Chiral lithium amide base-mediated rearrangement of meso-cyclohexene oxides: Asymmetric synthesis of amino- and aziridinocyclohexenols

Article abstract of DOI:10.1039/b210608f

Two different chiral lithium amide base routes for the synthesis of amino- and aziridino-containing cyclohexenols have been explored. The first strategy involved the diastereoselective preparation of novel meso-aziridinocyclohexene oxides and their subsequent enantioselective rearrangement using chiral bases. In this approach, the diphenyl-phosphinoyl nitrogen protecting group proved optimal and aziridinocyclohexenols of 47-68% ee were obtained. Of particular note was the smooth rearrangement of the epoxide to an allylic alcohol in the presence of an aziridine: under optimised chiral base conditions, the aziridine remained essentially unaffected. A second more straightforward strategy for introduction of an amino functionality was also investigated: (1S,4R,5S)- and (1R,4R,5S)-4,5-bis(tert-butyldimethylsilyloxy)cyclohex-2- enols, readily prepared in >95% ee using a chiral base approach, were subjected to Mitsunobu substitution using a sulfonamide and Overman rearrangement.

Full text of DOI:10.1039/b210608f

View Article Online / Journal Homepage / Table of Contents for this issue
Chiral lithium amide base-mediated rearrangement of
meso-cyclohexene oxides: asymmetric synthesis of
amino- and aziridinocyclohexenols
Peter O’Brien* and Christopher D. Pilgram
Department of Chemistry, University of York, Heslington, York, UK YO10 5DD.
E-mail: paob1@york.ac.uk; Fax: ϩ44 1904 432516; Tel: ϩ44 1904 432535
Received 28th October 2002, Accepted 28th November 2002
First published as an Advance Article on the web 20th January 2003
Two different chiral lithium amide base routes for the synthesis of amino- and aziridino-containing cyclohexenols
have been explored. The first strategy involved the diastereoselective preparation of novel meso-aziridinocyclohexene
oxides and their subsequent enantioselective rearrangement using chiral bases. In this approach, the diphenyl-
phosphinoyl nitrogen protecting group proved optimal and aziridinocyclohexenols of 47–68% ee were obtained.
Of particular note was the smooth rearrangement of the epoxide to an allylic alcohol in the presence of an aziridine:
under optimised chiral base conditions, the aziridine remained essentially unaffected. A second more straightforward
strategy for introduction of an amino functionality was also investigated: (1S,4R,5S)- and (1R,4R,5S)-4,5-bis(tert-
butyldimethylsilyloxy)cyclohex-2-enols, readily prepared in >95% ee using a chiral base approach, were subjected to
Mitsunobu substitution using a sulfonamide and Overman rearrangement.
Introduction
The chiral lithium amide base-mediated rearrangement of
meso-epoxides to the corresponding allylic alcohols is a widely
studied and useful reaction in asymmetric synthesis.1 Recent
highlights in this area have involved the use of sub-stoichio-
metric amounts of chiral lithium amide bases in the presence of
excess achiral (bulk) bases2–4 and the extension of the reaction
to functionalised epoxide substrates.5,6 Our interest in epoxide
rearrangement reactions, spanning the last six years, has
focused on the development of a new norephedrine-derived
chiral base7,8 and on extending the reaction to new types of
meso-epoxides. For example, we were the first group to report
the enantioselective rearrangement of bis-protected meso-4,5-
dihydroxycyclohexene oxides,7–10 meso-4-amino substituted
cyclopentene oxides7,11 and meso-aziridinocyclohexene oxides.12
In addition, we reported the first examples of diastereoselective
epoxide rearrangements of chiral bis-protected 4,5-dihydroxy-
cyclohexene oxides13 and we have also developed epoxide
rearrangement methodology suitable for use in synthesis.14–16
Since we had been successful in developing a general strategy
for the diastereo- and enantiocontrolled preparation of poly-
hydroxylated cyclohexanes, exemplified by a synthesis of con-
duritol F,15 we decided to investigate whether related chiral base
methods could be used to synthesise amino conduritol ana-
logues. Amino conduritols have proven to be popular targets in
Scheme 1
recent times17 and two different strategies for introduction of
the amino functionality were envisaged (Scheme 1).
In strategy 1, we proposed to investigate the chiral base-
mediated rearrangement of meso-aziridino epoxides 1 to the
corresponding aziridino allylic alcohols 2. There were two key
issues to address in this new approach to amino conduritol
analogues: (i) methods for the diastereoselective synthesis of
cis- and trans-aziridino epoxides 1 needed to be developed and
trans-3 by chiral base-mediated desymmetrisation10) and an
investigation of ways of introducing the required nitrogen
functionality. Of the methods available, we selected a direct
Mitsunobu substitution with an appropriate amino source19,20
to give allylic amines 5 and an Overman rearrangement
approach21 to give allylic amines 6. As described in this paper,
both of these methods also proved successful for introducing
(ii) conditions for the chiral base reactions which would
the required amino functionality.18
rearrange the epoxide but leave the aziridine untouched needed
to be found. Full details of our studies in these areas are
Results and discussion
described in this paper: in particular, by careful choice of pro-
tecting group, reaction conditions and chiral base, it proved
possible to obtain diastereomerically pure aziridino allylic
alcohols 2 in good enantiomeric excess (47–68% ee).18
Strategy 1: synthesis and enantioselective rearrangement of
meso-aziridinocyclohexene oxides
A much simpler approach (strategy 2) involved use of allylic
alcohol 4 (which can be prepared in >95% ee from epoxide
Prior to our work, aziridinocyclohexene oxides 1 were an
unknown class of compound. However, since the corre-
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
523

View Article Online
Table 1 Diastereoselective epoxidation of aziridinocyclohexenes 9 and 10
Entry
R
Alkene
Epoxidation method
Epoxide
trans–cisa
Yield of trans (%)b
Yield of cis (%)b
1
2
3
4
Ph2PO
Ph2PO
Ts
9
9
10
10
MCPBAc
Dioxiraned
MCPBAc
Dioxiraned
11
11
12
12
10 : 90
64 : 36
42 : 58
91 : 9
9
56
40
83
81
31
55
9
Ts
1
a Ratio determined by H NMR spectroscopy on the crude product mixtures. b Isolated yield of pure epoxide after chromatography. c MCPBA,
NaHCO3, CH2Cl2, rt, 16 h. d Oxone®, trifluoroacetone, Na2ؒEDTA, MeCN–water, NaHCO3, 0 ЊC, 2.5 h.
sponding aziridinocyclohexenes had been prepared previ-
ously22,23 and we had an ongoing interest in stereoselective
epoxidation reactions,24,25 we elected to investigate an epoxid-
ation route to meso-epoxides 1. By varying the nitrogen protect-
ing group and the epoxidation conditions, we hoped to develop
stereoselective syntheses of cis- and trans-diastereomeric
epoxides 1.
Our preferred approaches for the synthesis of the required
N-protected aziridinocyclohexenes are summarised in Scheme
2. Using a method developed by Heathcock et al.,26 and utilised
Evans et al.,29 gave a low 20% yield of N-tosylaziridine 10 of
poor purity. In contrast, reaction of cyclohexa-1,4-diene with
Chloramine-T (TsNClؒ3H2O) and phenyltrimethylammonium
tribromide, Sharpless’s conditions,30 produced a good 63%
yield of the monoaziridine 10.
With N-protected aziridinocyclohexenes in hand, we were
ready to investigate their epoxidation, with particular emphasis
on the stereoselectivity. Surprisingly, we found that carbamate
(e.g. N–CO2Me, N–Cbz, N–Boc) and amide (N–Bz, N–Ac)
protecting groups were of no use in the epoxidation reactions
using either MCPBA or a dioxirane generated in situ:31 1H
NMR spectroscopy indicated the formation of several
products, none of which appeared to be the desired epoxides
and none of which could be isolated pure in acceptable yields.
We are still unable to offer an explanation for the complete
lack of success of these epoxidation reactions. In contrast,
N-diphenylphosphinoyl-protected cyclohexene 9 and N-tosyl-
protected cyclohexene 10 behaved in the expected fashion upon
exposure to MCPBA.
The full results of the epoxidation of alkenes 9 and 10 using
buffered MCPBA in dichloromethane and in situ generated
methyl(trifluoromethyl)dioxirane (Yang’s method31) are pre-
sented in Scheme 3 and Table 1. The stereoselectivity of each
Scheme 2 Reagents and conditions: i, (a) I2, AgOCN, THF, Ϫ20 ЊC,
6 h; (b) MeOH, reflux, 1.5 h; ii, KOH, water, reflux, 1 h; iii, Ph2P(O)Cl,
Et3N, DMAP, CH2Cl2, rt, 50 h; iv, TsCl, Et3N, DMAP, CH2Cl2, rt, 20 h;
v, Chloramine-Tؒ3H2O, 10 mol% PhMe3NϩBr3Ϫ, MeCN, rt, 18 h.
by Paquette et al. in the synthesis of a range of aziridinocyclo-
hexenes,22 treatment of cyclohexa-1,4-diene with in situ formed
iodine isocyanate followed by reaction with methanol generated
iodocarbamate 7 (51% recrystallised yield on a multi-gram
scale). Conversion of iodocarbamate 7 into the NH aziridine
8 (88% yield) was accomplished by reaction with aqueous
potassium hydroxide at reflux. Aziridine 8 is a relatively volatile
compound and must be carefully isolated by Kugelrohr distil-
lation. A direct synthesis of aziridine 8 from cyclohexa-1,4-diene
which did not require the intermediacy of iodocarbamate 7,
reported by Paquette et al.,22 was not reproducible in our hands.
The Heathcock methodology was a convenient way of prepar-
ing aziridine 8. In general, we preferred to store stocks of iodo-
carbamate 7 and convert it into aziridine 8 immediately prior to
N-protection.
Although we initially prepared aziridines with a wide
range of N-protecting groups (N–Ph2PO, N–Ts, N–CO2Me,
N–Cbz, N–Boc, N–Bz, N–Ac), it transpired that carbamate
and amido protecting groups were useless for the present work
(vide infra). Standard protection of aziridine 8 generated N-di-
phenylphosphinoyl and N-tosylaziridinocyclohexenes 9 and 10
in high yields (Scheme 2). The diphenylphosphinoyl N-protect-
ing group was originally introduced by Ramage and co-
workers27 and has recently gained popularity as an aziridine
protecting/activating group.28 The alternative, direct synthesis
of N-tosylaziridine from cyclohexa-1,4-diene by monoazirid-
Scheme 3
epoxidation reaction was determined by analysis of the 1H
NMR spectra of the crude product mixtures and the cis–trans
diastereoisomers were identified by independent syntheses of
aziridines trans-11 and trans-12 from a compound of known
stereochemistry (vide infra). In each case, the diastereomeric
cis- and trans-aziridino epoxides were easily separated by
chromatography and high yields of epoxides were obtained.
Although the MCPBA epoxidations could be carried out on a
multi-gram scale, the dioxirane reactions were a lot less robust
in terms of scale-up (>1 mmol of alkene). The dioxirane results
in Table 1 were obtained on a 0.7–0.9 mmol scale.
It is possible to identify some trends from the epoxidation
results presented in Table 1. Epoxidation of alkenes 9 and 10 is
cis-selective using MCPBA in dichloromethane (Table 1, entries
1 and 3) and the degree of cis-selectivity (90 : 10) is significantly
higher for the diphenylphosphinoyl protecting group (Table 1,
entry 1). In contrast, epoxidation using a dioxirane is trans-
selective (Table 1, entries 2 and 4) and a much higher degree of
ination was briefly investigated: use of iodinane PhI᎐NTs in the
᎐
presence of copper() triflate, † according to the method of
† The IUPAC name for triflate is trifluoromethanesulfonate.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
524

