Atom transfer radical cyclisations of activated and unactivated N-allylhaloacetamides and N-homoallylhaloacetamides using chiral and non-chiral copper complexes

Article abstract of DOI:10.1039/a909666c

Activated N-tosyl-2,2,2-trichloroacetamide 6a, N-benzyl-2,2,2-trichloroacetamide 6d, 2,2-dichloroacetamides 6b-c and 6e-f and 2-monohaloacetamides lla-g undergo efficient 5-exo atom transfer radical cyclisations at room temperature mediated by CuCl or CuBr in the presence of tris(N,N-dimethylaminoethylene)amine 3 (trien-Me₆). The efficiency and stereoselectivity of these cyclisations was found to be greater than existing published atom transfer procedures based upon CuCl(bipyridine), RuCl₂(PPh₃)₃ and CuCl(TMEDA)₂. The product distribution for the cyclisation onto alkyne 11g was found to be solvent dependent. Attempts to make larger ring sizes by endo cyclisation of N-tosylacetamides 19a-c led to a competing 5-exo ipso aromatic substitution into the N-tosyl group followed by re-aromatisation and loss of SO₂ to furnish an amidyl radical. Cyclisation of N-homoallylacetamides 25a-d proceeded smoothly to give δ-lactams with a range of catalysts based upon ligands 2 and 26. The stereoselectivity of cyclisation to give γ lactams could be somewhat influenced by using chiral enantiopure copper complexes 28-30 suggesting that the reactions may involve metal-complexed radicals.

Full text of DOI:10.1039/a909666c

View Article Online / Journal Homepage / Table of Contents for this issue
Atom transfer radical cyclisations of activated and unactivated
N-allylhaloacetamides and N-homoallylhaloacetamides using chiral
and non-chiral copper complexes
1
Andrew J. Clark,*a Floryan De Campo,b Robert J. Deeth,a Robert P. Filik,a Sylvain Gatard,b
Nicola A. Hunt,a Dominique Lastécouères,b Gerard H. Thomas,c Jean-Baptiste Verlhac*b and
Hathaichanuk Wongtapa
a Department of Chemistry, University of Warwick, Coventry, UK CV4 7AL
b Laboratoire de Chimie Organique et Organométallique (UMR 5802 CNRS),
Université Bordeaux I, 351 cours de la Libération, 33405 Talence cedex, France
c Knoll Pharmaceuticals, Research and Development Department, Pennyfoot Street,
Nottingham, UK NG1 1G
Received (in Cambridge, UK) 8th December 1999, Accepted 25th January 2000
Activated N-tosyl-2,2,2-trichloroacetamide 6a, N-benzyl-2,2,2-trichloroacetamide 6d, 2,2-dichloroacetamides
6b–c and 6e–f and 2-monohaloacetamides 11a–g undergo efficient 5-exo atom transfer radical cyclisations at room
temperature mediated by CuCl or CuBr in the presence of tris(N,N-dimethylaminoethylene)amine 3 (trien-Me6).
The efficiency and stereoselectivity of these cyclisations was found to be greater than existing published atom
transfer procedures based upon CuCl(bipyridine), RuCl2(PPh3)3 and CuCl(TMEDA)2. The product distribution
for the cyclisation onto alkyne 11g was found to be solvent dependent. Attempts to make larger ring sizes by endo
cyclisation of N-tosylacetamides 19a–c led to a competing 5-exo ipso aromatic substitution into the N-tosyl group
followed by re-aromatisation and loss of SO2 to furnish an amidyl radical. Cyclisation of N-homoallylacetamides
25a–d proceeded smoothly to give δ-lactams with a range of catalysts based upon ligands 2 and 26. The stereo-
selectivity of cyclisation to give γ lactams could be somewhat influenced by using chiral enantiopure copper
complexes 28–30 suggesting that the reactions may involve metal-complexed radicals.
2-monohaloacetamide derivatives at room temperature. Both
Introduction
the yields and stereoselectivities are superior to those reported
for similar substrates with either CuCl(bipy),7 homogeneous
and silica supported CuCl(N-pentyl-2-pyridylmethanimine)9 or
CuCl(TMEDA)2 5 at room temperature. In addition, cyclisation
onto alkynes and 6-exo cyclisations to give δ-lactams are
reported as well as attempts to mediate 8 to 12-endo cyclis-
ations at room temperature. The effect of chiral catalysts on the
stereoselectivity of cyclisation of a range of substrates is also
reported.
In recent years, transition metal mediated free radical processes
have gained in importance.1 In particular the atom transfer
radical cyclisation reactions (ATRC) of 2,2,2-trichlorinated
carbonyl compounds have been reported with a range of metal
catalysts, e.g. RuCl2(PPh3)3,2 FeCl2(P(OEt)3)3,3 CuCl(bipy),4
CuCl(TMEDA)2,5 and CuCl(N,N,NЈ,NЈ,NЉ-pentamethyl-
diethylenetriamine).6 However, even with these catalysts both
high temperatures 60–160 ЊC and activated carbon–halogen
bonds (e.g. 2,2,2-trihaloacetyl or 2,2-dihaloacetyl groups) are
generally required as initiators. The cyclisation of N-allyl-N-(4-
tolylsulfonyl)-2,2,2-trichloroacetamides by CuCl(bipy)7 (bipy =
2,2Ј-bipyridine) has been shown to be an efficient process
occurring at room temperature and has been recently extended
to the sequencing of both intramolecular and intermolecular
reactions.8 Ghelfi and co-workers5 reported the use of CuCl-
(TMEDA)2 [e.g. CuCl(1)2] in the cyclisation of N-allyl-N-
benzyl-2,2-dichloro-2-alkylacetamides at room temperature
and claimed it to be superior to that of RuCl2(PPh3)3 and
CuCl(bipy). While it was possible to mediate cyclisations using
CuCl(TMEDA)2 at room temperature, conversions were often
low and diastereoselectivity was relatively poor.5c In addi-
tion cyclisations onto triple bonds failed5b and only cyclisations
of 2,2,2-trihaloacetamides or 2,2-dihaloacetamides were
described. We recently reported that N-alkyl-2-pyridylmethan-
imines could act as versatile tuneable alternative to bipyridine
as ligands in ATRC9 and ATRP10 reactions. In fact a range of
2,2,2-trihaloacetamides or 2,2-dihaloacetamides could be
reacted with high efficiency at room temperature with 2-halo-
acetamides requiring elevated temperatures. We report in this
paper a versatile improved procedure which allows for the gen-
eration of radicals not only from 2,2,2-trihaloacetamide and
2,2-dihaloacetamide derivatives but also from the less activated
Results and discussion
Screening of multidentate amine ligands
We speculated that the origin of the reported improvement in
the activity of CuCl(TMEDA)2 relative to CuCl(bipy) in ATRC
reactions was due to the fact that simple copper(amine) com-
plexes have lower redox potentials than copper(bipyridine)
complexes.11 Ghelfi and co-workers reported that the optimum
ratio of 1 to copper halide was 2:1, indicating that two
bidentate ligands were required for the preparation of the active
catalyst.5 As a consequence, we recently screened a variety of
multivalent amine ligands (1–3) in atom transfer reactions and
discovered that CuBr(2) and CuBr(3) were substantially more
active than CuBr(1)2 in the cyclisation reaction of the bromide
DOI: 10.1039/a909666c
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680
This journal is © The Royal Society of Chemistry 2000
671

View Article Online
Scheme 1 Reagents and conditions: i, 30 mol% CuBr, ligand (see Table
1), CH2Cl2, 30 minutes, rt.
Scheme 2 Reagents and conditions: i, 30 mol% CuCl, 30 mol% 3,
CH2Cl2, rt.
4 (Table 1, Scheme 1).9b Repeating the reactions with CuBr(2)
and CuBr(3) at lower catalyst loadings and lower concen-
trations (for two hours) allowed us to determine that the tetra-
dentate ligand (trien-Me6) 3 was at least ten times faster at
mediating the cyclisation of 4 than the tridentate ligand
(PMDETA) 2. A similar conclusion was reached recently by
Matyjaszewski and co-workers12 who compared the series of
multidentate ligands 1–3 with bipyridine in copper mediated
atom transfer radical polymerisation of styrene. He discovered
that catalysts derived from tridentate 2 or tetradentate 3 ligands
were more active than those derived from TMEDA 1 or bipyrid-
ine ligands. With this information in hand we examined the
cyclisation reactions of a variety of substrates in order to
compare the efficiency of the new ligand system to that of the
published systems using CuCl(1) and RuCl2(PPh3)3.
with only products arising from cyclisation being detected.
While cyclisations of both 6a and 6d using CuCl(bipyridine)7 at
room temperature have been reported to take 15 minutes and 1
hour respectively, with our catalyst system the reactions were
over in less than 30 seconds and 5 minutes respectively (Table 2).
No advantage was found in running the reactions at low con-
centrations, and all reactions were consequently run at 0.12 M
in substrate (identical to that reported for the RuCl2(PPh3)3
mediated cyclisation of 6a).2b,c While the reactions of the 2,2,2-
trichloroacetamide derivatives 6a and 6d were over rapidly at
room temperature, the less activated dichloroacetamide sub-
strates 6b and 6c took approximately 2 h and 30 minutes
respectively. Both reactions furnished mixtures of diastereo-
mers with 6b giving the cis isomer 8b as the major product (ratio
7b:8b = 17:83) while 6c gave the trans isomer 7c as the major
product (ratio 7c:8c = 85:15). Stereochemical assignments
were confirmed by comparison of the NMR data of the prod-
ucts with those of authentic samples already published.2b The
selectivities of these processes were greater than those reported
for the RuCl2(PPh3)3 mediated reactions with the added advan-
tage that the reactions were carried out at room temperature
while the ruthenium mediated processes required 80–100 ЊC.
