Catalytic asymmetric heterogeneous aziridination of alkenes using zeolite CuHY with [N-(p-tolylsulfonyl)imino]phenyliodinane as nitrene donor

Article abstract of DOI:10.1039/a806409a

Copper-exchanged zeolite Y (CuHY) is found to be a highly effective heterogeneous catalyst for the aziridination of alkenes using [N-(p-tolylsulfonyl)imino]phenyliodinane (PhI=NTs) as the nitrogen source. Exchange of zeolite Y with other cations (Ag⁺

Full text of DOI:10.1039/a806409a

View Article Online / Journal Homepage / Table of Contents for this issue
Catalytic asymmetric heterogeneous aziridination of alkenes
using zeolite CuHY with [N-(p-tolylsulfonyl)imino]phenyliodinane
as nitrene donor
Christopher Langham,a Sophia Taylor,a Donald Bethell,b Paul McMorn,a Philip C. Bulman
Page,c David J. Willock,a Chris Sly,d Frederick E. Hancock,e Frank Kinge and
Graham J. Hutchings*a
a Department of Chemistry, University of Wales Cardiff, PO Box 912, Cardiff, UK CF1 3TB
b Leverhulme Centre for Innovative Catalysis, Department of Chemistry, University of
Liverpool, Liverpool, UK L69 3BX
c Department of Chemistry, Loughborough University, Loughborough, Leicestershire,
UK LE11 3TU
d Robinson Brothers Ltd., Phoenix Street, West Bromwich, West Midlands, UK B70 OAH
e ICI Katalco, R&T Division, PO Box 1, Billingham, Teesside, UK TS23 1LB
Received (in Cambridge) 13th August 1998, Accepted 16th March 1999
Copper-exchanged zeolite Y (CuHY) is found to be a highly effective heterogeneous catalyst for the aziridination
of alkenes using [N-(p-tolylsulfonyl)imino]phenyliodinane (PhI᎐NTs) as the nitrogen source. Exchange of zeolite Y
᎐
with other cations (Agϩ, Co2ϩ, Fe3ϩ, Mg2ϩ, Ni2ϩ, Zn2ϩ) was found to be ineffective. This is considered to be due to the
ability of these metals to catalyse the breakdown of the PhI᎐NTs reagent into iodobenzene and toluene sulfonamide.
᎐
Modification of the CuHY catalyst with bis(oxazolines) leads to preparation of the first heterogeneous enantio-
selective aziridination catalyst and the results showing the effect of temperature and modifier concentration are
described and discussed. A pyridine-bridged bis(oxazoline) was observed to give the highest enantioselectivity
of 61% ee for the aziridination of styrene using acetonitrile as solvent and at Ϫ10 ЊC.
Introduction
commercialized. A central problem with all these homogeneous
The identification of new commercial catalysts for the synthesis
of the building blocks for pharmaceuticals and agrochemicals
continues to attract public interest. In this respect, there is a
need for catalysts to be designed that are stable, give high yields
of the desired products, whether achiral or enantiomerically
enriched, and can be readily re-used. Most attention has
focused on the use of transition metal complexes in homo-
geneous solution processes, and much progress has been made
with these systems. The identification, however, of a general
approach to the design of suitable heterogeneous catalysts,
has until now proved elusive since most examples tend to be sub-
strate specific. Herein we report an approach based on zeolites
that can in principle be generalized to the design of hetero-
geneous catalysts that have a potential to become commercially
viable. We illustrate it by the asymmetric aziridination of
alkenes, the first example of a heterogeneous catalyst reported
for this reaction.
The requirement for new routes to racemic or enantio-
merically pure intermediates for fine chemicals synthesis has
bought a resurgence in interest in homogeneous catalysis in
recent years.1 The reason for this is that metal complexes that
are used as catalysts can be located in well defined geometries
that are necessary for the achievement of high specificity for the
formation of the desired product. This approach to catalyst
design has been thoroughly studied for a range of cations and
reaction types; for example, titanium tartrate complexes for the
enantioselective epoxidation of allylic alcohols (the Sharpless
epoxidation reaction),2 Cu2ϩ complexes for the cyclopro-
panation of alkenes3 and aziridination of alkenes,4 manganese
salen complexes for the epoxidation of alkenes,5 and ruthenium
BINAP complexes for the hydrogenation of alkenes,6 are all
well established, and indeed the Sharpless epoxidation has been
systems relates to the separation and reuse of the catalyst
which is often very difficult and expensive. To overcome this
disadvantage considerable effort has been devoted to the
heterogenization of these catalysts. To date there have been a
number of approaches to heterogenization of homogeneous
catalysts7 but two have predominated. The first involves the use
of a thin layer of a suitable solvent containing the homo-
geneous catalyst that is supported on the surface of a high area
solid, an approach which can retain the performance of the
homogeneous catalyst.6 The second approach involves encap-
sulation of the homogeneous catalyst in the pores of a solid,
typically a zeolite, and this is often designated as a ‘ship in a
bottle’ method since the catalyst is constructed within the
pores.8 Both these approaches are general, but suffer the
disadvantage that the active catalyst usually degrades with time
and often is removed into the reaction medium through solu-
bilization. In contrast to these approaches we have used the
observations that zeolites are well known for their cation
exchange properties and that the cations are in many cases
known to be located in precise sites with the zeolite acting as a
ligand.9 We believe that such metal-exchanged zeolites could,
with suitable modification, be useful heterogeneous catalysts for
the reactions listed above. Zeolites have been used as catalysts
for fine chemicals synthesis for many years10 but the main
emphasis has been on their Brønsted acidity/basicity and their
well known shape selective properties. As an initial demon-
stration of our approach we focus on the aziridination of
alkenes, since this was known to be homogeneously catalysed
by a number of relatively simple copper compounds,4 and
no heterogeneous counterparts had yet been identified. Pre-
liminary accounts of this and related work using the same
catalyst design principles have already appeared.11,12
J. Chem. Soc., Perkin Trans. 2, 1999, 1043–1049
1043

View Article Online
Table 1 CuHY-catalysed aziridination of representative alkenes
Yieldb (%)
TsNH2
Entry
Alkenea
[Cu] (mol%)
PhI
Aziridine
1
2
3
4
5
6
7
8
9
Styrene
25
25
5
25
25
25
25
25
25
25
10
13
38
77
24
34
50
16
100
56
100
100
100
100
100
100
100
100
100
100
90 (92)
87 (35)
62
33
76
Styrenec
Styrene
α-Methylstyrene
p-Chlorostyrene
p-Methylstyrene
Cyclohexane
Methyl (E)-cinnamate
(E)-Stilbene
(E)-Hex-2-ene
66
50 (60)
84 (73)
0d (52)
44
10
a Unless otherwise specified reaction conditions were: solvent MeCN, 25 ЊC, styrene:PhI᎐NTs = 5:1 molar ratio. b Isolated yield of aziridine based
᎐
on PhI᎐NTs. c Styrene:PhI᎐NTs = 1:1 molar ratio. d No product observed by TLC. Values in parentheses indicate yields obtained from comparable
᎐
᎐
homogeneous reactions.
