An investigative study of kinetic resolutions by the Sharpless asymmetric dihydroxylation reaction

Article abstract of DOI:10.1016/S0040-4020(01)00948-6

The requirements for a highly selective kinetic resolution with the Sharpless asymmetric dihydroxylation (AD) reaction were investigated with a number of alkene substrates. It was found that with 1-phenyl-4-tert-butylcyclohexene enantioselectivity is very high, yet diastereoselectivity is poor and kinetic resolution is ineffective. With 5-methyl-2-phenylbicyclo[3.2.0]heptan-2-ene both diastereoselectivity and enantioselectivity are high and kinetic resolution is effective. It was discovered that the transition state for the product-determining step in the Sharpless AD reaction is not product-like, and effective kinetic resolution can occur when one face of a chiral alkene is hindered.

Full text of DOI:10.1016/S0040-4020(01)00948-6

TETRAHEDRON
Pergamon
Tetrahedron 57 -2001) 9499±9508
An investigative study of kinetic resolutions by the Sharpless
asymmetric dihydroxylation reaction
David P. G. Hamon, Kellie L. Tuckp and Hamish S. Christie
The Department of Chemistry, The University of Adelaide, Adelaide, SA 5005, Australia
Received 18 June 2001; accepted 27 September 2001
AbstractÐThe requirements for a highly selective kinetic resolution with the Sharpless asymmetric dihydroxylation -AD) reaction were
investigated with a number of alkene substrates. It was found that with 1-phenyl-4-tert-butylcyclohexene enantioselectivity is very high, yet
diastereoselectivity is poor and kinetic resolution is ineffective. With 5-methyl-2-phenylbicyclo[3.2.0]heptan-2-ene both diastereoselectivity
and enantioselectivity are high and kinetic resolution is effective. It was discovered that the transition state for the product-determining step
in the Sharpless AD reaction is not product-like, and effective kinetic resolution can occur when one face of a chiral alkene is hindered.
q 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
The Sharpless asymmetric dihydroxylation -AD)1 and
asymmetric epoxidation -AE)2 reactions have proved to be
very effective means whereby asymmetry can be introduced
into molecules starting from prochiral alkenes and allylic
alcohols, respectively. The AE reaction has also proved
useful for effecting kinetic resolution of chiral allylic
alcohols. However, the AD reaction has only been
Figure 1.
successfully applied to effect kinetic resolutions in a limited
number of cases and the reasons for this are not well
understood.1
accelerating ligand) gave exclusively the diastereomer 4
which resulted from attack on the face trans to the two
methyl groups.4 When the chiral ligand, -DHQD)2-PHAL
Investigations by Sharpless and coworkers demonstrated
was used the diol 4 formed in an 86% ee, after a 23%
conversion. Its enantiomer was obtained in greater than a
95% ee, after 40% conversion, when the chiral ligand was
-DHQ2)-PHAL4 -Scheme 1).
that kinetic resolution of some axially chiral alkenes is
possible through application of the asymmetric dihydroxyl-
ation technology.3 Reaction of the alkenes 1 and 2 with the
commercially available a and b AD-mixes resulted in
kinetic resolution. The krel values for the enantiomers of 1
with AD mix-b and AD mix-a was 9.7 and 5.0, respectively
and those for the enantiomers of 2 were 32.0 and 26.5,
respectively. It was reasoned that the kinetic resolution
observed in these experiments is the result of dihydroxyl-
ation of the faster reacting ole®n from the intrinsically
disfavoured diastereoface -Fig. 1).
In 1996 we explored5 the kinetic resolution of 1-phenyl-4-
tert-butylcyclohexene 5 as a method to make chiral
auxiliaries and found that it was ineffective. This led us to
explore this area in more detail and the results obtained were
communicated earlier.6 Herein we describe the full
experimental details for those studies.
It has also been discovered recently that the dihydroxylation
reaction of the alkene 3 with quinuclidine -an achiral
Keywords: Sharpless asymmetric dihydroxylation; kinetic resolutions; tran-
sition state.
Corresponding author. Present address: School of Pharmaceutical,
p
Molecular and Biomedical Sciences, University of South Australia,
Adelaide, 5000, Australia. Tel.: 161-8-8302-2301; fax: 161-8-8302-
2389;
Scheme 1.
e-mail: kellie.tuck@unisa.edu.au
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020-01)00948-6

9500
D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
2. Results and discussion
drawn) of enantiomer 5a whereas it will preferentially
attack the bottom face of enantiomer 5b -Scheme 3). It is
therefore expected that, if there is very high facial
selectivity of the reagent for the alkene and the reaction is
allowed to go to completion, that both the diols 6 and 7
would be produced in equal quantities.10 Under these condi-
tions the ratio of the diols 6 and 7 indicates nothing about
the relative rates of attack of the AD mix from the two sides
of the alkene 5. Initially when the reaction was carried to
completion at 08C equal qualities of the diols 6 and 7 were
2.1. Kinetic resolution of the alkene 5
The AD reaction of 1-phenylcyclohexene proceeds with
high asymmetric induction,3 99% ee with AD mix-b and
97% ee with AD mix-a. It was therefore reasoned that
reaction of these reagents with 1-phenyl-4-tert-butylcyclo-
hexene 5 should also proceed with high asymmetric induc-
tion. However, in this case two diastereomeric products
would form. For example with AD mix-b the enantio-
enriched diastereomers 6 and 7 would be expected -Scheme 2).
1
produced, as determined by integration of the H NMR
resonances of the protons on the carbon bearing the second-
ary hydroxyl group. The diols 6 and 7 were obtained in an
84% yield after separation by ¯ash column chromatography,
with care taken so as to not optically enrich the diols.
1H NMR chiral shift reagent experiments on the correspond-
ing racemic mono acetates of the separated diols 6 and 7 was
not useful. However, similar experiments with the racemic
diols themselves gave baseline separation of the tert-butyl
Scheme 2.
²
resonances. This observation was useful for two reasons;
each tert-butyl resonance is a singlet, and this resonance is
1
the largest peak for each enantiomer in the H NMR spec-
If these molecules exist predominately in the expected chair
conformations, then due to the ring locking effect of the tert-
butyl substituent it can be seen that diastereomer 6 has the
phenyl group in an equatorial position whereas the phenyl
group in the other diastereomer 7 has an axial orientation.
Molecular modeling -gaussian94) revealed that diastereo-
mer 7 was 15.9 kJ mol21 higher in energy than the
diastereomer 6. It was thus reasoned that if the transition
state for the product-determining step in the AD reaction is
product-like, there would be a considerable difference in the
rate of formation of the two diastereomers, and an effective
kinetic resolution should be possible.
trum. Therefore, with the optically enriched diols the ratio
of the enantiomers can be integrated accurately. In this way
it was determined that the optical purity of 6 was greater
than 99% ee since the tert-butyl resonance due to the minor
enantiomer was not visible in the spectrum whereas one
13C±H satellite of the tert-butyl resonance of the major
isomer was. The addition of 1% of the racemic diol in this
experiment gave a discernible peak for the minor enantio-
mer which was of comparable intensity to this 13C±H
satellite. The optical purity of diol 7 could only be deter-
mined to be greater than 95% ee due to distortion of the
baseline. Addition of 5% of the racemate was required
before a discernible peak for the tert-butyl resonance of
The required racemic diols7 were formed by two methods;
by reaction of 5 with OsO4 and NMO, and also by reaction
with a modi®ed Sharpless dihydroxylation reaction where
quinuclidine was used as an achiral accelerating ligand.8
Both methods produced a mixture of approximately 4:1
ratio of the diols 6 to 7.
the minor enantiomer could be readily observed. By the
Â
10
arguments of Gutte and Horeau, if two diastereomers are
formed in equal quantities then the optical purity of each
diastereomer must be equal. Within the experimental limits,
these chiral shift reagent 1H NMR experiments con®rm this
expectation.
