Stereocontrolled Synthesis of Highly Substituted Proline Esters via [3 + 2] Cycloaddition between N-Metalated Azomethine Ylides and Nitroalkenes. Origins of the Metal Effect on the Stereochemical Outcome

Article abstract of DOI:10.1021/jo971212q

The [3 + 2] cycloaddition reaction between several N-metalated azomethine ylides and nitroalkenes has been studied using AgOAc and LiClO₄ as metalating reagents in the presence of triethylamine. The reaction is found to be very versatile and can be extended to homochiral nitroalkenes. In general, lithium and silver salts promote preferential formation of the endo and exo cycloadducts, respectively. The presence of a phenol moiety induces a shift toward the exo-cycloadduct even when lithium is used. A model based upon the experimental results obtained and SCF-MO calculations is proposed to explain the variable stereochemical outcome of these reactions.

Full text of DOI:10.1021/jo971212q

J . Org. Chem. 1998, 63, 1795-1805
1795
Ster eocon tr olled Syn th esis of High ly Su bstitu ted P r olin e Ester s
via [3 + 2] Cycloa d d ition betw een N-Meta la ted Azom eth in e Ylid es
a n d Nitr oa lk en es. Or igin s of th e Meta l Effect on th e
Ster eoch em ica l Ou tcom e
Mirari Ayerbe, Ana Arrieta, and Fernando P. Coss´ıo*
Kimika Fakultatea, Euskal Herriko Unibertsitatea, P.K. 1072, San Sebastia´n-Donostia, Spain
Anthony Linden
Organisch-chemisches Insitut der Universita¨t Zurich,Winterthurestrasse 190,
CH-8057, Zurich, Switzerland
Received J uly 2, 1997 (Revised Manuscript Received J anuary 5, 1998 )
The [3 + 2] cycloaddition reaction between several N-metalated azomethine ylides and nitroalkenes
has been studied using AgOAc and LiClO4 as metalating reagents in the presence of triethylamine.
The reaction is found to be very versatile and can be extended to homochiral nitroalkenes. In
general, lithium and silver salts promote preferential formation of the endo and exo cycloadducts,
respectively. The presence of a phenol moiety induces a shift toward the exo-cycloadduct even
when lithium is used. A model based upon the experimental results obtained and SCF-MO
calculations is proposed to explain the variable stereochemical outcome of these reactions.
In tr od u ction
interesting feature of this reaction is that the stereocon-
trol observed is usually satisfactory. Thus, the reaction
between R,â-unsaturated carbonyl compounds and N-
metalated azomethine ylides derived from amino esters
and aldehydes yields endo-cycloadducts9 as the major
stereoisomers.10 However, To¨ke et al.11 have observed a
reversal of stereoselectivity in the 1,3-dipolar cycloaddi-
tion between several azomethine ylides and aryl nitro
olefins. These authors reported that lithium metalation
promotes the preferential formation of endo cycloadducts,
whereas the use of silver salts favors the formation of
exo-pyrrolidines as major stereoisomers.
The importance of substituted prolines1 in the design
of new catalysts2 or in the chemical synthesis of phar-
macologically3 or biologically interesting4 molecules is
well recognized. The [3 + 2] reaction of azomethine
ylides and alkenes is one of the most useful methods for
the preparation of these molecules in a convergent
manner.5 Among the different versions of this reaction,6
the interaction between N-metalated azomethine ylides
and π-defficient alkenes is specially promising, since, in
general, it allows the synthesis of proline nuclei under
mild conditions and with good chemical yields.7 In