Interestingly, the sense of induction in the cyclisation of 6b was
opposite to that observed for the related RuCl2(PPh3)3 mediated
reaction. The reason for this outcome is unclear.
The effect of catalyst loading on the efficiency of the cyclis-
ation of 6c was briefly investigated. Hence, 6c was reacted with
either 10, 5, 1 or 0.5 mol% of catalyst at rt over a 24 h period.
While the reactions with 10 and 5 mol% of catalyst proceeded
to completion (by NMR) within the 24 h period, the reactions
with loadings of 1 or 0.5 mol% of catalyst proceeded to give 57
and 33% conversions respectively. While the results indicated
that it was possible to mediate the cyclisations at room temper-
ature with lower catalyst loadings for the rest of the work, we
continued to utilise 30 mol% of catalyst in order to keep the
reaction times conveniently low. Having shown that the CuCl(3)
catalyst was superior to RuCl2(PPh3)3 for the cyclisation of N-
tosylacetamides, we next compared its efficiency to that of
CuCl(TMEDA)2 in the cyclisation of the N-benzyl compounds
6e–f. While Ghelfi reported that the cyclisation of 6f with
CuCl(TMEDA)2 required 20 hours at room temperature and
proceeded to give a 51:49 mixture of cis:trans isomers, we were
delighted to find that using CuCl(3) the reaction was over in 2
hours giving a superior 9:1 ratio of products. For 6e, selectivity
was marginally greater using the CuCl(3) catalyst system. The
sense of the diastereoselectivity in both examples was identical
to that reported by Ghelfi and co-workers.5a–c
We initially chose to evaluate the efficiency and stereoselec-
tivity of the 1:1 complex of CuCl and 3 in atom transfer radical
cyclisation of acetamides 6a–f. Initial experiments involved
comparison of the efficiency of this catalyst CuCl(3) with
that of the previously reported catalysts for the cyclisation
of acetamides 6a–f, those being CuCl(bipyridine),7 CuCl-
5
(TMEDA)2 and RuCl2(PPh3)2.2 The substrates 6a–f were
prepared by the previously reported literature methods.2b,5a,b
Ligand 3 was synthesised according to the standard literature
procedure.13 The active catalyst, CuCl(3) was prepared by the
reaction of a 1:1 ratio of CuCl with the ligand 3 in CH2Cl2.
The catalyst (30 mol%) was then added to the trichloroacet-
amide substrates 6a and 6d (Scheme 2). It was not necessary to
use rigorously dried solvents or glassware or to use an inert
atmosphere in order to carry out the reactions. After the reac-
tions were complete the crude reaction mixture was passed
through a short silica plug (eluted with CH2Cl2) and the solvent
was evaporated to furnish the atom transfer products in high
yield (92–98%). In both cases the reactions proceeded cleanly
Table 1 Screening of ligands
Conversion
(%)a
Mass balance
(%)
Ligand
Equiv.
1
1
2
3
2
3
1
2
1
1
1
1
5
37
100
100
<2b
20b
98
98
92
92
90
94
a 30 mol% CuBr, CH2Cl2, 0.12 M, rt, 30 min. b 10 mol% CuBr, CH2Cl2,
0.03 M, rt, 30 min.
Table 2 Cyclisation of substrates 6a–f
6
R1
R2
R3
Time/min
<0.5
<120
<30
<5
Yield (%)
Diastereoselectivity (%)a
a
b
c
d
e
f
Cl
Cl
Cl
Cl
Cl
Cl
Cl
H
Me
Cl
H
Ts
Ts
Ts
Bn
Bn
Bn
92
96
98
94
90
88
—
66 (56)b
70 (46)b
—
240
120
62 (44)c
80 (2)d
Me
a Ratio determined by 250 MHz 1H NMR. b Ratio in brackets from RuCl2(PPh3)3 cyclisation at 100 ЊC.2b,c c Ratio from CuCl(TMEDA)2 cyclisation at
80 ЊC.5b d Ratio from CuCl(TMEDA)2 cyclisation at rt.5c
672
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680

View Article Online
Table 3 Cyclisation of substrates 11a–11f
Cyclisation of monohaloacetamide substrates
Having shown that activated trihaloacetamides and dihalo-
acetamides underwent atom transfer reactions at room tem-
perature in a more efficient manner than CuCl(TMEDA)2, we
next investigated the reactions of the less activated mono-
haloacetamides 11a–g. Cyclisation of these unactivated sys-
tems with CuBr(bipyridine), CuBr(TMEDA)2 or RuCl2(PPh3)3
had not been reported previously. However we recently
reported the cyclisation of 11a,b and 11e,f at elevated temper-
atures using a silica supported N-pentyl-2-pyridylmethanimine
catalyst.9c Hence, reaction of the lithium amide of 9 with the
various acid bromides 10a–d at Ϫ78 ЊC for 2 hours furnished
the 2-haloacetamides 11a–d in good yields (Scheme 3). The
remaining cyclisation precursors 11e–g were prepared from
their corresponding lithium amides using the same approach.
Yield (%)
Entry
1
Substrate
Product
(cis:trans)a,b
11a
92
2
3
11b
11c
92 (12:88)c
18d
4
11d
95 (6:94)
Scheme 3 Reagents and conditions: i, a) BuLi, THF, Ϫ78 ЊC, 30 min;
b) 10a–d, 2 h.
5
6
11e
11f
86 (6:94)e
96
a 30 mol% CuBr, 30 mol% ligand 3, in CH2Cl2 at rt (0.12 M).
b Determined by 300 MHz 1H NMR of the crude mixture. c Ratio using
silica supported N-pentyl-2-pyridylmethanimine at 80 ЊC (18:82).
d Reaction carried out at 100 ЊC in ClCH2CH2Cl in a sealed tube for
24 h. e Ratio using silica supported N-pentyl-2-pyridylmethanimine at
80 ЊC (17:83).
The five bromo precursors 11a–d and 11f underwent cyclis-
ation with 30 mol% of CuBr(3) to furnish the expected cyclis-
ation products 13a–d and 13f respectively, (Table 3). It was
discovered that as the degree of substitution at the α-carbon
decreased, the rate of the cyclisation reactions slowed markedly.
Hence, cyclisation of the tertiary precursors 11a or 11f pro-
ceeded with the fastest rate and were over after a few hours at
room temperature while cyclisation of the primary halide 11c
required heating (100 ЊC, sealed tube) over an extended period
of time (24 hours). Under these conditions 11c furnished a mix-
ture of products with the cyclised product 13c being obtained in
low yield (18%). A significant amount of deacetylated product
N-allyl-N-toluene-4-sulfonamide 9 was also detected. In this
case cleavage of the amide bond was the major reaction path-
way indicating that the use of high temperatures is not applic-
able to this methodology. While the reaction of 11c was not
very efficient the result is significant in that Nagashima et al.7
reported that CuCl(bipyridine) failed to cyclise the related
N-allyl-N-benzyliodoacetamide.7 Atom transfer cyclisation of
the secondary bromoacetamides 11b, 11d and 11e furnished
the expected 5-exo products 13b, 13d and 13e as mixtures of
diastereomers. The major products in all cases were determined
to be the trans diastereomers based upon comparison with
authentic samples9c,14 and NOE evidence. The high trans selec-
tivity can be rationalised by examining the potential transition
states of cyclisation (Fig. 1). Transition states (TS) for pathways
leading to cis and trans products were computed using
MOPAC.15 The calculations indicated that cyclisation via the
trans pathway TS was lower in energy than that for the cis
pathway (∆E = 4.0 kJ molϪ1). Cyclisation of the N-(2-methyl-
allyl)-N-(4-tolylsulfonyl)amide derived precursor 11f furnished
the 5-exo cyclisation product 13f exclusively with no 6-endo
product being detected in the crude NMR spectrum.
Fig. 1 Transition states for pathways leading to a) cis and b) trans
compounds.
Cyclisation onto alkynes
The report that CuCl(TMEDA)2 failed to mediate the cyclis-
ation of radicals onto alkynes5b prompted us to investigate the
cyclisation of the propargylic† acetamide 11g. Reaction with 30
mol% CuBr(3) in CH2Cl2 at room temperature gave a mixture
of products (Scheme 4). Analysis of the crude reaction mixture
indicated that the expected bromoalkene derivatives 14 (3:1
mixture of (E)- and (Z)-isomers) had been formed along with a
significant amount of the reduced alkene 15 (ratio 14:15 =
1:1). In addition, a trace amount (<2%) of a mixture of
† The IUPAC name for propargylic is prop-2-ynyl.