View Article Online
trans-selectivity (91 : 9) is obtained with a tosyl protecting
group (Table 1, entry 4). Taken together, the epoxidation results
indicate that by suitable choice of aziridine protecting group
and reaction conditions, it is possible to selectively synthesise
aziridino epoxides with cis or trans stereochemistry: an 81%
isolated yield of epoxide cis-11 (Table 1, entry 1) and an 83%
yield of epoxide trans-12 (Table 1, entry 4) were obtained. Thus,
multi-gram quantities of epoxide cis-11 could be generated
easily but it must be stressed that the dioxirane reaction used for
the preparation of epoxide trans-12 could not be carried out
effectively on a scale >1 mmol.
Much is known about diastereoselective epoxidation reac-
tions in general32 and the epoxidation of bicyclic-fused alkenes
related to 9 and 10 has been described. For example, epoxid-
ation of 4,5-epoxycyclohexene 1333 and norcar-3-enes 14–1634
using MCPBA are generally trans-selective. Although the exact
conformation of these cyclohexenes is unknown, epoxidation
trans to the three-membered ring appears to be the norm. This
is supported by the fact that epoxidation of sterically hindered
norcar-3-enes 15 and 16 are completely trans-selective whereas
reaction of norcar-3-ene 14 gave a 62 : 38 mixture of trans- and
cis-epoxides.34
epoxy group are trans relative to each other. Hence, epoxide 18
furnished aziridino epoxides trans-11 and trans-12 after a
Staudinger reaction39 and subsequent N-protection. Com-
parison of 1H and 13C NMR spectroscopic data allowed
assignment of the cis–trans stereochemistry in the aziridino
epoxides generated from the previously described epoxidation
reactions. We were unable to obtain a pure sample of epoxide
trans-11 via this sequence as it was always contaminated with
triphenylphosphine oxide (even after careful chromatography).
Next, we initiated our study into the enantioselective
rearrangement of both diastereomers of aziridino epoxides 11
and 12 to the corresponding allylic alcohols. Over the last
few years, a range of chiral diamines has been developed
and used for epoxide rearrangements. To date, the most useful
chiral bases for rearranging epoxides are derived from diamines
19 and 20 (introduced by Asami40,41), diamine 21 (developed
by Andersson2,42), diamine 22 (introduced by Singh43) and
diamine 23 (developed independently by ourselves7,8 and
Ahlberg3,44). In the present study, we have utilised Singh’s
diamine 22 and our own norephedrine-derived diamine 23.
Diamines rac-22, (R)-22 and (1R,2S)-23 were prepared in
multi-gram quantities using the procedures previously
developed in our group.8,45
Set against these literature results, it is surprising to note
that MCPBA epoxidation of aziridinocyclohexenes 9 and 10
is cis-selective (Table 1, entries 1 and 3). We suspect that the
N-diphenylphosphinoyl and N-tosyl groups hydrogen bond
to the MCPBA leading to some degree of cis-direction. The
N-diphenylphosphinoyl groups is clearly the better hydrogen
bonding group (Table 1, entry 1) and this has some literature
precedent.35 Apparently, the trans-selective epoxidation of
aziridinocyclohexenes 9 and 10 using a dioxirane reagent (Table
1, entries 2 and 4) are more in line with the literature precedent
in related systems. As we have noted previously,24,25 we believe
that Yang’s in situ generated methyl(trifluoromethyl)dioxirane
(acetonitrile–water solvent system) overturns any cis-directing
effects observed with MCPBA.
Our usual conditions for carrying out epoxide rearrangement
reactions involve (i) generation of two equivalents of the chiral
base (relative to the epoxide) from the diamine and n-butyl-
lithium in THF at 0 ЊC; (ii) reaction of the chiral base with the
epoxide at 0 ЊC for four hours and then at room temperature for
a further 12 hours and (iii) aqueous work-up including washing
with 2% hydrochloric acid in order to remove and recover the
diamine. To establish whether these conditions were suitable for
rearranging aziridino epoxides, initial optimisation studies were
carried out with N-diphenylphosphinoyl protected epoxide
cis-11 as this was the easiest substrate to prepare on a large
scale. Reaction of epoxide cis-11 with the lithium amide base
derived from diamine rac-22 under the standard conditions did
not lead to isolation of any of the desired allylic alcohol 24.
Instead, the only product obtained (25% recovery) was tenta-
tively ascribed a structure in which ring opening of the allylic
alcohol 24 with chloride had occurred. We presume that the
hydrochloric acid in the work-up was the source of the chloride
since omitting the hydrochloric acid from the work-up enabled
quantities of racemic allylic alcohol 24 to be isolated (Scheme 5
and Table 2).
The relative stereochemistry of all the epoxides described in
this paper was established as outlined in Scheme 4. Thus,
Scheme 4 Reagents and conditions: i, NaN3, NH4Cl, reflux, 8 : 1
t
MeOCH2CH2OH–water, 24 h; ii, BuOOH, Mo(CO)6, benzene, reflux,
2.5 h; iii, (a) Ph3P, THF, reflux, 20 h; (b) Ph2P(O)Cl, Et3N, DMAP,
CH2Cl2, rt, 20 h; (a) Ph3P, THF, reflux, 20 h; (b) p-TsCl, Et3N, DMAP,
CH2Cl2, rt, 20 h.
known23 azido alcohol 17 (prepared by reaction of the
monoepoxide of cyclohexa-1,4-diene36 with sodium azide) was
subjected to a molybdenum-catalysed, hydroxy-directed epox-
idation reaction37,38 to generate epoxide 18 in 96% yield.
Epoxide 18 was obtained as the sole diastereomeric product
and as a result of the hydroxy direction, the azido group and the
Scheme 5
Encouraged by the formation of allylic alcohol 24, we
varied the amount of lithium amide base used (Scheme 5,
Table 2) and found that the highest yield was obtained with 1.2
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
525

View Article Online
Table 2 Optimisation of conditions for rearrangement of epoxide
cis-11
Entry
rac-22 (eq.)
BuLi (eq.)
Yield of rac-24 (%)a
1
2
3
4
5
1.0
1.2
1.3
1.5
2.0
1.0
1.2
1.3
1.5
2.0
46b
62
46
32
42
Scheme 7
(1R,2S)-23 gave good yields of the enantioenriched allylic
alcohol (1R,3S,6S)-24 (Table 3, entries 2 and 3, 53% and 60%
yields respectively). In each case, the enantiomeric excess
was significantly lower than that obtained for cyclohexene
oxide43,44 which presumably reflects the conformational differ-
ences between cyclohexene oxide and aziridino epoxide cis-11.
With epoxide cis-11, the less sterically encumbered diamine
(R)-22 gave the highest enantiomeric excess (45% ee) of allylic
alcohol (1R,3S,6S)-24 (Table 3, entry 2).
In contrast to the N-diphenylphosphinoyl protected aziridino
epoxides, the corresponding N-tosyl protected epoxides cis-12
and trans-12 were much less robust as substrates and quite
varied and non-reproducible results were obtained (Table 3,
entries 4–6 and Table 4, entry 4). For example, diamine rac-22
converted aziridino epoxide cis-12 into allylic alcohol rac-25
in 41% yield (Table 3, entry 4) but use of diamine (R)-22
under otherwise identical conditions afforded allylic alcohol
(1R,3S,6S)-25 in only 14% yield (Table 3, entry 5). Although
the enantioselective rearrangements of epoxide cis-12 pro-
ceeded in low yields, the same trend as in the N-diphenyl-
phosphinoyl protected examples was observed: higher enantio-
selectivity (allylic alcohol of 66% ee, Table 3, entry 5) was
observed with diamine (R)-22 compared to that with diamine
(1R,2S)-23 (allylic alcohol of 10% ee, Table 3, entry 6). Allylic
alcohol rac-27 was obtained from epoxide trans-12 in only 26%
yield (Table 4, entry 4) and no enantioselective rearrangements
with epoxide trans-12 were attempted due to the lack of
availability of this epoxide.
a Isolated yield of pure allylic alcohol rac-24. b 34% yield of recovered
starting material also obtained.
equivalents of base (Table 2, entry 2): if 1.0 equivalents of
lithium amide was used, the reaction did not go to completion
(Table 2, entry 1, 34% recovered starting material) and if 2.0
equivalents (our generally preferred conditions) were used
(Table 2, entry 5), we presume that the excess base destroys the
aziridine in either the starting material or product, possibly via
α-lithiation.
The modified conditions for the rearrangement of aziridino
epoxide cis-11 using the lithium amide base derived from
diamine rac-22 are: (i) use of 1.2 equivalents of the lithium
amide under our usual conditions (THF, 0 ЊC for four hours
and then room temperature for 12 hours) and (ii) quenching the
reaction with aqueous ammonium chloride solution and separ-
ating the allylic alcohol from diamine using chromatography.
Under these conditions, generally good yields of allylic alcohols
were obtained. Crucially, the N-diphenylphosphinoyl protected
aziridine remained unscathed during these lithium amide base
conditions.
Attention was then turned to all of the aziridino epoxide
substrates and the use of 1.2 equivalents of enantiomerically
pure chiral lithium amide bases. The full results of all of our
studies are presented in Scheme 6 and Table 3 (for cis aziridino
epoxides) and in Scheme 7 and Table 4 (for trans aziridino
epoxides).
Rearrangement of N-diphenylphosphinoyl protected azirid-
ino epoxide trans-11 was also well behaved (Table 4, entries
1–3). In this case, generally higher yields (up to 82%, Table 4,
entry 3) and higher enantioselectivity (58–68% ee of allylic
alcohol, Table 4, entries 2 and 3) was obtained for the reactions
using diamines (R)-22 and (1R,2S)-23. There was also less dif-
ference in the enantioselectivity obtained using diamines (R)-22
and (1R,2S)-23 since the aziridine is now on the opposite side
of the ring to the epoxide functionality. With epoxide trans-11,
diamine (1R,2S)-23 gave allylic alcohol (1R,3S,6R)-26 with the
highest enantiomeric excess (68% ee, Table 4, entry 3); diamine
Rearrangement of N-diphenylphosphinoyl protected azirid-
ino epoxide cis-11 with chiral bases from diamines (R)-22 and
Scheme 6
Table 3 Rearrangement of cis-aziridino epoxides
Entry
SM
Diamine
Product
Yield (%)a
ee (%)
1
2
3
4
5
6
cis-11
cis-11
cis-11
cis-12
cis-12
cis-12
rac-22
rac-24
62
53
60
41
14
23
—
(R)-22
(1R,3S,6S)-24
(1R,3S,6S)-24
rac-25
45b
25b
—
(1R,2S)-23
rac-22
(R)-22
(1R,2S)-23
(1R,3S,6S)-25
(1R,3S,6S)-25
66c
10c
1
a Isolated yield of pure allylic alcohol. b Enantiomeric excess determined by H NMR spectroscopy in the presence of (R)-(Ϫ)-2,2,2-trifluoro-1-
(9-anthryl)ethanol. c Enantiomeric excess determined by chiral HPLC.
Table 4 Rearrangement of trans aziridino epoxides
Entry
SM
Diamine
Product
Yield (%)a
ee (%)
1
2
3
4
trans-11
trans-11
trans-11
trans-12
rac-22
rac-26
72
62
82
26
—
(R)-22
(1R,3S,6R)-26
(1R,3S,6R)-26
rac-27
58b
68b
—
(1R,2S)-23
rac-22
a Isolated yield of pure allylic alcohol. b Enantiomeric excess determined by Mosher’s ester formation.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
526