The AD reaction on the alkene 5 was performed following
Sharpless's recommended procedure9 for tri-substituted
alkenes. It is expected, from Sharpless's mnemonic8 that
the AD mix-b will preferentially attack the top face -as
A kinetic resolution with the AD reaction was then studied
under the normal conditions and the relative rates for the
enantiomers -E or krel) were determined following a litera-
ture procedure11 which is related to the relationship derived
by Kagan.12 During the reaction, the percentage conversion
of the racemic alkene 5 was determined by NMR spectro-
scopy. This was done by removing aliquots, which were
quenched with a solution of sodium sul®te. The disappear-
ance, in the NMR spectrum, of the vinyl proton of alkene 5
with the appearance of the protons on the carbon bearing the
secondary hydroxyl group of the diols 6 and 7 were used to
make this determination. The alkene remaining was
separated from the diols 6 and 7 and this alkene 5 was
converted, by the achiral dihydroxylation reaction, with
quinuclidine, into the diols 6 and 7. The enantiopurity of
the major diol 6 from this reaction, and hence that of the
²
Contrary to our expectation and previous observations that tightly
complexed compounds give too much line broadening in high ®eld
NMR to be useful.
Scheme 3. Conditions: -a) AD mix-b, MeSO2NH2, tert-BuOH, H2O, 08C,
24 h, 84%.

D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
9501
alkene 5, was then determined as described earlier. Two
separate experiments were conducted at 08C and the
reactions were stopped at 15 conversion and 94% conver-
sion, respectively. For the former reaction the ratio of
diasteremores 6 and 7 was ,2:1, for the later, ,52:42.
The ee for the remaining alkene 5, for the former reaction,
was determined as ,4% and, for the latter, as ,86%. This
computes to an E of ,2in each case. This small E difference
means that the kinetic resolution is ineffective as a means of
obtaining the pure enantiomers.
mer 8b would be expected to react faster and it was expected
that a kinetic resolution should be effective -Scheme 4).
It is known13 that rearrangement of the racemic diol 6 with
boron tri¯uoride etherate gives the ketone 10 as the major
product. Treatment of the mixture with acid -HBr, acetic
acid) changes the ratio to 1:10 now in favour of the more
stable equatorial alkene 11. We ®nd that pure ketone 10 can
be obtained in reasonable quantities directly after rearrange-
ment by puri®cation of the mixture by `¯ash' column
chromatography. The time that the compound spends on
the column must be kept to a minimum to obviate isomer-
isation to the more stable isomer.
The small E value obtained shows that the energy difference
of the products is not re¯ected in the transition state of the
product-determining step. Therefore, this transition state is
not product like. Consequently an early -the starting alkenes
are energetically identical) rather than a late transition is
indicated for the product-determining step.
The stereochemistry for each of the two products 10 and 11
was con®rmed by comparison of the coupling constants for
1
the benzylic protons in the H NMR spectrum. The dia-
stereomer with the axial phenyl group 10 has the resonance
due to the benzylic proton at d 3.71 ppm -t, Jˆ4.5 Hz), the
small couplings suggesting an equatorial proton. The isomer
11 has the benzylic resonance at d 3.54 ppm -dd, Jˆ5 and
8 Hz), the larger coupling being consistent with an axial
orientation for that proton.
It can be reasoned that, if the product-determining step in
the AD reaction is reactant-like and if the alkene is highly
enantiofacially directing, then a kinetic resolution should be
effective when there is considerable difference in the ease of
approach to the two faces of a chiral alkene. This reasoning
can be applied to the example where such an effect has been
observed recently4 for a molecule 3 which displays axial
chirality. We therefore proposed to investigate this further,
®rst with the alkene 8 and then, as things turned out, with the
alkene 9 -Fig. 2).
Reaction of the ketone 10 with phenylmagnesium bromide
gave mostly one alcohol product -Scheme 5), which was
tentatively assigned the structure 12 by the following
reasoning. It was anticipated that the alcohol 12 might
form preferentially to the alcohol 13 due to the steric block-
1
ing effect of the axial phenyl substituent. The H NMR
spectrum has a resonance due to the benzylic proton at d
3.47 ppm -dd, Jˆ4 and 10 Hz). The coupling constant of
1
10 Hz implies that this proton is axial. The H NMR spec-
trum of the crude reaction mixture showed a small amount
of the starting ketone 10 but no sign of the isomeric ketone
11. This suggested that no isomerisation of the ketone 10
occurred under the Grignard conditions. The presence of
axial benzylic proton is best rationalized by assuming a
twist boat conformation -Fig. 3) which would allow all
Figure 2.
2.2. Synthesis of the alkene 8
With the diol 6 in hand it was expected that alkene 8 might
easily be synthesized. Alkene 8 was expected to be very
effectively resolved by the Sharpless AD reaction. The
tert-butyl group locks one phenyl group in the axial position
and an inspection of a molecular model reveals that the
alkene 8 is effectively blocked from attack adjacent to the
axial phenyl group. Therefore, with AD mix-b the enantio-
Scheme 5. Conditions: -a) BF3´Et2O, benzene, 258C, 3 h, 54% -b) PhMgBr,
Et2O, 258C, 40 min, 83% -c) MgSO4, CH2Cl2, H2SO4, 2 58C, 20 min, 80%.
Scheme 4.
Figure 3.

9502
D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
1
epimerisation of the ketone 10 to the ketone 11. The H
NMR spectrum of the major product has a benzylic
resonance at d 3.03 ppm -dd, Jˆ13 and 4 Hz). The large
coupling -13 Hz) suggests that the proton is axial and hence
the phenyl group has remained equatorial. The rest of the
spectrum does not indicate whether the new phenyl
group has entered cis -15) or trans -16) -Scheme 7). As
the alcohol was only an intermediate, its con®guration
was not pursued.
Scheme 6. Conditions: -a) MgSO4, CH2Cl2, H2SO4, 2 58C, 20 min, 80%.
three large substituents to occupy equatorial positions. This
conformation puts the benzylic proton in a pseudo axial
position, which would account for the observed large
coupling.
Dehydration of this alcohol 15 or 16 gave mainly
three alkenes -Scheme 7). The ratio of these alkenes was
14/17/18 -3:1:1) -determined by integration of the tert-butyl
proton signals in the 1H NMR spectra). The two alkenes 14
and 17 were identi®ed with some certainty, however the
structure of 18 was only tentatively assigned. In the spec-
trum of the crude mixture it was observed that there were
two different ole®nic signals, a broad singlet at d 6.23 ppm
which was assigned to 17 and a dd at d 6.46 ppm -Jˆ3 and
5 Hz) which by deductive reasoning was assigned to 18. The
synthesis of the stilbene 14 in a larger amount allowed its
full spectral characterisation and con®rmed that it was the
minor product formed by dehydration of the alcohol 12.
This initial mixture of alcohols 12 and 13 could be puri®ed,
with dif®culty, by ¯ash chromatography to give pure 12,
however the crude mixture could be used in the next
reaction, dehydration, without puri®cation.
