Asymmetric reduction of prochiral cycloalkenones. The influence of exocyclic alkene geometry

Article abstract of DOI:10.1039/b004540n

The asymmetric reduction of a series of prochiral enones of general structure 1 using the Corey oxazaborolidine 2, leading to enantiomerically enriched allylic cycloalkanols 3 is described. The influence of alkene geometry on both the sense (R vs. S) and

Full text of DOI:10.1039/b004540n

View Article Online / Journal Homepage / Table of Contents for this issue
Asymmetric reduction of prochiral cycloalkenones. The influence
of exocyclic alkene geometry
Alison F. Simpson,a Corinna D. Bodkin,a Craig P. Butts,ab Mark A. Armitagec and
Timothy Gallagher*a
1
a School of Chemistry, University of Bristol, Bristol, UK BS8 1TS
b School of Chemistry, University of Exeter, Exeter, UK EX4 4QD
c SB Pharmaceuticals, Leigh, Tonbridge, Kent, UK TN11 9AN
Received (in Cambridge, UK) 7th June 2000, Accepted 20th July 2000
Published on the Web 23rd August 2000
The asymmetric reduction of a series of prochiral enones of general structure 1 using the Corey oxazaborolidine
2, leading to enantiomerically enriched allylic cycloalkanols 3 is described. The influence of alkene geometry on
both the sense (R vs. S) and efficiency (% ee) of the asymmetric reduction process has been probed for two
systems, (E)- and (Z)-4 and (E)- and (Z)-7, based on cyclohexanone and cyclopentanone respectively. The
absolute stereochemistry of the cyclopentyl derivative (E)-8 has been established by X-ray crystallographic
analysis of carbamate 10. The ability to assign an absolute configuration to allylic alcohols 3, based on the
NMR methods described earlier by Riguera, has been evaluated.
The combination of oxazaborolidines and borane to mediate
the asymmetric reduction of prochiral ketones represents a
highly efficient process that has become one of the benchmarks
of contemporary asymmetric synthesis.1,2 Over recent years,
extensive efforts have been applied towards identifying more
efficient oxazaborolidine catalysts3 and, following a series of
detailed studies, a number of the critical features associated
with the catalytic cycle and the influence of additives have also
been elucidated.4,5
A series of prochiral enones have been evaluated as sub-
strates for this asymmetric reduction, which is outlined in
Scheme 1, and our preliminary results have been reported.13b ‡ A
Understanding the basic mechanism of an asymmetric pro-
cess is important, particularly because this information can be
applied towards developing a model capable of predicting
both the sense and also (though more difficult) the level of
asymmetric induction that might be anticipated. Several differ-
ent models6–10 have been proposed to explain the course of
oxazaborolidine-mediated reductions and, in addition, elec-
tronic factors associated with the ketone substituents have also
been shown to be important in determining the stereochemical
outcome.11 The latter, while clearly of significance, is more a
difficult property to incorporate within a computational model.
There are some broader issues to be considered since in a prac-
tical sense, an ability to define the types of ketones that repre-
sent viable substrates for particular reagents is key to exploiting
the potential associated with individual asymmetric reductants.
Most work in this area has, however, focussed on the “reagent”
rather than “substrate” side of the problem.
Scheme 1
number of specific examples that are relevant to this discus-
sion, as well as results obtained for several new substrates
using our optimised reaction conditions, are shown in Table
1. This work did, however, raise two issues. Firstly, defining
the stereochemical outcome of the reduction step by estab-
lishing the absolute configuration of the product allylic alco-
hol was considered important; the value of most mechanistic
models is as a predictive tool and any such predictions must,
at some point, be validated. A second and potentially more
intriging aspect of these molecules, which we did not examine
in our initial work, is the variable associated with enones 1
i.e. (E)- vs. (Z)-alkene geometry, and the impact that this
makes on the stereochemical outcome of the reduction: (1R)
vs. (1S) allylic alcohol. We considered this latter feature
unusual given that most “standard” prochiral ketones, which
find use as test substrates for asymmetric reductions, do not
incorporate such stereochemical “complications”, and it was
therefore of interest to establish the role, if any, that exocyclic
As part of a broader synthetic project, we have examined the
capability of the Corey oxazaborolidine–borane complex 22a,12
to mediate the asymmetric reduction of a cyclic ketone 1, where
prochirality is due to the location of an adjacent exocyclic C᎐C
᎐
associated with the enone.13 It is interesting to note that enones
based on general structure 1, although readily prepared, have
attracted little attention as substrates for asymmetric reduction
despite the synthetic potential inherent in the corresponding
allylic alcohols 3.†
‡ We have also evaluated the use of catalytic (20 mol%) amounts of 2
(using BH3ؒSMe2 as the stoichiometric reductant), and the results of
this study have already been described.13b In this present work, we used
one equivalent of 2 in order to minimize the impact made by competing
pathways that have been shown to participate under catalytic reduction
conditions.5d,12
† The bioreduction process reported by Fuganti and co-workers16a–c
provides an alternative entry into cyclic allylic alcohols related to 3. A
more lengthy procedure to one example of an allylic alcohol of this type
has been described by Warren and Harmat.16d
DOI: 10.1039/b004540n
J. Chem. Soc., Perkin Trans. 1, 2000, 3047–3054
This journal is © The Royal Society of Chemistry 2000
3047

View Article Online
Table 1 Asymmetric reduction of prochiral enones using (S)-2
In this paper we provide evidence to support the assignment
of absolute configuration to allylic alcohols 3 based on X-ray
crystallographic analysis and H NMR. In addition, the influ-
ee (%)a
1
Enone 1
Allylic alcohol 3
Yield (%)
ence of alkene geometry on the asymmetric reduction of repre-
sentative cyclopentyl and cyclohexyl enone variants has been
established.
84
88b
Results and discussion
Asymmetric reduction of (E)-4 and (Z)-4
89
98
94b
88b
The reduction of (E)-4 and (Z)-4 was carried out using a
stoichiometric amount of oxazaborolidine-borane complex 2
under our optimised conditions (addition of the enone sub-
strate to the reductant at Ϫ20 ЊC over 35 min) and the enantio-
meric excess of the allylic alcohol products (E)-5 (isolated in
96% ee) and (Z)-5 (isolated in 8% ee) was determined by chiral
phase HPLC (Scheme 2). The corresponding racemic alcohols
81
87
98
> 95b,c
a For optimised reaction conditions using stoichiometric 2, see Experi-
mental. The % ee of alcohols 3a–3e were determined via the corre-
sponding Moshers ester derivative. b Absolute stereochemistry has not
been established in these cases. In the case of 3d, see text. c Reduction of
the (Z)-isomer of 1e gave the corresponding α,β-unsaturated lactone in
36% ee, but this reaction did not go to completion. The (Z)-isomer of
1e was prone toward enolisation–cyclisation to give i.15a Asymmetric
reduction of the (Z)-isomer of 1e led to the known unsaturated lactone
ii,15b and the enantiomeric excess obtained (36% ee, absolute configur-
ation unknown) was determined by a chiral shift NMR experiment
using Eu[(ϩ)-hfc]3 [hfc = (heptafluoropropylhydroxymethylene)cam-
phorate]. This reaction was, however, slow and did not reach
completion.
