined by comparing the measured optical
rotations of the R-hydroxy ketones with the literature (see: refs 10
3 3
gel, YbCl , and AlMe were used (Table 2). In most cases the
9
isomer with inverted configuration was the major product;
however, in two cases (Table 2, entries 6 and 7) the rearrangement
proceeded with retention of configuration. The preference for
pathway a with these benzylic epoxides is probably due to a
stabilized carbocation intermediate 5.
h
and 11). The absolute configuration was tentatively assigned by
i
analogy. The absolute configuration was determined by comparing
j
the measured optical rotation with the literature (see: ref 5a). Isolated
yield. k Conversion determined by the H NMR of the crude reaction
1
l
mixture. No reaction was found with silica gel.
In summary, a detailed study of enantiomerically enriched enol
ester epoxides has shown that the acid-catalyzed rearrangement
of these epoxides operates through two distinct pathways, one
with retention of configuration and the other with inversion.
Acidity of the catalyst is one important factor, with strong acids
favoring retention of configuration and weak acids favoring
inversion. In addition to the mechanistic significance, the elucida-
tion of the dual pathways for the acid-catalyzed enol ester epoxide
rearrangement also proVides the flexibility to synthesize either
enantiomer of R-acyloxy ketone from one enantiomer of an enol
ester epoxide by judicious choice of reaction conditions (Scheme
Scheme 3
Acknowledgment. We are grateful to the generous financial support
from the General Medical Sciences of the National Institutes of Health
(GM55704-01), the Beckman Young Investigator Award Program, the
3).
Camille and Henry Dreyfus New Faculty Award Program, and Colorado
State University.
(
9) We found that the thermal rearrangement of these epoxides could also
lead to the formation of the inverted product with high stereospecificity when
the rearrangements were carried out in sealed tubes at high temperature (160-
90 °C) in vacuo. Since the Lewis acid-catalyzed rearrangement can be
1
Supporting Information Available: The characterization of enol ester
epoxides and R-acyloxy ketones, the NMR spectral and HPLC data for
the determination of the enantiomeric excess of these compounds, and
the crossover experimental data (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
conveniently carried out at room temperature, it provides a mild method to
obtain the inverted product, particularly for systems which are thermally
sensitive.
(
10) Nicolosi, G.; Patti, A.; Piattelli, M.; Sanfilippo, C. Tetrahedron:
Asymmetry 1995, 6, 519-524.
11) Naemura, K. Wakebe, T.; Hirose, K.; Tobe, Y. Tetrahedron: Asym-
metry 1997, 8, 2585-2595.
(
JA990124F