different from that found in homogeneous copper()-containing
acetonitrile solutions (see below).
Results and discussion
Heterogeneous aziridination of alkenes with CuHY
CuHY was found to be successful in catalysing the aziridin-
ation of a range of alkenes (Table 1) in addition to styrene. It is
observed that the catalyst gives best results with phenyl-
substituted alkenes, lower yields being observed with cyclohex-
ene and (E)-hex-2-ene. Interestingly, for the aziridination of
(E)-stilbene no product was observed, although in this case the
analysis was carried out by TLC.
Evans et al.4 have shown that copper triflate is a homogeneous
catalyst for the aziridination of a range of alkenes. Based on
this previous study we investigated copper-exchanged zeolites
as catalysts for this reaction and we have found that CuHY
zeolite is successful in catalysing the aziridination of a range of
alkenes employing [N-(p-tolylsulfonyl)imino]phenyliodinane
(PhI᎐NTs) as the nitrogen source. Reactions were carried out
Stirring PhINTs in an acetonitrile solution of a five-fold
excess of styrene at 25 ЊC slowly leads to the formation of
aziridine (22%) and toluene-p-sulfonamide (78%) over a period
of 72 hours. We believe, however, that our results indicate that
᎐
at 25 ЊC in stirred flasks containing a mixture of PhI᎐NTs
᎐
(typically 0.52 g), the CuHY (0.071 g, the amount being
adjusted for the copper content of the zeolite so that the total
amount of copper in the reaction mixture was 0.25 equivalents
the aziridination of alkenes by PhI᎐NTs in the presence of
᎐
per mol of PhI᎐NTs), and an acetonitrile solution of the
zeolite CuHY is almost wholly heterogeneous and takes place
within the pores of the zeolite. Our reasons are as follows.
1. The catalysed reaction is much faster; at the time when
it reaches completion, the uncatalysed pathway would have
᎐
alkene, usually in five-fold excess over PhI᎐NTs. Analysis of the
᎐
reaction mixtures for aziridine, toluenesulfonamide and iodo-
benzene was by HPLC. Reactions were followed until all the
PhINTs (which is only sparingly soluble in acetonitrile) had
disappeared and the yield of PhI by HPLC was constant (typic-
ally after 3 to 6 hours). The results are shown in Table 1. CuHY
was initially screened in the aziridination of styrene (Table 1,
entries 1–3), since this alkene affords good yields of aziridine
when copper triflate is used as a homogeneous catalyst.4 Using
a five-fold molar excess of styrene, i.e. similar conditions to the
homogeneous process, the desired N-tosylaziridine was
obtained in 90% yield (entry 1). These initial results confirmed
our contention that cations within zeolites could be used as
heterogeneous counterparts of known homogeneous catalysts,
and, to our knowledge, this is the first example of an aziridin-
ation reaction catalysed heterogeneously.
contributed less than 5% of the consumption of PhI᎐NTs.
᎐
2. The failure of (E)-stilbene to undergo aziridination in the
presence of the zeolite although reaction occurs with the rather
similarly sized methyl (E)-cinnamate (84% yield), and for
stilbene in the homogeneous copper catalysed process (52%
yield) indicates that the active catalytic sites are within the
zeolite pores. Catalysis of aziridination by the external surface
of the zeolite is clearly insignificant for (E)-stilbene. Molecular
modelling studies, based on a systematic evaluation of free
volume available to each molecule, indicate that the aziridine
formed from the ester can adopt a conformation that fits within
the zeolite pores and is free to diffuse through the pore
structure, whereas the aziridine from (E)-stilbene, although it
can be constructed within the supercages of the zeolite, is too
bulky to diffuse through the interconnecting channels in the
structure.
In order to confirm that this process was wholly hetero-
geneous, following reaction the zeolite catalyst was recovered
by filtration, and another aliquot of reactants (styrene:
PhI᎐NTs = 5:1 molar ratio) was added to the recovered filtrate;
3. For the zeolite Y crystallites used in this study only 0.5% of
the available acid sites are present on the external crystallite
surface13 and hence any reaction resulting from these non-
microporous sites can be expected to be negligible. The effect of
the external sites was investigated in more detail by examining
the yield of aziridine formed by zeolites containing different
exchange levels of Cu2ϩ. A series of CuHY zeolites, containing
2.4, 3.1, 4.2 and 4.7% Cu, were prepared using Cu(OAc)2 solu-
tions with different concentrations (0.025–0.2 mol lϪ1). These
zeolites were then used in the aziridination of styrene and the
amount of catalyst was adjusted so that in each reaction the
same mass of Cu was utilized. The same yield of aziridine and
rate of formation was observed in all four cases although the
ratio of external to internal sites is significantly different for
these catalysts. These results indicate that the sites on the
external crystallite surface are not playing a disproportionately
significant role in the reaction.
᎐
no further product was observed. Further, the removed catalyst
was reused with fresh reagents and solvent, and the zeolite
demonstrated similar activity to when it was used initially.