Scheme 2 Reagents and conditions: i, 2, CH2Cl2, Ϫ20 ЊC.
served as analytical standards and were prepared by direct
reduction of the corresponding enone with NaBH4.
The key stereochemical issue to be established was whether
alkene geometry influenced the sense (R vs. S) of asymmetric
induction, and clearly alkene geometry does make a significant
impact on the level of asymmetric induction (96 vs. 8% ee). We
have correlated the structures of (E)-5 and (Z)-5 using the
sequence shown in Scheme 3. Hydrogenation of (E)-5 gave a
Scheme 3 Reagents and conditions: i, H2, 10% Pd on C, EtOH (*% ee
not determined).
mixture of the 2-benzylcyclohexanols cis-6 and trans-6 which
were separable by chromatography, and chiral phase HPLC of
cis-6 confirmed that no significant loss of enantiomeric purity
had occurred. Once again, the racemic variants were available
using the same reduction process and provided the correspond-
ing racemic analytical standards.¶
alkene geometry plays in determining both the sense and level
of asymmetric induction.§
Our study has concentrated on (E)- and (Z)-exocyclic enones
derived from cyclohexanone (and PhCHO) and cyclopenta-
none (and PhCHO), (E)-4 and (Z)-4, and (E)-7 and (Z)-7
respectively.
A similar result was obtained when hydrogenation of (Z)-5
was carried out to give a mixture of cis- and trans-6. Again, no
§ A number of acyclic enones have been used as substrates for
oxazaborolidine-mediated reductions, and the presence of α and αЈ sub-
stituents does influence the sense of asymmetric induction.14 These sub-
strates were the subject of a parallel computational study, but this used
the s-trans enone arrangement, a conformational option that is not
available to enones 1.
¶ cis- and trans 6 and cis- and trans- 9 have been reported by Fuganti.16a–c
We have not carried out an independent assignment of relative stereo-
chemistry since the precise structure of the isomer involved (cis vs.
trans) was not relevant to the correlations that we have drawn.
3048
J. Chem. Soc., Perkin Trans. 1, 2000, 3047–3054

View Article Online
significant erosion of enantiomeric purity was observed during
the reduction step and chiral HPLC confirmed that cis-6
derived from (E)-5 (96% ee) has the same absolute configur-
ation (R) as cis-6 derived from (Z)-5 (10% ee); the absolute
configuration of cis-6 is based on the earlier work of Fuganti et
al (see below).16a–c In other words, within the cyclohexyl series
based on (E)- and (Z)-4, the exocyclic alkene geometry does
not have a direct impact on the sense of asymmetric induction
available from oxazolidinone reduction. However, while the
(E)-isomer is reduced with a high degree of asymmetric induc-
tion, the corresponding (Z)-isomer is a very poor substrate for
the oxazaborolidine–borane reagent 2.
Asymmetric reduction of (E)-7 and (Z)-7
Asymmetric reduction of the cyclopentyl based enones (E)-7
and (Z)-7 was also carried out and, once again, we were able to
correlate the resulting allylic alcohols (E)-8 (isolated in 92% ee)
and (Z)-8 (isolated in 73% ee) following alkene hydrogenation
(Scheme 4).
Fig. 1 Crystal structure of carbamate 10.
the cyclopentyl derivative trans-9, we were able to prepare the
corresponding (S)-α-methylbenzylcarbamate 10 (Scheme 5), the
Scheme 5 Reagents and conditions: i, (S)-α-methylbenzyl isocyanate,
DMAP, CH2Cl2 (96%).
structure of which was determined by X-ray crystallographic
analysis (Fig. 1).|| This established the absolute stereochemistry
of (E)- and (Z)-8 (and cis- and trans-9) as (1R).
Application of NMR methods for the determination of absolute
configuration
A major challenge within asymmetric synthesis is the determin-
ation of absolute configuration. The use of X-ray crystallo-
graphic analysis, while often a very secure means of assigning
configuration, is a hit-or-miss affair since obtaining a suitable
crystalline derivative can be time consuming. Spectroscopic
methods based on NMR approaches have gained in popularity
as a convenient and rapid means of analysis that are also well
suited to microscale methodologies.17 Recently, Riguera and co-
workers have reported a simplified method for the determin-
ation of absolute configuration of secondary alcohols based on
use of esters derived from either (R)- or (S)-α-methoxyphenyl-
acetic acid (MPA).18 Until recently, this was achieved by prep-
aration of both diastereomeric esters, using the alcohol and
(R)- and (S)-MPA, followed by comparison of the two products
by NMR. Riguera has now demonstrated that it is possible to
elucidate an absolute configuration using only one diastereo-
Scheme 4 Reagents and conditions: i, 2, CH2Cl2, Ϫ20 ЊC; ii, H2, 10%
Pd on C, EtOH (* cis isomer was obtained in trace amounts and % ee
was not determined).
We have carried out hydrogenation of both (E)- and (Z)-8,
and in both cases, alkene reduction led to the trans diastereo-
mer trans-9 as the predominate product; cis-9 was only detected
in trace amounts. Analysis of trans-9 (by chiral phase HPLC)
demonstrated that alkene geometry did not alter the sense of
asymmetric induction and (1R)-trans-2-benzylcyclopentan-1-ol
9 was produced from both (E)-8 and (Z)-8; the absolute
configuration of trans-9 has also been established (see below).
In this cyclopentyl series, the level of asymmetric induction
is influenced to a lesser extent by alkene geometry, although it
is interesting that some loss of enantiomeric integrity was
detected during the conversion of both (E)-8 and (Z)-8 to
trans-9.
|| X-Ray diffraction measurements were determined on a Siemens
SMART area detector diffractometer, using graphite monochromated
Mo-Kα radiation. No absorption correction was applied. The struc-
ture was solved by direct-methods (SHELXS) and full-matrix least
squares refinement carried out using SHELXL-93 (SHELXTL version
5.03, Siemens Analytical X-ray, Madison WI, 1994). Absolute con-
figuration was assigned by using the known stereochemistry of the (S)-
α-methylbenzyl moiety incorporated within carbamate 10. C21H23NO2,
M = 321.40, orthorhombic, a = 5.1199(13), b = 13.998(4), c = 24.286(6)
Å, U = 1740.6(8) Å3, T = 173 K, space group P212221 (no. 19), Z = 4,
µ(Mo-Kα) = 0.078 mmϪ1, 18250 reflections measured, 3980 unique,
(Rint = 0.0765) reflections used in all calculations, wR2 = 10.3%
(all data), R1 = 5.0% [2526 I > 4σ(I) data]. CCDC reference number
207/459. See http://www.rsc.org/suppdata/p1/b0/b004540n for crystallo-
graphic files in .cif format.
