Electrophile-induced ring expansions of β-lactams toward γ-lactams

Article abstract of DOI:10.1021/jo0509556

An efficient and straightforward route toward 3,4-cis-4- isopropenylazetidin-2-ones was developed from 4-(1-chloroalkyl)azetidin-2-ones. Starting from the latter β-lactams, a new synthesis of pyrrolidin-2-ones was achieved. When 4-isopropenylazetidin-2-ones were treated with bromine in dichloromethane, diastereoselective electrophile-induced ring expansions toward 5-bromopyrrolidin-2-ones were performed. Further oxidation of 3-benzyloxypyrrolidin-2-ones with bromine toward 3-bromopyrrolidin-2-ones was also established. When 4-isopropenyl-β-lactams were added to a mixture of NBS and TMSN₃, 5-azidopyrrolidin-2-ones were obtained in moderate to high yields.

Full text of DOI:10.1021/jo0509556

Electrophile-Induced Ring Expansions of â-Lactams toward
γ-Lactams
Willem Van Brabandt and Norbert De Kimpe*
Department of Organic Chemistry, Faculty of Bioscience Engineering, Ghent University,
Coupure links 653, B-9000 Ghent, Belgium
norbert.dekimpe@ugent.be
Received May 12, 2005
An efficient and straightforward route toward 3,4-cis-4-isopropenylazetidin-2-ones was developed
from 4-(1-chloroalkyl)azetidin-2-ones. Starting from the latter â-lactams, a new synthesis of
pyrrolidin-2-ones was achieved. When 4-isopropenylazetidin-2-ones were treated with bromine in
dichloromethane, diastereoselective electrophile-induced ring expansions toward 5-bromopyrrolidin-
2
3
-ones were performed. Further oxidation of 3-benzyloxypyrrolidin-2-ones with bromine toward
-bromopyrrolidin-2-ones was also established. When 4-isopropenyl-â-lactams were added to a
mixture of NBS and TMSN
3
, 5-azidopyrrolidin-2-ones were obtained in moderate to high yields.
Introduction
SCHEME 1
Recently, diastereoselective ring expansions of â-lac-
tams toward γ-lactams via N-acyliminium intermediates
have been described.1 During these reactions, 4-(1-
bromoalkyl)-2-azetidinones 1 were converted into the
corresponding carbenium ions 2 via dissociation of the
halide, and a subsequent ring transformation converted
these electron-deficient species into N-acyliminium ions
-3
use of the diastereoselective ring expansions of â-lactams
toward γ-lactams.
3
, which are susceptible to attack of oxygen, nitrogen,
and carbon nucleophiles. In this way, a variety of
-hydroxy-, 5-alkoxy-, 5-allylamino-, 5-azido-, and 5-cy-
Results and Discussion
5
ano-2-pyrrolidinones 4 were synthesized (Scheme 1). It
4
-(1-Chloroalkyl)azetidin-2-ones 6 were synthesized by
was found that dehydrobromination of 4-(1-bromoalkyl)-
cyclocondensation reactions of R-chloroisobutylidene-
2
-azetidinones 1 constituted an important side reaction
amines 54 with various ketenes, prepared in situ by
competing with the ring enlargements.
dehydrohalogenation of the corresponding acetyl chlo-
rides with triethylamine in benzene. This method is
These new developments in the chemistry of 2-azeti-
dinones encouraged us to investigate whether these ring
expansions could be induced by attack of electrophiles
onto 4-isopropenylazetidin-2-ones. In this way, a new
pathway via N-acyliminium intermediates can give rise
to the formation of a wide variety of γ-lactams, without
the occurrence of side reactions (e.g., dehydrohalogena-
tion reactions of 4-(1-bromoalkyl)-2-azetidinones 1), which
would be a great improvement concerning the synthetic
generally known as the Staudinger reaction (Scheme
5-7
2
).
It was found that the obtained 3,4-cis-4-(1-chloro-
alkyl)azetidin-2-ones 6 underwent a clean dehydrochlo-
rination at 160 °C in dry DMSO for 4 h, affording 3,4-
cis-4-isopropenylazetidin-2-ones 7 in 62-82% yields.