It was noted in earlier studies that, in the homogeneously
catalysed reaction,4 the yield of aziridine decreased to 37%
when a 1:1 molar ratio of styrene:PhI᎐NTs was employed,
᎐
presumably due to the competing breakdown of the PhI᎐NTs
᎐
reagent to yield toluene-p-sulfonamide. This decrease was
found to be dramatically less significant using our hetero-
geneous catalyst, where an 87% yield of aziridine was obtained
when a styrene:PhI᎐NTs molar ratio of 1:1 was used (Table 1,
᎐
entry 2), as compared to a yield of 90% when a styrene:
PhI᎐NTs molar ratio of 5:1 was used (Table 1, entry 1). This is
᎐
a particularly important observation since the procedure can be
applied to the aziridination of expensive alkenes. Moreover
it indicates that the environment of the catalytic sites is very
1044
J. Chem. Soc., Perkin Trans. 2, 1999, 1043–1049

View Article Online
Fig. 1 Effect of metal-exchanged zeolite Y on the aziridination of
Fig. 2 Effect of temperature on aziridination of styrene. Isolated
yields based on PhI᎐NTs, ee determined by chiral HPLC. Reaction
styrene. Reaction conditions: PhI᎐NTs:styrene molar ratio = 1:5,
᎐
᎐
metal-exchanged zeolite Y in CH3CN at 25 ЊC, ‘none’ indicates the
yield obtained when no catalyst is added. Reaction time for CuHY
catalysed reaction was five hours. All other reactions were left for five
conditions: solvent CH3CN, styrene:PhI᎐NTs:bis(oxazoline) =
᎐
5:1:0.05 molar ratio.
days without complete dissolution of PhI᎐NTs reagent.
᎐
reaction, the use of lower reaction temperatures was found to
give the highest enantioselectivities. The temperature variation
of yield and ee requires comment. The yield of the aziridine is
determined by competing processes for the tosylnitrenoid active
species within the zeolite, whereas the ee will determined by
competing processes via the diastereoisomeric transition states.
In both situations, competing reactions will in general have
different activation parameters and so a temperature variation
is to be expected. That both yield an ee showing a maximum is
not unique and a number of possible reasons could account for
our observations. One is the possibility that there is reaction of
copper uncomplexed by the bis(oxazoline) giving racemic
aziridine that is slow at higher temperatures but becomes
important at low temperatures because of a lower enthalpy
of activation than the reaction catalysed by the copper–
bis(oxazoline) complex; such a situation would then oppose the
increase in ee that is expected as the temperature falls leading to
a maximum being observed in the ee versus temperature plot.
We have found that a temperature of Ϫ10 ЊC provides the
highest enantioselectivity of 42% ee without compromising
yield when using acetonitrile solvent and 2,2-bis[((4R)-4-
phenyl-1,3-oxazolin-2-yl)]propane (1) as the chiral modifier
simply added to the reaction mixture (Fig. 2). Evans et al.4
observed that benzene was the preferred solvent for the homo-
geneous copper catalyst. In contrast, we do not observe
improved enantioselection or activity when benzene is used as
the solvent with the heterogeneous CuHY catalyst (Fig. 3).
It would be reasonable to assume that one chiral modifier
would be required per zeolite supercage to obtain maximum
enantioselectivity. We have, however, found that very low levels
of the expensive modifier can be used without resulting in
decreases in yield or enantioselectivity (Fig. 4). An excess of
bis(oxazoline) significantly reduces the yield of aziridine, due
to pore-blocking, but both yield and enantioselectivity
It should be noted that, given that the reaction takes place in
the intracrystalline space of the zeolite, the almost identical
aziridine yields in reactions with styrene:PhI᎐NTs ratios of 1
᎐
and 5 suggest that the alkene is preferentially bound within the
zeolite pores compared with solvent molecules (which are weak
ligands for copper). Thus the reaction of PhI᎐NTs at the
catalytic site, which we presume to give rise to a copper–
tosylnitrenoid intermediate (which we represent as TsN᎐
᎐
᎐
CuHY), occurs in an environment in which trapping to form
aziridine is favoured by the presence of the alkene in close
proximity and at an effective concentration substantially higher
than in the bulk liquid phase. Alkenes, particularly ones capable
of chelation, are known to form copper complexes. While this
might be thought to be an explanation of the better aziridine
yields from phenyl-bearing alkenes, such complexation takes
place only to copper in its ϩ1 oxidation state. It should be
borne in mind, however, that Evans has offered no compelling
evidence for the actual oxidation state of copper in the homo-
geneously catalysed reaction, although he expressed the belief
that it was Cu().
Several other metal exchanged zeolites were also screened,
but CuHY was found to be the most active. The results are
shown in Fig. 1. In most cases when zeolite Y was exchanged
with other metals, e.g. Agϩ, Co2ϩ, Fe3ϩ, Mg2ϩ, Ni2ϩ, Zn2ϩ, the
yield of the aziridine was lower than that obtained when com-
pared with the uncatalysed and the CuHY catalysed reactions.
We consider that this is due to the ability of these metals to
catalyse the breakdown of the PhI᎐NTs reagent into iodo-
᎐
benzene and toluene-p-sulfonamide. This catalysed break-
down has also been noted by Burrows et al.14 and was
confirmed in our experiments by the observation of high yields
of iodobenzene.