Assignment of absolute configuration
In the cyclohexyl series, the R absolute configuration of (E)-5
can be assigned based on the earlier work of Fuganti et al. who
have prepared the corresponding (S)-enantiomer. In the case of
J. Chem. Soc., Perkin Trans. 1, 2000, 3047–3054
3049

View Article Online
1
1
Table 2 Variable temperature H NMR dataa for the (R)-MPA ester
Table 4 Variable temperature H NMR dataa for the (R)-MPA ester
of (E)-3d
of (E)-8
Temperature/ЊC
δ H(1)
δ Hvinyl
δ H(3)
δ H(3)b
Temperature/ЊC
δ H(1)
δ Hvinyl
22
Ϫ40
Ϫ60
Ϫ80
5.32
5.26
5.24
5.22
5.67
5.32
5.13
4.93
2.35
2.42
2.46
2.53
2.18
2.05
1.97
1.90
22
Ϫ40
Ϫ60
Ϫ80
5.65
5.59
5.58
5.55
6.18
5.97
5.88
5.77
Hvinyl undergoes an upfield shift of 0.76 ppm on cooling from 22 to
Ϫ80 ЊC. One of the C(3) methylene protons undergoes an upfield shift
(by 0.28 ppm) and the other C(3) proton shifted downfield (by 0.18
ppm). Small upfield and downfield shifts associated with H(5) and H(6)
were observed, but these signals were overlapping and unambiguous
assignments were not possible. a 400 MHz, CD2Cl2. b Assignment based
on a COSY analysis. The C(3) methylene appears as two separate
multiplets, but these have not been individually assigned.
Hvinyl undergoes an upfield shift of 0.41 ppm on cooling from 22 to
Ϫ80 ЊC. Small changes were observed for some of the methylene signals
between δ 1.60 and 2.63, but we have been unable to assign the (over-
lapping) signals involved. a 400 MHz, CD2Cl2.
Table 5 Variable temperature 1H NMR dataa for the (S)-MPA ester of
(E)-8
δ H(6)b
1
Table 3 Variable temperature H NMR dataa for the (R)-MPA ester
Temperature/ЊC
δ H(1)
δ Hvinyl
of (E)-5
22
Ϫ40
Ϫ60
Ϫ80
5.67
5.62
5.59
5.57
6.56
6.55
6.55
6.55
1.59
1.50
1.48
1.45
Temperature/ЊC
δ H(1)
δ Hvinyl
δ H(3)
δ H(3)b
22
Ϫ40
Ϫ60
Ϫ80
5.32
5.32
5.32
5.32
6.04
5.73
5.57
5.32
2.33
2.38
2.15
1.98
1.93
1.94
One signal associated with the C(6) methylene undergoes an upfield
shift of 0.14 ppm on cooling from 22 to Ϫ80 ЊC. Hvinyl also shifts upfield
by 0.01 ppm. a 400 MHz, CD2Cl2. b This assignment is based on a
COSY analysis but we cannot assign the stereochemistry of the H(6)
proton involved.
2.43
c
Hvinyl undergoes an upfield shift of 0.72 ppm on cooling from 22 to
Ϫ80 ЊC. One of the C(3) methylene protons undergoes an upfield shift
(by 0.21 ppm) and the other proton shifts downfield (by ≤0.1 ppm). a 400
MHz, CD2Cl2. b Assignment is based on a COSY analysis. The C(3)
methylene appears as two separate multiplets (not individually
assigned). c The lower field signal was very broad at this temperature.
on the alkene. The shifts associated with H(6) (H(5) in the case
of (E)-8) were both less marked and less clear cut. The difficulty
here was that the ring methylene protons overlapped and it was
not possible to track and assign individual signals with con-
fidence at the different temperatures used. Shifts associated
with H(1) are included in Tables 2–5, but these do not appear to
have a diagnostic value.
In the case of the other diastereomeric series, the (S)-MPA
ester of (E)-8, then much smaller shifts (≤0.15 ppm) were
observed for all protons. Interestingly, when (R)- and (S)-
indan-1-ol was used as a test substrate by Riguera, the “mis-
matched” diastereomer combination (in Riguera’s work this
was the (R)-MPA ester of (S)-indan-1-ol) showed only small
(<0.05 ppm) shift changes. The “matched” diastereomer—the
(S)-MPA ester of (S)-indan-1-ol—displayed somewhat larger
(0.05 to 0.1 ppm) shift changes, with H(7) (which we can com-
pare to Hvinyl in MPA derivatives of (E)-8) showing a significant
upfield shift. With the (S)-MPA ester of (R,E)-8, H(6) moved to
higher field (as expected), but a very small (insignificant?) upfield
shift (0.01 ppm) was also associated with the Hvinyl signal.
Generally, the shift associated with Hvinyl in the “mis-
matched” (S, R) diastereomer was much less pronounced than
in the “matched” (R, R) combination, and this signal appears to
be the most diagnostic when applying the simplified version of
the Riguera procedure to allylic alcohols of this type. Using
the data shown in Tables 2–5, we can confirm the absolute
configuration of both (E)-5 and (E)-8 as (1R) and assign the
configuration of (E)-3d as (1R).
In summary, by using prochiral exocyclic enones we have
determined a series of potentially important factors associated
with oxazaborolidine-mediated asymmetric reduction. We must
acknowledge that the stereochemical outcome described
for enones 4 and 7 is predicted by existing transition state
models6–10 (based on a large vs. small substituent effect), but
these models do not recognise the subtleties associated with
enone geometry and ring size. Three key points can be made.
1. For the limited range of substrates we have studied,
alkene geometry does not alter (i.e. reverse) the sense of asym-
metric induction, although (E)-alkenes are reduced with a
higher degree of asymmetric induction than is observed with
the corresponding (Z)-isomers.
Fig. 2 sp and ap Conformations corresponding to the (R)-MPA ester
of (1R,E)-5 (see ref. 18).
mer, thereby obviating the need to prepare and compare two
derivatives.18b
Since this method has not yet been applied to allylic alcohols
of the type described in this paper, and given that we have firm
experimental data available, it was of interest to apply the
simplified Riguera protocol within this context. The process
relies on the thermodynamic preference for the sp rather than
ap conformation associated with MPA esters (Fig. 2). At low
temperatures the population of the sp conformer is increased
resulting in either a downfield shift (for those protons not
shielded by the aryl ring) or an upfield shift (for those protons
which are shielded) relative to the observed shift of the same
1
proton(s) at room temperature. By acquiring H NMR data
at various temperatures, the direction of shift can be used to
deduce the absolute configuration of the secondary alcohol.