Working under dry conditions while performing the
dehydrohalogenation step turned out to be important. In
(
4) (a) De Kimpe, N.; Verh e´ , R.; De Buyck, L.; Schamp, N. Can. J.
(
1) Van Brabandt, W.; De Kimpe, N. J. Org. Chem. 2005, 70, 3369-
Chem. 1984, 62, 1812-1816. (b) De Kimpe, N.; Stanoeva, E.; Verh e´ ,
R.; Schamp, N. Synthesis 1988, 587-591.
3
3
4
374.
(
2) Speckamp, W. N.; Moolenaar, M. J. Tetrahedron 2000, 56, 3817-
(5) Palomo, P.; Aizpura, J. M.; Ganboa, I.; Oiarbide, M. Eur. J. Org.
Chem. 1999, 3223-3235.
(6) Singh, G. S. Tetrahedron 2003, 59, 7631-7649.
(7) Dejaegher, Y.; De Kimpe, N. J. Org. Chem. 2004, 69, 5974-5985.
856.
(
3) Speckamp, W. N.; Hiemstra, H. Tetrahedron 1985, 41, 4367-
416.
1
0.1021/jo0509556 CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/27/2005
J. Org. Chem. 2005, 70, 8717-8722
8717

Van Brabandt and De Kimpe
SCHEME 2
SCHEME 3
commercial DMSO that was not dry, ring expansions
toward 5-hydroxy-γ-lactams were observed as side reac-
tions. Performing the dehydrochlorination of 3,4-cis-4-
(1-chloroalkyl)azetidin-2-ones 6 with 1 equiv of sodium
azide suppressed these side reactions in commercial (not
dried) DMSO, probably because of the water-withdrawing
effects due to the solvation of this salt. Potassium tertiary
butoxide could not be used as a base because dehydro-
halogenation reactions were always accompanied by cis-
trans isomerization reactions of the azetidin-2-ones 7 (via
deprotonation-protonation at C3). Potassium carbonate
also could not be used because of the release of water
during the experiment via this reagent. Triethylamine,
dissolved in DMSO, did not enhance the dehydrochlori-
nation reactions and required high temperatures.
The cycloaddition of N-isobutenylideneamines 8 with
ketenes is proposed as an alternative pathway towards
the synthesis of 3,4-cis-4-isopropenylazetidin-2-ones. Al-
though these reactions could be performed in comparable
yields as the yields obtained after dehydrohalogenation
of 3,4-cis-4-(1-chloroalkyl)azetidin-2-ones 6 (Scheme 2),
some considerable drawbacks concerning this method
should be stressed. A first synthesis toward N-isobu-
tenylideneamines concerns the condensation of 2-meth-
ylacrylaldehyde with different amines. Because of the
high reactive terminal double bond in 2-methylacrylal-
halogenation when treated with base to give 2-azadiene
a (in 91% yield).8a Also, N-allyl R-chloroisobutylidene-
9
amine 5e suffered from 1,4-dehydrohalogenation when
treated with potassium t-butoxide in ether.8b Because of
this 1,4-dehydrohalogenation reaction, no N-benzylated
or N-allylated 3,4-cis-4-isopropenylazetidin-2-ones can be
synthesized starting from R-chloroisobutylideneamines.
The capability of electrophiles to induce ring expan-
sions of 4-isopropenylazetidin-2-ones 7 toward pyrrolidin-
2
1
-ones was investigated by reacting the â-lactams 7 with
equiv of bromine in dichloromethane at 0 °C. Im-
mediately after adding bromine, the solvent was evapo-
rated and 5-bromopyrrolidin-2-ones 10 were obtained
dehyde, these reactions toward R,â-unsaturated imines
(
Scheme 3). Due to the high reactivity of the resulting
4
8
proceed in low yields and show many side products. A
compounds 10, no further purification was possible, as
they led to decomposition products. The stereochemical
second synthesis toward N-isobutenylideneamines 8 is
shown in Scheme 2. 1,2-Dehydrohalogention of R-chloro-
isobutylideneamines proceeds in low yields when volatile
imines (e.g., 8a) are obtained. Furthermore, N-benzyl
R-chloroisobutylideneamine 5c undergoes 1,4-dehydro-
(
8) (a) De Kimpe, N.; Nagy, M.; Boeykens, M.; Van der Scheuren,
D. J. Org. Chem. 1992, 57, 5761-5764. (b) De Kimpe, N.; Yao, Z.;
Boeykens, M.; Nagy, M. Tetrahedron Lett. 1990, 19, 2771-2774.