View Article Online
(1R,2S)-23 is often the base of choice with other epoxide
substrates.7,10,11
Mosher’s esters obtained from allylic alcohol 26 established the
absolute stereochemistry of all of the N-diphenylphosphinoyl
protected allylic alcohols. The absolute stereochemistry of
N-tosyl protected allylic alcohol (1S,3S,6S)-25 was assigned by
analogy but has not been established unequivocally.
There are some important observations from the results of
rearranging aziridino epoxides using lithium amide bases. First
of all, it is possible to rearrange epoxides to allylic alcohols in
the presence of aziridines and to obtain good yields (up to 82%)
of allylic alcohols. Of particular note is the fact that the azirid-
ine survives these reactions. Indeed, recent results suggest that
more forcing conditions (e.g. sec-butyllithium–(Ϫ)-sparteine46
or superbases47) are required in order to convert aziridines into
the corresponding allylic amines (presumably proceeding via
α-lithiation and a carbene mechanism). Secondly, based on
the isolated yields, we suggest that the N-tosyl protected azirid-
ines are more susceptible to decomposition (via α-lithiation of
the aziridine either in the starting aziridino epoxide or in the
allylic alcohol product) than the corresponding N-diphenyl-
phosphinoyl protected aziridines. In addition, cyclic vinyl
aziridines, especially if activated by a N-tosyl substituent are
relatively reactive16,48 and it may be that the lower yields with
N-tosyl aziridino epoxides cis- and trans-12 is due to decom-
position of the allylic alcohol product as it is also a vinyl-
aziridine. Thirdly, the variation in yields obtained with racemic
and enantiopure diamine 22 (cf. Table 3, entries 1 and 2;
Table 3, entries 4 and 5; Table 4, entries 1 and 2) highlights
the capricious nature of these reactions and that different
extents of decomposition can occur under seemingly identical
reaction conditions. Finally, good yields of allylic alcohols with
moderate to good enantiomeric excess can be obtained using
this chiral base route: use of diamine (R)-22 with epoxide
cis-11 gave allylic alcohol (1R,3S,6S)-24 in 53% yield and 45%
ee (Table 3, entry 2) and use of diamine (1R,2S)-23 with
epoxide trans-11 gave allylic alcohol (1R,3S,6R)-26 in 82%
yield and 68% ee (Table 4, entry 3). Both of these allylic alcohol
products could prove to be useful building blocks for
synthesis.
Strategy 2: Mitsunobu and Overman approaches to
aminodeoxyconduritols
An alternative approach to the synthesis of amino-containing
cyclohexenols has been investigated. The plan was to make use
of allylic alcohol 4 and its diastereomeric allylic alcohol 29
as we have already reported their preparation in >95% ee.10
For example, chiral base-mediated rearrangement of epoxide
trans-3 using diamine (1R,2S)-23 gave allylic alcohol 4 in 93%
yield and >95% ee and simple oxidation–reduction of 4 gener-
ated allylic alcohol 29 (81% yield over the two steps) in >95% ee
(Scheme 9). With straightforward access to multi-gram
quantities of 4 and 29, we envisaged converting them into
allylic amines using direct Mitsunobu substitution19,20 and/or
Overman rearrangement.21
A range of amines has been developed for use under
Mitsunobu reaction conditions.19,20 The key to the success of
these reagents is the pKa of the amine proton which requires
an appropriate electron withdrawing N-substituent. For the
Mitsunobu substitution of allylic alcohol 4 with an amine, we
screened a few of the commonly employed reagents19,20 and
found that the known,50 Fukuyama-like20 N-(4-methoxyben-
zyl)-4-methylbenzenesulfonamide (TsNHPMB) was optimal.
The Mitsunobu reagent TsNHPMB was prepared by tosylation
of p-methoxybenzylamine. Under typical Mitsunobu condi-
tions, allylic alcohol 4 gave allylic amine 30 in 91% yield whilst
allylic alcohol 29 gave allylic amine 31 in 60% yield (Scheme 9).
Although the relative stereochemistry in 30 and 31 has not been
proved by other means, the absence of the other diastereo-
isomer in each of the Mitsunobu reactions strongly suggests
that the substitutions proceed with stereospecific inversion
which is not always the case in cyclic allylic alcohols.51 Thus, the
Mitsunobu route was a simple and convenient way of preparing
allylic amines.
In all rearrangements of the cyclohexene oxide system
reported to date in the literature, the chiral bases derived from
diamines (R)-22 and (1R,2S)-23 generate an allylic alcohol
chiral centre of (S)-configuration. A comparison of the chem-
ical shifts of some of the resonances in the 1H NMR spectra of
the Mosher’s esters49 obtained from allylic alcohol 26 con-
firmed that this was the case for allylic alcohol 26. Unfortu-
nately, the same approach was not successful with allylic
alcohol 24 since the Mosher’s ester formation did not go to
completion. Instead, synthesis (Scheme 8) coupled with a
Allylic alcohols 4 and 29 were also utilised in another syn-
thetic sequence, the Overman rearrangement, which generates
allylic amines with transposed amino functionality compared
to that present in the previously prepared 30 and 31. Thus,
allylic alcohol 4 was converted into trichloroacetimidate 32 in
essentially quantitative yield upon treatment with DBU and
trichloroacetonitrile. Then, without isolation, the crude acetim-
idate 32 was heated overnight in xylene with potassium carbon-
ate (Isobe’s modification52) to give the rearranged allylic amine
33 in 95% yield (Scheme 9). However, carrying out exactly the
same sequence with allylic alcohol 29 did not furnish any of the
hoped-for allylic amine 35. Instead, a 35% yield of trichloro-
acetimidate 34 was isolated (and not a 35% yield of 35, as we
had originally reported in our communication15). Two of the
1
diagnostic signals in the H NMR spectrum of the product
confirm that trichloroacetimidate 34 has been isolated:
δH(270 MHz; CDCl3) 8.3 (1 H, br s, NH) and 5.4 (1 H, m,
CHO). These signals clearly point to a trichloroacetimidate
product when compared with those for isolated trichloro-
acetimidates reported by Overman21 and Isobe et al.52 In con-
trast, in allylic amine 33, the NH resonance appears at δH 6.45–
6.42 and the CHN resonance appears at δH 4.34, both of which
are similar magnitudes to those of related allylic amines
reported in the literature.21,52 Thus, it was not possible to syn-
thesise allylic amine 35 using the Overman rearrangement.
Apparently, the silyloxy groups present too much steric bulk to
allow the trichloroacetimidate to rearrange towards them in
compound 34 which has all of the groups on the same face. The
diastereomeric system 32 encounters no such problems and a
high 95% yield of allylic amine 33 was obtained (Scheme 9).
Scheme 8 Reagents and conditions: i, Ph3P, diisopropylazodicarb-
oxylate (DIAD), AcOH, toluene, rt, 72 h; ii, AcCl, Et3N, DMAP,
CH2Cl2, rt, 72 h.
comparison of the sign of optical rotations was used to estab-
lish the absolute stereochemistry of allylic alcohol 24. Thus,
allylic alcohol (1R,3S,6S)-24 of 45% ee was converted into
allylic acetate (1R,3R,6S)-28 by reaction with acetic acid under
Mitsunobu conditions. Although this allylic acetate could
not be obtained free from triphenylphosphine oxide (even after
repeated chromatography), it had the opposite sign of optical
rotation to its enantiomer, allylic acetate (1S,3S,6R)-28, pre-
pared by simple acetylation of allylic alcohol (1S,3S,6R)-24 (of
68% ee). These interconversions, together with analysis of the
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
527

View Article Online
Scheme 9 Reagents and conditions: i, 2 eq. nBuLi, 2 eq. diamine (1R,2S)-23, THF, 0 ЊC
CeCl3ؒ7H2O, MeOH, 0 ЊC, 15 min; iii, 1. 5 eq. TsNHPMB, 3 eq. Ph3P, 1 eq. DIAD, THF, rt, 16 h; iv DBU, CCl3CN, Ϫ20 ЊC
TsNHPMB, 3 eq. Ph3P, 2.5 eq. DIAD, THF, rt, 16 h; vi, K2CO3, xylene, reflux, 18 h.
rt, 16 h; ii, (a) PCC, CH2Cl2, rt, 16 h; (b) NaBH4,
rt, 20 h; v, 1.5 eq.
Attempted elaboration of aziridinocyclohexenols and
amino-cyclohexenols: use in synthesis
Patterson used as a key intermediate in their synthesis of
(Ϫ)-swainsonine.56 Thus, conversion of 33 into 37 was
investigated.
Having developed good synthetic routes to different amino-
and aziridinocyclohexenols, it was our intention to use these
compounds in synthesis. In particular, we attempted to ring
open the aziridine in aziridinocyclohexenol rac-24 using benzyl
alcohol–boron trifluoride–diethyl ether,53 sodium azide in
refluxing ethanol–water54 and lithium thiophenolate in THF.55
However, in each case, decomposition occurred and we were
never able to isolate any amounts of ring opened products. Due
to this lack of success, it appears that much more work is
required to discover suitable conditions/reagents for further
elaboration of the aziridinocyclohexenols generated in this
work.
Treatment of allylic amine 33 with MCPBA under standard
conditions afforded a 93% isolated yield of epoxide 36. Epoxide
36 was the only diastereomer detected. Based on literature pre-
cedent,32,57 we assume that the epoxidation is cis-directed due to
hydrogen bonding between MCPBA and the amide (and this
sense of epoxidation is trans to the bulky silyloxy groups).
However, we have no proof of this. Nevertheless, we continued
with our synthetic plan with the intention of establishing the
relative stereochemistry in 36 at a later stage.
There are a number of methods available for the conversion
of epoxides into allylic alcohols. Given our interest in lithium
amide-mediated epoxide rearrangements, we attempted the
reaction of epoxide 36 with 3 equivalents of LDA or the lithium
amide from rac-22 under typical conditions (THF, room tem-
perature, 16–72 hours).58 However, in all cases, decomposition
of the starting epoxide 36 was observed and none of allylic
alcohol 37 was obtained. Ring opening of the epoxide with
lithium phenylselenide (with a view to a subsequent oxidation–
elimination sequence for the generation of the allylic alcohol)
was also unsuccessful.59 Finally, the conversion of epoxides into
silyl protected allylic alcohols using tert-butyldimethylsilyl tri-
fluoromethanesulfonate and DBU has been reported.60 Even
using these conditions at reflux for prolonged times (up to
72 hours), we were unable to isolate any protected allylic alco-
hol from the reaction of epoxide 36. Thus, we have been unable
to satisfactorily utilise the amino- or aziridino-cyclohexenols in
synthetic endeavours to date.
We envisaged that allylic amine 33, readily prepared in multi-
gram quantities, would be a useful synthetic intermediate. One
sequence that was attempted is summarised in Scheme 10.
Conclusions
In summary, two different chiral lithium amide base routes for
the synthesis of amino- and aziridino-containing cyclohexenols
have been explored. The first strategy involved the diastereo-
selective synthesis of meso-aziridinocyclohexene oxides and
their subsequent enantioselective rearrangement using chiral
bases. Here, several interesting observations on the epoxidation
Scheme 10
Epoxidation of the alkene in 33 should give 36 which could be
converted into allylic alcohol 37. Allylic alcohol 37 has the
same regio- and stereochemical arrangement of functionality
(different protecting groups) as compound 38 which Trost and
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
528