In Tables 2–5 we have shown data for three MPA esters
derived from allylic alcohols (E)-3d, (E)-5 and (E)-8. In the
case of (E)-8 (for which we have X-ray crystallographic data
as the basis of our assignment of the (R)-configuration), esters
derived from both (R)- and (S)-MPA were prepared and evalu-
ated. 1H NMR studies were carried out in CD2Cl2 solution and
on cooling from ϩ22 to Ϫ80 ЊC, Hvinyl showed very significant
upfield shifts (between 0.41 and 0.76 ppm), which is fully con-
sistent with Riguera’s model. The individual methylene protons
at C(3) of these three derivatives shifted in different directions
to one another, but here one cannot exclude an influence
associated with the anisotropic influence of the aryl substituent
3050
J. Chem. Soc., Perkin Trans. 1, 2000, 3047–3054

View Article Online
2. Alkene geometry is more influential in the cyclohexyl
series than in the corresponding five-membered ring: while (E)-
4 is a good substrate, (Z)-4 gives the corresponding allylic
alcohol in very low% ee.
2 × CH2), 2.51 (2 H, t, J 7, 6-H2), 2.73 (2 H, td, J 7 and 2, 3-H2),
6.59 (1H, d, J 1, 4Ј-H), 7.42 (1 H, m, C᎐CHAr), 7.46 (1 H, m,
᎐
5Ј-H) and 7.66 (1 H, m, 2Ј-H); δC (75.5 MHz; CDCl3) 22.9
(CH2), 23.3 (CH2), 28.8 (CH2), 38.8 (CH2), 111.3 (CH), 121.9
(C), 126.9 (CH), 134.3 (C), 143.4 (CH), 144.72 (CH) and 200.5
3. Clearly, when predicting the level of asymmetric induc-
tion, one cannot consider only alkene geometry and ignore ring
size (6- vs. 5-ring). We had anticipated that there might be a
significant decrease in% ee associated with presence of a (Z)-
alkene, given that this would place the alkenyl substituent in
close proximity to ketone carbonyl which itself provides a co-
ordination site for the reductant. While this is true in the cyclo-
hexyl series, this argument does not hold to the same degree in
the case of the cyclopentyl substrate. In this latter case, alkene
geometry makes a less significant impact, which may reflect the
modified (more accessible) environment associated with the
carbonyl residue in the five-membered ring.
That alkene geometry can influence the stereochemical out-
come of the oxazaborolidine reduction process is clear. How-
ever, while it is tempting to speculate on the reasons behind this,
any attempt to apply current predictive models would also
involve assuming that other key mechanistic features of the
reaction remained the same. Alkene geometry will influence the
environment around the carbonyl residue, which may, in turn,
have an impact on co-ordination and/or hydride delivery.
Significant structural variations within the ketone substrate
may also have a bearing on the stability of the intermediates
involved; it is known that oxazaborolidine 2 is not the only
reductant present in the reaction mixture.5d We would hope that
future work directed towards modelling the process of hydride
transfer involved in this important asymmetric process would
take account of and benefit from the results described in this
paper.
(C᎐O); m/z (EI) 176 (Mϩ, 71%), 148 (20), 91 (45).
᎐
(Z)-2-Benzylidenecyclohexanone (Z)-4.24 Benzylidenecyclo-
hexanone (Z)-4 was prepared in 18% yield by photoisomeris-
ation of (E)-2-benzylidene cyclohexanone (E)-4 as described
previously.24 δH (270 MHz; CDCl3) 1.87–2.03 (4 H, m, 2 ×
CH2), 2.55 (2 H, t, J 7, CH2), 2.63 (2 H, t, J 7, CH2), 6.39 (1 H, s,
C᎐CHAr) and 7.20–7.35 (5 H, m, Ar).
᎐
(Z)-2-Benzylidenecyclopentanone (Z)-7.25 Benzylidenecyclo-
pentanone (Z)-7 was prepared in 18% yield by photoisomeris-
ation of (E)-2-benzylidenecyclopentanone (E)-7 as described
previously.25 δH (270 MHz; CDCl3) 1.98 (2 H, quintet, J 7.5,
4-H2), 2.42 (2 H, t, J 7.5, 5-H2), 2.83 (2 H, d t, J 7.5 and 2, 3-H2),
6.73 (1 H, t, J 2, C᎐CHAr), 7.28–7.43 (3 H, m, 3 × ArH) and
᎐
7.80–7.90 (2 H, m, 2 × ArH).
General procedure for reduction of exocyclic enones with NaBH4
To a solution of the enone (1 mmol) in MeOH (2 cm3) was
added NaBH4 (1 mmol) and the mixture was stirred at room
temperature until the reaction was judged complete by TLC.
The reaction was then quenched with H2O (5 cm3) and
the aqueous phase extracted with EtOAc (3 × 5 cm3). The
combined organic extracts were dried (Na2SO4), filtered and
concentrated in vacuo. The resultant residue was purified by
flash column chromatography (petroleum ether–EtOAc) to
afford the corresponding racemic allylic alcohol. 1H NMR data
for individual compounds are provided below and the race-
mates were all resolved using the analytical chiral phase HPLC
conditions described below.
Experimental
General
Infrared spectra (νmax) were recorded using a Perkin-Elmer 1715
FTIR spectrometer, in the range of 4000–600 cmϪ1 either as a
neat film on NaCl plates or as a solution in CH2Cl2. Mass
spectra m/z (EI, CI, FAB) were obtained using a Fisons/VG
Analytical Autospec System. Nuclear magnetic resonance
(NMR) spectra were recorded at the field strength and in
CDCl3, unless otherwise indicated using standard pulse
sequences on an Alpha 500, JEOL GX400 or a Lambda 300
spectrometer, with proton and carbon assignments made
General procedure for asymmetric reduction of enones
To a solution of (S)-oxazaborolidine–borane complex (S)-2 (58
mg, 0.2 mmol) in dry CH2Cl2 (0.3 cm3) at Ϫ20 ЊC was added a
solution of the enone (0.2 mmol) in CH2Cl2 (0.6 cm3) over a
period of 35 min. After the reaction was complete, as judged
by TLC, cold (0 ЊC) MeOH (3 cm3) was carefully added, and
the cooling bath was removed and stirring continued for 1 h.