were maximized with only a molar ratio of PhI᎐NTs:bis-
᎐
Heterogeneous asymmetric aziridination of alkenes with CuHY
(oxazoline) = 1:0.05. This corresponds to a molar ratio of
Cu2ϩ :bis(oxazoline) of 2:1, indicating, remarkably, that not all
the Cu2ϩ cations are modified in our experiments. In a sub-
sequent experiment, an excess of bis(oxazoline) 1 was stirred
with CuHY [molar ratio Cu2ϩ :bis(oxazoline) = 1:5] in aceto-
nitrile. The zeolite was filtered and washed with acetonitrile,
then used as the catalyst in fresh solvent and reactants. Both
yield and enantioselectivity observed were identical to those
obtained when the bis(oxazoline) was added directly to the
Evans et al.4 have shown that modification of the copper homo-
geneous catalysts using chiral bis(oxazoline) ligands induces
enantioselectivity in the aziridination reaction. We have
examined modification of the CuHY catalyst with a range of
oxazolines and have observed N-tosylaziridine products with
up to 61% enantiomeric excess. The racemic heterogeneous
aziridination of alkene, in the absence of bis(oxazoline), gave
yields of 80% in the temperature range Ϫ20 to ϩ25 ЊC using
acetonitrile as solvent. However, the higher temperatures gave
shorter reaction times (e.g. 2 h at ϩ25 ЊC) and hence 25 ЊC
was considered optimal from a practical viewpoint for
non-enantioselective aziridination. For the enantioselective
reaction mixture (molar ratio of PhI᎐NTs: bis(oxazoline) =
᎐
1:0.05). It is clear that very low levels of the modifier are
required to obtain the enantioselectivities reported. It is
possible that different copper sites are present within the zeolite
J. Chem. Soc., Perkin Trans. 2, 1999, 1043–1049
1045

View Article Online
homogeneous reactions. However, for the heterogeneously
catalysed reaction, using CuHY as catalyst, the bulky diphenyl
bis(oxazoline) 4 gave racemic product, despite inducing 15% ee
for the equivalent homogeneously catalysed reaction. This is
again evidence that the reaction is truly heterogeneous and is
occurring within the pores of the zeolite. Further modification
of the bis(oxazoline) has shown that the pyridine-bridged
bis(oxazoline) 5 gives the highest enantioselectivity of 61% ee
for the aziridination of styrene (Table 2). We consider these
initial results to be encouraging and that careful optimization
of choice and concentration of the chiral ligand might result in
further improvements in ee and yield.
The major advantage of the use of CuHY as a catalyst for
this reaction is the ease with which it can be recovered from the
reaction mixture by simple filtration if used in a batch reactor
(alternatively it could be used in a continuous flow fixed bed
reactor). We have carried out the heterogeneous asymmetric
aziridination of styrene until completion, filtered and washed
the zeolite then added fresh styrene, PhI᎐NTs and solvent,
᎐
without further addition of bis(oxazoline) 1, for several con-
secutive experiments; the results are shown in Fig. 5. After each
consecutive experiment a portion of CuHY was retained to
determine the concentration of copper still present in the
zeolite. For each experiment we found that only traces of the
copper were removed from the catalyst (<0.5% of the total Cu2ϩ
is lost from the catalyst). The filtrate containing the trace con-
centration of Cu2ϩ has been used in a reaction and was not
found to catalyse aziridination. The yield and the enantio-
selectivity decline on reuse; we have found that adsorbed water
can build up within the pores of the zeolite on continued use
and we believe that this is the cause of loss of activity and
enantioselection. However, full enantioselectivity and yield can
be recovered if the catalyst is simply dried in air prior to reuse
(Fig. 5) or alternatively the catalyst can be recalcined and fresh
oxazoline ligand added. We are therefore confident that this
catalyst system can form the basis of a commercial hetero-
geneous catalyst for the aziridination of alkenes.
Fig. 3 Effect of solvent on the aziridination of styrene. Isolated yields
based on PhI᎐NTs, ee determined by chiral HPLC. Reaction condi-
᎐
tions: 25 ЊC, styrene:PhI᎐NTs:bis(oxazoline) = 5:1:0.05 molar ratio.
᎐
Heterogeneous asymmetric aziridination of alkenes with Cu-
MCM41
Fig. 4 Dependence of the aziridination of styrene on the amount of
bis(oxazoline) measured as molar equivalents based on PhI᎐NTs. Isol-
᎐
To demonstrate that alternative types of silicate framework can
be used for this reaction, experiments were carried out using
Cu2ϩ-exchanged MCM-41, a mesoporous material with pores
of 3.5 nm in diameter.15 Yields of up to 87% of the aziridine
and ee of 37% were obtained. Using this mesoporous material
greatly increases the versatility of heterogeneous aziridination
since larger alkenes and chiral modifiers can be used, while
retaining all the advantages of the heterogeneous catalyst
system.
ated yields based on PhI᎐NTs, ee determined by chiral HPLC. Reaction
᎐
conditions: solvent CH3CN, Ϫ20 ЊC, styrene:PhI᎐NTs = 5:1 molar
᎐
ratio.
and that only a fraction of the total are active in the activation
of PhI᎐NTs,15 or that the bulky oxazoline molecules diffuse
᎐
slowly into the zeolite pore structure and only sites close to the
surface are modified. Further detailed spectroscopy will be
needed to clarify this point.
With the optimum conditions established for the enantiose-
lective aziridination of styrene, other bis(oxazolines) and
alkenes were screened with CuHY as the catalyst (Table 2). (E)-
β-Methylstyrene was found to show similar degrees of enantio-
selectivity to styrene. However, methyl (E)-cinnamate gave a
much superior result of 61% ee, albeit in poorer yield. The tert-
butyl substituted bis(oxazoline) 2, when used in acetonitrile
solvent, gave racemic aziridine. We suggest that this is because
acetonitrile, a ligand for Cu2ϩ, competes more effectively with
this bis(oxazoline) for the active sites. Support for this inter-
pretation comes from our observation that, by carrying out the
reaction using styrene as the solvent, enantioselectivity was
restored, although both yield and enantiomeric excess were
then lower than for the homogeneous reaction. To provide
further evidence that the reaction is proceeding within the
supercages of the zeolite, reactions were carried out using a
simple phenyl substituted bis(oxazoline) 3, which is known to fit
inside the zeolite, and using a diphenyl substituted analogue 4,
which was considered as the result of molecular simulations to
be too bulky to fit inside the zeolite pores. At 25 ЊC the smaller
bis(oxazoline) 3 gave 10% ee for both the heterogeneous and
Conclusions
In this paper we have presented results for the first hetero-
geneously catalysed aziridination reaction. Furthermore we
have shown that this catalyst can be modified to induce enantio-
selectivity. The design ideas upon which this catalyst discovery
is based are the detailed understanding that cations can be
located at defined locations in zeolite pores and that the zeolite
acts as part of the coordination sphere, and that the remaining
coordination sites on the metal can be filled by the addition of
further ligands that can be chiral and hence (or in other ways)
control the reactivity. We consider that this approach can have
general applicability and will be of value in the design of
improved asymmetric catalysts.
Experimental
Apparatus
1
a) H and 13C NMR spectra were recorded on Bruker AC300
and AMX 400 spectrometers, equipped with an X32 computer.