View Article Online
and rearrangement reactions were made. In particular, the
N-diphenylphosphinoyl protecting group proved crucial
and aziridinocyclohexenols of 47–68% ee were obtained. Thus,
under optimised conditions, it was possible to rearrange the
epoxide to an allylic alcohol in the presence of an aziridine: the
aziridine remained essentially unaffected. Unfortunately, fur-
ther synthetic elaboration of the aziridinocyclohexenols was
not possible. A second more straightforward strategy for intro-
duction of an amino functionality was also investigated. Allylic
alcohols 4 and 29, prepared in >95% ee by established methods,
were converted into regioisomeric allylic amines via Mitsunobu
substitution using a sulfonamide and Overman rearrangement.
The Mitsunobu approach was more successful. Further work is
required to make use of chiral building blocks such as 24, 26,
30, 31, 33 and 36 in total synthesis studies.
Method C: azido alcohol cyclisation and aziridine protection
Triphenylphosphine (1.0 equiv.) was added in one portion to
a stirred solution of azido alcohol epoxide 18 (0.65 mmol) in
THF (5 cm3) at room temperature under nitrogen. The resulting
solution was heated at reflux for 20 h. After cooling to room
temperature, the solvent was evaporated under reduced pres-
sure to give the crude product. To the crude product in CH2Cl2
(4 cm3) at room temperature under nitrogen was added toluene-
p-sulfonyl chloride or diphenylphosphinic chloride (1.2 equiv.),
triethylamine (2.0 equiv.) and DMAP (“catalytic” amount).
The resulting solution was stirred at room temperature for 20 h.
Water (15 cm3) and CH2Cl2 (15 cm3) were added and the two
layers were separated. The organic layer was washed with water
(2 × 20 cm3) and the combined aqueous washings were
extracted with CH2Cl2 (2 × 20 cm3). The combined organic
extracts were dried (MgSO4) and evaporated under reduced
pressure to give the crude product.
Method D: enantioselective rearrangement of aziridino
epoxides
Experimental
General
n-Butyllithium (1.4–1.6 M solution in hexane, 1.2 equiv.) was
added dropwise to a stirred solution of diamine (1.2 equiv.) in
THF (2.5 cm3) at room temperature under nitrogen. After stir-
ring for 30 min at 0 ЊC, a solution of the epoxide (0.64 mmol) in
THF (2.5 cm3) was added dropwise via a cannula and the mix-
ture was warmed to room temperature over 4 h. After stirring at
room temperature for 12 h, saturated ammonium chloride solu-
tion (3 cm3) was added followed by Et2O (10 cm3) and the two
layers were separated. The aqueous layer was extracted with
Et2O (3 × 20 cm3) and the combined organic extracts were
washed with water (15 cm3) and saturated aqueous brine solu-
tion (15 cm3), dried (MgSO4) and evaporated under reduced
pressure to give the crude product.
General details have been described previously.10 Et2O was
dried over sodium–benzophenone and distilled before use.
Triethylamine was stored over potassium hydroxide pellets.
MCPBA (approx. 70% pure) was used as supplied. n-Butyl-
lithium was titrated against N-benzylbenzamide before use.61 In
1
the H NMR spectra, the symbol * indicates that the signal
disappears after a D2O shake.
For Kugelrohr distillations, the temperatures quoted corre-
spond to the oven temperatures. Microanalyses were carried out
at the University of Newcastle on a Carlo Erba 1106 elemental
analyser and weighed using a Mettler MT 5 microbalance.
Analytical HPLC was carried out on a Chiralcel AS column
and the compounds (detected at 215 nm) were eluted using a
solution of 25% iPrOH in heptane as the mobile phase at a flow
rate of 1.0 cm3 minϪ1. Optical rotations are given in 10Ϫ1 deg
cm2 gϪ1 and were recorded at 15–20 ЊC.
Toluene-p-sulfonyl chloride was purified before use.
Anhydrous tert-butyl hydroperoxide was prepared using the
literature procedure.62 The preparation of diamines (R)-2224
and (1R,2S)-238 as well as allylic alcohols (1S,4R,5S)-410
(>95% ee) and (1R,4R,5S)-2910 (>95% ee) has been described
previously.
Method E: Mitsunobu substitution of allylic alcohols
Allylic alcohol (0.28 mmol) and then triphenylphosphine
(1.5 equiv.) were added sequentially to a stirred solution of
N-(4-methoxybenzyl)-4-methylbenzenesulfonamide (1.5 equiv.)
in THF (5 cm3) at room temperature under nitrogen. After stir-
ring for 30 min, diisopropylazodicarboxylate (1.0 or 2.5 equiv.)
was added dropwise and the resulting mixture was stirred at
room temperature for 16 h. Then, the solvent was evaporated
under reduced pressure to give the crude product.
Methyl 6-iodocyclohex-3-en-1-yl carbamate 7
General methods
Iodine (12.0 g, 47 mmol) was added to a stirred slurry of cyclo-
hexa-1,4-diene (4.0 g, 50 mmol) and silver cyanate (9.5 g,
63 mmol) in THF (100 cm3) at Ϫ20 ЊC under nitrogen. After
stirring for 4 h at Ϫ20 ЊC, the salts were removed by filtration
and the filtrate was evaporated under reduced pressure to a
volume of approximately 20 cm3. MeOH (100 cm3) was added
and the solution was heated at reflux for 1.5 h. After cooling to
room temperature, the solvent was evaporated under reduced
pressure and the residue dissolved in Et2O (60 cm3). The organic
layer was washed with 20% aqueous sodium sulfite solution
(20 cm3) and the aqueous layer was then extracted with Et2O
(2 × 20 cm3). The combined organic extracts were dried
(MgSO4) and evaporated under reduced pressure to give the
crude product. Recrystallisation from 10 : 1 CH2Cl2–hexane
(12 cm3) gave iodocarbamate 7 (7.2 g, 51%) as a white powder,
mp 96–98 ЊC (from 10 : 1 CH2Cl2–hexane)(lit.,22 99.5–100.5 ЊC);
RF(2 : 1 petrol–Et2O) 0.4; data identical to those reported
previously.22
Method A: epoxidation using MCPBA
Sodium hydrogen carbonate (2 equiv.) and MCPBA
(2 equiv., approx. 70% pure material) were added in portions to
a stirred solution of the alkene (0.8 mmol) in CH2Cl2 (10 cm3)
at room temperature under nitrogen. After stirring for 16 h
at room temperature, 20% aqueous sodium sulfite solution
(10 cm3) was added and the mixture stirred for a further 20 min.
The two layers were separated and the aqueous layer was
extracted with CH2Cl2 (3 × 20 cm3). The combined organic
extracts were washed with 20% aqueous sodium sulfite solution
(20 cm3), saturated aqueous sodium hydrogen carbonate solu-
tion (20 cm3) and water (20 cm3), dried (MgSO4) and evapor-
ated under reduced pressure to give the crude product.
Method B: epoxidation using methyl(trifluoromethyl)-
dioxirane
1,1,1-Trifluoroacetone (1.1 equiv.) was added via a pre-
cooled syringe to a stirred solution of alkene (0.68 mmol) and
Na2EDTA (0.002 equiv. of a 4 × 10Ϫ4 M aqueous solution)
in acetonitrile (6 cm3) at 0 ЊC. Then, a mixture of Oxone®
(5.0 equiv.) and sodium hydrogen carbonate (8.0 equiv.) was
added in portions over 1 h. After stirring at 0 ЊC for a further
1.5 h, the reaction mixture was poured into water (20 cm3) and
extracted with CH2Cl2 (3 × 20 cm3). The combined organic
extracts were dried (MgSO4) and evaporated under reduced
pressure to give the crude product.
7-Azabicyclo[4.1.0]hept-3-ene 8
Iodocarbamate 7 (2.45 g, 8.7 mmol) was added in one portion
to a stirred solution of potassium hydroxide (10 g) in water
(50 cm3). The resulting solution was heated at reflux for
1 h. After cooling to room temperature, the aqueous layer
was extracted with Et2O (4 × 40 cm3). The combined
organic extracts were dried (MgSO4) and evaporated at room
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
529

View Article Online
temperature under reduced pressure to give the crude product.
Purification by Kugelrohr distillation gave aziridine 8 (730 mg,
88%) as a colourless oil, bp 60–80 ЊC/10 mmHg (lit.,23 60.5–61.5
ЊC/14 mmHg); data identical to those reported previously.23
Aziridine 8 should be used immediately or stored under
nitrogen in the refrigerator.
was evaporated under reduced pressure. Purification by fil-
tration through a plug of silica with petrol–EtOAc (10 : 1) as
eluent gave N-tosylaziridine 10 (783 mg, 63%) as a white crys-
talline solid, mp 109–110 ЊC (from 10 : 1 petrol–EtOAc) (lit.,23
112–113 ЊC); data identical to those reported above.
(1R*,3S*,5R*,7S*)-8-(Diphenylphosphoryl)-4-oxa-8-aza-
tricyclo[5.1.0.03,5]octane cis-11
7-(Diphenylphosphoryl)-7-azabicyclo[4.1.0]hept-3-ene 9
Diphenylphosphinic chloride (0.96 cm3, 5.0 mmol) was added
dropwise to a stirred solution of aziridine 8 (400 mg, 4.2 mmol),
triethylamine (1.17 cm3, 8.4 mmol) and DMAP (“catalytic”
amount) in CH2Cl2 (25 cm3) at room temperature under nitro-
gen. The resulting mixture was stirred at room temperature for
50 h. Then, water (20 cm3) was added and the two layers were
separated and the organic layer was washed with water (2 ×
20 cm3). The combined aqueous layers were extracted with
CH2Cl2 (3 × 20 cm3) and the combined organic extracts were
dried (MgSO4) and evaporated under reduced pressure to give
the crude product. Purification by flash column chromato-
graphy on silica with CHCl3–MeOH (10 : 1) as eluent gave
N-Ph2PO aziridine 9 (1.10 g, 90%) as a white crystalline solid,
mp 117–120 ЊC (from 1 : 1 petrol–EtOAc) (lit.,22 119–120.5 ЊC);
RF(10 : 1 CHCl3–MeOH) 0.6; νmax(CHCl3)/cmϪ1 2987, 2904,
Using general method A, MCPBA (242 mg of 70% pure
material, 0.98 mmol), sodium hydrogen carbonate (118 mg,
1.4 mmol) and alkene 9 (207 mg, 0.70 mmol) in CH2Cl2 (10 cm3)
gave the crude product which contained a 10 : 90 mixture of
1
epoxides trans- and cis-11 (by H NMR spectroscopy). Purifi-
cation by flash column chromatography on silica with EtOAc–
MeOH (20 : 1) as eluent gave epoxide trans-11 (9 mg, 9%) as an
off-white solid identical to that obtained below and epoxide cis-
11 (170 mg, 81%) as a white crystalline solid, mp 164–166 ЊC
(from 20 : 1 EtOAc–MeOH); RF(20 : 1 EtOAc–MeOH) 0.2;
νmax(CHCl3)/cmϪ1 2989, 1439 (P–Ph), 1350, 1184 (P᎐O), 1126,
᎐
1093, 1017, 894, 739 and 668; δH(270 MHz; CDCl3) 8.02–7.94
(4 H, m, o-Ph2PO), 7.49–7.41 (6 H, m, m- and p-Ph2PO), 3.14
(2 H, s, CHO), 2.81 (2 H, dd, J 2.5 and 17.0, CHN), 2.50 (2 H,
d, J 16.5, CHAHB) and 2.20 (2 H, dd, J 2.5 and 16.5 (app. br d),
CHAHB); δC(67.9 MHz; CDCl3) 133.3 (d, JCP 127.0, ipso-
Ph2PO), 131.7 (d, JCP 9.5), 131.6, 128.3 (d, JCP 12.5), 49.4
(CHO), 31.0 (d, JCP 7.0, CHN) and 22.9 (d, JCP 5.5, CH2); m/z
(CI; NH3) 312 [100%, (M ϩ H)ϩ][Found: (M ϩ H)ϩ, 312.1150.
C18H18NO2P requires M ϩ H, 312.1153]; Found: C, 69.1; H,
5.5; N, 4.5%; C18H18NO2P requires C, 69.5; H, 5.8; N, 4.5%.
1439 (P–Ph), 1216, 1186 (P᎐O), 1126, 1028, 970, 896, 800 and
᎐
729; δH(270 MHz; CDCl3) 7.95–7.88 (4 H, m, o-Ph2PO), 7.52–
7.39 (6 H, m, m- and p-Ph PO), 5.53 (2 H, s, ᎐CH), 3.01 (2 H, d,
᎐
2
J 16.5, CHN) and 2.36 (4 H, s, CH2); δC(67.9 MHz; CDCl3)
133.5 (d, JCP 127.0, ipso-Ph2PO), 131.6, 131.5 (d, JCP 8.0), 128.3
(d, JCP 13.0), 121.9 (C᎐C), 33.5 (d, JCP 5.5, CHN) and 24.4 (d,
᎐
JCP 5.5, CH2); m/z (CI; NH3) 296 [100%, (M ϩ H)ϩ][Found:
(M ϩ H)ϩ, 296.1204. C18H18NOP requires M ϩ H, 296.1204];
Found: C, 73.0; H, 6.0; N, 4.7%; C18H18NOP requires C, 73.2;
H, 6.1; N, 4.7%.
(1R*,3R*,5S*,7S*)-8-(Diphenylphosphoryl)-4-oxa-8-aza-
tricyclo[5.1.0.03,5]octane trans-11
Using general method B, 1,1,1-trifluoroacetone (0.7 cm3,
8.0 mmol), alkene 9 (200 mg, 0.68 mmol), Na2EDTA (3.5 cm3
of a 4 × 10Ϫ4 M aqueous solution, 0.0014 mmol), Oxone®
(2.17 g, 3.5 mmol) and sodium hydrogen carbonate (460 mg,
5.4 mmol) in acetonitrile (6 cm3) gave the crude product which
contained a 64 : 36 mixture of epoxides trans- and cis-11 (by 1H
NMR spectroscopy). Purification by flash column chromato-
graphy on silica with EtOAc–MeOH (20 : 1) as eluent gave
epoxide trans-11 (119 mg, 56%) as a white crystalline solid, mp
132–135 ЊC (from 20 : 1 EtOAc–MeOH); RF(20 : 1 EtOAc–
MeOH) 0.35; νmax(CHCl3)/cmϪ1 2993, 1438 (P–Ph), 1337, 1239,
1189 (P᎐O), 1126, 1054, 981, 827, 751 and 666; δ (270 MHz;
7-[(4-Methylphenyl)sulfonyl]-7-azabicyclo[4.1.0]hept-3-ene 10
Toluene-p-sulfonyl chloride (1.00 g, 8.2 mmol) was added in
one portion to a stirred solution of aziridine 8 (650 mg,
6.8 mmol), triethylamine (1.9 cm3, 13.7 mmol) and DMAP
(“catalytic” amount) in CH2Cl2 (25 cm3) at room temperature
under nitrogen. The resulting mixture was stirred at room tem-
perature for 20 h. Then, water (20 cm3) was added and the two
layers were separated and the organic layer was washed with
water (2 × 20 cm3). The combined aqueous layers were
extracted with CH2Cl2 (2 × 20 cm3) and the combined organic
extracts were dried (MgSO4) and evaporated under reduced
pressure to give the crude product. Purification by flash column
chromatography on silica with petrol–EtOAc (1 : 1) as eluent
gave N-tosylaziridine 10 (1.32 g, 78%) as a white crystalline
solid, mp 111–113 ЊC (from 1 : 1 petrol–EtOAc)(lit.,23 112–113
ЊC); RF(1 : 1 petrol–EtOAc) 0.5; νmax(CHCl3)/cmϪ1 3033, 2902,
1600, 1408, 1320, 1232, 1157, 1092, 998, 947, 812 and 672;
δH(270 MHz; CDCl3) 7.83 (2 H, d, J 8.0, o-C6H4SO2), 7.32 (2 H,
᎐
H
CDCl3) 7.93–7.85 (4 H, m, o-Ph2PO), 7.55–7.41 (6 H, m, m- and
p-Ph2PO), 3.15 (2 H, d, J 2.0, CHO), 2.81 (2 H, dd, J 3.0 and
16.5, CHN), 2.25 (2 H, dd, J 2.0 and 16.5, CHAHB) and 2.12
(2 H, dd, J 3.0 and 16.5, CHAHB); δC(67.9 MHz; CDCl3) 132.9
(d, JCP 127.0, ipso-Ph2PO), 131.8 (d, JCP 2.5), 131.4 (d, JCP 8.0),
128.4 (d, JCP 12.5), 49.2 (CHO), 31.3 (d, JCP 7.0, CHN) and 23.3
(d, JCP 5.5, CH2); m/z (CI; NH3) 312 [100%, (M ϩ H)ϩ][Found:
(M ϩ H)ϩ, 312.1151. C18H18NO2P requires M ϩ H, 312.1153];
Found: C, 69.2; H, 5.9; N, 4.5%; C18H18NO2P requires C, 69.5;
H, 5.8; N, 4.5% and epoxide cis-11 (66 mg, 31%) as a white
crystalline solid identical to that obtained above.
d, J 8.0, m-C H SO ), 5.44–5.43 (2 H, m, ᎐CH), 3.11 (2 H, d,
᎐
6
4
2
J 0.5, CHN), 2.44 (3 H, s, Me) and 2.37–2.36 (4 H, m, CH2);
δC(67.9 MHz; CDCl3) 144.2 (ipso-C6H4SO2), 135.8 (ipso-
C H Me), 129.6, 127.7, 121.6 (C᎐C), 38.6 (CHN), 23.1 (CH )
᎐
6
4
2
and 21.6 (Me); m/z (CI; NH3) 250 [100%, (M ϩ H)ϩ][Found: (M
ϩ H)ϩ, 250.0905. C13H15NO2S requires M ϩ H, 250.0902];
Found: C, 62.8; H, 6.1; N, 5.6%; C13H15NO2S requires C, 62.6;
H, 6.1; N, 5.6%.
(1R*,3S*,5R*,7S*)-8-[(4-Methylphenyl)sulfonyl]-4-oxa-8-aza-
tricyclo[5.1.0.03,5]octane cis-12
Using general method A, MCPBA (300 mg of 70% pure
material, 1.20 mmol), sodium hydrogen carbonate (100 mg,
1.28 mmol) and alkene 10 (200 mg, 0.80 mmol) in CH2Cl2
(10 cm3) gave the crude product which contained a 42 : 58 mix-
7-[(4-Methylphenyl)sulfonyl]-7-azabicyclo[4.1.0]hept-3-ene 10
1
A solution of phenyl trimethylammonium tribromide (188 mg,
0.5 mmol) in acetonitrile (8 cm3) was added dropwise to a
stirred mixture of cyclohexa-1,4-diene (400 mg, 5.0 mmol) and
Chloramine-T hydrate (1.55 g, 5.5 mmol) in acetonitrile (16
cm3) at room temperature under nitrogen. The resulting yellow
suspension was stirred vigorously for 18 h and then the solvent
ture of epoxides trans- and cis-12 (by H NMR spectroscopy).
Purification by flash column chromatography on silica with
EtOAc–petrol (1 : 1) as eluent gave epoxide trans-12 (85 mg,
40%) as white needles identical to that obtained below and
epoxide cis-12 (116 mg, 55%) as a white crystalline solid,
mp 109–112 ЊC (from EtOAc); RF(1 : 1 petrol–EtOAc) 0.1;
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
530