Removal of solvents in vacuo and purification of the residue
by flash column chromatography (petroleum ether–EtOAc)
yielded the allylic alcohol product. The enantiomeric excess of
the product was then determined either (i) by preparation and
1H NMR analysis of the Mosher ester derivative 3a–e, see
below) or (ii) by chiral phase HPLC (used for products derived
from 4 and 7). The latter required use of a Chiracel OD column
[250 mm length, 4.6 mm internal diameter, with a sample size of
0.01 cm3 (substrate concentration of 1 mg cmϪ3) through a loop
of 0.02 cm3 with detection at 254 nm] eluting with 90% hexane–
10% propan-2-ol (unless otherwise stated) at a flow rate of 1
cm3 minϪ1. Retention times are quoted in minutes.
1
1
using a combination of H/1H and H/13C correlation spectra.
Coupling constants are reported in hertz (Hz).
(S)-Oxazaborolidine–borane complex (S)-2 (as well as the
corresponding (R)-enantiomer) were kindly provided by Dr
David Mathre (Merck).
Enones 1a,19 1b,19 1c20,21 and 1e15,22,23 were prepared using
literature procedures.
2-(3-Furylmethylidene)cyclohexanone (1d)
To a solution of sodium hydroxide (6.52 g) in water (1500 cm3)
was added cyclohexanone (33.8 cm3, 326 mmol). After stirring
for 5 min at room temperature, 3-furaldehyde (9.75 cm3, 112
mmol) was added and the solution stirred for a further 72
hours. The reaction mixture was then quenched with acetic acid
to pH 7 and extracted with toluene (4 × 40 cm3). The organic
extracts were washed with brine (3 × 300 cm3), dried (Na2-
SO4), filtered and concentrated in vacuo to give an orange
oil which was purified by distillation under reduced pressure to
yield a yellow oil, bp 138–140 ЊC/2 mmHg which crystallised on
cooling. Recrystallisation gave the title compound (10.9 g, 55%)
as a pale yellow solid, mp 51–52 ЊC (from petroleum ether)
(Found: C, 75.1; H, 6.7; C13H16O requires C, 75.0; H, 6.9%.
Found: Mϩ, 176.0835. C11H12O2 requires 176.0837); νmax (Nujol
mull)/cmϪ1 1719 (C᎐O); δ (400 MHz; CDCl3) 1.87 (4 H, m,
General procedure for preparation of Mosher ester derivatives
To a solution of the allylic alcohol (0.09 mmol) in dry CH2Cl2
(1 cm3) was added DMAP (ca. 6 mg), Et3N (0.039 cm3) and (R)-
(MTPA) chloride (0.14 mmol) and the mixture stirred at room
temperature until the reaction was complete, as judged by TLC.
The solution was then filtered through a short pad of silica and
the residue washed with 50:50 petroleum ether–EtOAc. The
filtrate was concentrated in vacuo to give the Mosher ester
derivative which was not purified further, and in each case, the
Mosher ester of the racemic alcohol was prepared and used an
1
analytical standard. Analysis was carried out by H NMR but
the Mosher ester derivatives were not characterized further.
᎐
H
J. Chem. Soc., Perkin Trans. 1, 2000, 3047–3054
3051

View Article Online
( )- (E)-2-Isobutylidenecyclopentan-1-ol (E)-3a26
( )-2-[(Ethoxycarbonyl)methylidene]cyclohexan-1-ol (1R)-3e28
Using the general procedure, reduction of 1a gave the title com-
pound as a colourless syrup in 72% yield; δH (270 MHz; CDCl3)
0.97 (6 H, dd, J 7 and 1.5, 2 × Me), 1.51–1.92 (5 H, m), 2.13–
2.45 (3 H, m), 4.36 (1 H, br t, J 5, 1-H) and 5.36 (1 H, dq, J 9
Using the general procedure, reduction of 1e gave the title com-
pound as a colourless oil in 57% yield; δH (270 MHz; CDCl3)
1.20 (3 H, t, J 7) and 1.42–2.20 (8 H, m), 3.65 (1 H, m), 4.15
(4 H, m) and 5.95 (1 H, br s, 1-H).
and 1.5, C᎐CHCHMe ).
᎐
2
(1R*)-2-[(Ethoxycarbonyl)methylidene]cyclohexan-1-ol (1R*)-
(E)-3e
(1R*,E)-2-Isobutylidenecyclopentan-1-ol (1R*,E)-3a
Using the general procedure, reduction of (E)-1e gave the title
compound as a colourless oil in 76% chemical yield and >95%
ee. The enantiomeric excess was determined by 1H NMR
spectroscopy by conversion to the Mosher’s ester using the
method described above. δH (270 MHz; CDCl3) 5.80 (1 H, s,
Using the general procedure, reduction of (E)-1a gave the title
compound as a colourless oil in 84% chemical yield and 88% ee.
The enantiomeric excess was determined by 1H NMR spectro-
scopy by conversion to the Mosher’s ester using the method
described above. δ (270 MHz; CDCl ) 5.55 (0.94 H, m, C᎐CH-
᎐
H
3
C᎐CHCO Et) (the signal associated with the other enantiomer
᎐
2
CHMe ) and 5.63 (0.06 H, m, C᎐CHCHMe ).
᎐
2
2
of 3e appeared at 5.60 ppm but was not detected).
( )-(E)-2-Octylidenecyclopentan-1-ol ( )-(E)-3b
(1R,E)-2-Benzylidenecyclohexan-1-ol (1R,E)-5
Using the general procedure, reduction of 1b gave the title com-
pound in quantitative yield as a colourless oil (Found: M ϩ Hϩ,
197.1901. C13H24O requires M, 197.1905); νmax (liquid film)/
cmϪ1 3382, 1651; δH (270 MHz; CDCl3) 0.86–0.91 (3 H, m,
CH3), 1.12–2.40 (19 H, m), 4.38 (1 H, m, 1-H) and 5.53 (1 H, m,
Using the general procedure, reduction of (E)-4 gave the title
compound as a colourless crystalline solid in 76% chemical yield
and in 96% ee. (E)-5 [α]D22 ϩ35.8 (c 1.2, CHCl3), (1S)-isomer
lit.,16a [α]D22 Ϫ35.2 (c 1.2, CHCl3). Retention times: major
enantiomer 5.93 min (98.2% area); minor enantiomer 7.45 min
(1.8% area); δH (270 MHz; CDCl3) 1.40–2.16 (8 H, m, 4 × CH2),
C᎐CHC H ); m/z (CI) 197 (M ϩ Hϩ).
᎐
7
15
2.71 (1 H, m, OH), 4.23 (1 H, m, 1-H), 6.52 (1 H, s, C᎐CHAr)
and 7.17–7.34 (5 H, m, Ar).
