A simple copper catalyst for both aziridination of alkenes and amination of activated hydrocarbons with chloramine-T trihydrate

Full text of DOI:10.1021/jo981335q

J . Org. Chem. 1998, 63, 9569-9571
9569
the nitro analogue NsNdIPh,6 the synthesis of which is
more straightforward, and this is the favored nitrene-
transfer agent in the promising, but to date more limited,
rhodium-catalyzed aziridination developed by the group
of Mu¨ller.7
A Sim p le Cop p er Ca ta lyst for Both
Azir id in a tion of Alk en es a n d Am in a tion of
Activa ted Hyd r oca r bon s w ith
Ch lor a m in e-T Tr ih yd r a te
David P. Albone, Pavandeep S. Aujla, and
Paul C. Taylor*
Department of Chemistry, University of Warwick,
Coventry, CV4 7AL, UK
Stephen Challenger and Andrew M. Derrick
Pfizer Central Research, Sandwich, Kent, CT13 9NJ , UK
Received J uly 9, 1998
Very recently, Mu¨ller has published a study of the
amination of hydrocarbons with the same rhodium-NsNd
IPh systems. High yields of aminated products were
observed with cycloalkenes (eq 3) and with benzylic
substrates (eq 4), but, as with the rhodium-catalyzed
aziridination, very large excesses of hydrocarbon are
required.8 Another very recent paper, from the Katsuki
group, describes a copper-catalyzed benzylic and allylic
amination procedure using t-BuOO(CdO)NHTs (one-step
synthesis) as the nitrogen source.9 Benzylic amination
was efficient, but for allylic amination 1 equiv of catalyst
was required for an acceptable yield. Although the
products are consistent with a nitrene-transfer mecha-
nism, this reaction is believed to be a radical process.9
Our interest in nitrene transfer sulfimidation10 led us
to investigate cheap, commercially available chloram-
ine-T hydrate (TsNClNa‚(H2O)3) as a nitrene-transfer
agent.11 In fact, chloramine-T has very recently been
shown by Komatsu’s group to aziridinate alkenes with
copper(I) triflate as catalyst,12 but this process has the
drawback of requiring prior dehydration of chloramine-
T, a procedure with a significant explosion hazard.13 To
avoid this hazard Cenini’s group exchanged sodium for
tetraalkylammonium. Using iron and manganese ca-
talysis (but surprisingly not copper or rhodium) and huge
excesses of cycloalkene substrates, they obtained mix-
tures of products, including aziridines and allylic ami-
nation products.14
In tr od u ction
In 1967 Kwart and Kahn reported the copper-catalyzed
decomposition of benzenesulfonyl azide in the presence
of cyclohexene to yield products consistent with a nitrene-
transfer mechanism (eq 1).1 Both the aziridine 1 and the
allylic amination 2 products are synthetically very in-
teresting, and throughout the 1980s the groups of
Breslow and of Mansuy sought catalysts which led to
higher yields and selectivities for each of these two
classes of product.2
It was not until 1991 that an efficient and general
nitrene-transfer aziridination procedure was developed
by the group of Evans.3 Simple copper(I) and copper(II)
salts were found to give high yields of aziridines from a
wide range of alkenes and TsNdIPh (eq 2), with only
isolated examples of allylic insertions. In 1993 the
groups of J acobsen and of Evans simultaneously reported
an asymmetric variant of this reaction, with very high
Herein we describe a mechanism-based approach to
selection of an efficient catalyst for nitrene-transfer from
ees being recorded for some classes of substrate.4
A
nitrene-transfer mechanism was formally proposed by
J acobsen in 1995.5 The disadvantage of the Evans
aziridination procedure is the necessity of using TsNd
IPh as the nitrene-transfer agent, because its two-step
synthesis is notoriously capricious and iodobenzene is
liberated as a byproduct. Andersson recommends instead
(6) So¨dergren, M. J .; Alonso, D. A.; Andersson, P. G. Tetrahedron:
Asymmetry 1997, 8, 3563.
(7) Mu¨ller, P.; Baud, C.; J acquier, Y. Tetrahedron 1996, 52, 1543.
(8) Na¨geli, I.; Baud, C.; Bernadelli, G.; J acquier, Y.; Moran, M.;
Mu¨ller, P. Helv. Chim. Acta 1997, 80, 1087.
(9) Kohmura, Y.; Kawasaki, K.; Katsuki, T. Synlett 1997, 1456.