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680
673

View Article Online
large amount of starting material was recovered (53%). The
unexpected formation of the amide 20a can be explained by
a competing 5-exo ipso aromatic radical substitution of initial
radical 22 to give 23 (Scheme 6). This new radical can then
Scheme 4 Reagents and conditions: i, 30 mol% CuBr, 30 mol% 3,
solvent, rt.
compounds, tentatively assigned as the chloroalkene derivatives
16, was also detected. The products 16 and 15 presumably arise
from chlorine atom and hydrogen atom abstraction from the
solvent (CH2Cl2) respectively. Repeating the reaction using
tetrahydrofuran as solvent (a better hydrogen atom donor than
CH2Cl2) furnished the reduced product 15 almost exclusively in
high yield (90%) even though a catalytic amount (30 mol%) of
CuBr(3) was used. In this case it is unclear whether the tetra-
hydrofuranyl radical formed by the reduction of the intermedi-
ate vinyl radical can facilitate cleavage of the carbon–bromine
bond in the precursor 11g thus completing the chain reaction.
These competing abstraction reactions were not observed for
any of the other cyclisations reported here. This can be rational-
ised in terms of the greater reactivity of the intermediate vinyl
radical arising from the reaction of 11g with respect to the
primary radicals arising from the cyclisation of the substrates
11a–f.
Scheme 6 Reagents and conditions: 30 mol% CuCl, 30 mol% 3,
CH2Cl2, rt.
undergo re-aromatisation followed by C–S bond cleavage and
ultimately loss of SO2 to furnish the observed product 20a. The
intermediate amidyl radical 24 in theory could undergo a 5-exo
cyclisation reaction. However, no product arising from this
pathway was detected in the crude mixture. We have recently
reported that the cyclisations of amidyl radicals of type 24 with
hindered secondary or tertiary groups appended to the carb-
onyl are extremely slow16 and, as a consequence, in this example
the amidyl radical is likely to be competitively trapped by a
H atom from the solvent to furnish the observed amide. The
competitive migration of arylsulfonyl groups in relatively slow
radical cyclisations mediated by tributyltin hydride has been
observed before17 and has been exploited by Motherwell
and co-workers to develop a new approach to substituted
biphenyls.18 In our case the relatively slow rate of 8-endo cyclis-
ation allows competitive migration to be observed under atom
transfer conditions. The observed yield of only 30% of the
rearranged product 20a suggests that the copper complex is not
regenerated in the reaction and that it is not acting as a
“catalyst”. Attempts to mediate macrocyclisation of the other
two precursors 19b,c using stoichiometric amounts of copper
reagents (100 mol%) also led to no observable macrocyclisation
products. As before, varying amounts of the products tentatively
assigned as 20b,c were detected (20b = 49%, 20c = 37%, based
upon NMR spectra) although these could not be isolated pure
from the crude reaction mixture.
Attempts to mediate macrocyclisations
Recently we have shown that medium ring lactones could be
prepared by 8-, 9- and 10-endo atom transfer radical macro-
cyclisation of trichloro esters 17 at elevated temperatures.6 No
reports of the cyclisation to furnish medium ring lactams were
reported in this work. In order to evaluate the tetradentate
catalyst system 3 with respect to the synthesis of larger ring
lactams we prepared the acetamides 19a–c (Scheme 5). Acet-
Cyclisation to give ꢀ-lactams
Functionalised δ-lactams are valuable intermediates for the
synthesis of six-membered heterocyclic compounds. During the
last ten years there has been much interest in their synthesis due
to their biological activity against ophthalmic infections,19
gastric carcinogenesis,20 blood contamination21 or digestive
tract cancer.22 Surprisingly, only a few methods are reported in
the literature for the synthesis of these compounds. Most
approaches have relied on ionic chemistry that involves
nitrogen–carbon bond formation. Radical cyclisation involving
carbon–carbon bond formation is an alternative route, but this
methodology has received little attention in contrast with the
manifold applications for the synthesis of γ-lactams. Con-
sequently, we prepared the trichloroacetamides 25a–d in 40–
93% yields from the corresponding N-substituted alkenyl
amines, by direct trichloroacetylation at 0 ЊC in dichloro-
methane.
Scheme 5 Reagents and conditions: i, TsCl, Et3N, CH2Cl2; ii, BuLi,
THF, Ϫ78 ЊC CCl3COCl; iii, 30 mol% CuCl, 30 mol% 3, CH2Cl2, rt.
amides 19a–c were prepared in good overall yield by tosylation
of the corresponding amines 18a–c followed by acylation with
trichloroacetyl chloride.
Attempts to facilitate 8-endo cyclisation of acetamide 19a at
room temperature failed with the main products being the
rearranged product 20a (30%) and the toluenesulfonamide
21a (16%) formed by amide bond cleavage. In addition, a
We screened a variety of copper complexes including the
674
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680

View Article Online
Table 4 Synthesis of γ- and δ-lactams using ligands 2 and 26
Entry
Substrate
Catalyst (mol%)
Time/h
Temp/ЊC
Product
Yield (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
6a
6a
CuClؒbipy (30)
CuClؒ2 (30)
CuClؒ26 (30)
CuClؒ2 (5)
CuClؒ26 (30)
CuClؒbipy (30)
CuClؒ2 (30)
CuClؒ26 (30)
CuClؒ2 (30)
CuClؒ26 (30)
CuClؒbipy (10)
CuClؒ2 (10)
CuClؒ26 (10)
2
2
3
18
18
72
72
72
2
25
25
25
80
80
25
25
25
80
80
80
80
80
7
95
99
99
92
97
60
30
40
92
90
99
98
96
7
6a
7
6a
6a
7
7
25a
25a
25a
25a
25a
25d
25d
25d
27a
27a
27a
27a
27a
27d
27d
27d
2
18
18
18
Table 5 Effect of chiral catalysts on diastereoselectivity of cyclisation
Diastereomer
Entry
Substrate
Ligand
Time/h
Temp/ЊC
Product
ratio (Yield %)a
1
2
3
4
5
6
7
8
9
31
2
28
29
30
2
28
29
30
2
18
18
18
18
96
96
18
18
18
18
18
18
80
80
80
80
25
25
25
25
80
80
25
80
32
43:57 (97)
42:58 (86)
44:56 (85)
49:51 (85)
42:58 (78)
27:73 (21)
40:60 (61)
43:57 (77)
54:46 (78)
56:44 (95)
54:46 (6)
31
32
31
32
31
32
31
32
31
32
31
32
31
32
25c
25c
25c
25c
27c
27c
27c
27c
10
11
12
29
29
30
55:45 (96)
a Diastereomeric ratio determined by GC, stereochemistry of major isomer not determined.
Modification of stereoselectivity of cyclisations mediated by
chiral copper(I) complexes
We next prepared a range of chiral enantiopure ligands (28–
30)23–25 and investigated their effect on the diastereoselectivity
of cyclisation of a range of substrates. Hence, initial work
focussed on the cyclisation of the trichloracetamides 31 and 25c
(Table 5, Scheme 8).
Scheme 7 Reagents and conditions: i, 30 mol% CuCl, 30 mol% ligand 2
or 26, 25–28 ЊC, 1,2-dichloroethane.
complexes derived from CuCl and the ligands 2 and 266 in the
cyclisation both of 25a and 25d (Scheme 7) as well as of 6a.
Cyclisation proceeded at 25–80 ЊC in moderate to excellent
yields (see Table 4) in 1,2-dichloroethane (0.1 M) under an
argon atmosphere. We observed that 1:1 Cu() complexes with
ligands 2 and 26 are as active as the previously reported
CuCl(bipy) catalyst in the case of the cyclisation of substrates
6a (entries 1–5) and 25d (entries 11–13). Surprisingly, the intro-
duction of a tert-butyl protecting group onto the nitrogen of
25d does not impede the reaction. Whichever ligand was
employed, only exo cyclisation products were obtained.
Scheme 8 Reagents: 10 mol% CuCl, 10 mol% ligand, ClCH2CH2Cl.
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680
675

View Article Online
Table 6 Effect of chiral catalysts on diastereoselectivity of cyclisation
Diastereomer
Entry
Substrate
Ligand
Time/h
Temp/ЊC
Product
ratio (Yield %)a
1
2
3
4
5
6
7
8
9
33
bpy
2
28
29
30
2
29
30
2
18
18
18
18
18
72
96
18
96
96
96
18
25
80
80
80
80
25
25
25
80
25
80
80
34
10:90 (54)
14:86 (99)
25:75 (94)
20:80 (92)
16:84 (100)
8:92 (97)
19:81 (100)
10:90 (65)
15:85 (88)
5:95 (15)
33
34
33
34
33
34
33
34
33
34
33
34
33
34
25b
25b
25b
25b
27b
27b
27b
27b
10
11
12
2
29
30
15:85 (83)
15:85 (87)
a Diastereomeric ratio determined by GC, stereochemistry of major isomer not determined.
As the majority of the enantiopure copper() complexes used
in this study were very oxygen sensitive, their reactions were
Conclusions
In conclusion, we have shown that the use of CuX [X = Br or
Cl] with ligands 2, 3 or 26 will successfully mediate atom
transfer radical cyclisation reactions of a range of halo-
acetamides in excellent yields at room temperature to give γ-
or δ-lactams. The selectivity of these processes was greater
than that reported for related RuCl2(PPh3)3 and Cu(TMEDA)2
mediated cyclisations. While the previous procedures have
described the cyclisation of activated 2,2,2-trichloroacetamides
or 2,2-dichloroacetamides at elevated temperatures we have
shown that it is possible to cyclise not only these activated
systems but also less activated 2-bromoacetamides at ambient
temperatures. Cyclisation onto alkynes is also possible, in con-
trast to the reported CuCl(TMEDA)2 procedure. For relatively
slow cyclisations, N-(4-tolylsulfonyl)acetamides undergo com-
peting rearrangement reactions under the reaction conditions.