1046
J. Chem. Soc., Perkin Trans. 2, 1999, 1043–1049

View Article Online
Table 2 Representative bis(oxazolines) for the enantioselective aziridination of alkenes
Temp./
ЊC
Yield
(%)
Oxazoline
Alkenea
ee (%)b
Methyl (E)-cinnamate
(E)-β-Methylstyrene
Styrene
Ϫ10
Ϫ10
Ϫ10
25
8 (21)
74
82
61 (70)
36
44
Me
N
Me
O
O
N
Styrene
87
29
Ph
Ph
1c
Me
Me
Styrene
Ϫ20
Ϫ20
64
15 (89)
0
O
O
Styrened
18 (63)
N
N
tBu
tBu
2
O
O
Styrene
Styrene
25
25
78 (75)
73 (74)
10 (10)
0 (15)
N
N
Ph
Ph
3
O
O
Ph
Ph
N
N
Ph
Ph
4
Styrene
Ϫ10
4
61
O
O
N
N
N
5
a Unless otherwise specified reaction conditions were: solvent CH3CN, alkene:PhI᎐NTs = 5:1 molar ratio. b Enantioselectivity determined by chiral
᎐
HPLC. Absolute configurations of major products, determined by optical rotation, are (S) for (E)-β-methylstyrene and methyl (E)-cinnamate, (R)
for styrene. Values in parentheses indicate yields obtained from homogeneous reactions. c The absolute configuration shown in ref. 11 as the
S-enantiomer is incorrect, the R-enantiomer of I was used in all the reported studies. d Styrene was used as solvent.
d) Microanalyses were performed by the University of Liver-
pool, Department of Chemistry, Microanalytical Laboratory.
e) Flash column chromatography was performed on Merck
Kieselgel 60 (230–400 mesh) and analytical TLC on silica gel 60
F-254 plates.
f) HPLC analysis was recorded on a Varian 5000 liquid
chromatograph, using C18 Apex Octadecyl 5µ reverse-phase
columns for analytical work. The eluent system was water–
acetonitrile (50:50), 235 nm, flow rate 0.6 ml minϪ1. Baseline
separation was achieved for all reagents and products. For
chiral HPLC analysis a 25 cm chiralcel OJ column was used.
The eluent system was hexane–isopropan-2-ol (95:5), 235 nm,
flow rate 1.0 ml minϪ1. Baseline separation was achieved for
both enantiomers. Absolute configuration was confirmed by
optical polarimetry and comparison with the literature.3
Materials
Fig. 5 Reusability of CuHY as a catalyst for the aziridination of styr-
ene. Following experiment 12 the catalyst was recovered by filtration,
recalcined and fresh oxazoline was added. Reaction time increased from
1 h for initial use to 35 h for run 12; reaction times for runs 13 and 14
were 1 h.
(E)-Stilbene and bis-oxazolines were obtained from Aldrich.
Synthesis of [N-(p-tolylsulfonyl)imino]phenyliodinane (PhI᎐
᎐
NTs) was carried out following the method of Yamada et al.16
Cu-MCM-41
Unless otherwise stated deuteriochloroform was used as the
solvent. Spectra were recorded on the d scale and signals
quoted in the form: chemical shift measured in ppm (no. of
protons, multiplicity, assignment).
b) Infrared spectra were recorded in the range 4000–600
cmϪ1, using Perkin-Elmer 1320 and 883 infrared spectrometers
with a polystyrene reference. Spectra of the liquids were taken
as films, or as solutions in chloroform, using sodium chloride
cells. Spectra of solids were taken as nujol mulls.
AlMCM-41 was prepared according to the method of Beck
et al.17 Calcined AlMCM-41 (3.0 g) was stirred in copper()
acetate solution (100 ml, 0.2 M) for 24 h. The material was then
filtered, washed, dried and then stirred again in fresh mangane-
se() acetate solution (100 ml) for another 24 h. This process
was repeated a further two times. Finally the exchanged zeolite
was calcined (550 ЊC) for eight hours prior to use; Cu content
5.0% by weight.
c) Mass spectra were obtained on VG analytical 7070E and
Fisons TRIO1000 double focusing magnetic sector spectro-
meters, using electron impact (E.I.) or chemical ionisation (C.I.)
techniques with ammonia as the carrier gas.
CuHY ion-exchange
ϩ
Ultrastabilised NH4 Y zeolite (5.0 g), obtained from Union
Carbide, was calcined (550 ЊC) for five hours, then stirred in 0.5
J. Chem. Soc., Perkin Trans. 2, 1999, 1043–1049
1047

View Article Online
molar solution of copper() acetate solution (100 ml) for 24
hours at room temperature. The mixture was then centrifuged
and washed with distilled water. This was repeated until all
unexchanged copper was removed. The zeolite was then dried
at 100 ЊC for 24 hours then recalcined (550 ЊC) for five hours;
Cu content 4.0% by weight.
acetonitrile (15 cm3). Reaction time: 1 h at room temperature.
Flash column chromatography (1.5 × 20 cm silica, 10:1.5
petroleum ether (40/60)–ethyl acetate) afforded 0.31 g (76%) of
the aziridine as a crystalline solid: mp 115–116 ЊC; νmax/cmϪ1
(CH2Cl2) 3055, 1600, 1494, 1462, 1376, 1325, 1185, 1160, 1190,
1014, 980, 908, 812, 688, 667; δH (300 MHz, CDCl3) 7.85 (d,
2H, J = 8.3 Hz, Ar-H), 7.33 (d, 2H, J = 7.8 Hz, Ar-H), 7.25 (d,
2H, J = 8.5 Hz, Ar-H), 7.15 (d, 2H, J = 8.5 Hz, Ar-H), 3.72
(dd, 1H, Jcis = 7.3 Hz, Jtrans = 4.3 Hz, CHPh-aziridine), 2.29 (d,
1H, J = 7.3 Hz, trans-CH-aziridine), 2.42 (s, 3H, Ar-CH3), 2.35
(d, 1H, J = 4.3 Hz, CH-aziridine); δC (75 MHz, CDCl3) 144.8,
134.8, 134.1, 133.6, 129.8, 128.8, 128.0, 40.2, 36.1, 21.1;
HRMS (FAB,MNBA) exact mass calcd. for C15H14ClNO2S
(M ϩ Na)ϩ 307.04337, found 307.04348. Anal. calcd. for
C15H14ClNO2S: C, 58.62; H, 4.60; N, 4.56. Found: C, 59.29;
H, 5.07; N, 4.21%.