(10) Takada, H.; Nishibayashi, Y.; Ohe, K.; Uemura, S.; Baird, C.
P.; Sparey, T. J .; Taylor, P. C. J . Org. Chem., 1997, 62, 6512. Takada,
H.; Nishibayashi, Y.; Ohe, K.; Uemura, S. J . Chem. Soc., Chem.
Commun. 1996, 931.
* Corresponding author. E-mail p.c.taylor@warwick.ac.uk. Fax +44
1203 524112.
(1) Kwart, H.; Kahn, A. A. J . Am. Chem. Soc. 1967, 89, 1951.
(2) Breslow, R.; Gellman, S. H. J . Chem. Soc., Chem. Commun. 1982,
1400. Mahy, J .-P.; Bedi, G.; Battioni, P.; Mansuy, D. Tetrahedron Lett.
1988, 29, 1927.
(3) Evans, D. A.; Faul, M. M.; Bilodeau, M. T. J . Org. Chem. 1991,
56, 6744. Evans, D. A.; Faul, M. M.; Bilodeau, M. T. J . Am. Chem.
Soc. 1994, 116, 2742.
(4) Evans, D. A.; Faul, M. M.; Bilodeau, M. T.; Anderson, B. A.;
Barnes, D. M. J . Am. Chem. Soc. 1993, 115, 5328. Li, Z.; Conser, K.
R.; J acobsen, E. N. J . Am. Chem. Soc. 1993, 115, 5326.
(5) Li, Z.; Quan, W.; J acobsen, E. N. J . Am. Chem. Soc. 1995, 117,
5889.
(11) (a) Campbell, M. M.; J ohnson, G. Chem. Rev. 1978, 78, 65. (b)
Barton, D. H. R.; Hay-Motherwell, R. S.; Motherwell, W. B. J . Chem.
Soc., Perkin Trans. 1 1983, 445. (c) Aujla, P. S.; Baird, C. P.; Taylor,
P. C.; Mauger, H.; Valle´e, Y. Tetrahedron Lett. 1997, 38, 7453. (d) For
a very recent example of efficient bromine-catalyzed aziridination with
chloramine-T, see J eong, J . U.; Tao, B.; Sagasser, I.; Henniges, H.;
Sharpless, K. B. J . Am. Chem. Soc. 1998, 120, 6844.
(12) Ando, T.; Minakata, S.; Ryu, I.; Komatsu, M. Tetrahedron Lett.
1998, 39, 309.
(13) Klundt, I. L. Chem. Eng. News 1977, 55, 5 (49), 56.
(14) Cenini, S.; Penoni, A.; Tollari, S. J . Mol. Catal. A 1997, 124,
109.
10.1021/jo981335q CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/14/1998

9570 J . Org. Chem., Vol. 63, No. 25, 1998
Notes
Sch em e 1
Sch em e 2
chloramine-T trihydrate. Such a catalyst is desirable
since (i) chloramine-T is a very cheap nitrene source, (ii)
use of the trihydrate obviates the need for the hazardous
dehydration, (iii) a large excess of alkene should not be
necessary, and (iv) it offers potential for asymmetric
modification.
Ta ble 1. Azir id in a tion of Ar yl Alk en es
Resu lts a n d Discu ssion
In our earlier work on sulfimidation we noted that
copper-catalyzed imidation of sulfides with chloramine-T
appeared to have two alternative mechanisms, a Lewis
acid mechanism and a nitrene-transfer mechanism.11c
The proposed Lewis acid mechanism is similar to the
traditional mechanism assumed for reaction of chloram-
ine-T hydrate with sulfides, with the copper catalyst
accelerating at least one of the steps. Importantly for
the current study, in the absence of molecular sieves, the
proposed sulfonium intermediates, e.g. 3, are intercepted
by water to produce sulfoxides. This was observed in
reactions where no other nitrogen ligand was present.
However, when ligands such as pyridine or bipy were
added, no sulfoxide was observed even in the absence of
sieves. This is consistent with the operation of a nitrene-
transfer mechanism, which cannot lead to sulfoxide.
a
5 equiv reactions were run at 5-fold dilution (see Experimental
b
Section). Reaction may not have gone to completion.
ing led us to select catalyst 7 with which no epoxide was
observed, even without added sieves, suggesting a ni-
trene-transfer mechanism (Scheme 2).