8718 J. Org. Chem., Vol. 70, No. 22, 2005

Electrophile-Induced Ring Expansions
SCHEME 4
FIGURE 1. nOe effects observed in compound 10b.
course of the reaction is highly selective, as only com-
pounds with a 3,4-trans and 4,5-cis relationship were
obtained.
The mechanism of this ring expansion of â-lactams 7
can be explained by an initial activation of the carbon-
carbon double bond of the 4-isopropenyl group, when
treated with bromine (Scheme 4). A subsequent ring
expansion converts compounds 11 into N-acyliminium
ions 12, which are susceptible to attack of bromide,
affording the corresponding 5-bromopyrrolidin-2-ones 10.
The stereochemistry of the 5-bromopyrrolidin-2-ones
FIGURE 2. Reaction of 4-isopropenylazetidin-2-ones 7 with
bromine.
trans oriented, according to the δ-values around 3.9 ppm
of the C3 protons of these compounds (CDCl ). These
3
findings, in combination with the performed nOe experi-
ments, shown in Figure 1, proved the exact stereochem-
istry of pyrrolidin-2-ones 10 as shown in Scheme 4.
To explain the high diastereoselectivity observed in the
title reaction, the following proposition can be made. On
the basis of allylic strain concepts,12 it can be predicted
that the double bond of 4-isopropenylazetidin-2-ones 7
will be syn or eclipsed with the C4-H bond, as shown in
Figure 2. Attack of bromine at this double bond will occur
via the B-plane, because in this way a better orbital
overlap toward the ring expansion is possible. Re-
arrangement via this mechanism gives rise to the forma-
tion of N-acyliminium ion 12. These results are in
accordance with literature data where the same relative
C3,C4 stereochemistry was obtained after an intramo-
lecular ring expansion of 3-methoxy-4-(2-methyl-2-oxira-
nyl)azetidin-2-ones toward γ-lactams.13 In the second step
of the reaction, the outcome can be explained by external
ion pair return concepts. After ring expansion, the
bromide will attack from the same side where the
bromomethyl group is formed, because it is at this side
that the bromide is formed after the ring expansion. The
same concepts concerning external ion pair return were
1
0 was determined based on both nOe experiments
(
Figure 1) and the δ-values of the protons at C3. The
former technique indicated clearly the cis relationship
between the proton at C3 and the bromomethyl group,
and between the methyl group at C4 and the proton at
C5, as nOe effects of more than 2% were observed
between these groups in both cases.
The latter attribution technique was based on the
observation that δ-values of protons at C3 of 3-alkoxy-
4
,4-dimethylpyrrolidin-2-ones are mainly influenced by
the cis-trans stereochemistry of the substituents at C3
and C5. The 3,5-cis derivatives showed a δ-value for the
1
proton at C3 around 3.3 ppm (CDCl
in 3,5-trans isomers appeared around 3.9 ppm (CDCl
3
), while this proton
).
3
To compare these data with the present new results,
γ-lactam 10a (obtained as the reaction mixture) was
poured into water and extracted with dichloromethane,
affording a mixture of 3,5-cis- and 3,5-trans-5-hydroxy-
pyrrolidin-2-ones (3S*,4R*,5R*)-13 and (3S*,4R*,5S*)-
1
3 in a 1:4 ratio, respectively (Scheme 5). One of the two
stereoisomers showed a W-coupling of 1.1 Hz between
the two protons of the γ-lactam ring. Since this long range
coupling exists only when the bonds between the two
protons form a W-pattern in the same plane, this
W-coupling can only appear in the 3,5-cis form
already proven to be responsible for the stereochemical
control of “memory effects”.1
4-22
This term is used in
reactions of carbocations in which the latter “remember”
how they were formed before they go on to give the second
step. Also in these cases, counterions attacked the
rearranged carbocations from the side from which they
were expelled (during ionization).