The greater activity of the new catalyst system CuX(3) relative
to CuCl(TMEDA)2 should allow studies into chiral induction
by chiral N-groups in cyclisations to be optimised by carrying
out these reactions at low temperatures. Finally, we have shown
that by using chiral ligands in conjunction with chiral sub-
strates it is possible to alter slightly the diastereoselectivity of
cyclisation (matched or mismatched) suggesting that enantio-
selective cyclisations may be possible using chiral copper com-
plexes. This suggests that the cyclisations are not truly “free
radical” in character and that the copper complex may be
intimately involved in the TS for cyclisation either by a templat-
ing effect4 or by the direct involvement of metal complexed
radicals.
performed after rapid degassing of the reaction vessel with
nitrogen. The effect of the α-methylbenzylamine substituent on
the outcome of cyclisation of both 31 and 25c remained low
even at room temperature. However, the diastereoselectivity for
31 was dependent upon the chirality of the ligand, particularly
at room temperature (compare entries 5 and 6), although the
low yield for this reaction means the result should be treated
with caution. However, it does provide some evidence that it
should be possible to induce stereochemical control in a cyclis-
ation by utilising chiral copper complexes (i.e. by matching the
ligand with the substrate). In addition it suggests that enantio-
selective cyclisations may be possible using chiral copper com-
plexes (i.e. the chiral copper complex is involved in the TS for
cyclisation perhaps by complexing to the radical and thus the
reactions may not be truly “free radical” in nature).
Finally, we investigated the effect of the different chiral
ligands on the stereochemical outcome of cyclisation of the
substrates 337 and 25b where the stereochemistry is provided by
the alkenyl side-chain (Schemes 9, 10). The use of the ligands 2
Scheme 9 Reagents: 10 mol% CuCl, 10 mol ligand, ClCH2CH2Cl.
Experimental
Melting points were recorded on a Stuart Scientific SMP1 melt-
ing point apparatus and are uncorrected. Accurate mass
determinations were performed either on a Kratos MS80 at the
University of Warwick or on a LC-MS SSS at Knoll Pharm-
aceuticals. Microanalyses were recorded on a Leeman Labs Inc.
CE440 Elemental Analyser. Infra-red spectra were recorded in
a solution cell, as Nujol mulls or neat, as stated in the text, on a
Perkin-Elmer 1720X Fourier transform spectrometer. 1H NMR
spectra were recorded at either 250, 300, or 400 MHz on a
Bruker ACF250, Bruker DPS300 or Bruker ACP400 instru-
ment respectively. Chemical shifts are quoted in parts per
million (ppm) and referenced to the appropriate solvent peak;
coupling constants are given in Hz. 13C NMR spectra were
recorded at 62.9, 75 and 100.6 MHz. Chemicals used were
obtained from either Lancaster or Sigma–Aldrich at the highest
grade available. All solvents were purchased from Fisons
Scientific Equipment at SLR grade and purified, when needed,
by literature methods. Flash chromatography was carried out
on silica gel (Merck Kieselgel 60F254, 230–400 mesh). TLC was
Scheme 10 Reagents: 10 mol% CuCl, 10 mol ligand, ClCH2CH2Cl.
and 28–30 resulted in the formation of trans γ- and δ-lactams
predominantly with very high conversions. Substrate 25b was
converted quantitatively even at ambient temperature in
the presence of 10 mol% of catalyst derived from ligand 29.
In all the experiments the resulting diastereomeric ratio
(Table 6) is as good as previously reported by Nagashima et al.7
with the 2,2Ј-bipyridine complex and it seems to be slightly
dependent upon the ligand at low temperature but not as
greatly as for 31.
676
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680

View Article Online
carried out using aluminium backed plates precoated with silica
d, J 6.7, CHMeMe) and 0.86 (3 H, d, J 6.7, CHMeMe); δC(75
(0.2 mm, 60F254). [α]D has units of 10Ϫ1 deg cm2 gϪ1
.
MHz; CDCl3) 169.3 (s), 145.6 (s) 136.2 (s), 132.9 (d), 130.1
(2 × d), 128.5 (2 × d), 118.6 (t), 53.2 (d), 49.1 (t), 36.6 (d), 22.1
(q), 20.9 (q) and 20.2 (q); m/z (EI) 373.0347 (Mϩ, C15H20-
Br79NO3S requires 373.0347), 373 (8%, Mϩ), 294 (13), 204 (86),
155 (80), 91 (100) and 65 (45).
Preparation of substrates
Tris(N,N-dimethylaminoethylene)amine 3, N-allyltoluene-
4-sulfonamide, N-(2-methylprop-2-enyl)toluene-4-sulfonamide,
N-allyl-N-benzylamine,
N-allyl-N-(4-tolylsulfonyl)-2,2,2-tri-
N-Prop-2-enyl-N-(4-tolylsulfonyl)-2-chloro-2-phenylacet-
amide 11e. Yield 65%, as a white crystalline solid, mp 70–71 ЊC
(from ethyl acetate) (Found: C, 59.2; H, 5.0; N, 3.75.
C18H18ClNO3S requires C, 59.4; H, 5.0; N, 3.85%); νmax (CHCl3)/
cmϪ1 2925, 1702, 1593, 1375, 1161 and 919; δH(250 MHz;
CDCl3) 7.69 (2 H, d, J 8.2, Ar), 7.37–7.25 (7 H, m, Ar), 6.13
chloroacetamide 6a, N-allyl-N-(4-tolylsulfonyl)-2,2-dichloro-
acetamide 6b, and N-allyl-N-(4-tolylsulfonyl)-2,2-dichloro-2-
methylacetamide 6c, N-allyl-N-benzyl-2,2,2-trichloroacetamide
6d, N-allyl-N-benzyl-2,2-dichloroacetamide 6e, and N-allyl-N-
benzyl-2,2-dichloro-2-methylacetamide 6f were prepared
according to literature procedures.2b,c,5a–c N-(4-tolylsulfonyl)-4-
chloromethyl-3,3-dichloropyrrolidin-2-one 7a, cis- and trans-N-
(4-tolylsulfonyl)-4-chloromethyl-3-chloropyrrolidin-2-one 7b
and 8b, cis- and trans-N-(4-tolylsulfonyl)-4-chloromethyl-3-
chloro-3-methylpyrrolidin-2-one 7c and 8c, N-benzyl-4-
chloromethyl-3,3-dichloropyrrolidin-2-one 7d, cis- and trans-
N-benzyl-4-chloromethyl-3-chloropyrrolidin-2-one 7e and 8e,
and cis- and trans-N-benzyl-4-chloromethyl-3-chloro-3-
methylpyrrolidin-2-one 7f and 8f exhibited spectroscopic
data identical to those previously published.2b,c,5a–c N-Allyl-N-
(4-tolylsulfonyl)-2-bromo-2-methylpropionamide 11a, N-allyl-
(1 H, s, CHPh), 5.86–5.71 (1 H, m, CH᎐CH ), 5.24–5.16 (2 H,
᎐
2
m, CH᎐CH ), 4.50–4.27 (2 H, m, NCH ) and 2.43 (3 H, s, Me);
᎐
2
2
δC(75 MHz; CDCl3) 162.7 (s), 145.6 (s) 135.9 (s), 132.3 (d),
130.7 (d), 129.8 (d), 129.4 (2 × d), 128.9 (2 × d), 128.6 (d), 119.0
(t), 58.8 (d), 49.1 (t) and 22.1 (q); m/z (EI) 363.0684 (Mϩ,
C18H18Cl35NO3S requires 363.0696), 362 (5%, Mϩ), 328 (34),
155 (100), 125 (82), 91 (83) and 69 (20).
N-(2-Methylprop-2-enyl)-N-(4-tolylsulfonyl)-2-bromo-2-
methylpropanamide 11f. Yield 75%, as a white crystalline solid,
mp 112–113 ЊC (from ethyl acetate) (Found: C, 48.2; H, 5.4;
N, 3.4. C15H20BrNO3S requires C, 48.1; H, 5.35; N, 3.7%);
νmax (CHCl3)/cmϪ1 2936, 1680, 1597, 1353, 1169 and
924; δH(300 MHz; CDCl3) 7.78 (2 H, d, J 8.5, Ar), 7.21 (2 H, d,
N-(4-tolylsulfonyl)-2-bromopropionamide
methyl-3,3-dimethyl-1-(4-tolylsulfonyl)pyrrolidin-2-one
and trans-4-bromomethyl-3-methyl-1-(4-tolylsulfonyl)pyrrol-
idin-2-one 13b exhibited spectroscopic data identical to those
previously published.9c,14
11b,
4-bromo-
13a,
J 8.5, Ar), 4.95 (1 H, br s, CMe᎐CH ), 4.73 (2 H, br s, CH ),
᎐
2
2
2.32 (3 H, s, Me) and 1.72 (6 H, s, 2 × Me); δC(75 MHz; CDCl3)
170.8 (s), 145.1 (s) 140.9 (s), 136.2 (s), 129.5 (2 × d), 129.1
(2 × d), 112.4 (t), 58.0 (s), 53.7 (t), 32.4 (2 × q), 22.1 (q) and 20.7
(q); m/z (EI) 373 (2%, Mϩ), 309 (34), 202 (82), 155 (82) and 91
(100).