᎐
(1R*,E)-2-Octylidenecyclopentan-1-ol (1R*,E)-3b
Using the general procedure, reduction of (E)-1b gave the title
compound as a colourless solid in 89% chemical yield and
94% ee. The enantiomeric excess was determined by H NMR
spectroscopy by conversion to the Mosher’s ester using the
method described above. δH (270 MHz; C6D6) 3.15 (2.91 H, m,
OMe) and 3.21 (0.09 H, m, OMe).
(1R,Z)-2-Benzylidenecyclohexan-1-ol (1R,Z)-5
1
Using the general procedure, reduction of (Z)-4 gave the title
compound as a colourless crystalline solid in 71% yield and in
8% ee. Retention times: minor enantiomer 5.71 min (45.7%
area); major enantiomer 6.91 min (53.4% area); δH (270 MHz;
CDCl3) 1.36–1.57 (4 H, m), 1.78–1.95 (3 H, m), 2.18 (1 H, m),
2.61 (1 H, tdd, J 16, 6 and 2), 4.83 (1 H, s, CHOH), 6.35 (1 H, s,
( )-2-Cyclopentylidenecyclopentan-1-ol ( )-3c27
C᎐CHAr) and 7.20–7.35 (5 H, m, Ar). The racemic alcohol
᎐
Using the general procedure, reduction of 1c gave the title com-
pound as a colourless crystalline solid in 91% yield, mp 58–
59 ЊC (from petroleum ether) (lit.,27 57–59 ЊC); δH (270 MHz;
CDCl3) 1.25–2.36 (15 H, m) and 4.57 (1 H, br s, 1-H).
has been described and is prepared by photoisomerisation of
(E)-5.29
(1R,E)-2-Benzylidenecyclopentan-1-ol (1R,E)-8
Using the general procedure, reduction of (E)-7 gave the title
compound as a colourless crystalline solid in 67% yield and 92%
ee. Retention times (95% hexane–5% propan-2-ol): major
enantiomer 13.01 min (95.8% area); minor enantiomer 14.63
min (4.2% area); δH (270 MHz; CDCl3) 1.58–2.04 (5 H, m),
2.50–2.80 (2 H, m), 4.58 (1 H, m, 1-H), 6.57 (1 H, q, J 2.5,
(1R*)-2-Cyclopentylidenecyclopentan-1-ol (1R*)-3c
Using the general procedure, reduction of (E)-1c gave the title
compound as a colourless solid in 98% chemical yield and 88%
ee. The enantiomeric excess was determined by 1H NMR
spectroscopy by conversion to the Mosher’s ester using the
method described above. δH (270 MHz; CDCl3) 5.75 (0.94 H, m,
1-H) and 5.82 (0.06 H, m, 1-H).
C᎐CHAr) and 7.15–7.40 (5 H, m, Ar). The racemic alcohol has
᎐
been described.30
( )-2-(3-Furylmethylidene)cyclohexan-1-ol ( )-3d
(1R,Z)-2-Benzylidenecyclopentan-1-ol (1R,Z)-8
Using the general procedure, reduction of 1d gave the title com-
pound as a colourless syrup in 94% yield (Found: Mϩ, 178.0990.
C11H14O2 requires 178.0994); νmax (liquid film)/cmϪ1 3385;
δH (400 MHz; CDCl3) 1.45–1.66 (5 H, m), 1.80–1.99 (2 H, m),
2.18 (1 H, M), 2.74 (1 H, m), 4.21 (1 H, m, CHOH), 6.21 (1 H, s,
Using the general procedure, reduction of (Z)-7 gave the title
compound as a pale yellow oil in 72% yield and 73% ee. Reten-
tion times (95% hexane–5% propan-2-ol): major enantiomer
8.25 min (86.7% area); minor enantiomer 9.11 min (13.3%
area); δH (270 MHz; CDCl3) 1.58–2.0 (5 H, m), 2.37–2.76 (2 H,
C᎐CHAr), 6.42 (1 H, s, 3Ј-H) and 7.38–7.41 (2 H, m, 2Ј-H and
᎐
m), 4.87 (1 H, m, CHOH), 6.49 (1 H, s, C᎐CHAr) and 7.20–7.50
᎐
(5 H, m, Ar). The racemic alcohol has been described.31
5Ј-H); δC (75.5 MHz; CDCl3) 23.0 (CH2), 27.0 (CH2), 27.5
(CH2), 36.4 (CH2), 73.7 (CH), 110.8 (CH), 111.2 (CH), 121.8
(CH), 140.6 (CH), 142.6 (CH) and 143.9 (CH).
General procedure for hydrogenation of allylic alcohols
To a solution of the allylic alcohol (0.15 mmol) in EtOH (5 cm3)
was added a catalytic amount of 10% palladium on carbon and
the reaction vessel was evacuated and purged with hydrogen
(balloon) several times. After stirring for 18 h under an atmos-
phere of hydrogen, the reaction mixture was filtered through a
short pad of Celite and the residue washed with EtOH
(5 × 10 cm3). The filtrate was concentrated in vacuo and purified
by flash column chromatography (petroleum ether–EtOAc) to
afford the reduced secondary alcohol.
(1R)-2-(3-Furylmethylidene)cyclohexan-1-ol (1R)-3d
Using the general procedure, reduction of (E)-1d gave the title
compound as a colourless oil in 81% chemical yield and 87% ee.
The enantiomeric excess was determined by 1H NMR spectro-
scopy by conversion to the Mosher’s ester using the method
described above. δH (270 MHz; CDCl3) 6.12 (0.94 H, s,
C᎐CHAr), 6.25 (0.06 H, s, C᎐CHAr), 6.36 (0.93 H, m, 4Ј-H)
᎐
᎐
and 6.42 (0.07 H, m, 4Ј-H).
3052
J. Chem. Soc., Perkin Trans. 1, 2000, 3047–3054

View Article Online
cis- and trans-( )-2-Benzylcyclohexanol ( )-6
Reduction of (1R,Z)-2-benzylidenecyclopentan-1-ol (1R,Z)-8
Method A. Reduction of ( )-(E)-2-benzylidenecyclohexan-1-
ol 5 using the general procedure, followed by flash column
chromatography (hexane–EtOAc) gave (i) ( )-cis-2-benzyl-
cyclohexan-1-ol cis-6 in 36% yield as a colourless crystalline
solid, mp 68–69 ЊC (from hexane) (lit.,16b 67–70 ЊC). Retention
times for the enantiomers (Chiralcel OD, 95%, hexane–55%
propan-2-ol) 6.28 min and 8.71 min; δH (270 MHz; CDCl3)
1.15–1.80 (10 H, m, 4 × CH2, OH and 2-H), 2.55 (1 H, dd,
J 12.5 and 7.5, CHAr), 2.70 (1 H, d d, J 12.5 and 7.5, CHAr),
3.8 (1 H, br s, CHOH) and 7.16–7.32 (5 H, m, Ar).