When 4-isopropenylazetidin-2-ones 7b and 7d were
used to perform electrophile-induced ring expansions,
addition of 5 equiv of bromine instead of 1 equiv gave
(
3
3S*,4R*,5R*)-13.9-11 It was found that C3 protons of
,5-cis- and 3,5-trans-3-alkoxy-4-bromomethyl-4-meth-
ylpyrrolidin-2-ones (3S*,4R*,5R*)-13 and (3S*,4R*,5S*)-
1
3 showed equal δ-values as C3 protons of 3,5-cis- and
3
,5-trans-3-alkoxy-4,4-dimethylpyrrolidin-2-ones, respec-
tively. In this way, it was proven that the C3 and C5
substituents of the 5-bromopyrrolidin-2-ones 10 were
(
12) (a) Ohmori, Y.; Yamashita, A.; Tsujita, R.; Yamamoto, T.;
(
9) Viladomat, F.; Codina, C.; Bastida, J.; Mathee, S.; Campbell, W.
Phytochemistry 1995, 40, 961-965.
10) Kam, T.-S.; Choo, Y.-M.; Chen, W.; Yao, J.-X. Phytochemistry
999, 52, 959-963.
11) Machocho, A.; Chhabra, S. C.; Viladomat, F.; Codina, C.;
Bastida, J. Phytochemistry 1999, 51, 1185-1191.
Taniuchi, K.; Matsuda, A.; Shuto, S. J. Med. Chem. 2004, 46, 5326-
5333 and references therein. (b) Dorigo, A. E.; Pratt, D. W.; Houk, K.
N. J. Am. Chem. Soc. 1987, 109, 6591-6600. (c) Kenneth, B. W.;
Martin, E. J. Am. Chem. Soc. 1985, 107, 5035-5041.
(
1
(
(13) Anaya, J.; Fernandez-Mateos, A.; Grande, M.; Martianez, J.;
Ruano, G.; Rubio-Gonzalez, R. Tetrahedron 2003, 59, 241-248.
J. Org. Chem, Vol. 70, No. 22, 2005 8719

Van Brabandt and De Kimpe
SCHEME 5
SCHEME 6
SCHEME 7
unexpected results when the reaction mixture was stirred
for 4 h at 0 °C. After the ring expansion, a further
reaction toward 3-bromopyrrolidin-2-ones 16 was ob-
served, together with the formation of benzaldehyde. The
mechanism of this reaction can be explained by attack
of oxygen at bromine, followed by a dehydrobromination
of the resulting oxonium ion 14 and a second-order
nucleophilic substitution of “benzaldehyde” by bromide
2
3-25
(Scheme 6).
In a second approach toward electrophile-induced ring
expansions, 4-isopropenylazetidin-2-ones 7 were added
to a mixture of 1.3 equiv of N-bromosuccinimide (NBS),
compounds gave 5-azidopyrrolidin-2-ones in moderate to
high yields (Scheme 7). The stereochemistry of the
resulting γ-lactams 18 was determined in the same way
as described for the 5-bromopyrrolidin-2-ones 10.
When the same reactions were performed with tri-
methylsilyl cyanide instead of azidotrimethylsilane or
with N-iodosuccinimide (NIS) instead of NBS, starting
materials 7 were recovered.
1
3
.3 equiv of azidotrimethylsilane (TMSN ) and dichlo-
romethane and nitromethane in a ratio of 3:1 by volume
at room temperature.26 Azidobromination of the starting
(
14) Berson, J. A.; Gajewski, J. J.; Donald, D. S. J. Am. Chem. Soc.
969, 91, 5550-5566.
15) Berson, J. A.; Poonian, M. S.; Libbey, W. J. J. Am. Chem. Soc.
969, 91, 5567-5579.
16) Berson, J. A.; Donald, D. S.; Libbey, W. J. J. Am. Chem. Soc.
969, 91, 5580-5593.
17) Berson, J. A.; Wege, D.; Clarke, G. M.; Bergman, R. G. J. Am.