addition, the reaction is very versatile and consequently
is well suited for combinatorial chemistry.8 Another
Within this context and in connection to our ongoing
research program on nitroalkene chemistry,12 we present
(8) (a) Murphy, M. M.; Schullek, J . R.; Gordon, E. M.; Gallop, M. A.
J . Am. Chem. Soc. 1995, 117, 7029. (b) Hamper, B. C.; Dukesherer, D.
R.; South, M. S. Tetrahedron Lett. 1996, 37, 3671. (c) Marx, M. A.;
Grillot, A.-L.; Louer, C. T.; Beaver, K. A.; Bartlett, P. A. J . Am. Chem.
Soc. 1997, 119, 6153.
(9) We denote the endo and exo stereoisomers as those in which the
electron-withdrawing group at C4 and the substituent at the C5
position are in a cis and trans relationship, respectively.
(1) (a) Pichon, M.; Figade`re, B. Tetrahedron: Asymmetry 1996, 7,
927. (b) Sharma, R.; Lubell, W. D. J . Org. Chem. 1996, 61, 202. (c)
Collado, I.; Ezquerra, J .; Pedregal, C. J . Org. Chem. 1995, 60, 5011.
(d) Mulzer, J .; Meier, A.; Buschmann, J .; Luger, P. Synthesis 1996,
123.
(2) (a) Catalytic Asymmetric Synthesis; Ojima, I. Ed.; VCH Publish-
ers: New York, 1993. (b) Noyori, R. Asymmetric Catalysis in Organic
Synthesis; Wiley: New York, 1994.
(3) See, for example: (a) Waid, P. P.; Flynn, G. A.; Huber, E. W.;
Sabol, J . S. Tetrahedron Lett. 1996, 37, 4091. (b) Khav, V. V.;
Martinelli, M. J . Tetrahedron Lett. 1996, 37, 4323. (c) Kolodziej, S. A.;
Nikiforovich, G. V.; Skeean, R.; Lignon, M.-F.; Martinez, J .; Marshall,
G. R. J . Med. Chem. 1995, 38, 137.
(4) See for example: (a) Fishwick, C. W. G.; Foster, R. J .; Carr, R.
E. Tetrahedron Lett. 1996, 37, 3915. (b) Bianco, A.; Maggini, M.;
Scorrano, G.; Toniolo, C.; Marconi, G.; Villani, C.; Prato, M. J . Am.
Chem. Soc. 1996, 118, 4072. (c) Hirschmann, R. Angew. Chem., Int.
Ed. Engl. 1991, 30, 1278. (d) Giannis, A.; Kolter, T. Angew. Chem.,
Int. Ed. Engl. 1993, 32, 1244.
(10) See for example: (a) Tsuge, O.; Kanemasa, S.; Yoshioka, M. J .
Org. Chem. 1988, 53, 1384. (b) Kanemasa, S.; Yoshiora, M.; Tsuge, O.
Bull. Chem. Soc. J pn. 1989, 62, 2196. (c) Grigg, R.; Kemp, J . J . Chem.
Soc. Chem. Commun. 1978, 2885. (d) Tsuge, O.; Kanemasa, S.; Ohe,
M.; Yorozu, K.; Takenaka, S.; Veno, K. Chem. Lett. 1986, 1271. (e)
Tsuge, O.; Kanemasa, S.; Ohe, M.; Yorozu, K.; Takenaka, S.; Veno, K.
Bull. Chem. Soc. J pn. 1987, 60, 4067.
(11) (a) Nyerges, M.; Bala´zs, L.; Bitter, I.; Ka´das, I.; Ko¨vesdi, I.;
To¨ke, L. Tetrahedron 1995, 51, 6783. (b) Nyerges, M.; Rudas, M.; To´th,
G.; Here´nyi, B.; Ka´das, I.; Bitter, I.; To¨ke, L. Tetrahedron 1995, 51,
13321. (c) Nyerges, M.; Bitter, I.; Ka´das, I.; To´th, G.; To¨ke, L.
Tetrahedron Lett. 1994, 35, 4413.
(5) (a) Huisgen, R. Angew. Chem. 1983, 2, 565. (b) Huisgen, R. J .
Org. Chem. 1976, 41, 403.
(6) (a) Grigg, R. Chem. Soc. Rev. 1987, 16, 89. (b) Lowin, J . W. In
1,3-Dipolar Cycloaddition Chemistry; Padwa, A., Ed.; J . Wiley &
Sons: New York, 1984; Vol. 1, pp 653-732. (c) Tsuge, O.; Kanemasa,
S. In Advances in Heterocyclic Chemistry; Katritzky, A. R., Ed.;
Academic Press: San Diego, 1989; Vol. 45, pp 232-349. (d) Vedejs, E.
In Advances in Cycloaddition; Curran, D. P., Ed.; J ai Press: Green-
wich, 1988; Vol. 1, pp 33-51.
(12) (a) Ayerbe, M.; Coss´ıo, F. P. Tetrahedron Lett. 1995, 36, 4447.
(b) Ayerbe, M.; Morao, I.; Arrieta, A.; Linden, A.; Cossı´o, F. P.
Tetrahedron Lett. 1996, 37, 2311.
(7) Kanemasa, S.; Tsuge, O. In Advances in Cycloaddition; Curran,
D. P., Ed.; J ai Press: Greenwich, 1993; Vol. 3, pp 99-159.
S0022-3263(97)01212-7 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/24/1998