We then screened catalyst 7 with a number of aryl
alkenes (Table 1), using 5 mol % copper(I) triflate and 6
mol % ligand in acetonitrile. Pleasingly, even with only
1.1 equiv of alkene, respectable yields of aziridine re-
sulted (entries 1, 3 and 5). With 5 equiv of alkene and
at higher dilution (in entries 2, 4, and 6, the concentration
of alkene was unchanged from the 1.1 equiv examples),
the percentage of p-toluenesulfonamide (TsNH2) byprod-
uct decreased in all cases. However, with the two more
electron-rich alkenes (entries 4 and 6), the percentage of
other byproducts increased, presumably due to competing
ionic additions. Nevertheless, for styrene and dihy-
dronaphthalene (entries 2 and 6), good yields of aziridine
were obtained.
We thus carried out a similar study with styrene as
substrate. Interestingly, the copper (I) triflate-bipy
catalyst appeared to act as a Lewis acid in this case,
giving mixtures of aziridine 5 and epoxide 6. We believe
these two products arise from a common chloronium
intermediate 4 (Scheme 1); indeed, chloramine-T has
been used to prepare chlorohydrins.15 In this work we
aimed to promote the aziridine product, but we note that
this reaction has the potential to be a useful epoxidation
procedure. Inspired by other work at Warwick,16 we
reasoned that n-alkyl imines of pyridine-2-carboxalde-
hyde might favor aziridination, as they are known to be
better ligands for iron(II) than bipy.17 An initial screen-
(15) Damin, B.; Garapon, J .; Sillion, B. Synthesis 1981, 362.
(16) Haddleton, D. M., J asieczek, C. B., Hannon, M. J ., Shooter, A.
J . Macromolecules 1997, 30, 2190.
When a representative alkyl alkene, cyclohexene, was
used, very little aziridine was observed either with 1.1
(17) Ba¨hr, G.; Do¨ge, H.-G. Z. Anorg. Allg. Chem. 1957, 292, 119.

Notes
Ta ble 2. Am in a tion of Cycloh exen e a n d Tetr a lin
J . Org. Chem., Vol. 63, No. 25, 1998 9571
mmol) in ether (10 mL) containing 4 Å molecular sieves (0.5 g)
at 0 °C. The solution was left to stir overnight at ambient
temperature before drying over anhydrous magnesium sulfate,
filtration, and evaporation in vacuo to yield N-(2-pyridinyl-
methylene)-1-pentanamine (4.70 g, 83%). 1H NMR δ 8.63 (m,
1H, ArH), 8.35 (s, 1H, imine-H), 7.97 (m, 1H, ArH), 7.72 (m,
1H, ArH), 7.29 (m, 1H, ArH), 3.65 (td, 2H, NCH2), 1.71 (m, 2H,
CH2), 1.34 (m, 4H, 2 × CH2), 0.90 (m, 3H, CH3).
Gen er a l P r oced u r e for Azir id in a tion a n d Am in a tion .
(CuOTf)2‚C6H6 (46 mg, 0.09 mmol for 1.1 equiv of substrate, 10
mg, 0.02 mmol for 5 equiv of substrate) and N-(2-pyridinyl-
methylene)-1-pentanamine (19 mg, 0.1 mmol for 1.1 equiv of
substrate, 4 mg, 0.02 mmol for 5 equiv of substrate) were added
to acetonitrile (37 mL) in a Schlenk tube, using acetonitrile (2
× 1 mL) to aid transfer. The resulting yellow solution was
stirred for 30 min under nitrogen. Chloramine-T trihydrate (510
mg, 1.8 mmol for 1.1 equiv of substrate, 113 mg, 0.4 mmol for 5
equiv of substrate) was added, aided with acetonitrile (1 mL).
The resulting green solution was left stirring for a further 5 min
before hydrocarbon substrate (2 mmol for 1.1 equiv of substrate,
2 mmol for 5 equiv of substrate) was added. The solution was
stirred under nitrogen for 3 days. The resulting reaction mixture
was quenched with 50% hexanes-ethyl acetate (25 mL) and
filtered through a plug of silica gel (2.5 cm) and Celite 535 (0.5
cm). The silica gel was then washed with additional portions
of 50% hexanes-ethyl acetate (3 × 25 mL), and the combined
filtrates were concentrated in vacuo.
a
5 equiv reactions were run at 5-fold dilution (see Experimental
Section).
or 5 equiv of alkene (Table 2), and the major product was
p-toluenesulfonamide. However, we were excited to
observe a significant amount of allylic amination product
8 (entry 1). This encouraged us to use a benzylic
substrate; with tetralin (Table 2, entry 3) a reasonable
yield of benzylic amination product 9 resulted. In the
amination reactions, using more substrate (entries 2 and
4) favored production of even higher percentages of
p-toluenesulfonamide and was thus detrimental. It
should be noted that these reactions did not appear to
have gone to completion within the standard three-day
reaction time. Hence, ratios of products rather than
yields are given.