General procedure for the preparation of toluene-4-sulfonamide
cyclisation precursors
A solution of BuLi (2.24 cm3, 2.5 M in hexanes, 5.6 mmol)
was added dropwise over 5 minutes to a stirred solution of
N-alkyltoluene-4-sulfonamide (2.4 mmol) in dry tetrahydro-
furan (30 cm3) at Ϫ78 ЊC under nitrogen and the mixture was
stirred for 30 minutes at this temperature. The acid halide (6.2
mmol) was added and the mixture stirred for 2 h at Ϫ78 ЊC. The
reaction was quenched with ammonium chloride (5 cm3) and
allowed to warm to room temperature. The mixture was
extracted with diethyl ether (2 × 30 cm3) and washed with
saturated sodium bicarbonate (2 × 30 cm3). The organic extracts
were dried over MgSO4 and the solvent evaporated under
reduced pressure to give the products. The crude compounds
were purified by column chromatography on silica using
hexane–ethyl acetate (4:1) as eluent.
N-Prop-2-ynyl-N-(4-tolylsulfonyl)-2-bromo-2-methylpropion-
amide 11g. Yield 76%, as a white crystalline solid, mp 92–93 ЊC
(from ethyl acetate) (lit.,26 92–93 ЊC) (Found: C, 47.3; H, 4.55;
N, 3.9. C14H16BrNO3S requires C, 46.9; H, 4.5; N, 3.9%); νmax
(CHCl3)/cmϪ1 2925, 1692, 1593, 1376 and 1165; δH(300 MHz;
CDCl3) 7.91 (2 H, d, J 8.5, Ar), 7.23 (2 H, d, J 8.5, Ar), 5.05
(2 H, d, J 2.5, CH CH᎐CH ), 2.40 (1 H, t, J 2.5, CH), 2.35 (3 H,
᎐
2
2
s, Me) and 1.87 (6 H, s, 2 × Me); δC(75 MHz; CDCl3) 170.0 (s),
145.3 (s) 136.0 (s), 129.6 (4 × d), 79.1 (s), 74.3 (d) 57.1 (s), 38.2
(t), 32.1 (2 × q) and 22.1 (q); m/z (CI) 358.0127 (MHϩ
C14H17Br79NO3S requires 358.0113), 358 (39%, MHϩ), 280
(100), 216 (45), 189 (32) and 124 (55).
N-Allyl-N-(4-tolylsulfonyl)bromoacetamide 11c. Yield 62%,
as a white crystalline solid, mp 89–90 ЊC (from ethyl acetate)
(Found: C, 43.7; H, 4.1; N, 3.9. C12H14BrNO3S requires C, 43.4;
H, 4.25; N, 4.2%); νmax (CHCl3)/cmϪ1 2923, 1694, 1648, 1593,
1356, 1167, 838 and 717; δH(250 MHz; CDCl3) 7.80 (2 H, d,
N-(4-tolylsulfonyl)-pent-4-enamine 21a. Yield 16% (Found:
C, 60.45; H, 7.0; N, 5.8. C12H17NO2S requires C, 60.2; H, 7.2; N,
5.85%); νmax (Nujol)/cmϪ1 3280, 2927, 1640, 1598, 1162, 1095
and 914; δH(400 MHz; CDCl3) 7.72 (2 H, d, J 8.2, Ar), 7.26
(2 H, d, J 8.2, Ar), 5.66 (1 H, m, CH᎐CH ), 5.05 (1 H, br s,
᎐
2
J 8.5, Ar), 7.32 (2 H, d, J 8.5, Ar), 5.89–5.73 (1 H, m, CH᎐CH ),
᎐
2
NH), 4.91 (2 H, m, CH᎐CH ), 2.89 (2 H, t, J 7.2, CH NHTs),
᎐
2
2
5.26–5.16 (2 H, m, CH᎐CH ), 4.43 (1 H, dt, J 5.4 and 1.5,
᎐
2
2.38 (3 H, s, Me), 1.99 (2 H, m, CH CH᎐CH ) and 1.52 (2 H,
᎐
2
2
CH CH᎐CH ), 4.18 (2 H, s, CH Br), and 2.41 (3 H, s, Me);
᎐
2
2
2
quintet, J 7.2, CH2CH2CH2); δC(75 MHz; CDCl3) 143.7 (s),
137.6 (d), 137.3 (s), 130.1 (2 × d), 127.5 (2 × d), 115.9 (t), 43.0
(t), 31.0 (t), 29.0 (t) and 21.9 (q); m/z (CI) 240 (30%, MHϩ), 184
(65), 155 (100) and 91 (94).
δC(67.8 MHz; CDCl3) 166.1 (s), 146.0 (s) 136.0 (s), 132.4
(d), 130.3 (2 × d), 128.5 (2 × d), 119.1 (t), 49.6 (t), 29.4 (t) and
22.1 (q); m/z (EI) 330.9887 (Mϩ, C12H14Br79NO3S requires
330.9877), 331 (4%, Mϩ), 162 (80), 155 (65), 91 (100) and 65
(70).
N-Pent-4-enyl-N-(4-tolylsulfonyl)-2,2,2-trichloroacetamide
19a. Yield 77%, as a white crystalline solid, mp 70–71 ЊC (from
ethyl acetate) (Found: C, 43.7; H, 4.2; N, 3.5. C14H16Cl3NO3S
requires C, 43.7; H, 4.2; N, 3.6%); νmax (Nujol)/cmϪ1 2924, 1713,
1661, 1368, 1371, 1173 and 1086; δH(400 MHz; CDCl3) 7.89
(2 H, d, J 8.4, Ar), 7.32 (2 H, d, J 8.4, Ar), 5.79 (1 H, m,
N-Allyl-N-(4-tolylsulfonyl)-2-bromo-3-methylbutanamide
11d. Yield 64%, as a clear viscous oil (Found: C, 48.55; H, 5.5;
N, 3.7. C15H20BrNO3S requires C, 48.1; H, 5.4; N, 3.7%); νmax
(neat)/cmϪ1 2922, 1706, 1597, 1462, 1376, 1172 and 928; δH(300
MHz; CDCl3) 7.82 (2 H, d, J 8.5, Ar), 7.32 (2 H, d, J 8.5, Ar),
5.90–5.80 (1 H, m, CH᎐CH ), 5.32–5.22 (2 H, m, CH᎐CH ),
CH᎐CH ), 5.06 (2 H, m, CH᎐CH ), 4.18 (2 H, m, CH N), 2.43
᎐ ᎐
2 2 2
᎐
᎐
2
2
4.65 (1 H, ddt, J 17.1, 5.2 and 1.5, CHHCH᎐CH ), 4.48 (1 H, d,
J 9.2, CHBr), 4.35 (1 H, ddt, J 17.1, 5.5 and 1.2, CHH-
(3 H, s, Me) and 2.11 (4 H, m, CH2CH2); m/z (EI) 383.9975
(MHϩ, C14H16Cl3NO3S requires 383.9994), 382 (<1%, Mϩ), 321
(5), 214 (27), 155 (95) and 91 (100).
᎐
2
CH᎐CH ), 2.42 (3 H, s, Me), 2.26 (1 H, m, CHMe ), 1.06 (3 H,
᎐
2
2
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680
677

View Article Online
N-Hex-5-enyl-N-(4-tolylsulfonyl)-2,2,2-trichloroacetamide
19b. Yield 81%, as a white crystalline solid, mp 73–74 ЊC (from
ethyl acetate) (Found: C, 44.8; H, 4.6; N, 3.15. C15H18Cl3NO3S
requires C, 45.2; H, 4.55; N, 3.50%); νmax (Nujol)/cmϪ1 2924,
1712, 1551, 1368, 1371, 1173 and 1086; δH(400 MHz; CDCl3)
7.89 (2 H, d, J 8.4, Ar), 7.32 (2 H, d, J 8.4, Ar), 5.78 (1 H, m,
CHPh), 3.50 (1 H, dd, J 11.6 and 6.6, CHHCl), 2.75–2.84 (1 H,
m, CHCH2Cl) and 2.44 (3 H, s, Me); δC(75 MHz, CDCl3) 172.3
(s), 145.9 (s), 135.3 (s), 133.2 (s), 130.2 (2 × d), 129.9 (d), 129.0
(2 × d), 128.9 (2 × d), 128.8 (2 × d), 53.0 (t), 48.4 (s), 44.3 (s),
43.2 (t) and 22.1 (q); m/z (EI) 363.0683 (Mϩ, C18H18ClNO3S
requires 363.0696), 363 (5%, Mϩ), 328 (23), 265 (17), 155 (80)
and 91 (100).
CH᎐CH ), 5.06–4.96 (2 H, m, CH᎐CH ), 4.18 (2 H, m, CH N),
᎐
᎐
2
2
2
2.44 (3 H, s, Me), 2.14 (2 H, m, CH2), 2.00 (2 H, m, CH2) and
1.49 (2 H, m, CH2); m/z (EI) 397.0085 (Mϩ, C15H18Cl3NO3S
requires 397.0090).