N-(p-Tolylsulfonyl)-2-phenylaziridine
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and styrene (0.69 g, 6.7 mmol) was stirred in acetonitrile
(15 cm3). Reaction time: 1 h at room temperature. Flash column
chromatography (1.5 × 20 cm silica, 10:1.5 petroleum ether
(40/60)–ethyl acetate) afforded 0.33 g (90%) of the aziridine as a
crystalline solid. (for asymmetric aziridination of styrene
[α]2D0 = Ϫ29.2 (25 ЊC, chloroform)). mp 89–90 ЊC; νmax/cmϪ1
(CHCl3) 3015, 1325, 1215, 1160, 915, 785, 770, 715, 695, 665; δH
(300 MHz, CDCl3) 7.86 (d, 2H, J = 8.1 Hz, Ar-H), 7.32 (d, 2H,
J = 8.1 Hz, Ar-H), 7.10 (s, 5H, Ar-H), 3.74 (dd, 1H, Jcis = 7.0,
Jtrans = 4.8 Hz, CHPh), 2.96 (d, 1H, 7.0 Hz, cis-CH-aziridine),
2.42 (s, 3H, Ar-CH3), 2.37 (d, 1H, J = 4.8 Hz, trans-CH-
aziridine); δC (75 MHz, CDCl3) 144.7, 135.2, 129.8, 128.6,
128.4, 128.0, 126.6, 41.1, 35.9, 21.6; HRMS (FAB,MNBA)
exact mass calcd. for C15H15NO2S (M ϩ Na)ϩ 273.08234, found
273.08251. Anal. calcd. for C15H15NO2S: C, 65.91; H, 5.53;
N, 5.12. Found: C, 65.31; H, 5.59; N, 4.93%.
trans-N-(p-Tolylsulfonyl)-2,3-diphenylaziridine
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and (E)-stilbene (1.21 g, 6.7 mmol) was stirred in
dichloromethane (15 cm3). Reaction time: 1 h at room temper-
ature. Flash column chromatography (1.5 × 20 cm silica, 10:1.5
petroleum ether (40/60)–ethyl acetate) afforded 0.25 g (52%) of
the aziridine as a crystalline solid: mp 139–140 ЊC; νmax/cmϪ1
(CHCl3) 3031, 3019, 1327, 1306, 1215, 1161, 1088, 930, 910,
814, 745, 741, 708, 698, 691; δH (300 MHz, CDCl3) 7.61 (d, 2H,
J = 8.4 Hz, Ar-H), 7.45–7.36 (m, 10H, Ar-H), 7.32 (d, 2H,
J = 8.4 Hz, Ar-H), 4.26 (s, 2H, CH-aziridine), 2.41 (s, 3H,
Ar-CH3); δC (75 MHz, CDCl3) 170.2, 144.1, 134.9, 132.2, 131.4,
129.6, 129.2, 128.2, 50.3, 21.6; HRMS (FAB,MNBA) exact
mass calcd. for C21H19NO2S (M ϩ Na)ϩ 350.1215, found
350.1227. Anal. calcd. for: C, 72.18; H, 5.48; N, 4.01. Found: C,
72.19; H, 5.46; N, 4.05%.
N-(p-Tolylsulfonyl)-2-methyl-3-n-propylaziridine
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and (E)-hex-2-ene (0.56 g, 6.7 mmol) was stirred in
acetonitrile (15 cm3). Reaction time: 1.5 h at room temperature.
Flash column chromatography (1.5 × 20 cm silica, 10:1.5
petroleum ether (40/60)–ethyl acetate) afforded 0.17 g (44%) of
the aziridine as a colourless oil: νmax/cmϪ1 (thin film) 2959, 2932,
1456, 1325, 1306, 1291, 1233, 1163, 1092, 957, 943, 924, 866,
816, 716, 694, 662; δH (300 MHz, CDCl3) 7.80 (d, 2H, J = 8.3
Hz, Ar-H), 7.27 (d, 2H, J = 8.3 Hz, Ar-H), 2.67 (m, 1H, CH-
aziridine), 2.61 (d, 1H, J = 7.0 Hz, cis-CH-aziridine), 2.38 (s,
3H, Ar-CH3), 1.54 (m, 1H, aliphatic), 1.31–1.22 (m, 6H,
aliphatic), 0.83 (t, 3H, J = 7.1 Hz, CH3); δC (75 MHz, CDCl3)
143.8, 138.3, 129.5, 127.3, 49.5, 45.7, 32.3, 21.5, 20.4, 14.7, 13.6;
HRMS (FAB,MNBA) exact mass calcd. for C16H17NO2S
(M ϩ Na)ϩ 287.09799, found 287.09799. Anal. calcd. for
C16H17NO2S: C, 66.87; H, 5.96; N, 4.87. Found: C, 65.26;
H, 5.61; N, 4.31%.
cis-N-(p-Tolylsulfonyl)-2,3-diphenylaziridine
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and (Z)-stilbene (1.2 g, 6.7 mmol) was stirred in
acetonitrile (15 cm3). Reaction time: 1 h at room temperature.
Flash column chromatography (1.5 × 20 cm silica, 10:1.5
petroleum ether (40/60)–ethyl acetate) afforded 0.21 g (45%) of
the aziridine as a crystalline solid: mp 154-155 ЊC; νmax/cmϪ1
(CHCl3) 3034, 1599, 1495, 1329, 1306, 1186, 1161, 1092, 1026,
909, 814, 801, 698, 675; δH (300 MHz, CDCl3) 7.97 (d, 2H,
J = Ar-H), 7.36 (d, 2H, J = 8.2 Hz, Ar-H), 7.12-7.03 (m, 10H,
Ar-H), 4.21 (s, 2H, CH-aziridine), 2.43 (s, 3H, Ar-CH3); δC (75
MHz, CDCl3) 144.6, 134.8, 132.1, 129.8, 128.0, 127.8, 127.7,
127.6, 47.4, 21.6; HRMS (FAB,MNBA) exact mass calcd. for
C21H19NO2S (M ϩ Na)ϩ 350.1215, found 350.1219. Anal.
calcd. for C21H19NO2S: C, 72.18; H, 5.48; N, 4.01. Found: C,
71.83; H, 5.49; N, 3.95%.