Continued elution gave (ii) ( )-trans-2-benzylcyclohexan-1-
ol trans-6 in 25% yield as a colourless crystalline solid, mp 45–
47 ЊC (from hexane) (lit.,16b 46–48 ЊC); δH (270 MHz; CDCl3)
0.85–2.02 (10 H, m, 4 × CH2, OH and 2-H), 2.35 (1 H, dd, J 14
and 8, CHAr), 3.18 (1 H, dd, J 14 and 4, CHAr), 3.30 (1 H, m,
CHOH) and 7.15–7.33 (5 H, m, Ar).
Reduction of (1R,Z)-2-benzylidenecyclopentan-1-ol (Z)-8
(with 73% ee) was carried out using the general procedure.
Following isolation of the trans-9, HPLC analysis demon-
strated that the trans compound was obtained in 69% ee.
(1R,E)-(2-Benzylidenecyclopentan-1-yl) N-[(S)-1-(phenylethyl)]-
carbamate 10
A solution of alcohol (E)-8 (50 mg, 0.29 mmol, 92% ee) in
CH2Cl2 (1 cm3) was treated with DMAP (a few crystals) fol-
lowed by (S)-α-methylbenzyl isocyanate (45 mm3, 0.32 mmol).
The solution was allowed to stand for 20 h, and water (10 cm3)
was added. The mixture was extracted with CH2Cl2 (3 × 10
cm3) and the combined extracts were washed with water, fol-
lowed by brine and then dried (Na2SO4). Removal of solvent
followed by chromatography (EtOAc–hexanes) gave carbamate
10 (90 mg, 96%) as a colourless solid (Found: M 321.1730.
C21H23NO2 requires M, 321.1728); νmax (CHCl3)/cmϪ1 3440,
1717; δH (300 MHz; CDCl3) 1.48 (3 H, d, J 7), 1.51–2.10 (4 H,
m), 2.50–2.75 (2 H, m), 4.87 (1 H, m), 4.98 (1 H, br s), 5.55 (1 H,
m), 6.58 (1 H, br s) and 7.20–7.40 (10 H, m) (broadening/poor
resolution of most signals was observed); δC (75.5 MHz;
CDCl3) 22.6 (CH3)*, 23.3 (CH2), 29.6 (CH2), 32.5 (CH2), 51.0
(CH)*, 79.8 (CH)*, 126.0 (CH), 126.1 (CH), 126.8 (CH), 127.3
(CH), 128.3 (CH), 128.6 (CH), 128.7 (CH), 137.7 (C), 143.3 (C),
143.6 (C)* and 155.8 (C) (* these signals were broadened).
Crystals suitable for X-ray crystallographic analysis were
obtained by recrystallisation from MeOH–H2O.
Method B. Reduction of ( )-(Z)-2-benzylidenecyclohexan-1-
ol ( )-(Z)-5 using the general procedure, followed by flash
column chromatography gave: (i) cis-( )-2-benzylcyclohexan-1-
ol cis-6 in 28% yield and (ii) trans-( )-2-benzylcyclohexan-1-ol
trans-6 in 26% yield. Both compounds were identical (1H NMR
and HPLC characteristics) to that obtained from the (E)-
isomer.
Reduction of (1R,E)-2-benzylidenecyclohexan-1-ol (1R,E)-5
Reduction of (1R,E)-2-benzylidenecyclohexan-1-ol (1R,E)-5
(with 96% ee) was carried out using the general procedure.
Following isolation of the cis-isomer, HPLC analysis demon-
strated that the cis-6 was obtained in 96% ee.
Acknowledgements
We thank Dr David Mathre (Merck) for a generous supply of
both (S)-and (R)-2, SB Pharmaceuticals for financial support,
and Professor Ian H. Williams and Dr Lynda P. Linney
(University of Bath) for advice.
Reduction of (1R,Z)-2-benzylidenecyclohexan-1-ol (1R,Z)-5
Reduction of (1R,Z)-2-benzylidenecyclohexan-1-ol (1R,Z)-5
(with 8% ee) was carried out using the general procedure.
Following isolation of the cis-isomer, HPLC analysis demon-
strated that the cis-6 was obtained in 10% ee.
References
1 (a) A. Hirao, S. Itsuno, S. Nakahama and N. Yamazaki, J. Chem.
Soc., Chem. Commun., 1981, 315; (b) S. Itsuno, A. Hirao,
S. Nakahama and N. Yamazaki, J. Chem. Soc., Perkin Trans. 1,
1983, 1673; (c) S. Itsuno, Y. Sakurai, K. Ito, A. Hirao and
S. Nakahama, Bull. Chem. Soc. Jpn., 1987, 60, 395.
2 (a) E. J. Corey, R. K. Bakshi and S. Shibata, J. Am. Chem. Soc.,
1987, 109, 5551; (b) E. J. Corey and C. J. Helal, Angew. Chem., Int.
Ed.., 1998, 37, 1987.
3 (a) S. Wallbaum and J. Martens, Tetrahedron: Asymmetry, 1992, 3,
1475; (b) V. K. Singh, Synthesis, 1992, 605; (c) S. Itsuno, Org. React.,
1998, 52, 395.
4 E. J. Corey, M. Azimioara and S. Sarshar, Tetrahedron Lett., 1992,
33, 3429.
5 (a) D. Cai, D. Tschaen, Y. J. Shi, T. R. Verhoeven, R. A. Reamer and
A. W. Douglas, Tetrahedron Lett., 1993, 34, 3243; (b) Y. J. Shi,
D. Cai, U. H. Dolling, A. W. Douglas, D. M. Tschaen and T. R.
Verhoeven, Tetrahedron Lett., 1994, 35, 6409; (c) D. M. Tschaen,
L. Abramson, D. Cai, R. Desmond, U. H. Dolling, L. Frey,
S. Karady, Y. J. Shi and T. R. Verhoeven, J. Org. Chem., 1995, 60,
4324; (d) A. W. Douglas, D. M. Tschaen, R. A. Reamer and Y. J. Shi,
Tetrahedron: Asymmetry, 1996, 7, 1303.
cis- and trans-( )-2-Benzylcyclopentanol ( )-916b
Method A. Reduction of ( )-(E)-2-benzylidenecyclopentan-
1-ol ( )-(E)-8 using the general procedure, followed by flash
column chromatography (petroleum ether–EtOAc) gave (i)
cis-( )-2-benzylcyclopentan-1-ol cis-9 in <5% yield as a colour-
less oil. δH (270 MHz; CDCl3) 1.22–2.1 (8 H, m, 3 × CH2, OH
and 2-H), 2.55 (1 H, dd, J 10 and 6, CHAr), 2.76 (1 H, dd, J 10
and 6, CHAr), 3.91 (1 H, m, CHOH) and 7.18–7.32 (5 H, m,
Ar).