Chem. Soc. 1969, 91, 5594-5601.
18) Berson, J. A.; Wege, D.; Clarke, G. M.; Bergman, R. G. J. Am.
Chem. Soc. 1969, 91, 5601-5613.
19) Berson, J. A.; Luibrand, R. T.; Kundu, N. G.; Morris, D. G. J.
Am. Chem. Soc. 1971, 93, 3075-3077.
20) Collins, J. A.; Glover, I. T.; Eckart, M. D.; Raaen, V. F.;
1
1
1
(
(
(
Conclusions
(
Dehydrochlorination of 3,4-cis-4-(1-chloroalkyl)azeti-
din-2-ones 6 constitutes an efficient and straightforward
pathway toward 3,4-cis-4-isopropenylazetidin-2-ones 7.
Starting from the latter â-lactams, a new synthesis of
pyrrolidin-2-ones was achieved. When 4-isopropenylaze-
tidin-2-ones 7 were treated with bromine in dichlo-
romethane, diastereoselective electrophile-induced ring
expansions toward 5-bromopyrrolidin-2-ones 10 were
performed. Further oxidation of 3-benzyloxypyrrolidin-
2-ones toward 3-bromopyrrolidin-2-ones 16 was also
established. When 4-isopropenyl â-lactams 7 were added
(
(
Benjamin, B. M.; Benjaminov, B. S. J. Am. Chem. Soc. 1972, 94, 899-
9
08.
(
21) Collins, C. J. Chem. Soc. Rev. 1975, 4, 251-262.
22) Kirmse, W.; G u¨ nther, B. J. Am. Chem. Soc. 1978, 100, 3619-
(
3
3
3
3
4
620.
(
23) Deno, N. C.; Potter, N. H. J. Am. Chem. Soc. 1967, 89, 3550-
554.
(
24) Deno, N. C.; Potter, N. H. J. Am. Chem. Soc. 1967, 89, 3555-
556.
(
25) Deno, N. C.; Fruit, R. E. Jr. J. Am. Chem. Soc. 1968, 90, 3502-
506.
3
to a mixture of NBS and TMSN , 5-azidopyrrolidin-2-ones
(
26) Olah, G. A.; Wang, Q.; Li, X.; Surya Prakash, G. K. Synlett 1990,
87-489.
18 were obtained in moderate to high yields. These new
8720 J. Org. Chem., Vol. 70, No. 22, 2005

Electrophile-Induced Ring Expansions
findings in the chemistry of 2-azetidinones emphasize the
synthetic potential of diastereoselective ring expansions
of â-lactams toward γ-lactams.
(3S*,4R*,5S*)-5-Bromo-4-(bromomethyl)-1-tert-butyl-3-
1
methoxy-4-methylpyrrolidin-2-one 10a: Purity >98% ( H
1
NMR). H NMR (300 MHz, CDCl
3
): δ 1.25 (3H, s), 1.50 (9H,
s), 3.64 (1H, d, J ) 10.3 Hz), 3.70 (3H, s), 3.81 (1H, s), 4.03
1
3
(
3
1H, d, J ) 10.3 Hz), 5.83 (1H, s). C NMR (75 MHz, CDCl ):
1
-
Experimental Section
δ 17.1, 26.8, 40.5, 47.8, 56.3, 78.7, 83.2, 173.9. IR (NaCl, cm ):
CdO ) 1680; νmax ) 2977, 1458.
Synthesis of 5-Hydroxypyrrolidin-2-ones
3S*,4R*,5R*)-13 and (3S*,4R*,5S*)-13. 3,4-cis-4-Isopro-
ν
R-Chloroisobutylideneamines 5 were synthesized via a
4
.
3
known procedure.
(
1
. Synthesis of 3,4-cis-4-(1-Chloroalkyl)azetidin-2-ones
penyl-3-methoxy-1-tert-butylazetidin-2-one 10a (0.12 g, 0.5
mmol) was cooled at 0 °C in dichloromethane (10 mL). Under
continuous stirring, bromine (0.08 g, 0.5 mmol) was added.