N-(p-Tolylsulfonyl)-2-(p-methylphenyl)aziridine
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and p-methylstyrene (0.79 g, 6.7 mmol) was stirred in
acetonitrile (15 cm3). Reaction time: 1 h at room temperature.
Flash column chromatography (1.5 × 20 cm silica, 10:1.5
petroleum ether (40/60)–ethyl acetate) afforded 0.25 g (67%) of
the aziridine as a crystalline solid: mp 137–138 ЊC; νmax/cmϪ1
(CHCl3) 3034, 1600, 1516, 1380, 1322, 1158, 1092, 1018, 978,
911, 814, 716, 708, 692, 660; δH (300 MHz, CDCl3) 7.86 (d, 2H,
J = 8.2 Hz, Ar-H), 7.31 (d, 2H, J = 7.9 Hz, Ar-H), 7.09 (s, 4H,
Ar-H), 3.73 (dd, 1H, Jcis = 7.2 Hz, Jtrans = 4.5 Hz, CHPh-
aziridine), 2.96 (d, 1H, Ar-H, J = 7.2 Hz, cis-CH-aziridine),
2.42 (s, 3H, Ar-CH3), 2.36 (d, 1H, J = 4.5 Hz, trans-CH-
aziridine), 2.32 (s, 3H, Ar-CH3); δC (75 MHz, CDCl3) 144.3,
138.1, 135.1, 132.0, 129.8, 129.3, 128.0, 126.5, 41.8, 35.8, 21.5,
21.0; HRMS (FAB,MNBA) exact mass calcd. for C16H17NO2S
(M ϩ Na)ϩ 287.09799, found 287.09799. Anal. calcd. for
C16H17NO2S: C, 66.87; H, 5.96; N, 4.87. Found: C, 65.26;
H, 5.61; N, 4.31%.
N-(p-Tolylsulfonyl)-2-methyl-2-phenylaziridine
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and α-methylstyrene (0.79 g, 6.7 mmol) was stirred in
acetonitrile (15 cm3). Reaction time: 1 h at room temperature.
Flash column chromatography (1.5 × 20 cm silica, 10:1.5
petroleum ether (40/60)–ethyl acetate) afforded 0.14 g (33%) of
the aziridine as a crystalline solid: mp 84–85 ЊC; νmax/cmϪ1
(CHCl3) 3060, 3028, 2992, 2930, 1600, 1449, 1330, 1267, 1160,
1128, 1089, 1027, 939, 870, 715, 694, 680, 650; δH (300 MHz,
CDCl3) 7.65 (d, 2H, J = 8.3 Hz, Ar-H), 7.38–7.11 (m, 5H,
Ar-H), 7.08 (d, 2H, J = 8.3 Hz, Ar-H), 3.91 (s, 1H, CH-
aziridine), 2.40 (s, 3H, CH3), 2.02 (s, 1H, CH-aziridine), 1.58 (s,
3H, CH3); δC (75 MHz, CDCl3) 145.0, 143.5, 136.8, 129.8,
128.5, 127.3, 127.1, 124.9, 73.7, 53.8, 27.5, 21.5; HRMS
(FAB,MNBA) exact mass calcd. for C16H17NO2S (M ϩ Na)ϩ
310.0878, found 310.0897. Anal. calcd. for C16H17NO2S: C,
66.87; H, 5.96; N, 4.87. Found: C, 66.94; H, 5.93; N, 4.87%.
N-(p-Tolylsulfonyl)-2-(p-chlorophenyl)aziridine
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and p-chlorostyrene (0.92 g, 6.7 mmol) was stirred in
1048
J. Chem. Soc., Perkin Trans. 2, 1999, 1043–1049

View Article Online
trans-N-(p-Tolylsulfonyl)-2-methyl-3-phenylaziridine
(M ϩ Na)ϩ 274.0878, found 274.0872. Anal. calcd. for
C13H12INO2S: C, 41.84; H, 3.24; N, 3.75. Found: C, 41.68; H,
3.21; N, 3.71%.
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and (E)-β-methylstyrene (0.79 g, 6.7 mmol) was stirred
in acetonitrile (15 cm3). Reaction time: 1 h at room temperature.
Flash column chromatography (1.5 × 20 cm silica, 10:1.5
petroleum ether (40/60)–ethyl acetate) afforded 0.31 g (74%) of
the aziridine as a crystalline solid: mp 73–74 ЊC; νmax/cmϪ1
(CHCl3) 3031, 1458, 1321, 1306, 1291, 1217, 1159, 1090, 1038,
974, 891, 814, 772, 743, 710, 698, 687, 667; δH (300 MHz,
CDCl3) 7.85 (d, 2H, J = 8.3 Hz, Ar-H), 7.28–7.19 (m, 5H, Ar-
H), 7.19 (d, 2H, J = 8.0 Hz, Ar-H), 3.79 (d, 1H, J = 4.3 Hz,
CHPh), 2.91 (dq, 1H, J = 6.1, 4.5 Hz, CHCH3), 2.38 (s, 3H,
Ar-CH3), 1.83 (d, 3H, J = 6.1 Hz, CH3); δC (75 MHz, CDCl3)
143.9, 137.9, 135.5, 129.5, 128.4, 128.0, 127.1, 126.2, 49.1, 21.5,
14.1; HRMS (FAB,MNBA) exact mass calcd. for C16H17NO2S
(M ϩ Na)ϩ 310.0878, found 310.0889. Anal. calcd. for C16H17-
NO2S: C, 66.87; H, 5.96; N, 4.87. Found: C, 66.89; H, 5.98;
N, 4.85%.