Dehydration of the crude tertiary alcohol 12, under
conditions which could favour the kinetic product, was
accomplished with concentrated sulfuric acid absorbed in
anhydrous magnesium sulfate.14 In the reaction, two alkenes
are formed in a ratio of 10:1. The major product was
obtained after puri®cation of the mixture by ¯ash chromato-
graphy. The 1H NMR spectrum of the major product has two
characteristic resonances, a 1H broad singlet at d 6.47 ppm
assigned to the vinylic proton, and a 1H multiplet at d
4.00 ppm consistent with the benzylic proton. This benzylic
proton resonance has only small couplings, suggesting that
it is equatorial and this is consistent with the expected
structure 8 -Scheme 6).
The ratio of the three alkenes -14, 17 and 18) varied with
reaction time. After 40 min -the time required for the
dehydration reaction to be complete) the ratio of the alkenes
was 14/17/18 -3:1:1); however, after 5 days the ratio of the
alkenes had changed and now the alkene 8 had formed. The
ratio of the products was now 14/17/18/8 -4.5:1:1.7:1).
The AD reaction on the alkene 8 using Sharpless's recom-
mended procedure,8 at 08C, gave no reaction. After several
days a large number of low Rf products were observed on
The minor product was identi®ed as the stilbene 14 by the
lack of benzene or alkene resonances in its 1H NMR
spectrum. In order to con®rm the identity of the stilbene
14 it was formed by a different route. The ketone 11 was
carried through a similar series of reactions used to form the
alkene 8. One major alcohol product formed from the
Grignard reaction. The H NMR spectrum of this major
product is quite different to that of the alcohol 12. It is
clear that the alcohol 12 could not have formed by
³
TLC. It appeared that the alkene 8 was not soluble in the
reaction mixture, therefore, suf®cient dichloromethane was
added to ensure that the alkene 8 was dissolved, but still no
reaction occurred. Reaction of the alkene 8 under similar
conditions with quinuclidine, which has less steric bulk than
the AD ligands, gave similar results. This interesting new
alkene 8 seems to be largely unreactive under Sharpless's
standard dihydroxylation conditions. It is likely that this is
due to too much steric hindrance at the double bond.
1
2.3. Synthesis and kinetic resolution of the alkene 9
Recently we completed15 an asymmetric synthesis of
grandisol in which the key step is the kinetic resolution of
the primary allylic alcohol 19 using a Sharpless AE reaction.
The key feature, in that resolution, is that the molecule
presents both an open convex face and a more-hindered,
concave face to reagents. The reaction proceeds to give
products of attack on one face only, the convex face. This
cis 4,5 ring system appeared suitable to pursue the study of
the kinetic resolution by the AD reaction. Indeed, the alkene
9 appears ideal for this study -see Fig. 2for structures).
Control of product formation in the Sharpless AD reaction
should be determined mainly by the enantiofacial selectivity
of the styrene moiety. The concave face of this molecule is
³
Scheme 7. Conditions: -a) PhMgBr, Et2O, 258C, 1 h, 86% -b) MgSO4,
CH2Cl2, H2SO4, 2 58C, 1 h, 35% -14).
The pattern was similar to the decomposition of the alkene 8 on standing
alone.

D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
9503
be assigned to the other diol were observed in the 1H NMR
spectrum of the product.
A kinetic resolution with the AD reaction was then studied
under the normal conditions and the relative rates -E) of the
reaction for the enantiomers were determined following a
similar procedure to that described earlier.11 The kinetic
resolution of this substrate was examined in two separate
experiments with 26 and 50% oxidant, respectively.
However, because the alkene 9 is somewhat unstable, the
percentage conversion could not be determined accurately.
Therefore, in each case the ee for both the diol -presumed to
be enantiomer 23a, based on the application of Sharpless's
mnemonic) and the recovered alkene 9 were determined,
after separation in the following way. The diol 23a was
converted to the mono Mosher ester derivative and the ee
Figure 4.
hindered, so there should be a considerable difference in the
rate of reaction of the enantiomers of this molecule and, by
our reasoning a kinetic resolution should be effective.
1
Reaction of the bicyclic ketone 20 with phenylmagnesium
bromide gave a single racemic alcohol. However, the struc-
ture is presumed to be that of diastereomer 21, formed by
attack from the unhindered top face of the ketone 20. No
resonances corresponding to other diastereomers could be
seen in the 1H NMR spectrum. Unfortunately, crystals
suitable for X-ray analysis could not be obtained. Inspection
of a molecular model reveals that the bicyclic ketone 20 has
an open accessible convex face and also a much more
hindered concave face -Fig. 4). The observation that attack
of a reagent comes preferentially from the top face in this
system has been argued previously for attack of dimethyl-
sulfonium methylide --CH3)3S1I2)) on the bicyclic ketone
20 and for attack of MCPBA on the corresponding methyl-
ene compound 22.15
was determined by H NMR spectral analysis. It was ®rst
shown that the 1H NMR spectrum of the diastereomers from
the racemic diol gave base-line separation for the methoxy
resonances at d 3.37 and 3.53 ppm. The product from the
kinetic resolution showed the peak at d 3.37 ppm as the
major resonance. The alkene from the kinetic resolution
was converted to the diol, presumed to be mainly enantio-
mer 23b, by the achiral dihydroxylation reaction with
quinuclidine as the ligand. This diol was then converted to
the mono Mosher ester derivative from which the ee was
obtained. The peak at d 3.53 ppm was now the major
peak.
The ee for the diols in the two reactions were 88 and 80%
respectively, and for the alkenes they were 52and 80%,
respectively. From these results the enantiomeric ratios for
these two reactions calculate to Eˆ26 and 21, respec-
tively.16 These are the same within experimental error and
this result implies an effective kinetic resolution giving
.95% ee in the remaining alkene after 60% conversion.
Dehydration of the tertiary alcohol 21 gave the required
alkene 9 as an oil -scheme 8). The alkene 9 was somewhat
unstable and initially the yield for the elimination reaction
was only 50%. However, addition of the radical inhibitor
2,6,-di-tert-butyl-p-cresol during the work-up process,
increased the yield to greater than 80%, presumably by
preventing free-radical polymerisation of the alkene.
2.4. Kinetic resolution of the methyl alkene 24
The allylic alcohol 19 undergoes very effective kinetic reso-
lution under the conditions of the Sharpless AE reaction.15
The methyl alkene 24 presents very similar overall steric
requirements to those of the allylic alcohol 19, therefore, it
was of interest to compare the effectiveness of this alkene to
kinetic resolution in the Sharpless AD reaction. It has been
shown earlier15 that very high diastereoselection occurs in
the achiral dihydroxylation of the alkene 24. Two reactions
were conducted at 08C and stopped at 26 and 24% conver-
sion. The ee of the product was determined by conversion to
the corresponding Mosher esters. In the former reaction the
enantiomeric excess of the product was 23%, in the latter the
enantiomeric excess was 24%. These results translate to a E
value of only 1.8. These results show that a kinetic
resolution was ineffective. The methyl group is not very
enantioselectively enhancing.
Dihydroxylation of the alkene 9 with quinuclidine as the
ligand, gave only one diastereomer. This was presumed to
be the diol 23 formed from attack of the reagent from the
unhindered top face -scheme 9). No resonances which could
Scheme 8. Conditions: -a) PhMgBr, Et2O, re¯ux, 1 h, 88% -b) MeSO2Cl,
Et3N, CH2Cl2, 08C, 4 h, 81%.
The question arises, why the AE reaction is often effective
in the kinetic resolution whereas the AD reaction is usually
ineffective? The obvious difference between these reactions
of allylic alcohols and alkenes in the complexing ability of
the hydroxyl group in the alcohols, and the requirement for
prior co-ordination of this group with the Ti centre for the
AE reaction to proceed.2 That is, the diastereomeric
Scheme 9. Conditions: -a) K3Fe-CN)6, K2CO3, quinuclidine, OsO4,
MeSO2NH2, tert-BuOH, H2O, 258C, 24 h, 50%.