Continued elution gave (ii) trans-( )-2-benzylcyclopentan-1-
ol trans-9 in 20% yield as a colourless oil. Retention times for
both enantiomers (Chiralcel OD, 95% hexane–5% propan-2-ol)
6.07 min and 9.81 min; δH (270 MHz; CDCl3) 1.2–2.1 (8 H, m,
3 × CH2, OH and 2-H), 2.68 (1 H, dd, J 14 and 8, CHAr), 2.85
(1 H, dd, J 14 and 8, CHAr), 4.08 (1 H, m, CHOH) and 7.15–
7.35 (5 H, m, Ar).
6 E. J. Corey, R. K. Bakshi, S. Shibata, C. P. Chen and V. K. Singh,
J. Am. Chem. Soc., 1987, 109, 7925.
7 D. K. Jones, D. C. Liotta, I. Shinkai and D. J. Mathre, J. Org. Chem.,
1993, 58, 799.
8 L. P. Linney, C. R. Self and I. H. Williams, J. Chem. Soc., Chem.
Commun., 1994, 1651.
9 G. J. Quallich, J. F. Blake and T. M. Woodall, J. Am. Chem. Soc.,
1994, 116, 8516.
Method B. Reduction of ( )-(Z)-2-benzylidenecyclopentan-
1-ol ( )-(Z)-8 using the general procedure, followed by flash
column chromatography gave (i) cis-( )-2-benzylcyclopentan-
1-ol cis-9 in < 5% yield and (ii) trans-( )-2-benzylcyclopentan-
1-ol trans-9 in 40% yield which was identical (1H NMR and
HPLC characteristics) to that obtained from the (E)-isomer.
10 V. Nevalainen, Tetrahedron: Asymmetry, 1994, 5, 903 and references
therein.
Reduction of (1R,E)-2-benzylidenecyclopentan-1-ol (1R,E)-8
11 E. J. Corey and C. J. Helal, Tetrahedron Lett., 1995, 36, 9153.
12 (a) D. J. Mathre, T. K. Jones, L. C. Xavier, T. J. Blacklock,
R. A. Reamer, J. J. Mohan, E. T. Turner Jones, K. Hoogsteen,
M. W. Baum and E. J. J. Grabowski, J. Org. Chem., 1991, 56, 751;
(b) T. K. Jones, J. J. Mohan, L. C. Xavier, T. J. Blacklock,
Reduction of (1R,E)-2-benzylidenecyclopentan-1-ol (E)-8
(with 92% ee) was carried out using the general procedure.
Following isolation of trans-9, HPLC analysis demonstrated
that the trans compound was obtained in 83% ee.
J. Chem. Soc., Perkin Trans. 1, 2000, 3047–3054
3053

View Article Online
Chem., 1996, 61, 8569; (b) S. K. Latypov, J. M. Seco, E. Quiñoá and
R. Riguera, J. Am. Chem. Soc., 1998, 120, 877.
19 A. Ijima and K. Takahashi, Chem. Pharm. Bull., 1973, 21, 215.
20 W. Hückel, M. Maier, E. Jordan and W. Seeger, Liebigs Ann. Chem.,
1958, 616, 46.
D. J. Mathre, P. Sohar, E. T. Turner Jones, R. A. Reamer,
F. E. Roberts and E. J. J. Grabowski, J. Org. Chem., 1991, 56, 763;
(c) D. J. Mathre, A. S. Thompson, A. W. Douglas, K. Hoogsteen,
J. D. Carroll, E. G. Corley and E. J. J. Grabowski, J. Org. Chem.,
1993, 58, 2880.
13 (a) P. Szeto, D. C. Lathbury and T. Gallagher, Tetrahedron Lett.,
1995, 36, 6957; (b) A. F. Simpson, P. Szeto, D. C. Lathbury and
T. Gallagher, Tetrahedron: Asymmetry, 1997, 8, 673.
14 J. Bach, R. Berenguer, J. Farras, J. Garcia, J. Meseguer and
J. Vilarrasa, Tetrahedron: Asymmetry, 1995, 6, 2683.
15 (a) R. W. Saalfrank, P. Schierling and P. Schätzlein, Chem. Ber.,
1983, 116, 1463; (b) S. Kagabu, Y. Shimizu, C. Ito and K. Moriya,
Synthesis, 1992, 830.
21 D. Varech, C. Ouannes and J. Jacques, Bull. Soc. Chim. Fr., 1965,
1662.
22 T. Severin and H. Poehlmann, Chem. Ber., 1978, 111, 1564.
23 R. J. K. Taylor, Synthesis, 1977, 564.
24 P. J. Smith, J. R. Dimmock and W. A. Turner, Can. J. Chem., 1973,
51, 1458.
25 B. Furth, J. P. Morizur and J. Kossanyi, C. R. Seances Acad. Sci.,
Ser. C, 1970, 271, 691.
16 (a) G. Fronza, G. Fogliato, C. Fuganti, S. Lanati, R. Rallo and S.
Servi, Tetrahedron Lett., 1995, 36, 123; (b) G. Fogliato, G. Fronza,
C. Fuganti, S. Lanati, R. Rallo, R. Rigoni and S. Servi, Tetrahedron,
1995, 51, 10231; (c) C. Fuganti, P. Grasselli, M. Mendozza, S. Servi
and G. Zucchi, Tetrahedron, 1997, 53, 2617; (d) N. J. S. Harmat and
S. Warren, Tetrahedron Lett., 1990, 31, 2743.
17 G. Uray, Houben-Weyl Methods in Organic Chemistry, ed.
G. Helchen, R. W. Hoffmann, J. Mulzer and E. Schaumann,
Stuttgart, 1996, vol. 1, p. 253.
26 S. V. Kelkar, A. A. Arbale, G. S. Joshi and G. H. Kulkarni, Synth.
Commun., 1990, 20, 839.
27 S. Mitsui, Y. Senda and H. Saito, Bull. Chem. Soc. Jpn., 1966, 39,
694.
28 B. M. Trost and T. A. Runge, J. Am. Chem. Soc., 1981, 103,
7550.
29 W. J. G. M. Peijnenburg and H. M. Buck, Tetrahedron, 1988, 44,
4927.
30 A. P. Phillips and J. Mentha, J. Am. Chem. Soc., 1955, 78, 140.
31 A. Martinez-Grau and D. P. Curran, Tetrahedron, 1997, 53, 5679.
18 (a) S. K. Latypov, J. M. Seco, E. Quiñoá and R. Riguera, J. Org.
3054
J. Chem. Soc., Perkin Trans. 1, 2000, 3047–3054