Homogeneous aziridination reactions catalysed by Cu(OTf)2
The synthesis was carried out following the method of Evans
et al.4
To anhydrous copper() triflate (0.024 g, 0.067 mmol, 5.0
mol%) and 15 cm3 of HPLC grade acetonitrile, in a 25 cm3
round bottom flask equipped with a stirrer bar, were added
alkene (6.7 mmol, 5.0 equiv.) and [N-(p-tolylsulfonyl)imino]-
phenyliodinane, PhI᎐NTs (0.52 g, 1.34 mmol). The reaction was
᎐
stirred in air at room temperature until complete dissolution of
PhI᎐NTs. When all the PhI᎐NTs had been drawn into solution,
᎐
᎐
the reaction was filtered through a plug of silica with ethyl
acetate (50 cm3) as eluent. The solvent was removed in vacuo to
give an oil that was purified by flash column chromatography.
Acknowledgements
trans-N-(p-Tolylsulfonyl)-2-(methoxycarboyl)-3-phenylaziridine
We thank ICI Katalco, Robinson Brothers, the EPSRC and the
DTI LINK programme on asymmetric synthesis for funding.
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and (E)-β-methyl cinnamate (1.09 g, 6.7 mmol) was
stirred in acetonitrile (15 cm3). Reaction time: 1 h at room tem-
perature. Flash column chromatography (1.5 × 20 cm silica,
10:1.5 petroleum ether (40/60)–ethyl acetate) afforded 0.37 g
References
1 R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley, New
York, 1994.
(84%) of the aziridine as a crystalline solid: mp 45–46 ЊC; νmax
/
cmϪ1 (CHCl3) 3068, 3021, 2960, 1750, 1600, 1498, 1456, 1440,
1412, 1338, 1306, 1292, 1161, 1086, 1081, 1004, 938, 906, 834,
812, 709, 692, 646; δH (300 MHz, CDCl3) 7.78 (d, 2H, J = 8.4
Hz, Ar-H), 7.32–7.22 (m, 7H, Ar-H), 4.42 (d, 1H, J = 4.0 Hz,
CH-aziridine), 3.85 (s, 3H, OCH3), 3.53 (d, 1H, J = CH-
aziridine), 2.41 (s, 3H, Ar-CH3); δC (75 MHz, CDCl3) 166.1,
144.2, 136.5, 132.3, 129.6, 128.7, 128.6, 127.6, 127.1, 53.1, 47.7,
21.6; HRMS (FAB,MNBA) exact mass calcd. for C17H17NO4S
(M ϩ Na)ϩ 331.08783, found 331.08735. Anal. calcd. for
C17H17NO4S: C, 61.62; H, 5.17; N, 4.23. Found: C, 61.06;
H, 5.08; N, 3.99%.
2 T. Katsuki and K. B. Sharpless, J. Am. Chem. Soc., 1980, 102, 5974.
3 D. A. Evans, K. A. Woerpel, M. M. Hinman and M. M. Faul, J. Am.
Chem. Soc., 1991, 113, 726.
4 D. A. Evans, M. M. Faul and M. T. Bilodeau, J. Am. Chem. Soc.,
1994, 116, 2742.
5 (a) K. Srinivasan, P. Michaud and J. K. Kochi, J. Am. Chem. Soc.,
1986, 108, 2309; (b) W. Zhang, J. L. Loebach, S. R. W. Wilson and E.
N. Jacobsen, J. Am. Chem. Soc., 1990, 112, 2801.
6 K. T. Wan and M. E. Davis, Nature, 1994, 370, 449.
7 H. U. Blasser, Tetrahedron: Asymmetry, 1997, 8, 1693.
8 P. P. Knops-Gerrits, D. De Vos, F. Thibault-Starzyk and P. A.
Jacobs, Nature, 1994, 369, 543.
9 K. Klier, Langmuir, 1998, 4, 13.
10 C. B. Dartt and M. E. Davis, Catal. Today, 1994, 19, 151.
11 C. Langham, P. Piaggio, D. Bethell, D. F. Lee, P. McMorn, P. C. B.
Page, D. J. Willock, C. Sly, F. E. Hancock, F. King and G. J.
Hutchings, Chem. Commun., 1998, 1601.
12 P. Piaggio, P. McMorn, C. Langham, D. Bethell, P. C. B. Page, F. E.
Hancock and G. J. Hutchings, New J. Chem., 1998, 1167.
13 D. A. Whan, Chem. Br., 1981, 17, 532.
14 C. J. Burrows, K. J. O’Conner and S.-J. Wey, Tetrahedron Lett.,
1992, 33, 1001.
15 T. Maschmeyer, F. Rey, G. Sankar and J. M. Thomas, Nature, 1995,
378, 159.
16 Y. Yamada, Chem. Lett., 1975, 361.
N-(p-Tolylsulfonyl)-7-azabicyclo[4.1.0]heptane
A solution of PhI᎐NTs (0.52 g, 1.34 mmol), CuHY (71 mg, 25
᎐
mol%) and cyclohexene (0.55 g, 6.7 mmol) was stirred in
acetonitrile (15 cm3). Reaction time: 1 h at room temperature.
Flash column chromatography (1.5 × 20 cm silica, 10:1.5 pet-
roleum ether (40/60)–ethyl acetate) afforded 0.18 g (50%) of the
aziridine as a crystalline solid: mp 55–56 ЊC; νmax/cmϪ1 (CHCl3)
3020, 2950, 2860, 1600, 1440, 1395, 1315, 1305, 1155, 1090, 965,
920; δH (300 MHz, CDCl3) 7.81 (d, 2H, J = 8.2 Hz, Ar-H), 7.34
(d, 2H, J = 8.5 Hz, Ar-H), 2.99 (t, 2H, J = 1.4 Hz, CH-
aziridine), 2.44 (s, 3H, Ar-CH3), 1.80 (m, 4H, ring-CH), 1.44–
1.36 (m, 4H, ring-CH), 1.26–1.19 (m, 2H, ring-CH); δC (75
MHz, CDCl3) 144.0, 136.0, 129.6, 127.7, 39.8, 22.8, 21.6, 19.4;
HRMS (FAB,MNBA) exact mass calcd. for C17H17NO4S
17 J. S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leonowicz, C. T. Kresge,
K. D. Schmitt, C. T.-W. Chu, D. H. Olson, E. W. Shepherd, S. B.
McCullen, J. B. Higgins and J. L. Schlenker, J. Am. Chem. Soc.,
1992, 114, 10834.
Paper 8/06409A
J. Chem. Soc., Perkin Trans. 2, 1999, 1043–1049
1049