4-Bromomethyl-3,3,4-trimethyl-1-(4-tolylsulfonyl)pyrrolidin-
2-one 13f. Yield 96%, as a white crystalline solid, mp 175–
176 ЊC (from ethyl acetate) (Found: C, 48.4; H, 5.5; N, 3.6.
C15H20BrNO3S requires C, 48.1; H, 5.35; N, 3.7%); νmax (CHCl3)/
cmϪ1 2936, 1738, 1367 and 1174; δH(300 MHz; CDCl3) 7.88
(2 H, d, J 8.3, Ar), 7.31 (2 H, d, J 8.3, Ar), 3.83 (1 H, d, J 10.7,
CHHN), 3.54 (1 H, d, J 10.7, CHHN), 3.06–3.27 (2 H, m,
CH2Br), 2.41 (3 H, s, Me), 1.06 (3 H, s, Me), 1.02 (3 H, s, Me)
and 0.94 (3 H, s, Me); δC(75 MHz, CDCl3); 177.4 (s), 145.7
(s), 135.3 (s), 130.1 (2 × d), 128.3 (2 × d), 54.5 (t), 48.9 (s), 43.0
(s), 38.9 (t), 22.1 (q), 21.3 (q), 19.5 (q) and 18.9 (q); m/z (EI)
374 (1%, Mϩ), 309 (100), 230 (46), 155 (51), 133 (70) and 91
(86).
N-Oct-7-enyl-N-(4-tolylsulfonyl)-2,2,2-trichloroacetamide
19c. Yield 81%, (Found: C, 48.0; H, 5.2; N, 3.0. C17H22Cl3NO3S
requires C, 47.8; H, 5.2; N, 3.2%); νmax (Nujol)/cmϪ1 2924, 1714,
1610, 1551, 1368, 1371, 1173 and 1086; δH(400 MHz; CDCl3)
7.48 (2 H, d, J 8.4, Ar), 7.32 (2 H, d, J 8.4, Ar), 5.78 (1 H, m,
CH᎐CH ), 5.02–4.92 (2 H, m, CH᎐CH ), 4.15 (2 H, m, CH N),
᎐
᎐
2
2
2
2.44 (3 H, s, Me), 2.00 (4 H, m, 2 × CH2) and 1.42–1.29 (6 H, m,
3 × CH2); m/z CI (426, 100%, MHϩ), 358 (47), 155 (90), 108
(83) and 91 (72).
General procedure for atom transfer cyclisation reactions
3,3-Dimethyl-4-methylene-1-(4-tolylsulfonyl)pyrrolidin-2-one
To the substrate (2 mmol) in dry CH2Cl2 (2 cm3) under nitrogen
at room temperature was added either CuCl (0.6 mmol) or
CuBr (0.6 mmol) and tri(N,N-dimethylaminoethylene)amine 3
(0.6 mmol). The reaction was followed by TLC. After complete
reaction the mixture was passed through a small silica plug. The
silica plug was washed with dichloromethane (50 cm3) and the
solvent removed in vacuo to give the crude products. Chroma-
tography using hexane–ethyl acetate furnished the pure cyclised
compounds.
15,26
(E)-4-bromomethylene-3,3-dimethyl-1-(4-tolylsulfonyl)-
pyrrolidin-2-one (E)-14,26 and (Z)-4-bromomethylene-3,3-di-
methyl-1-(4-tolylsulfonyl)pyrrolidin-2-one (Z)-14.26 Yield 95%,
as an inseparable mixture of compounds, νmax (Nujol)/cmϪ1
1740, 1670, 1365 and 1170; Discernible data for 15, δH(400
MHz; CDCl3) 7.92 (2 H, m, Ar), 7.33 (2 H, m, Ar), 5.08 (1 H, m,
C᎐CHH), 5.04 (1 H, m, C᎐CHH), 4.44–4.37 (2 H, m, CH ), 2.43
᎐
᎐
2
(3 H, s, Me) and 1.14 (6 H, s, 2 × Me); discernible data for
(E)-14 and (Z)-14 (ratio 3:1), 7.94 (2 H, m, Ar), 7.30 (2 H, m,
Ar), 6.17 (1 H (E), t, J 2.1, C᎐CHBr), 6.14 (1 H (Z), t, J 2.6,
᎐
4-Bromomethyl-1-(4-tolylsulfonyl)pyrrolidin-2-one13c.Cyclis-
ation gave an inseparable mixture of N-allyl-N-toluene-
C᎐CHBr), 4.44–4.37 (2H (E) and (Z), m, CH ), 2.37 (3 H (E)
᎐
2
and (Z), s, Me) 1.38 (3 H (E), s, Me) and 1.20 (3 H (Z), s, Me);
m/z LC-MS (APϩ) 4.70 minutes 280 (MHϩ) 280.1022 (MHϩ,
C14H18NO3S requires 280.1001), 5.20 minutes 358 (Mϩ),
358.0098 (Mϩ, C14H17Br79NO3S requires 358.0112), 215 (60%),
149 (46), 91 (100) and 81 (75).
4-sulfonamide5b
and
pyrrolidin-2-one 13c26 (ratio 1:3). Discernible data for
4-bromomethyl-1-(4-tolylsulfonyl)pyrrolidin-2-one: δH(250
4-bromomethyl-1-(4-tolylsulfonyl)-
MHz; CDCl3) 7.92 (2 H, d, J 8.5, Ar), 7.34 (2 H, d, J 8.5, Ar),
4.06 (1 H, dd, J 10.2 and 7.6, CHHN), 3.66 (1 H, dd, J 10.2 and
6.0, CHHN), 3.45 (2 H, m, CH2Br), 2.78, (1 H, m, CHCH2Br),
2.62, (1 H, dd, J 17.4 and 8.5, CHHCO), 2.43 (3 H, s, Me) and
2.35 (1 H, dd, J 17.4 and 6.8, CHHCO).
General procedure for trichloroacetylation of benzylamines
To a solution of trichloroacetyl chloride (0.037 mol) in dichloro-
methane (80 cm3) at 0 ЊC was added dropwise benzylamine
(0.034 mol). Triethylamine was then added dissolved in
dichloromethane (20 cm3) and the reaction left for two hours at
0 ЊC. At this temperature, dilute HCl was added (2 M, 50 cm3)
and the organic phase was washed with saturated brine and
water, dried over MgSO4 and evaporated in vacuo to give the
crude amide. Chromatography using petroleum ether–ethyl
acetate (80:20) furnished the purified compounds.
4-Bromomethyl-3-isopropyl-1-(4-tolylsulfonyl)pyrrolidin-2-
one 13d. Yield 95%, as an inseparable mixture of diastereomers
(ratio 87:13) (Found: C, 48.3; H, 5.4; N, 3.75. C15H20BrNO3S
requires C, 48.1; H, 5.4; N, 3.7%); νmax (CHCl3)/cmϪ1 for mix-
ture 2964, 1734, 1367, 1172 and 917; δH(250 MHz; CDCl3) 7.88
(2 H major ϩ 2 H minor, d, J 8.3, Ar), 7.31 (2 H major ϩ 2 H
minor, d, J 8.3, Ar), 3.98 (1 H major, dd, J 10.5 and 8.5,
CHHN), 3.62 (1 H major, dd, J 10.5 and 5.6, CHHN), 3.44
(1 H major, dd, J 10.2 and 4.9, CHHBr), 3.31 (1 H major, dd,
J 10.2 and 7.4, CHHBr), 2.78 (1 H minor, m, CHCH2Br), 2.54
(1 H major, m, CHCH2Br), 2.42 (3 H major, s, Me), 2.26 (1 H
major, dd, J 6.0 and 4.2, CH), 2.10 (1 H major, m, CHMe2),
1.84 (1 H minor, m, CHMe2), 0.99 (3 H minor, d, J 6.3, Me),
0.95 (3 H minor, d, J 6.6, Me), 0.92 (3 H major, d, J 6.7, Me)
and 0.8 (3 H major, d, J 4.5, Me); m/z (CI) 374.0422 (Mϩ ϩ H,
C15H21Br79NO3S requires 374.0425), 374 (11%, Mϩ), 296 (27),
279 (32) and 142 (10).
N-Benzyl-N-but-3-enyltrichloroacetamide 25a. Yield 41%, as
a clear oil (Found: C, 50.8; H, 4.6; N, 4.5; Cl, 34.2. C13H14-
Cl3NO requires C, 50.9; H, 4.6; N, 4.6; Cl, 34.7%); νmax (Nujol)/
cmϪ1 1678; δH(250 MHz; CDCl3) 7.29 (5 H, m, Ar), 5.88 (1 H,
m, CH᎐CH ), 4.98 (2 H, m, CH᎐CH ), 4.77 (2 H, s, CH Ph),
᎐
᎐
2
2
2
3.65 (2 H, m, CH2) and 2.41 (2 H, m, CH2); δC(62.9 MHz;
CDCl3) 160.4 (s), 137.3 (s), 134.5 (d), 128.8 (2 × d), 127.9
(2 × d), 126.8 (d), 115.5 (t), 92.3 (s), 53.4 (t), 47.7 (t) and 31.6
(t); m/z (EI) 305 (4%, Mϩ), 235 (12), 203 (100), 160 (27) and 90
(23).
4-Chloromethyl-3-phenyl-1-(4-tolylsulfonyl)pyrrolidin-2-one
13e. Yield 86%, as a clear oil (Found: C, 59.3; H, 4.9; N, 3.6.