View Article Online
νmax(CHCl3)/cmϪ1 3032, 3013, 1598, 1496, 1321, 1247, 1158,
1093, 960, 775, 715 and 665; δH(270 MHz; CDCl3) 7.83 (2 H, d,
J 8.5, o-C6H4SO2), 7.32 (2 H, d, J 8.5, m-C6H4SO2), 3.04 (2 H, s,
CHO), 2.94–2.83 (2 H, m, CHN), 2.53 (2 H, d, J 16.5, CHAHB),
2.44 (3 H, s, Me) and 2.23 (2 H, br d, J 16.5, CHAHB); δC(67.9
MHz; CDCl3) 144.2 (ipso-C6H4SO2), 135.6 (ipso-C6H4Me),
129.7, 127.7, 48.6 (CHO), 36.0 (CHN), 21.9 (Me) and 21.6
(CH2); m/z (CI; NH3) 266 [100%, (M ϩ H)ϩ][Found: (M ϩ H)ϩ,
266.0855. C13H15NO3S requires M ϩ H, 266.0851]; Found: C,
58.7; H, 5.8; N, 5.1%; C13H15NO3S requires C, 58.9; H, 5.7; N,
5.3%.
and 764; δH(270 MHz; CDCl3) 3.58–3.55 (2 H, m, CHOH and
CHN3), 3.26–3.21 (2 H, m, CHO), 2.63–2.33 (3 H, m) and 2.17–
1.88 (2 H, m); δC(67.9 MHz; CDCl3) 68.4, 59.5, 51.6 (CHO),
51.5 (CHO), 29.6 (CH2) and 27.8 (CH2); m/z (CI; NH3) 173
[13%, (M ϩ NH4)ϩ] and 128 (100)[Found: (M ϩ NH4)ϩ,
173.1039. C6H9N3O2 requires M ϩ NH4, 173.1039].
(1R*,3R*,5S*,7S*)-8-(Diphenylphosphoryl)-4-oxa-8-azatricyclo-
[5.1.0.03,5]octane trans-11
Using general method C, triphenylphosphine (170 mg, 0.65
mmol) and azido alcohol epoxide 18 (100 mg, 0.65 mmol) in
THF (5 cm3) followed by reaction with diphenylphosphinic
chloride (0.15 cm3, 0.77 mmol), triethylamine (0.18 cm3, 1.28
mmol) and DMAP (“catalytic” amount) in CH2Cl2 (4 cm3) gave
the crude product. Purification by flash column chromato-
graphy on silica with EtOAc–MeOH (20 : 1) as eluent gave an
inseparable mixture of epoxide trans-11 (identical to that
obtained above) and triphenylphosphine oxide (342 mg) as a
white foam.
(1R*,3R*,5S*,7S*)-8-[(4-Methylphenyl)sulfonyl]-4-oxa-8-aza-
tricyclo[5.1.0.03,5]octane trans-12
Using general method B, 1,1,1-trifluoroacetone (0.9 cm3,
10.0 mmol), alkene 10 (220 mg, 0.88 mmol), Na2EDTA (4.6 cm3
of a 4 × 10Ϫ4 M aqueous solution, 0.0018 mmol), Oxone®
(2.82 g, 4.6 mmol) and sodium hydrogen carbonate (600 mg,
7.1 mmol) in acetonitrile (6 cm3) gave the crude product which
contained a 91 : 9 mixture of epoxides trans- and cis-12 (by 1H
NMR spectroscopy). Purification by flash column chromato-
(1R*,3R*,5S*,7S*)-8-[(4-Methylphenyl)sulfonyl]-4-oxa-8-aza-
tricyclo[5.1.0.03,5]octane trans-12
graphy on silica with EtOAc–petrol (1 : 1)
EtOAc as eluent
Using general method C, triphenylphosphine (170 mg, 0.65
mmol) and azido alcohol epoxide 18 (100 mg, 0.65 mmol) in
Et2O (5 cm3) followed by reaction with toluene-p-sulfonyl chlor-
ide (146 mg, 0.77 mmol), triethylamine (0.18 cm3, 1.28 mmol)
and DMAP (“catalytic” amount) in CH2Cl2 (4 cm3) gave the
crude product. Purification by flash column chromatography
on silica with EtOAc–petrol (1 : 1) as eluent gave epoxide trans-
12 (70 mg, 41%) as an off-white solid identical to that obtained
above.
gave epoxide trans-12 (194 mg, 83%) as white needles, mp 118–
120 ЊC (from 1 : 1 EtOAc-petrol); RF(1 : 1 petrol–EtOAc) 0.3;
νmax(CHCl3)/cmϪ1 3034, 3007, 1325, 1243, 1159, 1093, 1040,
763, 731 and 634; δH(270 MHz; CDCl3) 7.81 (2 H, d, J 8.5,
o-C6H4SO2), 7.34 (2 H, d, J 8.5, m-C6H4SO2), 3.05–3.03 (2 H,
m, CHO), 2.89–2.87 (2 H, m, CHN), 2.46 (3 H, s, Me)
and 2.29–2.09 (4 H, m, CH2); δC(67.9 MHz; CDCl3) 144.6
(ipso-C6H4SO2), 135.1 (ipso-C6H4Me), 129.6, 127.7, 48.9
(CHO), 36.3 (CHN), 22.0 (Me) and 21.7 (CH2); m/z (CI;
NH3) 283 [18%, (M ϩ NH4)ϩ] and 266 [100, (M ϩ H)ϩ][Found:
(M ϩ H)ϩ, 266.0853. C13H15NO3S requires M ϩ H, 266.0851];
Found: C, 58.9; H, 5.7; N, 5.1%; C13H15NO3S requires C, 58.9;
H, 5.7; N, 5.3% and epoxide cis-12 (20 mg, 9%) as a white
crystalline solid identical to that obtained above.
(1R*,3S*,6S*)-7-Diphenylphosphoryl-7-azabicyclo[4.1.0]hept-
4-en-3-ol 24
Using general method D, n-butyllithium (0.51 cm3 of a 1.5 M
solution in hexane, 0.77 mmol), diamine rac-22 (157 mg, 0.77
mmol) and epoxide cis-11 (200 mg, 0.64 mmol) in THF (5 cm3)
gave the crude product. Purification by flash column chromato-
graphy on silica with EtOAc–MeOH (20 : 1) as eluent gave
allylic alcohol rac-24 (123 mg, 62%) as a white solid, mp 160–
162 ЊC (from 20 : 1 EtOAc–MeOH); RF(20 : 1 EtOAc–MeOH)
0.3; νmax(CHCl3)/cmϪ1 3683 (OH), 3020, 1523, 1423 (P–Ph),
(1R*,6R*)-6-Azidocyclohex-3-en-1-ol 17
Sodium azide (216 mg, 3.3 mmol) was added in portions over
10 min to a stirred solution of monoepoxide 13 (245 mg of
1
epoxide which contained 20% CH2Cl2 by H NMR spectro-
scopy, prepared according to a literature procedure,35 2.1 mmol)
and ammonium chloride (178 mg, 3.3 mmol) in 8 : 1 2-methoxy-
ethanol–water (18 cm3) at room temperature. The resulting
mixture was heated at reflux for 24 h. After cooling to room
temperature, water (20 cm3) was added and the aqueous layer
was extracted with Et2O (4 × 20 cm3). The combined organic
extracts were washed with saturated aqueous brine solution
(8 × 20 cm3), dried (MgSO4) and evaporated under reduced
pressure to give the crude product. Purification by flash column
chromatography on silica with EtOAc–petrol (1 : 1) as eluent
gave azido alcohol 17 (265 mg, 92%) as a yellow oil, RF(1 : 1
petrol–EtOAc) 0.5; data identical to those reported previously.23
1219 (P᎐O), 1034, 930, 756, 671 and 625; δ (270 MHz; CDCl )
᎐
H
3
7.87–7.77 (4 H, m, o-Ph2PO), 7.57–7.41 (6 H, m, m- and
p-Ph2PO), 6.30 (1 H, ddd, J 1.0, 6.0 and 9.5, H4), 6.21 (1 H, dd,
J 4.5 and 9.5, H5), 4.22–4.11 (1 H, m, H3), 3.47 (1 H, qdd, J 2.0,
6.0 and 15.0, H1), 3.36 (1 H, dddd, J 1.0, 4.5, 6.0 and 14.0, H6),
3.09* (1 H, d, J 11.5, OH), 2.33 (1 H, br d, J 15.5, H2) and 1.74
(1 H, tdd, J 2.0, 4.5 and 15.5, H2); δC(67.9 MHz; CDCl3) 134.4
(C4), 132.1 (d, JCP 125.5, ipso-Ph2PO), 132.1, 131.8 (d,
JCP 125.5, ipso-Ph2PO), 131.3 (d, JCP 9.5), 131.1 (d, JCP 9.5),
128.7 (d, JCP 12.0), 128.6 (d, JCP 12.0), 127.2 (d, JCP 5.5, C5),
63.0 (C3), 37.5 (d, JCP 7.0, C1), 33.1 (d, JCP 5.5,C6) and 28.6
(d, JCP 5.5, C5); m/z (CI; NH3) 312 [100%, (M ϩ H)ϩ][Found:
(M ϩ H)ϩ, 312.1155. C18H18NO2P requires M ϩ H, 312.1153].
The 1H and 13C NMR spectra were fully assigned using 1H–1H
and 1H–13C COSY experiments.
(1S*,3R*,4R*,6R*)-4-Azido-7-oxabicyclo[4.1.0]heptan-3-ol 18
tert-Butyl hydroperoxide (0.96 cm3 of a 3.0 M solution in tolu-
ene, 2.9 mmol) was added dropwise over 5 min to a stirred
refluxing solution of azido alcohol 17 (200 mg, 1.4 mmol) and
molybdenum hexacarbonyl (5 mg, 0.02 mmol) in benzene
(20 cm3) under nitrogen. The resulting solution was heated at
reflux for 2.5 h. After cooling to room temperature, the solvent
was evaporated under reduced pressure to give the crude prod-
uct. Direct purification by flash column chromatography on
silica with EtOAc–petrol (1 : 1) as eluent gave a single diastereo-
isomer (by 1H NMR spectroscopy) of azido alcohol epoxide 18
(214 mg, 96%) as a waxy solid (which decolourised on stand-
ing), RF(1 : 1 petrol–EtOAc) 0.2; νmax(CHCl3)/cmϪ1 3583 (OH),
3494 (OH), 3016, 2927, 2109, (N3), 1434, 1246, 1068, 798, 779
(1R,3S,6S )-7-Diphenylphosphoryl-7-azabicyclo[4.1.0]hept-4-en-
3-ol 24
Using general method D, n-butyllithium (0.38 cm3 of a 1.5 M
solution in hexane, 0.58 mmol), diamine (R)-22 (118 mg, 0.58
mmol) and epoxide cis-11 (150 mg, 0.48 mmol) in THF (5 cm3)
gave the crude product. Purification by flash column chromato-
graphy on silica with EtOAc–MeOH (20 : 1) as eluent gave
allylic alcohol (1R,3S,6S)-24 (80 mg, 53%, 45% ee by 1H NMR
spectroscopy in the presence of 2 equiv. of (R)-(Ϫ)-2,2,2-tri-
fluoro-1-(9-anthryl)ethanol as a white solid identical to that
obtained above; [α]D ϩ22.1 (c 1.0 in CHCl3).
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
531