9504
D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
transition states which control both the diastereoselection
and the enantioselection in the AE reaction are intra-
molecular. No such strong prior co-ordination occurs with
the AD reaction in these cases. In this case, if the reagent
can approach closely enough to the double bond, reaction
will take place and enantioselectivity will be determined
by the substituents on the double bond. That is, when
co-ordination to the double bond does occur it is the
environment local to the double bond, which dictates the
outcome. Diastereoselectivity is however, only controlled
by the initial ease of approach to each side of the double
bond and is controlled by the gross features of the molecule.
In broad terms the situation is somewhat analogous, but
more complicated, to the situation pertaining to epoxide
formation from the reaction of ketones with sulfur ylids.17
In those cases the diastereomeric products from the
reaction of the reactive ylid are controlled by the initial
approach of the reagent, whereas the products from the
reaction of the less reactive ylid are controlled by intra-
molcular interactions.
The product was distilled under reduced pressure -1008C/
0.4 mmHg) to give mainly the required alkene -83%)
and redistilled to give the pure alkene, 54% yield. m/z:
214 -M1, 65%), 57 -100%). dH -200 MHz): 0.92 -s, 9H),
1.2±1.6 -complex, 7H), 6.12 -m, 1H), 7.1±7.4 -complex,
5H). dC -50 MHz): 25.1, 27.9, 29.6, 31.5, 32.9, 44.5,
125.6, 126.0, 127.2, 128.8, 137.1, 142.9. nmax: 2690, 1395,
1365 cm21
.
3.1.2. 1-Phenyl-4-tert-butylcyclohexan-1,2-diols +6) and
+7). The general procedure outlined by Sharpless for phenyl-
cyclohexene20 was used as follows.
A mixture of AD mix-b -1.40 g), tert-BuOH -5 mL), H2O
-5 mL), and MeSO2NH2 -0.095 g) was cooled to 08C, where
upon some of the dissolved salts precipitated. Racemic 5
-139 mL, 1 mmol) was added and the reaction mixture stir-
red with a mechanical stirrer -500 rpm) for 24 h at 08C, after
which time TLC analysis indicated a lack of starting alkene.
Sodium sul®te -1.5 g) was added and the mixture stirred for
1 h while it came to room temperature. EtOAc -10 mL) was
added and after separation of the layers, the aqueous layer
was re-extracted with more EtOAc -3£5 mL). The
combined extracts were washed with 2N KOH -15 mL),
dried -MgSO4) and the solvent was removed in vacuo.
Separation of the diols from the ligand was effected by
¯ash chromatography -CH2Cl2/EtOAc, 90/10 v/v) to give
a mixture of the diols -0.209 g, 84% yield).
In conclusion we ®nd with the tert-butyl alkene 5 enantio-
selectivity to the double bond is very high due to the
enantioselectivity enhancing phenyl moiety but diastereo-
selectivity is poor -,2:1), attack can occur reasonably well
from either side of the double bond and kinetic resolution is
ineffective. With the methyl alkene 24 diastereoselectivity
is high, reaction occurs only from one side of the double
bond but enantioselectivity is poor, as the methyl group is
not very enantioselectivity enhancing, and kinetic resolution
is ineffective. However, with the phenyl alkene 9 both dia-
stereoselectivity and enantioselectivity are high and kinetic
resolution is effective. These results are consistent with an
early, rather than a late transition state for the product
determining step in the AD reaction.
$1R,2R,4R)-1-tert-Butyl-4-phenylcyclohexan-1,2-diol -6):
This diol was the ®rst eluted of the two diols from normal
phase chromatography. Recrystallized from hexanes/
CH2Cl2 mp 152±1538C. m/z: 248 -M1, 15%), 133
-100%); 105.-27%). dH -200 MHz): 0.93 -s, 9H), 1.0±2.0
-complex, 7H), 1.55 -s, 1H, OH), 2.58 -d, 1H, OH), 4.02
-ddd, 1H, Jˆ11, 4 and 3 Hz), 7.2±7.6 -m, 5H). Exchange
with D2O revealed the OH resonances and changed the ddd
at 4.02ppm to a dd -11, 4 Hz). dC -50 MHz): 22.1, 27.6,
30.6, 32.4, 38.5, 46.6, 75.3, 75.4, 125.1, 127.0, 128.5, 146.3.
3. Experimental
n
mzx: 3290 cm21 -OH). [a]Dˆ238 1/2 0.58 -cˆ5, EtOH).
3.1. General Procedure
Chiral shift analysis of the diol 6 -8 mg) with Eu-hfc)3
-10 mg) in 15% C6D6±CCl4 gave a 99% ee. Separation of
the But protons' resonance due to the minor enantiomer was
not visible by proton NMR, but the 13CH satellite of the But
protons' resonance of the major isomer was. Addition of 1%
of the racemate gave a discernible peak for the minor
enantiomer.
Melting points were taken on a Ko¯er hot stage apparatus
under a Reichert microscope and are uncorrected. 1H NMR
spectra were recorded in CDCl3 on a Varian spectrometer at
200, 300 or 600 MHz, as indicated. Optical rotations were
measured with a Perkin±Elmer 141MC Polarimeter. MS
were recorded on a VG ZAP 2HF mass spectrometer,
GC±MS were measured on a Finnigan MAT GCQ spectro-
meter operating at an ionization energy of 70 eV. Elemental
analyses were carried out at the University of Otago,
Dunedin, New Zealand. Flash chromatography was
performed with Merck Kieselgel 60 -230±400 mesh
ASTM). TLC was performed with Merck DC Alufolien
Kieselgel 60f254 which were visualized either with UV
light or by staining with acidic ammonium molybdate
solution. Organic solutions were dried over Na2SO4 or
MgSO4. Other anhydrous solvents and reagents were
prepared according to standard laboratory procedures.18
All calculations were performed with Gaussian 94 -revision
C), Gaussian, Inc., Pittsburgh, PA., 1995.
$1R,2R,4S)-1-tert-Butyl-4-phenylcyclohexan-1,2-diol -7):
eluted second from normal phase chromatography of the
mixture of two diols. Recrystallized from MeOH/H2O mp
101±1028C. m/z: 248 -M1, 5%); 133 -100%); 105.-38%).;
55.-98%). dH -300 MHz): 0.80 -s, 9H), 1.0±2.3 -complex,
2H), 1.55±1.70 -complex, 1H), 1.92 -m, 1H, Jˆ
14.1 Hz1smaller coupling), 2.07 -dt, 1H, Jˆ4.2, 13.5 Hz),
2.20±2.35 -complex, 2H), 2.77 -t, 1H, OH), 4.37 -br. s, 1H),
7.20±7.60 -complex, 5H). Exchange with D2O revealed the
OH resonances and changed the multiplet at 4.38 ppm
demonstrating that a small coupling to OH was present.
dC -75.5 MHz): 24.5, 27.5, 30.1, 31.9, 32.5, 39.6, 72.8,
74.5, 126.6, 127.8, 128.6, 143.3. nmax: 3290 cm21 -OH).