C18H18ClNO3S requires C, 59.4; H, 5.0; N, 3.85%); νmax
(CHCl3)/cmϪ1 for mixture 2923, 1737, 1596, 1453, 1361, 1171,
1088 and 963; δH(250 MHz; CDCl3) 7.96 (2 H, d, J 8.5, Ar),
7.34–7.24 (5 H, m, Ar), 7.06 (2 H, d, J 8.5, Ar), 4.20 (1 H, dd,
J 10.2 and 7.7, CHHN), 3.71 (1 H, dd, J 9.8 and 8.7, CHHN),
3.64 (1 H, dd, J 11.6 and 3.5, CHHCl), 3.58 (1 H, d, J 10.5,
N-Benzyl-N-(2-methylbut-3-enyl)trichloroacetamide
25b.
Yield 76% (Found: C, 52.9; H, 5.0; N, 4.35; Cl, 32.8. C14H16-
Cl3NO requires C, 52.4; H, 5.0; N, 4.4; Cl, 33.2%); νmax
(CH2Cl2)/cmϪ1 1678; δH(200 MHz; CDCl3) 7.26 (5 H, m, Ar),
5.63 (1 H, m, CH᎐CH ), 4.68–5.04 (4 H, m, CH᎐CH and
᎐
᎐
2
2
CH2Ph), 3.22 (2 H, m, CH2), 2.64 (1 H, m, CH) and 0.92 (3H,
d, J 7.0, Me); δC(50.3 MHz; CDCl3) 162.5 (s), 140.5 (s), 135.1
678
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680

View Article Online
(d), 128.4 (2 × d), 127.4 (2 × d), 126.6 (d), 115.2 (t), 93.5 (s),
(s), 128.7 (2 × d), 128.8 (2 × d), 127.7 (2 × d), 127.6 (2 × d),
53.4 (t), 52.7 (t), 35.8 (d) and 17.3 (q).
127.0 (d), 127.1 (d), 86.5 (s), 86.3 (s), 52.6 (d), 52.5 (d), 52.2 (d),
52.0 (d), 44.1 (t), 44.0 (t), 40.6 (t), 40.4 (t), 22.9 (t), 22.8 (t), 15.2
(q) and 14.8 (q); m/z (EI) for mixture 319 (100%, Mϩ), 284 (41),
249 (100), 214 (30) and 105 (88).
(S)-(N-ꢁ-Methylbenzyl)-N-but-3-enyltrichloroacetamide 25c.
Yield 67%, [α]D22 Ϫ20 (Found: C, 52.1; H, 4.9; N, 4.35; Cl, 33.0.
C10H16Cl3NO requires C, 52.4; H, 5.0; N, 4.4; Cl, 33.2%); νmax
(CH2Cl2)/cmϪ1 1673; δH(200 MHz; CDCl3) 7.23 (5 H, m, Ar),
N-tert-Butyl-3,3-dichloro-4-chloromethylpiperidin-2-one 27d.
Yield 99%, as a clear oil (Found: C, 44.05; H, 5.9; N, 5.05; Cl,
39.2. C10H16Cl3NO requires C, 44.1; H, 5.9; N, 5.1; Cl, 39.0%);
νmax (CH2Cl2)/cmϪ1 1674, δH(250 MHz; CDCl3) 4.14 (1 H, dd,
J 11.2 and 4.9, CHHCl), 3.52 (1 H, m, CHHN), 3.52 (1 H,
dd, J 11.2 and 10.1, CHHCl), 3.27 (1 H, m, CHHN), 2.63 (1 H,
m, CH), 2.35 (1 H, m, CHH), 1.75–1.95 (1 H, m, CHH) and
1.43 (9H, s, t-Bu); δC(62.9 MHz; CDCl3) 163.4 (s), 87.2 (s), 59.4
(s), 52.2 (d), 44.5 (t), 43.4 (t), 27.9 (3 × q) and 23.8 (t).
5.73 (1 H, m, CH᎐CH ), 5.48 (1 H, m, CH), 4.82 (2 H, m,
᎐
2
CH᎐CH ), 3.25 (1 H, m, NCHH), 2.74 (1 H, m, NCHH), 2.22
᎐
2
(2 H, m, CH2) and 1.60 (3H, d, J 7.0, Me); δC(50.3 MHz;
CDCl3) 161.2 (s), 140.0 (s), 135.7 (d), 129.6 (2 × d), 128.7
(2 × d), 127.9 (d), 117.7 (t), 95.3 (s), 57.6 (d), 47.2 (t), 33.1 (t)
and 18.7 (q); m/z (EI) 319 (4%, Mϩ), 284 (97), 214 (100) and 104
(10).
N-tert-Butyl-N-but-3-enyltrichloroacetamide 25d. Yield 62%,
as a clear oil (Found: C, 44.12; H, 6.0; N, 5.15; Cl, 38.6.
C10H16Cl3NO requires C, 44.0; H, 5.9; N, 5.5; Cl, 39.0%); νmax
(CH2Cl2)/cmϪ1 1684; δH(250 MHz; CDCl3) 5.42–5.25 (1 H, m,
References
1 J. Iqbal, B. Bhatia and N. K. Nayyar, Chem Rev., 1994, 94, 519.
2 (a) F. O. H. Pirrung, H. Hiemstra and W. N. Speckamp, Tetrahedron,
1994, 50, 12415; (b) M. A. Rachita and G. A. Slough, Tetrahedron
Lett., 1993, 43, 6821; (c) G. A. Slough, Tetrahedron Lett., 1993, 43,
6825; (d) T. K. Hayes, R. Villani and S. M. Weinreb, J. Am. Chem.
Soc., 1988, 110, 5533; (e) J. C. Phelps, D. E. Bergbreiter, G. M. Lee,
R. Villani and S. M. Weinreb, Tetrahedron Lett., 1989, 30, 3915;
( f ) G. M. Lee and S. M. Weinreb, J. Org. Chem., 1990, 55, 1281;
(g) H. Nagashima, H. Wakamatsu, N. Ozaki, M. Ishii, M.
Watanabe, T. Tajima and K. Itoh, J. Org. Chem., 1992, 57, 1682;
(h) H. Nagashima, K. Ara, H. Wakamatsu and K. Itoh, J. Chem.
Soc., Chem. Commun., 1985, 518.
CH᎐CH ), 4.80–4.60 (2 H, m, CH᎐CH ), 3.41 (2 H, m, CH ),
᎐
᎐
2
2
2
2.15 (2 H, m, CH2) and 1.23 (9H, s, t-Bu); δC(62.9 MHz; CDCl3)
161.2 (s), 133.2 (d), 116.9 (t), 95.1 (s), 59.8 (s), 45.7 (t), 35.4 (t)
and 27.8 (3 × q).
N-Benzyl-3,3-dichloro-4-chloromethylpiperidin-2-one
27a.
Yield 90%, as a clear oil (Found: C, 51.0; H, 4.5; N, 4.6; Cl,
34.4. C13H14Cl3NO requires C, 50.9; H, 4.6; N, 4.6; Cl, 34.7%);
νmax (CH2Cl2)/cmϪ1 1676; δH(250 MHz; CDCl3) 7.40–7.20 (5 H,
m, Ar), 4.80 (1 H, d, J 14.5, CHHPh), 4.45 (1 H, d, J 14.5,
CHHPh), 4.16 (1 H, dd, J 11.1 and 2.8, CHHCl), 3.58 (1 H, dd,
J 11.1 and 10.1, CHHCl), 3.33 (2 H, m, NCH2), 2.75 (1 H, m,
CH), 2.36 (1 H, ddd, J 14.2, 3.6 and 3.0, CHH) and 1.94 (1 H,
ddd, J 14.2, 12.3 and 9.0, CHH); δC(62.9 MHz; CDCl3) 165.7
(s), 134.5 (s), 129.1 (2 × d), 128.4 (2 × d), 128.2 (d), 84.3 (s),
51.6 (d), 47.9 (t), 47.3 (t) and 41.1 (t); m/z (EI) 305 (3%, Mϩ),
270 (3), 235 (14), 201 (100) and 90 (6).
3 G. M. Lee, M. Parvez and S. M. Weinreb, Tetrahedron, 1988, 44,
4671.
4 (a) J. H. Udding, K. C. J. M. Tuijp, M. N. A. Vanzanden,
H. Hiemstra and W. N. Speckamp, J. Org. Chem., 1994, 59, 1993;
(b) J. H. Udding, K. C. J. M. Tuijp, M. N. A. Vanzanden, H.
Hiemstra and W. N. Speckamp, Tetrahedron, 1994, 50, 1907;
(c) N. Baldovini, M.-P. Bertrand, A. Carrière, R. Nougier and J. M.
Plancher, J. Org. Chem., 1996, 61, 3205; (d) S. Iwamatsu, K.
Matsubara and H. Nagashima, J. Org. Chem., 1999, 64, 9625.
5 (a) L. Forti, F. Ghelfi and U. M. Pagnoni, Tetrahedron Lett., 1996,
37, 2077; (b) L. Forti, F. Ghelfi, E. Libertini, U. M. Pagnoni and
E. Soragni, Tetrahedron, 1997, 53, 17761; (c) F. Ghelfi, F. Bellesia,
L. Forti, G. Ghirardini, R. Grandi, E. Libertini, M. C. Monte-
maggi, U. M. Pagnoni, A. Pinetti, L. DeBuyck and A. F. Parsons,
Tetrahedron, 1999, 55, 1687.