View Article Online
(1R,3S,6S )-7-Diphenylphosphoryl-7-azabicyclo[4.1.0]hept-4-en-
3-ol 24
chromatography on silica with EtOAc–MeOH (20 : 1) as eluent
gave allylic alcohol rac-26 (144 mg, 72%) as a white solid, mp
187–190 ЊC (from 20 : 1 EtOAc–MeOH); RF(20 : 1 EtOAc–
MeOH) 0.25; νmax(CHCl3)/cmϪ1 3604 (OH), 3328 (OH), 2998,
Using general method D, n-butyllithium (0.38 cm3 of a 1.5 M
solution in hexane, 0.58 mmol), diamine (1R,2S)-23 (126 mg,
0.58 mmol) and epoxide cis-11 (150 mg, 0.48 mmol) in THF
(5 cm3) gave the crude product. Purification by flash column
chromatography on silica with EtOAc–MeOH (20 : 1) as eluent
1439 (P–Ph), 1189 (P᎐O), 1126, 1015, 779, 738, 705 and 666;
᎐
δH(270 MHz; CDCl3) 7.89–7.80 (4 H, m, o-Ph2PO), 7.54–7.38
(6 H, m, m- and p-Ph2PO), 5.96–5.93 (2 H, m, H4 and H5), 4.59–
4.48 (1 H, m, H3), 3.21 (1 H, tdd, J 2.0, 6.0 and 15.5, H1), 3.13
(1 H, tdd, J 3.0, 6.0 and 14.5, H6), 2.52 (1 H, ddd, J 2.0, 7.5 and
14.0, H2), 1.90* (1 H, d, J 6.5, OH) and 1.48 (1 H, tdd, J 2.0,
10.5 and 14.0, H2); δC(67.9 MHz; CDCl3) 137.7 (C4), 132.8 (d,
JCP 127.5, ipso-Ph2PO), 132.7 (d, JCP 127.5, ipso-Ph2PO), 131.9,
131.4 (d, JCP 9.5), 131.3 (d, JCP 9.5), 128.5 (d, JCP 12.5), 128.4 (d,
JCP 12.5), 123.0 (d, JCP 5.5, C5), 64.0 (C3), 33.6 (d, JCP 7.0, C1),
31.2 (d, JCP 5.5,C6) and 30.1 (d, JCP 5.5, C5); m/z (CI; NH3) 312
[100%, (M ϩ H)ϩ][Found: (M ϩ H)ϩ, 312.1143. C18H18NO2P
requires M ϩ H, 312.1153]. The 1H and 13C NMR spectra were
fully assigned using 1H–1H and 1H–13C COSY experiments.
1
gave allylic alcohol (1R,3S,6S)-24 (90 mg, 60%, 25% ee by H
NMR spectroscopy in the presence of 2 equiv. of (R)-(Ϫ)-2,2,2-
trifluoro-1-(9-anthryl)ethanol as a white solid identical to that
obtained above; [α]D ϩ12.5 (c 1.0 in CHCl3).
(1R*,3S*,6S*)-7-[(4-Methylphenyl)sulfonyl]-7-azabicyclo-
[4.1.0]hept-4-en-3-ol 25
Using general method D, n-butyllithium (0.43 cm3 of a 1.6 M
solution in hexane, 0.68 mmol), diamine rac-22 (139 mg, 0.68
mmol) and epoxide cis-12 (150 mg, 0.57 mmol) in THF (5 cm3)
gave the crude product. Purification by flash column chromato-
graphy on silica with EtOAc–petrol (1 : 1) as eluent gave allylic
alcohol rac-25 (61 mg, 41%) as a white solid, mp 82–84 ЊC (from
1 : 1 EtOAc–petrol); RF(1 : 1 EtOAc–petrol) 0.3; νmax(CHCl3)/
cmϪ1 3360 (OH), 3030, 1408, 1335, 1158, 1064, 814 and 663;
δH(270 MHz; CDCl3) 7.74 (2 H, d, J 8.0, o-C6H4SO2), 7.30 (2 H,
d, J 8.0, m-C6H4SO2), 6.06 (1 H, dd, J 4.0 and 9.5, H5), 5.78–
5.72 (1 H, m, H4), 5.06* (1 H, d, J 10.5, OH), 3.99–3.91 (1 H, m,
H3), 3.60–3.57 (1 H, m, H1), 3.35 (1 H, ddt, J 0.5, 1.5 and 4.0,
H6), 2.43 (3 H, s, Me), 2.31–2.25 (1 H, m, H2) and 1.72 (1 H, dd,
J 5.5 and 16.0, H2); δC(67.9 MHz; CDCl3) 143.1 (ipso-
C6H4SO2), 138.9 (ipso-C6H4Me), 132.3 (C4), 129.7, 127.1 (C5),
126.9, 55.2 (C1), 47.0 (C6), 45.9 (C3), 28.3 (C2) and 21.5 (Me);
m/z (CI; NH3) 283 [15%, (M ϩ NH4)ϩ], 266 [100, (M ϩ H)ϩ]
and 110 (15, M Ϫ Ts)[Found: (M ϩ H)ϩ, 266.0854. C13H15-
NO3S requires M ϩ H, 266.0851]. The 1H and 13C NMR
(1R,3S,6R)-7-Diphenylphosphoryl-7-azabicyclo[4.1.0]hept-4-
en-3-ol 26
Using general method D, n-butyllithium (0.36 cm3 of a 1.5 M
solution in hexane, 0.58 mmol), diamine (R)-22 (118 mg, 0.58
mmol) and epoxide trans-11 (150 mg, 0.48 mmol) in THF
(5 cm3) gave the crude product. Purification by flash column
chromatography on silica with EtOAc–MeOH (20 : 1) as eluent
gave allylic alcohol (1R,3S,6R)-26 (93 mg, 63%, 58% ee by
Mosher’s ester formation) as a white solid identical to that
obtained above; [α]D Ϫ81.0 (c 1.0 in CHCl3).
(1R,3S,6R)-7-Diphenylphosphoryl-7-azabicyclo[4.1.0]hept-4-
en-3-ol 26
Using general method D, n-butyllithium (0.36 cm3 of a 1.5 M
solution in hexane, 0.58 mmol), diamine (1R,2S)-23 (126 mg,
0.58 mmol) and epoxide trans-11 (150 mg, 0.48 mmol) in THF
(5 cm3) gave the crude product. Purification by flash column
chromatography on silica with EtOAc–MeOH (20 : 1) as eluent
gave allylic alcohol (1R,3S,6R)-26 (123 mg, 82%, 68% ee by
Mosher’s ester formation) as a white solid identical to that
obtained above; [α]D Ϫ93.3 (c 1.0 in CHCl3).
1
1
spectra were fully assigned using H–1H and H–13C COSY
experiments.
(1R,3S,6S )-7-[(4-Methylphenyl)sulfonyl]-7-azabicyclo[4.1.0]-
hept-4-en-3-ol 25
Using general method D, n-butyllithium (0.57 cm3 of a 1.6 M
solution in hexane, 0.91 mmol), diamine (R)-22 (185 mg, 0.91
mmol) and epoxide cis-11 (200 mg, 0.76 mmol) in THF (5 cm3)
gave the crude product. Purification by flash column chromato-
graphy on silica with EtOAc–petrol (1 : 1) as eluent gave allylic
alcohol (1R,3S,6S)-25 (28 mg, 14%, 66% ee chiral HPLC) as a
white solid identical to that obtained above; [α]D ϩ114.6 (c 1.0
in CHCl3); HPLC: Chiralcel AS, 25% iPrOH in heptane, 1.0 cm3
minϪ1, 215 nm, 22.2 min [(1S,3R,6R)-25], 24.9 min [(1R,3S,6S)-
25].
(1R*,3S*,6R*)-7-[(4-Methylphenyl)sulfonyl]-7-azabicyclo-
[4.1.0]hept-4-en-3-ol 27
Using general method D, n-butyllithium (0.43 cm3 of a 1.6 M
solution in hexane, 0.68 mmol), diamine rac-22 (139 mg, 0.68
mmol) and epoxide trans-12 (150 mg, 0.57 mmol) in THF
(5 cm3) gave the crude product. Purification by flash column
chromatography on silica with EtOAc–petrol (1 : 1) as eluent
gave allylic alcohol rac-27 (39 mg, 26%) as a white solid, mp
96–98 ЊC (from 1 : 1 EtOAc–petrol); RF(1 : 1 EtOAc–petrol) 0.3;
νmax(CHCl3)/cmϪ1 3380 (OH), 3030, 1599, 1414, 1335, 1160,
1081, 878, 764 and 663; δH(270 MHz; CDCl3) 7.76 (2 H, d,
J 8.0, o-C6H4SO2), 7.31 (2 H, d, J 8.0, m-C6H4SO2), 5.97 (1 H,
(1R,3S,6S )-7-[(4-Methylphenyl)sulfonyl]-7-azabicyclo[4.1.0]-
hept-4-en-3-ol 25
Using general method D, n-butyllithium (0.57 cm3 of a 1.6 M
solution in hexane, 0.91 mmol), diamine (1R,2S)-23 (197 mg,
0.91 mmol) and epoxide cis-11 (200 mg, 0.76 mmol) in THF
(5 cm3) gave the crude product. Purification by flash column
chromatography on silica with EtOAc–petrol (1 : 1) as eluent
gave allylic alcohol (1R,3S,6S)-25 (45 mg, 23%, 10% ee chiral
HPLC) as a white solid identical to that obtained above; [α]D
ddd, J 3.0, 4.0 and 10.0, ᎐CH), 5.58 (1 H, dd, J 1.0 and 10.0,
᎐
᎐CH), 4.65* (1 H, d, J 9.5, OH), 3.93–3.81 (1 H, m, H3), 3.42–
᎐
3.40 (1 H, m, CHN), 3.23 (1 H, dt, J 2.0 and 4.0, CHN), 2.53
(ddd, J 2.0, 8.0 and 14.5, H2), 2.44 (3 H, s, Me) and 1.42
(1 H, dd, J 11.5 and 14.5, H2); δC(67.9 MHz; CDCl3) 143.6
i
(ipso-C H SO ), 137.4 (ipso-C H Me), 134.8 (᎐CH), 129.7,
ϩ18.5 (c 1.0 in CHCl3); HPLC: Chiralcel AS, 25% PrOH in
᎐
6
4
2
6
4
127.1 (᎐CH), 125.2, 51.4 (CHN), 46.7 (CHN), 46.0 (C3), 29.5
heptane, 1.0 cm3 minϪ1, 215 nm, 22.1 min [(1S,3R,6R)-25], 25.3
min [(1R,3S,6S)-25].
᎐
(C2) and 21.5 (Me); m/z (CI; NH3) 283 [30%, (M ϩ NH4)ϩ], 266
[100, (M ϩ H)ϩ] and 110 (25, M Ϫ Ts)[Found: (M ϩ H)ϩ,
266.0854. C13H15NO3S requires M ϩ H, 266.0851].
(1R*,3S*,6R*)-7-Diphenylphosphoryl-7-azabicyclo[4.1.0]hept-
4-en-3-ol 26
N-((1R,4R,5S )-4,5-Bis{[tert-butyl(dimethyl)silyl]oxy}cyclohex-
2-en-1-yl)-N-(4-methoxybenzyl)-4-methylbenzenesulfonamide 30
Using general method D, n-butyllithium (0.51 cm3 of a 1.5 M
solution in hexane, 0.77 mmol), diamine rac-22 (157 mg, 0.77
mmol) and epoxide trans-11 (200 mg, 0.64 mmol) in THF
(5 cm3) gave the crude product. Purification by flash column
Using general method E, N-(4-methoxybenzyl)-4-methyl-
benzenesulfonamide (122 mg, 0.42 mmol), triphenylphosphine
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
532