Immediately after the addition of bromine, the mixture was
poured into water (50 mL), and the mixture was extracted two
6
. As a representative example, the synthesis of 3,4-cis-4-(1-
chloro-1-methylethyl)-1-tert-butyl-3-methoxy-azetidin-2-one 6a
is described. A solution of N-(2-chloro-2-methylpropylidene)-
-methyl-2-propylamine 5 (1.61 g, 10 mmol) and triethylamine
3.03 g, 30 mmol) in benzene (50 mL) was heated. A solution
2
(
4
times with dichloromethane (20 mL). After drying (MgSO ) of
of methoxyacetyl chloride (1.41 g, 13 mmol) in benzene (30
mL) was added dropwise to this refluxing solution. The
resulting solution was kept at reflux temperature for 30 min
and was subsequently stirred overnight at room temperature.
The reaction mixture was diluted with chloroform (70 mL) and
washed with a saturated sodium bicarbonate solution and
the combined organic fractions and evaporation of the sol-
vent in vacuo, a mixture of 19% 3,5-cis substituted 4-(bro-
momethyl)-5-hydroxy-1-tert-butyl-3-methoxy-4-methylpyrroli-
din-2-one (3S*,4R*,5R*)-13 and 81% 3,5-trans substituted
4-(bromomethyl)-5-hydroxy-1-tert-butyl-3-methoxy-4-methylpyr-
rolidin-2-one (3S*,4R*,5R*)-13 was obtained. NMR spectra
were obtained from the reaction mixture.
4
brine. After the solvent was dried (MgSO ) and evaporated,
the crude reaction product was obtained. Further purification
was performed by flash chromatography (1.75 g, 7.5 mmol).
For spectral data, see also ref 7.
(
3S*,4R*,5R*)-4-(Bromomethyl)-1-tert-butyl-5-hydroxy-
1
3
-methoxy-4-methylpyrrolidin-2-one (3S*,4R*,5R*)-13: H
3
NMR (300 MHz, CDCl ): δ 1.23 (3H, s); 1.47 (9H, s); 2.65 (1H,
(
3S*,4S*)-4-(1-Chloro-1-methylethyl)-1-tert-butyl-3-meth-
oxyazetidin-2-one 6a: White crystals, 75% yield, mp 77 °C,
TLC R
0.30 (petrol ether/ethyl acetate, 3:1). 1H NMR (300
MHz, CDCl
): δ 1.48 (9H, s), 1.70 and 1.71 (6H, 2 × s), 3.53
3H, s), 4.14 (1H, d, J ) 5.5 Hz), 4.33 (1H, d, J ) 5.5 Hz).
NMR (75 MHz, CDCl ): δ 27.2, 29.7, 28.7, 54.7, 59.7, 68.4,
d, J ) 12.4 Hz); 3.30 (1H, d, J ) 1.1 Hz); 3.56 (1H, d, J ) 10.5
Hz); 3.59 (3H, s); 3.84 (1H, d, J ) 10.5 Hz); 4.94 (1H, d × d, J
f
13
)
12.4 Hz, J ) 1.10 Hz). C NMR (75 MHz, CDCl
3
): δ 14.0,
3
-1
2
7.9, 38.8, 45.4, 54.5, 59.5, 84.3, 87.8, 172.1. IR (NaCl, cm ):
1
3
(
C
+
ν
CdO ) 1690; νOH ) 3335. MS (70 eV) m/z (%): 296/4 (M + 1,
3
1
00).
-
1
7
1.6, 82.0, 168.7. IR (NaCl, cm ): νCdO ) 1737, νmax ) 2987,
+
(3S*,4R*,5S*)-4-(Bromomethyl)-1-tert-butyl-5-hydroxy-
1
458. MS (70 eV) m/z (%): 236/4 (M + 1, 100); 180/78 (85).
20ClNO C, 56.52; H, 8.62; N, 5.99.
Found: C, 56.64; H, 8.76; N, 5.75.
. Synthesis of 3,4-cis-4-Isopropenylazetidin-2-ones 7.