[a]Dˆ21481/218 -cˆ2, EtOH). Chiral shift analysis of
the diol 7 -8 mg) with Eu-tfc)3 -6 mg) in 15% C6D6±CCl4
3.1.1. +4RS)-1-Phenyl-4-tert-butylcyclohexene +5). The
method of Garbisch was followed, on a smaller scale.19

D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
9505
gave .95% ee. Addition of 5% of the racemate was
required before a discernible peak for the tert-Bu protons'
resonance of the minor isomer could be detected.
measured using the chiral shift analysis experiment
described earlier.
For the 15% conversion experiment the ratio of 6 to 7 was
,2:1. The ee for the remaining alkene 5, was ,4%. The
calculated E was 1.8. For the 85% conversion the ratio was,
,52:42, the ee for the remaining alkene was ,86%. The
calculated E was 2.0.
3.1.3. Racemic diols +1RS,2RS,4RS)-1-tert-butyl-4-
phenylcyclohexan-1,2-diol +6) and +1RS,2RS,4SR)-1-
tert-butyl-4-phenylcyclohexan-1,2-diol +7). Quinuclidine
procedure: The procedure followed was similar to that
described earlier except that for 1 mmol of alkene
K3Fe-CN)6 -0.98 g, 3 equiv.), K2CO3 -0.41 g, 3 equiv.),
quinuclidine -0.001 g, 0.01 equiv.), and OsO4 -20 mL,
4 wt% solution 0.002equiv.) replaced the AD mix. A single
recrystallisation of the mixture from hexanes/CH2Cl2
yields pure diol 6. Alternatively, separation can be effected
using ¯ash chromatography. The ratio of the two diols was
3.1.4. +2RS,5SR)-5-tert-Butyl-2-phenylcyclohexanone +10)
and +2RS,5RS)-5-tert-butyl-2-phenylcyclohexanone +11).
A mixture composed mostly of these ketones was generated
1
by the method of Barili et al.22 Integration of the H NMR
singlet resonances for tert-Bu protons and also for the
benzylic protons gives an estimated ratio of 10:11 -2:1)
-variable between trials). After a trial using -1R,2R,4R)-1-
tert-butyl-4-phenylcyclohexan-1,2-diol 7 -1.42g), boron
tri¯uoride etherate -1.4 mL) in benzene -140 mL) puri®ca-
tion of the crude product -1.28 g) by ¯ash chromatography
-EtOAc/hexanes, 7:93 v/v) gave -2RS, 5RS)-5-tert-butyl-2-
phenylcyclohexanone -®rst eluting) 10 as a colourless oil
-583 mg, 54%). dH -200 MHz): 0.85 -s, 9H), 1.50±2.60
-complex, 7H), 3.71 -t, 1H, Jˆ4.2Hz), 7.20±7.40
-complex, 5H). dC -50 MHz): 22.2, 27.0, 28.7, 32.7, 41.2,
48.7, 53.3, 126.8, 127.3, 128.8, 138.1, 189.5.
1
6:7 -4.5:1), as determined by the H NMR spectrum and
mass recovery. dH -200 MHz), dC -50 MHz) and other
spectral properties are identical to the optically active 6
and 7. Mp 6 153±153.58C, 7 101±1028C, -lit7 150±
151.58C 6, 101±102.58C 7).
NMO procedure: With some modi®cations to the method of
Van Rheenen,21 alkene 5 -0.14 mL, 1.0 mmol) and an
aqueous solution of 41 mg/ mL OsO4 -0.2mL, 7 mg,
0.03 mmol) were added to NMO -0.206 g, 1.76 mmol) in
H2O -0.5 mL) and acetone -5 mL).and the reaction mixture
stirred vigorously until TLC analysis indicated no more
starting material -approx. 85 h). A slurry of ¯uorisil
-0.13 g) and sodium hydrosul®te -0.03 g) in water -2mL)
was added and the mixture stirred for 10 min at ambient
temperature. The mixture was ®ltered through a pad of
celite which was washed with additional acetone
-3£10 mL). Solvent was removed under reduced pressure
and the residue was taken up in water -10 mL). This solution
was extracted with CH2Cl2 -3£10 mL), the combined
organic extracts were dried and solvent was removed
under reduced pressure to give the diols -0.21 g). Separation
of the diols was achieved by ¯ash chromatography -EtOAc/
CH2Cl2, 10/90 v/v), the ratio of the two diols was 6:7
Second eluting is a mixture of -2RS, 5RS)-5-tert-butyl-2-
phenylcyclohexanone -10) and -2RS, 5SR)-5-tert-butyl-2-
phenylcyclohexanone -11) as an oily solid -494 mg, 46%).
This mixture was equilibrated following the method of
Barili et al.22 Recrystallisation of the equilibrated mixture
from hexanes yields pure -2RS, 5RS)-5-tert-butyl-2-phenyl-
cyclohexanone 11, mp 96±97.58C. m/z: 230 -M1 64%); 173
-29); 91 -100). dH -200 MHz): 0.94 -s, 9H), 1.5±2.7
-complex, 7H), 3.54 -dd, 1H, Jˆ5.1 and 8.1 Hz), 7.05±
7.40 -complex, 5H). dC -50 MHz): 27.6, 27.9, 33.5, 35.09,
44.8, 51.02, 57.9, 127.6, 129.0, 129.4, 139.5, 203.4. nmax
.
:
1714 -CvO) cm21
1
-3.8:1), as determined by H NMR spectroscopy and mass
recovery. All spectral properties were identical to those
listed earlier for the optically enriched isomers.
3.1.5.
+1RS,2RS,5RS)-5-tert-Butyl-1,2-diphenylcyclo-
hexan-1-ol +12). To a stirred solution of PhMgBr -1.5 mL,
0.85 M, 1.25 mmol) under an N2 atmosphere was added
-2RS,5RS)-5-tert-butylcyclohexanone 10 -238 mg 1.03
mmol) in Et2O -4 mL) via syringe over a period of 10 min
the mixture was stirred for a further 30 min. Excess sat.
NH4Cl-aq.) was added and the mixture stirred for 5 min.
The layers were separated and the ether washed again
with NH4Cl solution. The organic layer was dried over
CaCl2 and the solvent removed to yield the crude alcohol
product -325 mg) as a colourless oil. Puri®cation by ¯ash
chromatography -EtOAc/hexanes, 6/94 v/v) gave the title
compound as a colourless oil. m/z: 308 -M1). dH
-300 MHz): 0.89 -s, 9H), 1.60±1.90 -complex, 6H), 2.10±
2.30 -complex, 2H), 3.47 -dd, 1H, Jˆ3.6 and 9.9 Hz); 7.00±
7.50 -complex, 10H). dC -75.5 MHz): 23.1, 25.8, 27.2, 32.6,
39.7, 41.6, 48.5, 76.0, 125.2, 126.6, 126.6, 128.0, 128.0,
129.6, 141.1, 148.5. nmax -neat): 3558, 3466 -OH),
In general, these reactions were run as detailed earlier for
initial formation of 6 and 7 with the AD mix-b, however, the
scale and the time of quenching were varied. All reactions
1
were conducted at 08C. The reaction was monitored by H
NMR spectroscopy. Aliquots -0.2mL) were removed at
discrete time intervals and quenched with sodium su®te
solution -0.5 mL, 1.5 g/5 mL). The sample was extracted
with EtOAc -3£5 mL), the combined extracts washed
with 2N KOH-aq) -0.5 mL) and dried. Solvent was
1
removed in vacuo and the samples were analysed by H
NMR spectroscopy.
Once the desired degree of conversion had been reached
-determined from the 1H NMR spectrum), the reaction
was quenched and worked up as described earlier. The
alkene 5 was separated by short column chromatography
CH2Cl2/petroleum spirit -5:95, v/v), it was converted to
the corresponding diols 6 and 7 using the achiral
dihydroxylation reaction with quinuclidine as the achiral
ligand. After separation the ee of the diol 6 was then
1601 cm21
.