The product distributions from reactions of chloram-
ine-T catalyzed by 7 are markedly different to copper-
catalyzed reactions of ArSO2NdIPh species,3,6 in which
competing amination is only very rarely observed, even
with alkyl alkenes. In fact, the reactivity resembles
much more closely that of Mu¨ller’s rhodium-catalyzed
reactions,7,8 and we intend to pursue the mechanistic
significance of this observation. Komatsu’s group did not
report results with simple alkyl alkenes,12 so we are
unable to make this comparison with their work.
In summary, the advantages of our new procedure for
nitrene-transfer aziridination of aryl alkenes and ami-
nation of activated hydrocarbons are (i) that commercially
available chloramine-T trihydrate can be used as the
nitrene source without prior dehydration or addition of
molecular sieves and (ii) that large excesses of hydrocar-
bon substrate are not required. Extension of this process
to asymmetric aziridination and amination is underway.
N-(p-Tolu en esu lfon yl)-2-p h en yla zir id in e:3 1H NMR δ 7.86
(d, 2H, ArH), 7.27 (m, 7H, ArH), 3.77 (dd, 1H, CHPh), 2.98 (d,
1H, cis-CH-aziridine), 2.43 (s, 3H, Ar-Me), 2.38 (d, 1H, trans-
CH-aziridine).
t r a n s-N -(p -Tolu e n e su lfon yl)-2-m e t h yl-3-p h e n yla zir i-
d in e:3 1H NMR δ 7.82 (m, 2H, ArH), 7.26-7.20 (m, 5H, ArH),
7.13 (d, 2H, ArH), 3.79 (d, 1H, CHPh), 2.9 (dq, 1H, CHCH3),
2.37 (s, 3H, ArCH3), 1.83 (d, 3H, CH3).
N-[(p -Tolu en esu lfon yl)a m in o]-1,2,3,4-t et r a h yd r on a p h -
th a len -1,2-im in e:3 1H NMR δ 7.80 (m, 2H, ArH), 7.28-7.01 (m,
6H, ArH), 3.80 (d, 1H, CH-aziridine), 3.52 (d, 1H, CH-aziridine),
2.72 (dt, 1H, ArCH), 2.51 (dd, 1H, ArCH), 2.36 (s, 3H, ArCH3),
2.21 (dd, 1H, CH2CH), 1.62 (dt, 1H, CH2CH).
N-(p-Tolu en esu lfon yl)-1-a m in o-2-cycloh exen e:18 1H NMR
δ 7.75 (m, 2H, ArH), 7.30 (m, 2H, ArH), 5.81 (m, 1H, CHdCH),
5.35 (m, 1H, CHdCH), 4.41 (m, 1H, NH), 3.80 (m, 1H, CHd
CHCHNH), 2.43 (s, 3H, ArCH3), 2.00 (m, 2H, CH2CH), 1.85-
1.40 (m, 4H, CH2CH).
N-(p-Tolu en esu lfon yl)-1-a m in o-1,2,3,4-tetr a h yd r on a p h -
th a len e: 1H NMR δ 7.81 (m, 2H, ArH), 7.37-6.90 (m, 6H, ArH),
5.30 (m, 1H, CHNH), 4.69 (d, 1H, NH, absent after D2O shake),
3.10-1.45 (m, 6H, CH2CH).
Ack n ow led gm en t. Our thanks go to the EPSRC
and Pfizer UK for a CASE studentship (to D.P.A.) and
to Dr. David Haddleton, Dr. Rob Deeth, and Dr Andrew
Clark for their help in catalyst design.
Exp er im en ta l Section
All reagents were from commercial sources. 1H NMR spectra
were recorded at 250 MHz in CDCl3.
J O981335Q
N-(2-P yr id in ylm et h ylen e)-1-p en t a n a m in e.17 n-Pentyl-
amine (3.66 mL, 31.5 mmol) was added dropwise over 1 min to
a stirred solution of pyridine-2-carboxaldehyde (3.0 mL, 31.5
(18) McIntosh, M. C.; Weinreb, S. M. J . Org. Chem. 1993, 58, 4823.