1
3
-methoxy-4-methylpyrrolidin-2-one (3S*,4R*,5S*)-13: H
Anal. Calcd for C11
H
2
:
NMR (300 MHz, CDCl
3
): δ 1.10 (3H, s), 1.46 (9H, s), 2.60 (1H,
d, J ) 4.00 Hz), 3.55 (1H, d, J ) 9.5 Hz), 3.66 (3H, s), 3.84
2
13
(
1H, s), 3.84 (1H, d, J ) 9.5 Hz), 4.89 (1H, d, J ) 4.0 Hz).
NMR (75 MHz, CDCl ): δ 16.4, 27.8, 39.0, 46.4, 54.5, 60.5,
2.8, 85.5, 173.6. IR (NaCl, cm ): νCdO ) 1690; νOH ) 3335.
C
As a representative example, the synthesis of 3,4-cis-4-
isopropenyl-1-isopropyl-3-methoxyazetidin-2-one 7c is de-
scribed. 3,4-cis-4-(1-Chloro-1-methylethyl)-1-isopropyl-3-meth-
oxyazetidin-2-one 6c (2.20 g, 10 mmol) was heated at 160 °C
for 4 h in dry DMSO (30 mL). After cooling to room temper-
ature, the solution was poured into water (50 mL) and
extracted three times with ether (30 mL). The combined
organic fractions were washed three times with water (30 mL)
3
-
1
8
+
MS (70 eV) m/z (%): 296/4 (M + 1, 100).
. Synthesis of 5-Bromopyrrolidin-2-ones 16. As a
5
representative example, the synthesis of 3,5-dibromo-4-(bro-
momethyl)-1-tert-butyl-4-methylpyrrolidin-2-one 16a is de-
scribed. 3,4-cis-3-Benzyloxy-4-isopropenyl-1-tert-butylazetidin-
2
-one 7b (0.16 g, 0.5 mmol) was cooled at 0 °C in dichloro-
methane (10 mL). Under continuous stirring, bromine (0.4 g,
.5 mmol) was added. After 4 h at room temperature, the
4
and dried (MgSO ). After evaporation of the solvent in vacuo
the crude reaction mixture was obtained, which was further
purified by flash chromatography (1.36 g, 6.2 mmol). For
spectral data, see also ref 1.
2
solvent was evaporated, affording 3,5-dibromo-4-(bromo-
methyl)-1-tert-butyl-4-methylpyrrolidin-2-one 16a in 100%
yield with a purity of more than 90%. Mass spectra were taken,
but because of the use of aqueous solvents during this
technique, the corresponding mother ions of the 5-hydroxy-
pyrrolidin-2-ones were always detected.
(
3S*,4R*)-cis-4-Isopropenyl-1-isopropyl-3-methoxyaze-
tidin-2-one 7c: 62% yield, TLC R 0.15 (petrol ether/ethyl
acetate, 3:1). H NMR (300 MHz, CDCl ): δ 1.19 and 1.29 (6H,
× d, J ) 6.6 Hz), 1.84 (3H, d × d, J ) 0.8 Hz, J ) 1.4 Hz),
.44 (3H, s); 3.76 (1H, sept, J ) 6.6 Hz), 4.25 (1H, d, J ) 5.0
f
1
3
2
3
1
3
(3R*,4R*,5S*)-3,5-Dibromo-4-(bromomethyl)-1-tert-bu-
Hz), 4.48 (1H, d, J ) 5.0 Hz), 5.09-5.11 (2H, m). C NMR (75
MHz, CDCl ): δ 19.1, 20.1, 20.7, 45.0, 58.8, 62.2, 84.3, 116.2,
1
tyl-4-methylpyrrolidin-2-one 16a: Purity >90% ( H NMR).
3
-1
1
3
H NMR (300 MHz, CDCl ): δ 1.27 (3H, s), 1.51 (9H, s), 3.71
1
2
1
41.7, 166.9. IR (NaCl, cm ): νCdO ) 1756; νmax ) 3079, 2975,
934, 2835, 1454, 1385, 1369. MS (70 eV) m/z (%): 184 (M+
(1H, d, J ) 10.5 Hz), 3.98 (1H, d, J ) 10.5 Hz), 4.38 (1H, s),
+
13
, 85); 110 (100). Anal. Calcd for C10
H17NO
2
: C, 65.54; H, 9.35;
3
5.77 (1H, s). C NMR (75 MHz, CDCl ): δ 16.6, 26.8, 40.4,
-
1
N, 7.64. Found: C, 65.43; H, 9.31; N, 7.49.