3.1.6. +3RS,6RS)-3-tert-Butyl-1,6-diphenylcyclohex-1-ene
+8). To a stirred suspension of anhydrous MgSO4 -0.178 g,
4 equiv) in CH2Cl2 -1.5 mL) was added conc. H2SO4
-1 equiv., 20 mL).23 The suspension was allowed to stir for

9506
D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
20 min and then crude alcohol 12 -100 mg, 0.325 mmol)
was added in CH2Cl2 -1 mL). After 2h the MgSO 4/H2SO4
was ®ltered off and the solvent removed under reduced
pressure to yield a colourless oil. Integration of the tert-
Puri®cation by ¯ash chromatography -EtOAc/hexanes
0.3/99.7 v/v) gave the title compound 14 -eluted ®rst)
-23 mg 35%). The compound was further puri®ed by
recrystallisation -methanol) mp 108±1098C. Anal. Calcd
for C22H26: C, 90.98; H, 9.02. Found: C, 91.24; H, 9.31.
m/z: 290 -100% M1); 233 -±But); 115; 91; 41. dH
-300 MHz): 0.95 -s, 9H), 1.20±1.70 -complex, 3H), 1.90±
2.65 -complex, 7H), 6.90±7.20 -complex, 10H). dC
-75 MHz): 24.5, 27.2, 32.3, 33.4, 33.8, 44.7, 125.7, 125.8,
127.6, 127.7, 129.1, 129.1, 134.8, 135.2 143.7, 144.1. nmax
1
Bu region of the H NMR spectrum of the crude product
demonstrates a ratio of 90:5 of the peaks at d 1.01 -title
compound 8) and d 0.94 -stilbene 14). Puri®cation using
¯ash chromatography -EtOAc/hexanes 0.3/99.7, v/v). First
was a white solid -4 mg, 4%) having identical 1H NMR data
to stilbene 14 -see Section 3.1.8). This was followed by the
title compound as a colourless oil -75 mg 80%). Anal. Calcd
for C22H26: C, 90.98; H, 9.02. Found: C, 90.10; H, 9.03. m/z:
290 -M1); 233. dH -300 MHz): 1.02-s, 9H), 1.28 -m, 1H),
1.51 -m, 1H), 1.90±2.20 -complex, 3H), 3.99 -m, 1H, width
at 1/2height ,8 Hz), 6.47 -m, 1H, width at 1/2height
,1.5 Hz), 6.90±7.60 -complex, 10H). dC -75.5 MHz):
18.2, 27.5, 32.4, 33.5, 41.9, 47.2, 125.8, 125.9, 126.7,
128.1, 128.2, 128.8, 130.5, 138.2, 142.1, 145.2. nmax
-nujol mull): 3074; 1599 cm21
.
As the spectral data of stilbene 14 was known with some
certainty the spectra of the other two alkenes could be
assigned. The slower eluting fraction contained
2
compounds, GC±MS con®rmed that they were isomeric.
Careful column separation gave the alkene 17 with minimal
contamination.
-neat): 3058; 2958; 2866; 2815; 1600 cm21
.
$3RS, 6SR)-3-tert-butyl-1,6-diphenylcyclohex-1-ene -17):
GC MS -E.I.) 290 -M1), 233 -±But). dH -300 MHz): 0.97
-s, 9H), 1.40±1.90 -complex, 4H), 2.10±2.30 -complex,
2H), 3.93 -m, 1H, width at 1/2 height 36 Hz), 6.23 -br. s,
1H), 7.00±7.30 -complex, 10H). dC -75 MHz): 23.2, 27.5,
33.5, 34.7, 44.5, 46.6, 125.6, 126.0, 126.5, 127.8, 128.2,
128.2, 130.9, 140.6, 142.8, 146.0.
3.1.7. +1RS,2SR,5RS)-5-tert-Butyl-1,2-diphenylcyclo-
hexan-1-ol +15) and/or +1SR,2SR,5RS)-5-tert-butyl-1,2-
diphenylcyclohexan-1-ol +16). To a stirred solution of
PhMgBr -1.90 mL, 1.05 M, 2mmol) under an N 2
atmosphere was added 11 -300 mg, 1.30 mmol) dissolved
in Et2O -4 mL). After stirring for a further 1 h, excess aq.
NH4Cl-sat) -20 mL) was added and the mixture stirred for
5 min. The layers were separated and the aqueous layer
extracted with Et2O -10 mL). The combined organic phases
were dried over MgSO4 and the solvent removed under
reduced pressure to yield the crude alcohol product as a
3.1.9. Attempted kinetic resolution of +3RS, 6RS)-3-tert-
butyl-1,6-diphenylcyclohex-1-ene +8). The general pro-
cedure of Sharpless outlined previously was followed, all
trials were conducted at room temperature. In some trials
dichloromethane was added to help solubilize the alkene.
Addition of alkene resulted in a bright green colour of the
organic layer, which faded after several hours. TLC after
various lengths of time indicated that the starting alkene 8
was present in greatest amount with a large number of minor
lower Rf products present. No dihydroxylation reaction was
apparent.
1
colourless oil. Integration of the tert-Bu region of the H
NMR spectrum of the crude mixture revealed the peaks at d
0.94 and d 0.92were in the ratio of 1:12. Puri®cation by
¯ash chromatography -EtOAc/hexanes 6/94, v/v) gave the
title compound as a colourless oil. Recrystallisation by
concentration of a hexanes solution gave an analytically
pure sample, mp 116±1278C. Anal. Calcd for C22H28O: C,
85.66; H, 9.15. Found: C, 85.49; H, 9.18. m/z: 308 -M1). dH
-300 MHz): 0.91, -s, 9H), 1.20±1.40 -complex, 1H), 1.60±
2.10 -complex, 6H), 2.25 -dq, 1H, Jˆ3.6 and 12.9 Hz), 3.04
-dd, 1H, Jˆ3.9 and 12.9 Hz), 6.90±6.95 -m, 2H), 7.00±7.30
-complex, 8H). dC -75 MHz): 27.4, 27.5, 28.7, 32.2, 41.9,
42.9, 52.6, 76.5, 114.8, 126.2, 126.3, 127.7, 127.8, 129.0,
141.4, 148.1. nmax -neat): 3564; 3470 -OH); 2949; 2866;
3.1.10. +1SR,5SR)-5-Methylbicyclo[3.2.0]heptan-2-one
+20). Compound 20 was synthesized following the method
of Cargill24 as reported earlier.15
3.1.11. +1SR,2SR,5SR)-5-Methyl-2-phenylbicyclo[3.2.0]
heptan-2-ol +21). Bromobenzene -1.4 mL, 13.24 mmol)
was added dropwise to Mg -339 mg, 13.95 mmol), activated
with I2, in dry Et2O -6 mL). The mixture was then re¯uxed
for 1 h. The bicyclic ketone 20 -0.5 mL, 4.03 mmol) in Et2O
-3 mL) was added dropwise. After 2h, sat. NH 4Cl was
added, the organic phase was separated, washed with 10%
NaHCO3 and H2O, dried and solvent removed in vacuo.