View Article Online
(219 mg, 0.84 mmol), allylic alcohol (1S,4R,5S)-410 (100 mg,
0.28 mmol, >95% ee) and diisopropyl azodicarboxylate (50 µL,
0.28 mmol) in THF (3 cm3) gave the crude product. Purification
by flash column chromatography on silica with petrol–Et2O
(3 : 1) as eluent gave allylic amine (1R,4R,5S)-30 (160 mg, 91%,
>95% ee) as a white solid, mp 98–100 ЊC (from 3 : 1 petrol–
Et2O); RF(3 : 1 petrol–Et2O) 0.3; [α]D ϩ25.0 (c 1.0 in CHCl3);
νmax(CHCl3)/cmϪ1 3034, 2957, 2632, 1513, 1338, 1251, 1162,
1119, 1037 and 837; δH(270 MHz; CDCl3) 7.70 (2 H, d, J 8.5,
o-C6H4SO2), 7.32 (2 H, d, J 8.5, m-C6H4SO2), 7.28 (2 H, d, J 8.5,
m- C6H4OMe), 6.80 (2 H, d, J 8.5, o- C6H4OMe), 5.64 (1 H,
pressure. Baseline impurities were removed from the residue by
filtration through a plug of silica with petrol–Et2O (3 : 1) as
eluent. The solvent was evaporated under reduced pressure to
give the crude product. To a stirred solution of the crude
product in xylene (30 cm3) was added potassium carbonate
(“catalytic” amount). The resulting mixture was heated at reflux
for 18 h. After cooling to room temperature, the solvent was
evaporated under reduced pressure to give the crude prod-
uct. Purification by flash column chromatography on silica
with petrol–Et2O (10 : 1) as eluent gave trichloroacetamide
(1S,5S,6R)-33 (1.32 g, 95%, >95% ee) as a white solid, mp 69–
71 ЊC (from 10 : 1 petrol–Et2O); RF(10 : 1 petrol–Et2O) 0.35;
[α]D ϩ65.0 (c 1.0 in CHCl3); νmax(CHCl3)/cmϪ1 3246 (NH),
ddd, J 2.5, 6.0 and 10.0, ᎐CH), 5.13 (1 H, br d, J 10.0, ᎐CH),
᎐
᎐
4.59–4.51 (1 H, m, CHN), 4.39 (1 H, d, J 15.5, NCHAHB), 4.20
(1 H, d, J 15.5, NCHAHB), 3.89–3.85 (1 H, m, CHO), 3.79 (3 H,
s, OMe), 3.52–3.44 (1 H, m, CHO), 2.42 (3 H, s, Me), 2.07–1.94
(1 H, m, CHAHB), 1.36–1.25 (1 H, m, CHAHB), 0.86 (9 H, s,
CMe3), 0.82 (9 H, s, CMe3), 0.02 (6 H, s, 2 × SiMe), Ϫ0.01 (3 H,
s, SiMe) and Ϫ0.04 (3 H, s, SiMe); δC(67.9 MHz; CDCl3) 158.7
(ipso-C6H4OMe), 143.1 (ipso-C6H4SO2), 138.3 (ipso-C6H4Me),
130.7, 130.5, 129.7, 127.1, 113.4 (o-C6H4OMe), 69.7, 66.6, 55.7,
55.2 (OMe), 47.0 (NCH2), 29.4 (CH2), 26.0 (CMe3), 25.8
(CMe3), 21.5 (Me), 18.3 (CMe3), 18.2 (CMe3), Ϫ4.4 (SiMe),
Ϫ4.5 (SiMe), Ϫ4.6 (SiMe) and Ϫ4.8 (SiMe) (two aromatic
resonances not resolved); m/z (CI; NH3) 574 (10%, M Ϫ CMe3),
524 (10, M Ϫ p-MeOC6H4), 500 (100), 476 (35, M Ϫ Ts) and
121 (100). It was not possible to obtain HRMS data on this
compound.
2930, 2857, 1713 (C᎐O), 1494, 1256 and 837; δ (270 MHz;
᎐
H
CDCl ) 6.45–6.42 (1 H, m, NH), 5.88 (1 H, br d, J 9.0, ᎐CH),
᎐
3
5.54 (1 H, br d, J 9.0, ᎐CH), 4.34 (1 H, br s, CHN), 3.89–3.86
᎐
(1 H, m, CHO), 3.80 (1 H, br s, CHO), 2.43–2.30 (1 H, m,
CHAHB), 2.16 (1 H, br d, J 17.5, CHAHB), 0.89 (9 H, s, CMe3),
0.88 (9 H, s, CMe3), 0.17 (3 H, s, SiMe), 0.12 (3 H, s, SiMe), 0.07
(3 H, s, SiMe) and 0.05 (3 H, s, SiMe); δC(67.9 MHz; CDCl3)
161.0 (C᎐O), 131.2 (᎐CH), 122.1 (᎐CH), 92.5 (CCl ), 72.1, 68.5,
᎐
᎐
᎐
3
54.0, 30.8 (CH2), 25.9 (CMe3), 25.8 (CMe3), 18.2 (CMe3), 18.0
(CMe3), Ϫ4.4 (SiMe), Ϫ4.5 (SiMe), Ϫ4.7 (SiMe) and Ϫ4.8
(SiMe); m/z (CI; NH3) 521 [20%, (35,35,37M ϩ NH4)ϩ], 504 [55,
(
35,35,37M ϩ H)ϩ], 432 (100), 398 (85), 372 (40), 300 (40)
and 266[Found: (M ϩ H)ϩ, 502.1524. C20H38NO3Si235,35,35Cl3
requires M ϩ H, 502.1534]; Found: C, 47.9; H, 7.8; N, 2.8%;
C20H38NO3Si2Cl3 requires C, 47.8; H, 7.6; N, 2.8%.
N-((1S,4R,5S )-4,5-Bis{[tert-butyl(dimethyl)silyl]oxy}cyclohex-
2-en-1-yl)-N-(4-methoxybenzyl)-4-methylbenzenesulfonamide 31
N-((1R,2S,3R,4S,6S )-3,4-Bis{[tert-butyl(dimethyl)silyl]oxy}-7-
oxabicyclo[4.1.0]hept-2-yl)-2,2,2-trichloroacetamide 36
Using general method E, N-(4-methoxybenzyl)-4-methyl-
benzenesulfonamide (244 mg, 0.84 mmol), triphenylphosphine
(438 mg, 1.68 mmol), allylic alcohol (1R4R,5S)-2910 (200 mg,
0.56 mmol, >95% ee) and diisopropyl azodicarboxylate (250
µL, 1.4 mmol) in THF (6 cm3) gave the crude product. Purifi-
cation by flash column chromatography on silica with petrol–
Et2O (3 : 1) as eluent gave allylic amine (1R,4R,5S)-31 (210 mg,
60%, >95% ee) as a white solid, mp 84–86 ЊC (from 3 : 1 petrol–
Et2O); RF(3 : 1 petrol–Et2O) 0.3; [α]D Ϫ44.8 (c 1.0 in CHCl3);
νmax(CHCl3)/cmϪ1 3032, 2957, 2632, 1513, 1339, 1253, 1160,
1096, 837 and 748; δH(270 MHz; CDCl3) 7.70 (2 H, d, J 8.5,
o-C6H4SO2), 7.28 (4 H, d, J 8.5, m- C6H4SO2 and m- C6H4-
OMe), 6.84 (2 H, d, J 8.5, o- C6H4OMe), 5.44 (1 H, br d, J 10.0,
᎐CH), 5.06 (1 H, br d, J 10.0, ᎐CH), 4.75–4.67 (1 H, m, CHN),
Using general method A, MCPBA (700 mg of 70% pure
material, 4.0 mmol), sodium hydrogen carbonate (350 mg, 4.0
mmol) and alkene 33 (1.0 g, 2.0 mmol, >95% ee) in CH2Cl2
(30 cm3) gave the crude product. Purification by flash column
chromatography on silica with petrol–Et2O (10 : 1) as eluent
gave epoxide 36 (960 mg, 93%, >95% ee) as a white solid, mp
65–67 ЊC (from 10 : 1 petrol–Et2O); RF(10 : 1 petrol–Et2O) 0.3;
[α]D ϩ21.5 (c 1.0 in CHCl3); δH(270 MHz; CDCl3) 6.98–6.95
(1 H, m, NH), 4.29 (1 H, br s, CHN), 3.73–3.67 (1 H, m, CHO),
3.65–3.62 (1 H, m, CHO), 3.45 (1 H, br s, CHO), 3.36–3.31
(1 H, m, CHO), 2.22–2.06 (2 H, m, CH2), 0.91 (9 H, s, CMe3),
0.88 (9 H, s, CMe3), 0.17 (3 H, s, SiMe), 0.11 (3 H, s, SiMe) and
0.04 (6 H, s, SiMe); δ (67.9 MHz; CDCl ) 161.5 (C᎐O), 92.3
᎐
᎐
᎐
C
3
4.58 (1 H, d, J 15.5, NCHAHB), 4.07 (1 H, d, J 15.5, NCHAHB),
4.00–3.96 (1 H, m, CHO), 3.91–3.86 (1 H, m, CHO), 3.80 (3 H,
s, OMe), 2.42 (3 H, s, Me), 1.90–1.79 (1 H, m, CHAHB), 1.46–
1.36 (1 H, m, CHAHB), 0.87 (18 H, s, 2 × CMe3), 0.02 (6 H, s, 2
× SiMe), 0.01 (3 H, s, SiMe) and 0.00 (3 H, s, SiMe); δC(67.9
MHz; CDCl3) 158.9 (ipso-C6H4OMe), 143.1 (ipso-C6H4SO2),
137.9 (ipso-C6H4Me), 133.5, 130.7, 129.7, 129.2, 127.2, 127.1,
113.7 (o-C6H4OMe), 70.8, 69.4, 55.2, 53.0 (OMe), 47.7 (NCH2),
35.7 (CH2), 26.0 (CMe3), 25.8 (CMe3), 21.5 (Me), 18.3 (CMe3),
18.2 (CMe3), Ϫ4.4 (SiMe), Ϫ4.5 (SiMe), Ϫ4.8 (SiMe) and Ϫ5.0
(SiMe); m/z (CI; NH3) 574 (10%, M Ϫ CMe3), 500 (60), 476 (45,
M Ϫ Ts) and 121 (100). It was not possible to obtain HRMS
data on this compound.
(CCl3), 70.9, 65.8, 56.9, 52.3, 50.4, 28.5 (CH2), 25.8 (CMe3),
25.7 (CMe3), 18.1 (CMe3), 18.0 (CMe3), Ϫ4.5 (SiMe), Ϫ4.7
(SiMe), Ϫ4.8 (SiMe) and Ϫ4.9 (SiMe); m/z (CI; NH3) 537
[100%, (35,35,37M ϩ NH4)ϩ][Found: (M ϩ NH4)ϩ, 535.1747.
C20H38NO4Si235,35,35Cl3 requires M ϩ NH4, 535.1749]; Found:
C, 46.4; H, 7.4; N, 2.6%; C20H38NO3Si2Cl3 requires C, 46.6; H,
7.4; N, 2.7%.
Acknowledgements
We thank the EPSRC for the award of a project studentship
(to C. D. P. reference GR/L 58439) and GlaxoSmithKline’s
physical science department for the chiral HPLC results. We
also thank Laura Hester for carrying out the direct preparation
of aziridine 10 from cyclohexa-1,4-diene using the Sharpless
aziridination.
N-((1S,5S,6R)-5,6-Bis{[tert-butyl(dimethyl)silyl]oxy}cyclohex-
2-en-1-yl)-2,2,2-trichloroacetamide 33
DBU (0.5 cm3, 3.3 mmol) and then trichloroacetonitrile
(0.36 cm3, 3.6 mmol) were added sequentially to a stirred solu-
tion of the allylic alcohol (1S,4R,5S)-410 (1.0 g, 2.8 mmol,
>95% ee) in CH2Cl2 (30 cm3) at Ϫ20 ЊC under nitrogen. After
being allowed to warm to room temperature over 2 h, the result-
ing solution was stirred at room temperature for 18 h. Then,
Et2O (50 cm3) was added and the organic layer was washed
with saturated aqueous sodium hydrogen carbonate solution
(20 cm3), dried (Na2SO4) and evaporated under reduced
References
1 For reviews, see: A. Magnus, S. K. Bertilsson and P. G. Andersson,
Chem. Soc. Rev., 2002, 31, 223; J. Eames, Eur. J. Org. Chem., 2002,
393; P. O’Brien, J. Chem. Soc., Perkin Trans. 1, 1998, 1439; D. M.
Hodgson, A. R. Gibbs and G. P. Lee, Tetrahedron, 1996, 52, 14361;
P. J. Cox and N. S. Simpkins, Tetrahedron: Asymmetry, 1991, 2, 1.
2 A. Gayet, S. Bertilsson and P. G. Andersson, Org. Lett., 2002, 4,
3777; S. K. Bertilsson and P. G. Andersson, Tetrahedron, 2002, 58,
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
533