. Synthesis of 5-Bromopyrrolidin-2-ones 10. As a
48.3, 56.8, 74.8, 80.3, 176.4. IR (NaCl, cm ): νCdO ) 1692;
ν
max ) 2978; 1453; 1370.
3
representative example, the synthesis of 5-bromo-4-(bromo-
methyl)-1-tert-butyl-3-methoxy-4-methylpyrrolidin-2-one 10a
is described. 3,4-cis-4-Isopropenyl-3-methoxy-1-tert-butylaze-
tidin-2-one 7a (0.12 g, 0.6 mmol) was cooled at 0 °C in
dichloromethane (10 mL). Under continuous stirring, bromine
6. Synthesis of 5-Azidopyrrolidin-2-ones 18. As a rep-
resentative example, the synthesis of 5-azido-4-(bromomethyl)-
1-tert-butyl-3-methoxy-4-methylpyrrolidin-2-one 18a is de-
scribed. In a mixture of dichloromethane and nitromethane
at a ratio of 3:1 (30 mL), NBS (1.16 g, 6.5 mmol) and TMSN
3
(
0.10 g, 0.6 mmol) was added. Immediately after the addition
of bromine the solvent was evaporated, obtaining 3-benzyloxy-
-bromo-4-(bromomethyl)-1-tert-butyl-4-methylpyrrolidin-2-
(0.64 g, 6.5 mmol) were added. After being stirred at room
temperature for 1 h, 3,4-cis-1-tert-butyl-4-isopropenyl-3-meth-
oxyazetidin-2-one 7a (1.59 g, 5 mmol) was added, and the
mixture was stirred overnight at room temperature. After
pouring in water (50 mL), the water was extracted twice with
dichloromethane (20 mL). The resulting organic fractions were
5
one 10a in 100% yield with a purity of more than 95%. Mass
spectra were taken, but because of the use of aqueous solvents
during this technique, the corresponding mother ions of the
5-hydroxypyrrolidin-2-ones were always detected.
dried (MgSO
4
), and after evaporation of the solvent, the crude
J. Org. Chem, Vol. 70, No. 22, 2005 8721

Van Brabandt and De Kimpe
reaction mixture was obtained. Further purification was
performed by flash chromatography (1.39 g, 4.4 mmol).
Acknowledgment. We are indebted to Ghent Uni-
versity (GOA) and the Fund for Scientific Research
(
3S*,4R*,5S*)-5-Azido-4-(bromomethyl)-1-tert-butyl-3-
(
FWO-Flanders) for financial support.
methoxy-4-methylpyrrolidin-2-one 18a: 87% yield, TLC
1
R
f
0.15 (petrol ether/ethyl acetate, 4:1). H NMR (300 MHz,
CDCl ): δ 1.18 (3H, s), 1.47 (9H, s), 3.55 (1H, d, J ) 10.2 Hz),
.64 (3H, s), 3.66 (1H, s), 3.83 (1H, d, J ) 10.2 Hz), 4.78 (1H,
3
Supporting Information Available: General information
3
and all spectroscopic data for compounds 6b, 6d, 7e-h, 10b-
f, 16b, and 18b-e. This material is available free of charge
via the Internet at http://pubs.acs.org.
1
3
3
s). C NMR (75 MHz, CDCl ): δ 16.9, 27.7, 38.4, 47.4, 54.7,
-
1
6
2
0.4, 80.5, 82.4, 173.1. IR (NaCl, cm ): νCdO ) 1714; νN3
108. MS (70 eV) m/z (%): 321/19 (M+ + 1, 100). Anal. Calcd
: C, 41.39; H, 6.00; N, 25.03. Found: C, 41.53;
H, 6.14; N, 24.97.
)
4 2
for C11H19BrN O
JO0509556
8722 J. Org. Chem., Vol. 70, No. 22, 2005