6 F. De Campo, D Lastécouères and J.-B. Verlhac, Chem. Commun.,
1998, 2117.
7 H. Nagashima, N. Ozaki, M. Ishii, K. Seki, M. Washiyama and
K. Itoh, J. Org. Chem., 1993, 58, 464.
8 S.-I. Iwamatsu, H. Kondo, K. Matsubara and H. Nagashima,
Tetrahedron, 1999, 55, 1687.
9 (a) A. J. Clark, D. J. Duncalf, R. P. Filik, D. M. Haddleton, G. H.
Thomas and H. Wongtap, Tetrahedron Lett., 1999, 40, 3807; (b) A. J.
Clark, R. P. Filik and G. H. Thomas, Tetrahedron Lett., 1999, 40,
4885; (c) A. J. Clark, R. P. Filik, D. M. Haddleton, A. Radique,
C. J. Sanders, G. H. Thomas and M. E. Smith, J. Org. Chem., 1999,
64, 8954.
10 (a) D. M. Haddleton, D. J. Duncalf, A. J. Clark, M. C. Crossman
and D. Kukulj, New J. Chem., 1998, 315; (b) D. M. Haddleton, D. J.
Duncalf, D. Kukulj, M. C. Crossman, S. G. Jackson, S. A. F. Bon,
A. J. Clark and A. J. Shooter, Eur. J. Inorg. Chem., 1998, 1799;
(c) A. J. Clark, M. C. Crossman, D. J. Duncalf, D. M. Haddleton,
S. R. Morsley and A. J. Shooter, Chem Commun., 1997, 1734;
(d) D. M. Haddleton, A. J. Clark, D. J. Duncalf, A. M. Heming,
D. Kukulj and A. J. Shooter, J. Chem. Soc., Dalton Trans., 1998,
381; (e) D. M. Haddleton, D. J. Duncalf, D. Kukulj, A. J. Shooter
and A. J. Clark, J. Mater. Chem., 1998, 1525.
11 (a) K. D. Karlin and J. Zubieta, Copper Coordination Chemistry;
Biochemical and Inorganic Perspectives, Adenine Press, 1983; (b)
P. V. Bernhardt, J. Am. Chem. Soc., 1997, 119, 771.
12 (a) J. Xia and K. Matyjaszewski, Macromolecules, 1997, 30, 7697;
(b) J. Xia, S. G. Gaynor and K. Matyjaszewski, Macromolecules,
1998, 31, 5958.
13 M. Ciampolini and N. Nardi, Inorg. Chem., 1966, 5, 41.
14 S. E. Denmark and L. R. Marcin, J. Org. Chem., 1993, 58, 3857.
15 All calculations used the PM3 Hamiltonian as implemented in
MOPAC version 6.0 (JJP Stewart, MOPAC (QCPE 455)).
Approximate transition states were located by SADDLE
N-Benzyl-3,3-dichloro-4-chloromethyl-5-methylpiperidin-2-
one 27b. Yield 88% for mixture (Found: C, 52.85; H, 5.0; N, 4.4;
Cl, 32.8. C14H16Cl3NO requires C, 52.4; H, 5.0; N, 4.4; Cl,
33.2%); νmax (CH2Cl2)/cmϪ1 1677; data for trans compound,
δH(250 MHz; CDCl3) 7.40–7.20 (5 H, m, Ar), 4.82 (1 H, d,
J 14.5, CHHPh), 4.44 (1 H, d, J 14.5, CHHPh), 4.16 (1 H, dd,
J 12.1 and 1.0, CHHCl), 3.69 (1 H, dd, J 12.1 and 5.9,
CHHCl), 3.23 (1 H, dd, J 12.6 and 5.4, NCHH), 3.02 (1 H, dd,
J 12.6 and 10.6, NCHH), 2.56 (1 H, ddd, J 11.1, 6.2 and 1.7,
CHCH2Cl), 2.33 (1 H, m, CHMe) and 1.18 (3 H, d, J 6.9, Me);
δC(62.9 MHz; CDCl3) 163.9 (s), 135.9 (s), 128.9 (2 × d), 128.2
(2 × d), 128.1 (d), 87.6 (s), 57.9 (d), 52.8 (t), 51.5 (t), 43.2 (t), 30.8
(d) and 16.4 (q): data for cis compound, δH(250 MHz; CDCl3)
7.40–7.20 (5 H, m, Ar), 4.73 (1 H, d, J 14.5, CHHPh), 4.57 (1 H,
d, J 14.5, CHHPh), 4.05 (1 H, dd, J 11.6 and 3.1, CHHCl), 3.91
(1 H, dd, J 11.6 and 8.4, CHHCl), 3.45 (1 H, dd, J 12.6 and 5.1,
NCHH), 3.19 (1 H, dd, J 12.6 and 5.2, NCHH), 2.93 (1 H, m,
CHCH2Cl), 2.33 (1 H, m, CHMe) and 1.06 (3 H, d, J 7.4, Me);
δC(62.9 MHz; CDCl3) 163.8 (s), 135.7 (s), 128.4 (2 × d), 128.0
(2 × d), 127.9 (d), 87.3 (s), 56.2 (d), 53.8 (t), 52.8 (t), 41.9 (t), 26.7
(d) and 13.2 (q); m/z (EI) for mixture 319 (64%, Mϩ), 284 (35),
249 (100), 214 (22) and 91 (18).
N-(1-ꢁ-Methylbenzyl)-3,3-dichloro-4-chloromethylpiperidin-2-
one 27c. Yield 96%, as a clear oil, for mixture (Found: C, 53.0;
H, 5.1; N, 4.3. C14H16Cl3NO requires C, 52.4; H, 5.0; N, 4.4%);
νmax (CH2Cl2)/cmϪ1 1667; δH(250 MHz; CDCl3) 7.40 (5 H, m,
Ar), 6.03 (1 H, m, CHMe), 4.20 (0.5 H, m, CHHCl), 4.14 (0.5
H, m, CHHCl), 3.58 (1 H, m, CHHCl), 3.16–3.28 (1 H, m,
CHHN), 3.08 (0.5 H, m, CHHN), 2.80–2.60 (1 H, m, CH), 2.68
(0.5 H, m, CHHN), 2.24–2.41 (1.5 H, m, CHH), 2.24–2.41 (0.5
H, m, CHH), 1.56 (1.5 H, d, J 8.3, Me) and 1.60 (1.5 H, d, J 8.3,
Me); δC(62.9 MHz; CDCl3) 165.4 (s), 165.2 (s), 138.4 (s), 138.6
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680
679

View Article Online
calculations, refined using the TS option and verified with FORCE
calculations which yielded single negative frequencies of Ϫ821 and
Ϫ850 cmϪ1 for trans and cis pathways respectively.
21 K. Tsuzuki, T. Yokozuka, M. Murata, H. Tanaka and S. Omura,
J. Antibiot., 1989, 42, 727.
22 (a) T. Takasi and T. Yamane, JP 09071532/1997 (Chem. Abstr., 127,
9111); (b) S. M. Sowmimino-Coker, N. Yerram, R. P. Goodrich,
M. Platz and S. C. Park, WO 9639816/1997 (Chem. Abstr., 126,
135581).
16 A. J. Clark, R. J. Deeth, C. J. Samuel and H. Wongtap, Synlett, 1999,
444.
17 (a) J. J. Kohler and W. N. Speckamp, Tetrahedron Lett., 1977, 631;
(b) J. J. Kohler and W. N. Speckamp, Tetrahedron Lett., 1977, 635;
(c) J. J. Kohler and W. N. Speckamp, J. Chem. Soc., Chem. Commun.,
1978, 166; (d) J. J. Kohler and W. N. Speckamp, J. Chem. Soc.,
Chem. Commun., 1980, 142.
23 H. Nishiyama, S. Yamaguchi, S.-B. Park and K. Itoh, Tetrahedon:
Asymmetry, 1993, 4, 143.
24 D. A. Evans, K. A. Woerpel, M. M. Hinman and M. M. Faul,
J. Am. Chem. Soc., 1991, 113, 726.
18 (a) M. L. E. N. da Mata, W. B. Motherwell and F. Ujjainwalla,
Tetrahedron Lett., 1997, 38, 137; (b) M. L. E. N. da Mata, W. B.
Motherwell and F. Ujjainwalla, Tetrahedron Lett., 1997, 38, 141.
19 H. Urashima, Y. Takeji, H. Shinohara and S. Fujisawa, WO
9713515/1997 (Chem. Abstr., 126, 334426).
25 S. C. Benson, P. Cai, M. Colon, M. A. Haiza, M. Tokles and
J. K. Snyder, J. Org. Chem., 1988, 53, 5335.
26 S. Ozaki, H. Matsushita and H. Ohmori, J. Chem. Soc., Perkin
Trans. 1, 1993, 2339.
20 S. Kokura, T. Yoshikawa, Y. Naito, H. Ichikawa, H. Takano,
S. Takahashi, T. Tomii, N. Yoshida and M. Kondo, Dig. Dis. Sci.,
1997, 42, 2566.
Paper a909666c
680
J. Chem. Soc., Perkin Trans. 1, 2000, 671–680