View Article Online
4665; S. K. Bertilsson, M. J. Södergren and P. G. Andersson,
J. Org. Chem., 2002, 67, 1567.
3 D. Pettersen, M. Amedjkouh and P. Ahlberg, Tetrahedron, 2002, 58,
4669.
29 D. A. Evans, M. M. Faul and M. T. Bilodeau, J. Am. Chem. Soc.,
1994, 116, 2742.
30 J. U. Jeong, B. Tao, I. Sagasser, H. Henniges and K. B. Sharpless,
J. Am. Chem. Soc., 1998, 120, 6844.
4 A. Seki and M. Asami, Tetrahedron, 2002, 58, 4655.
5 D. Liu and S. A. Kozmin, Angew. Chem., Int. Ed., 2001, 40, 4757.
6 J. Leonard, J. D. Hewitt, D. Ouali, L. R. Bennett, A. Mahmood and
S. J. Simpson, Tetrahedron, 2002, 58, 4681.
7 S. E. de Sousa, P. O’Brien and H. C. Steffens, Tetrahedron Lett.,
1999, 40, 8423.
31 D. Yang, M.-K. Wong and Y.-C. Yip, J. Org. Chem., 1995, 60,
3887.
32 A. H. Hoveyda, D. A. Evans and G. C. Fu, Chem. Rev., 1993, 93,
1307; G. Berti, Top. Stereochem., 1973, 7, 93.
33 T. W. Craig, G. R. Harvey and G. A. Berchtold, J. Org. Chem., 1967,
32, 3743.
8 B. Colman, S. E. de Sousa, P. O’Brien, T. D. Towers and W. Watson,
Tetrahedron: Asymmetry, 1999, 10, 4175.
9 P. O’Brien and P. Poumellec, Tetrahedron Lett., 1996, 37, 8057;
P. O’Brien and P. Poumellec, J. Chem. Soc., Perkin Trans. 1, 1998,
2435.
34 L. A. Paquette, W. E. Fristad, C. A. Schuman, M. A. Beno and
G. G. Christoph, J. Am. Chem. Soc., 1979, 101, 4645.
35 J. Clayden, E. W. Collington, E. Egert, A. B. McElroy and
S. Warren, J. Chem. Soc., Perkin Trans. 1, 1994, 2801.
36 A. M. Costero and S. Rodriguez, Tetrahedron, 1992, 48, 6265.
37 K. B. Sharpless and R. C. Michaelson, J. Am. Chem. Soc., 1973, 95,
6136.
10 S. E. de Sousa, P. O’Brien and C. D. Pilgram, Tetrahedron, 2002, 58,
4643.
11 P. O’Brien, T. D. Towers and M. Voith, Tetrahedron Lett., 1998, 39,
8175; S. Barrett, P. O’Brien, H. C. Steffens, T. D. Towers and
M. Voith, Tetrahedron, 2000, 56, 9633.
12 P. O’Brien and C. D. Pilgram, Tetrahedron Lett., 1999, 40, 8427.
13 A. Kee, P. O’Brien, C. D. Pilgram and S. T. Watson, Chem.
Commun., 2000, 1521.
38 M. H. Wu and E. N. Jacobsen, Tetrahedron Lett., 1997, 38, 1693.
39 Y. G. Gololobov, I. N. Zhmurova and L. F. Kasukhin, Tetrahedron,
1981, 37, 437.
40 M. Asami, Bull. Chem. Soc. Jpn., 1990, 63, 721.
41 M. Asami, T. Suga, K. Honda and S. Inoue, Tetrahedron Lett., 1997,
38, 6425.
14 P. O’Brien and J. J. Tournayre, Tetrahedron, 1997, 53, 17527.
15 S. E. de Sousa, P. O’Brien and C. D. Pilgram, Tetrahedron Lett.,
2001, 42, 8081.
16 E. Baron, P. O’Brien and T. D. Towers, Tetrahedron Lett., 2002, 43,
723.
17 B. J. Paul, J. Willis, T. A. Martinot, I. Ghiviroga, K. A. Abboud and
T. Hudlicky, J. Am. Chem. Soc., 2002, 124, 10416; B. M. Trost,
J. Dudash and E. J. Hembre, Chem. Eur. J., 2001, 7, 1619;
A. Arecelli, V. Cerè, F. Peri, S. Pollicini and A. Ricci, Tetrahedron,
2001, 57, 3439; M. Desjardins, M. C. Lallemand, S. Freeman,
T. Hudlicky and K. A. Abboud, J. Chem. Soc., Perkin Trans. 1, 1999,
628; B. M. Trost, L. S. Chupak and T. Lübbers, J. Am. Chem. Soc.,
1998, 120, 17 32.
42 M. J. Södergren, S. K. Bertilsson and P. G. Andersson, J. Am. Chem.
Soc., 2000, 122, 6610.
43 D. Bhuniya, A. DattaGupta and V. K. Singh, J. Org. Chem., 1996,
61, 6108.
44 D. Pettersen, M. Amedjkouh, S. O. Nilsson Lill, K. Dahlén and
P. Ahlberg, J. Chem. Soc., Perkin Trans. 2, 2001, 1654.
45 S. E. de Sousa, P. O’Brien and P. Poumellec, Tetrahedron:
Asymmetry, 1997, 8, 2613; S. E. de Sousa, P. O’Brien and
P. Poumellec, J. Chem. Soc., Perkin Trans. 1, 1998, 1483.
46 P. Müller and P. Nury, Helv. Chim. Acta, 2001, 84, 662.
47 A. Mordini, L. Sbaragli, M. Valacchi, F. Russo and G. Reginato,
Chem. Commun., 2002, 778; A. Mordini, F. Russo, M. Valacchi,
L. Zani, A. Degl’Innocenti and G. Reginato, Tetrahedron, 2002, 58,
7153.
48 J. G. Knight and M. P. Muldowney, Synlett., 1995, 949.
49 I. Ohtani, T. Kusumi, Y. Kashman and H. Kakisawa, J. Am. Chem.
Soc., 1991, 113, 4092; J. A. Dale, D. L. Dull and H. S. Mosher,
J. Org. Chem., 1969, 34, 2543.
18 Some aspects of this work has previously been communicated: see
references 12 and 15.
19 D. L. Hughes, Org. React., 1992, 42, 335; K. E. Bell, D. W. Knight
and M. B. Gravestock, Tetrahedron Lett., 1995, 36, 8681;
C. P. Decicco and P. Grover, Synlett., 1997, 529; J. A. Campbell and
D. J. Hart, J. Org. Chem., 1993, 58, 2900; F. Berrée, G. Michelot
and M. Le Corre, Tetrahedron Lett., 1998, 39, 8275; J. R. Henry,
L. R. Marcin, M. C. McIntosh, P. M. Scola, G. D. Harris and
S. M. Weinreb, Tetrahedron Lett., 1989, 30, 5709.
50 W. R. McKay and G. R. Proctor, J. Chem. Soc., Perkin Trans. 1,
1981, 2443.
51 B. K. Shull, T. Sakai, J. B. Nichols and M. Koreeda, J. Org. Chem.,
1997, 62, 8294.
20 T. Fukuyama, C.-K. Jow and M. Cheung, Tetrahedron Lett., 1995,
36, 6373; T. Fukuyama, M. Cheung, C.-K. Jow, Y. Hidai and T. Kan,
Tetrahedron Lett., 1997, 38, 5831.
21 L. E. Overman, J. Am. Chem. Soc., 1976, 98, 2901.
22 L. A. Paquette, D. E. Kuhla, J. H. Barrett and R. J. Haluska,
J. Org. Chem., 1969, 34, 2866.
23 B. Zipperer, K.-H. Muller, B. Gallenkamp, R. Hildebrand,
M. Fletschinger, D. Burger, M. Pillat, D. Hunkler, L. Knothe,
H. Fritz and H. Prinzbach, Chem. Ber., 1988, 121, 757.
24 S. E. de Sousa, A. Kee, P. O’Brien and S. T. Watson, Tetrahedron
Lett., 1999, 40, 387.
25 S. E. de Sousa, P. O’Brien, C. D. Pilgram, D. Roder and T. D.
Towers, Tetrahedron Lett., 1999, 40, 391.
52 T. Nishikawa, M. Asai, N. Ohyabu and M. Isobe, J. Org. Chem.,
1998, 63, 188.
53 T. Hudlicky, K. A. Abboud, D. A. Entwistle, R. Fan, R. Maurya,
A. J. Thorpe, J. Bolonick and B. Myers, Synthesis, 1996, 897.
54 H. M. I. Osborn and J. B. Sweeney, Synlett., 1994, 145.
55 A. A. Cantrill and J. B. Sweeney, Synlett., 1995, 1227.
56 B. M. Trost and D. E. Patterson, Chem. Eur. J., 1999, 5, 3279.
57 G. Legler and R. Bollhayen, Carbohydr. Res., 1992, 233, 113;
W. J. Brouilette, A. Saeed, A. Abuelyaman, T. L. Hutichison,
P. E. Wolkowicz and J. B. McMillin, J. Org. Chem., 1994, 59,
4297.
58 J. K. Crandall and M. Apparu, Org. React. (N. Y.), 1983, 29,
345.
26 A. Hassner and C. Heathcock, Tetrahedron, 1964, 20, 1037.
27 R. Ramage, D. Hopton, M. J. Parrott, G. W. Kenner and
G. A. Moore, J. Chem. Soc., Perkin Trans. 1, 1984, 1357; R. Ramage,
B. Atrash, D. Hopton and M. J. Parrott, J. Chem. Soc., Perkin Trans.
1, 1985, 1217.
28 H. M. I. Osborn, J. Sweeney and B. Howson, Synlett., 1994, 145;
A. A. Cantrill, H. M. I. Osborn and J. Sweeney, Tetrahedron, 1998,
54, 2181.
59 K. B. Sharpless and R. F. Lauer, J. Am. Chem. Soc., 1973, 95, 2697;
T. Izuhara and T. Katoh, Tetrahedron Lett., 2000, 41, 7651.
60 S. Murata, M. Suzuki and R. Noyori, Bull. Chem. Soc. Jpn., 1982,
55, 247.
61 A. F. Burchat, J. M. Chong and N. Nielson, J. Organomet. Chem.,
1997, 542, 281.
62 J. G. Hill, K. B. Sharpless, C. M. Exon and R. Regenye, Org. Synth.,
1984, 63, 66.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 5 2 3 – 5 3 4
534