Flash chromatography -hexane/EtOAc, 70/30, v/v) gave
the title compound as a colourless liquid, which crystallized
in part on standing -719 mg, 88%), mp 56±588C m/z:
200-M1, 10%), 183-15), 128-47), 115-57), 105-82),
77-100). dH -600 MHz):¶ 1.22 -s, 3H, CH3), 1.29 -dt, 1H,
Jˆ6.5, 13.2Hz, H4 a), 1.50 -ddd, 1H, Jˆ1.8, 6.0, 13.2Hz,
H4b), 1.90 -m, 2H, H6), 2.0±2.1 -complex, 3H, H3a, H7),
2.43 -dt, 1H, Jˆ6.5, 13.2Hz, H3 b), 2.60 -dd, 1H, Jˆ5.4,
9.5 Hz, H1), 7.22 -d, 1H, Jˆ7.5 Hz, H40), 7.31 -dd, 2H,
1603 cm21
.
3.1.8. +4RS)-tert-Butyl-cyclohexyl-1,2-stilbene +14). To a
stirred suspension of anhydrous MgSO4 -0.350 g) in CH2Cl2
-1 mL) was added conc. H2SO4 -1 equiv, 35 mL).23 The
suspension was allowed to stir for 20 min and then the
crude alcohol 15 or 16 -70 mg, 0.228 mmol) was added
with the aid of CH2Cl2 -1 mL). When the reaction was
complete -TLC), MgSO4/H2SO4 was ®ltered off and the
solvent removed under reduced pressure to yield a colour-
less oil -66.5 mg). Integration of the tert-Bu region of the 1H
NMR spectrum of the crude product shows a ratio of 1.5:3:1
of the peaks at d 0.97 -17): d 0.95 -14): d 0.76 -18).
Resubjecting the crude mixture to the same conditions for
5 days then demonstrates a 1H NMR spectrum further
enriched in stilbene. The appearance of the tert-butyl
resonance for -3RS, 6RS)-3-tert-butyl-1,6-diphenyl-
cyclohex-1-ene 8 at d 1.02ppm was also observed.
Jˆ7.5, 8.4 Hz, H30), 7.42-d, H2,
Jˆ8.4 Hz, H20). dC
¶
COSY and HETCOR experiments were used to assign the resonances.

D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
9507
-150 MHz): 14.7-CH2), 26.5-CH3), 31.3-CH2), 37.8-CH2),
40.3-CH2), 44.4-q), 51.5-CH), 82.9-q), 125.6-CH),
126.8-CH), 128.0-CH), 147.9-q). nmax-neat): 3369-br.
of Hassner,25 the following reagents were combined; the
diol 23 -7 mg, 0.03 mmol), CH2Cl2 -0.5 mL), DCC
-22 mg, 0.11mmols), DMAP -5 mg, 0.04 mmol) and -R)-
-1)-a-methoxy-a--tri¯uoromethyl)phenylacetic acid -23 mg,
0.10 mmol). Flash chromatography -hexanes/EtOAC 95/5
v/v) eluant gave the title compound as a colourless liquid
s) cm21
.
3.1.12. +1RS,5SR)-5-Methyl-2-phenylbicyclo[3.2.0]hep-
tan-2-ene +9). To a solution of the alcohol 21 -688 mg,
3.4 mmol) in CH2Cl2 -5 mL) was added triethylamine
-0.4 mL). To the solution, cooled to 08C, was added
methanesulfonyl chloride -0.4 mL, 6.46 mmol). After the
reaction was stirred for 4 h, 2,6-di-tert-butyl-p-cresol was
added, the reaction mixture was washed with 15% HCl-aq)
-5 mL), extracted with CH2Cl2 -2£10 mL). The combined
organic layers were washed with sat. NaCl-aq) -5 mL), dried
and solvent removed in vacuo. Flash chromatography
-hexanes) gave the title compound as a colourless liquid
-507 mg, 81%). If the alkene 9 was not being used
immediately more 2,6-di-tert-butyl-p-cresol was added
and the alkene 9 was stored at 58C. m/z: 184-M1, 30%),
156-100), 115-45).dH -300 MHz): 1.20 -s, 3H, CH3), 1.6±
2.0 -complex, 3H, ring protons), 2.1±2.4 -complex, 3H, ring
protons), 3.05 -br. s, 1H, ring proton), 6.05 -br. s, 1H, ole®-
nic), 7.0±7.3 -m, 5H, aromatic protons). dC -75 MHz): 24.0,
25.5, 33.2, 43.6, 48.2, 50.8, 125.4, 125.7, 126.7, 128.1,
135.7, 145.2. nmax-neat): 3000-s), 1594-m), 1494-s),
1
-9 mg, 63%). From the H NMR spectrum of the products
from the racemic and optically active samples it was
possible to assign the following data. Diastereomer A: dH
-300 MHz): 1.33 -s, 3H), 1.3±2.5 -9H, complex), 3.37 -br. s,
3H), 6.26 -dd, 1H), 7.35±7.55 -m, 5H). Diastereomer B: dH
-300 MHz): 1.35 -s, 3H), 1.3±2.5 -9H, complex), 3.53 -br. s,
3H), 6.28 -dd, 1H), 7.35±7.55 -m, 5H).
3.1.16. Kinetic resolution of +1RS,5SR)-5-methyl-2-
phenylbicyclo[3.2.0]heptan-2-ene +9). Kinetic resolution
A: a mixture of K2CO3 -73 mg, 0.53 mmol), K3Fe-CN)6
-118 mg, 0.36 mmol), MeSO2NH2 -37 mg, 0.38 mmol),
-DHQD)2-PHAL -2mg, 0.01 mmol), tert-BuOH -3 mL),
H2O -3 mL), OsO4 -50 mL, 4 wt% solution, 2.0 mg,
0.01 mmol), were combined and cooled to 08C. The mixture
was stirred with the aid of a mechanical stirrer -350 rpm).
Once the reaction mixture had reached 08C, the stirring was
stopped and racemic alkene 9 -125 mg, 0.68 mmol) was
added. After stirring at 08C for 3 h the reaction
mixture was quenched with sodium sul®te solution -4 mL,
0.5 g/mL). The solution was then worked-up according to
the general procedure. The ee of the optically acitve diol
23a -30 mg recovered) was determined as for the racemic
diol -88% ee). The optical purity of the active alkene 9b
-52mg recovered) was determined by initially converting
the alkene 9b to the optically active diol 23b, and then to the
corresponding Mosher esters -52% ee).
1446-s), 754-s), 690-s) cm21
.
3.1.13. +1SR,2SR,3SR,5SR)-5-Methyl-2-phenylbicyclo-
[3.2.0] heptan-2,3-diol +23). The general procedure for
achiral dihydroxylation, was repeated with K3Fe-CN)6
-761 mg, 2.31 mmol), K2CO3 -165 mg, 1.19 mmol),
quinuclidine -2.9 mg, 0.03 mmol), OsO4 -50 mL, 4 wt%
solution, 2.0 mg, 0.01 mmol), MeSO2NH2 -67 mg,
0.71 mmol), tert-BuOH -3 mL), H2O -3 mL) and the alkene
9 -164 mg, 1.34 mmol). After stirring at ambient tempera-
ture for 24 h sodium sul®te -734 mg, 5.8 mmol) was added.
Puri®cation by ¯ash chromatography -EtOAc/hexanes, 30/
70 v/v) gave the title compound as white crystals -146 mg,
50%), mp 116±1188C. m/z: 218-M1, 25%), 189-55),
180-65), 149-75), 122-100). dH -300 MHz): 1.29 -s, 3H),
1.65±1.9 -complex, 5H), 2.0 -m, 1H), 2.45 -m, 1H), 4.95
-dd, 1H, Jˆ6.9 and 10.5 Hz), 7.2±7.5 -m, 5H). dC
-75 MHz): 16.2, 28.0, 29.6, 30.6, 40.1, 46.2, 53.0, 74.9,
83.6, 126.9, 127.3, 128.2, 142.0. nmax-neat): 3490-br. s,
OH), 3295-br. s, OH), 2800-s), 1500-w), 1270-m),
Kinetic resolution B: the above procedure was repeated with
K2CO3 -84 mg, 0.61 mmol), K3Fe-CN)6 -189 mg,
0.57 mmol), MeSO2NH2 -44 mg, 0.46 mmol), -DHQD)2-
PHAL -2mg, 0.01 mmol), tert-BuOH -3 mL), H2O
-3 mL), OsO4 -50 mL, 4 wt% solution, 2.0 mg,
0.01 mmol), and racemic alkene 9 -108 mg, 0.58 mmol).
The ee of both products was determined as described earlier
and found to be 80% for the diol -68 mg recovered) and 80%
for the alkene -32mg recovered).
3.1.17. Kinetic resolution of +1SR,5RS)-2,5-dimethyl-
bicyclo[3.2.0]hept-2-ene +24). The required alkene was
synthesized according to a method reported earlier.15
1150-m), 1100-s), 750-s), 700-s)cm21
.
3.1.14. +1R,2R,3R,5R)-5-Methyl-2-phenylbicyclo[3.2.0]-
heptan-2,3-diol +23b). In a similar manner; the optically
active alkene 9b was dihydroxylated with the following
reagents. K2CO3 -116 mg, 0.84 mmol), K3Fe-CN)6
-303 mg, 0.92mmol), MeSO 2NH2 -67 mg, 0.71 mL),
quinuclidine -2.9 mg, 0.03 mmol), tert-BuOH -3 mL),
H2O -3 mL), OsO4 -50 mL, 4 wt% solution, 2.0 mg,
0.01 mmol), and the optically active alkene 9b -52mg,
0.28 mmol). Upon work-up and puri®cation the title
compound was obtained as white crystals -45 mg, 74%).
The ee was determined as described later.
Kinetic resolution A: a mixture of K2CO3 -96 mg,
0.69 mmol), K3Fe-CN)6 -179 mg, 0.54 mmol), MeSO2NH2
-41 mg, 0.43 mmol), -DHQD)2-PHAL -2mg, 0.01 mmol),
tert-BuOH -4 mL), H2O -4 mL), OsO4 -50 mL, 4 wt%
solution, 2.0 mg, 0.01 mmol), were combined and cooled
to 08C. The mixture was stirred with the aid of a
mechanical stirrer -350 rpm). Once the reaction mixture
had reached 08C, the stirring was stopped and racemic
alkene 24 -128 mg, 1.05 mmol) was added. After stirring
at 08C for 3 h the reaction was quenched with sodium
sul®te solution -4 mL, 0.5 g mL21). The solution was
then worked-up according to the general procedure. The
ee of the optically active diol -21 mg recovered) was deter-
mined by conversion to the corresponding Mosher esters
-23% ee).
3.1.15. Formation of the racemic Mosher esters of the
phenyl diol: +1RS,3RS,4RS,5RS)-4-hydroxy-1-methyl-4-
phenylbicyclo[3.2.0]hept-3-yl +2R)-3,3,3-tri¯uoro-2-
methoxy-2-phenylpropanoate. According to the procedure

9508
D. P. G. Hamon et al. / Tetrahedron 57 $2001) 9499±9508
9. Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;
Hartung, J.; Jeong, K.-S.; Kwong, H.-L.; Morikawa, K.;
Wang, K.-M.; Xu, D.; Zhang, X.-L. J. Org. Chem. 1992, 57,
2768±2771.
Â
 Â
10. Guette, J.; Horeau, A. Bulletin de la Societe Chimique de
France 1967, 5, 1747±1752.
Kinetic resolution B: the above procedure was repeated with
K2CO3 -96 mg, 0.69 mmol), K3Fe-CN)6 -187 mg,
0.57 mmol), MeSO2NH2 -48 mg, 0.50 mmol), -DHQD)2-
PHAL -3 mg, 0.01 mmol), tert-BuOH -4 mL), H2O
-4 mL), OsO4 -50 mL, 4 wt% solution, 2.0 mg,
0.01 mmol), and racemic alkene 24 -125 mg, 1.02 mmol).
The enantiomeric excess of the optically active diol -22 mg
recovered) was determined by conversion to the correspond-
ing Mosher esters -23% ee).
11. Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.; Ikeda,
M.; Sharpless, K. B. J. Am. Chem. Soc. 1981, 103, 6237±6240.
12. Balavoine, G.; Moradpour, A.; Kagan, H. B. J. Am. Chem.
Soc. 1974, 96, 5152±5158.
13. Barili, P. L.; Berti, G.; Macchia, B.; Macchia, F.; Monti, L.
J. Chem. Soc. 1970, 1168±1172.
Acknowledgements
14. Parker, W.; Tranter, R. L.; Watt, C. I. F.; Chang, L. W. K.;
Schleyer, P. v. R. J. Am. Chem. Soc. 1974, 96, 7121.
15. Hamon, D. P. G.; Tuck, K. L. J. Org. Chem. 2000, 65, 7839±
7846.
K. L. T. and H. S. C. are grateful for the receipt of Australian
Post-graduate Awards.
16. The calculations were determined with Enantiomeric Ratio, a
freeware program available at http://www-orgc.tu-graz.ac.at/.
17. See Ref. 15 and references therein.
References
1. Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem.
Rev. 1994, 94, 2483±2547 and references therein.
2. Johnson, R. A.; Sharpless, K. B. In Catalytic Asymmetric
Synthesis, Ojima, I., Ed.; VCH: New York, 1993; pp. 103±
158.
18. Perrin, D. D.; Armarego, W. L. F. Puri®cation of Laboratory
Chemicals; 3rd ed; Pergamon: Oxford, 1988.
19. Garbisch Jr., E. W. J. J. Org. Chem. 1961, 26, 4165±4165.
20. King, S. B.; Sharpless, K. B. Tetrahedron Lett. 1994, 35,
5611±5612and references therein.
3. VanNieuwenhze, M. S.; Sharpless, K. B. J. Am. Chem. Soc.
1993, 115, 7864±7865.
21. Van Rheenen, V.; Cha, D. Y.; Hartley, W. M. Org. Synth.
1978, 58, 43±52.
4. Gardiner, J. M.; Nùrret, M.; Sadler, I. H. J. Chem. Soc., Chem.
Commun. 1996, 2709±2710.
22. Barili, P. L.; Berti, G.; Macchia, B.; Macchia, F.; Monti, L.
J. Chem. Soc. 1970, 1168±1172.
5. Unpublished observations, Christie, H. S. Honours Thesis,
University of Adelaide, Australia, 1996.
23. Wright, S. W.; Hageman, D. L.; Wright, A. S.; McClure, L. D.
Tetrahedron Lett. 1997, 38, 7345±7348.
6. Christie, H. S.; Hamon, D. P. G.; Tuck, K. L. J. Chem. Soc.,
Chem. Commun. 1999, 1989±1990.
24. Cargill, R. L.; Dalton, J. R.; Morton, G. H.; Caldwell, W. E.
Org. Synth. 1984, 62, 118±124.
7. Berti, G.; Macchia, B.; Macchia, F. Tetrahedron 1968, 24,
1755±1764.
25. Hassner, A.; Alexanian, V. Tetrahedron Lett. 1978, 4475±
4478.
8. VanNieuwenhze, M. S.; Sharpless, K. B. J. Am. Chem. Soc.
1993, 115, 7864±7865.