1796 J . Org. Chem., Vol. 63, No. 6, 1998
Ayerbe et al.
Ch a r t 1
Sch em e 2a
Sch em e 1a
a
Only one enantiomer of the cycloadducts 4 is drawn.
1, entry 2). In contrast, the same reaction in the presence
of 0.15 equiv of LiClO4 yields 4b in a 30% yield with a
endo:exo ratio of 63:37. When 5 equiv of LiClO4 were
used, the cycloadduct was obtained in a 47% yield and
with higher stereocontrol (see Table 1, entry 2). This
effect is more pronounced with more substituted sub-
strates. Thus, when the 1c+2b f 4e transformation was
performed in the presence of 0.15 equiv of LiClO4, a
reaction time of 20 h was required to obtain a 56:44 endo/
exo mixture of cycloadducts 4e in a 30% yield. When the
same reaction was carried out with 5 equiv of LiClO4,
only 4 h were required to achieve a higher yield of 4e
(see Table 1, entry 5). Similar results were observed in
other cases. Therefore, we decided to use 5 equiv of this
salt in all cases, because of the lower reaction times
required to achieve acceptable yields and/or the higher
stereocontrol observed.
a
The substituents at the different positions are not specified.
herein our results on the reaction between nitro olefins
and azomethines. We have extended the reaction to
highly substituted azomethine ylides and nitroalkenes,
and we provide an explanation to the metal dependence
of the stereochemical outcome of this important reaction.
Resu lts a n d Discu ssion
To obtain structurally diverse cycloadducts, imines
1a -g and nitroalkenes 2a -d were selected (see Chart
1). In particular, compounds 1b-f were studied because
of the potential of the corresponding cycloadducts as
tridentate ligands.2 Nitroalkene 2d was also selected in
order to study the synthetic potential of the reaction
when homochiral nitroalkenes are used as dipolarophiles.
Lithium and silver azomethine ylides (see Scheme 2)
derived from imines 1a -g were generated from LiClO4
and AgOAc, respectively, in the presence of triethylamine
as base and acetonitrile as solvent. LiClO4 has not been
previously described as a source of N-lithiated azo-
methines. These reactive intermediates are usually
generated by means of either the LiBr/NEt3 or the LDA/
NEt3 systems.7 We have observed that LiClO4 usually
leads to cleaner crude reaction mixtures with respect to
the above-mentioned lithium salts. The results obtained
are collected in Table 1. We have found that when
AgAcO is used to promote the formation of azomethine
ylides, the reaction takes place faster than when forma-
tion of the dipoles is assisted by LiClO4. For example,
reaction between 1a and 2b in the presence of 0.15 equiv
of AgOAc yields a mixture of endo- and exo-4b after 3 h
of reaction at room temperature in a 69% yield (see Table
We have observed a preferential formation of the exo-
cycloadduct when a silver azomethine ylide possessing a
noncoordinating aromatic substituent reacts with an
aromatic nitroalkene. In the case of the corresponding
lithium azomethine, the endo-stereoisomer is formed
preferentially under similar conditions. For instance,
exo-4a is the major product in the reaction between 1a
and 2a in the presence of AgOAc, whereas endo-4a is
preferentially formed when LiClO4 is used (see Table 1,
entry 1). Similar results were obtained in the reaction
between 1a and 2b (entry 2). When the same experi-
ments were repeated using N,N-diisopropylethylamine
as base instead of triethylamine, the same stereochemical
outcome was observed. These results are in agreement
with the observations of To¨ke et al.11 for structurally
related compounds.
We have found that the above-mentioned trend can be
modified when azomethines incorporating certain groups
at the aromatic moiety are used. Thus, when imine 1b
derived from salicylaldehyde and glycine was allowed to

Synthesis of Highly Substituted Proline Esters
J . Org. Chem., Vol. 63, No. 6, 1998 1797
Ta ble 1. [3 + 2] Cycloa d d ition betw een N-Meta la ted Azom eth in es Der ived fr om Im in es 1a -g a n d Nitr oa lk en es 2a -c
Usin g Aceton itr ile a s Solven t a n d in th e P r esen ce of AgOAc a n d LiClO4
method A (AgOAc)
method B (LiClO4)
entry product
R1
R2
R3
R4
time (h) yielda (%) endo:exob time (h) yielda (%) endo:exob
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
4o
C6H5
C6H5
2-HO-C6H4
2-HO-C6H4
2-HO-C6H4
H
H
H
H
CH3
H
H
4-Cl-C6H4
4-CH3OC6H4
4-Cl-C6H4
4-CH3OC6H4
4-CH3OC6H4
3
3
2.5
2.5
48
17
40
2
60
69
27:73
30:70
2:98
3
3
2.5
2.5
4
17
7.5
2
66
47
76:24
83:17
14:86
2:98
H
H
H
H
29c
41c
49c
24c
30c
62d
57
24c
45c
37c
24c
27c
38e
61
2:98
42:58
30:70
30:70
33:67
37:63
36:64
33:67
33:67
22:78
36:64
30:79
58:42
55:45
65:35
43:57
39:61
73:27
34:66
65:35
68:32
23:77
24:76
2-HO-C6H4
2-HO-C6H4
CH3 4-CH3OC6H4
CH3 CH3 4-CH3OC6H4
H
H
CH3
2-H3COC6H4
2-CH3OC6H4
2-CH3OC6H4
2-CH3OC6H4
2-pyridyl
2-pyridyl
4-HO-3-CH3OC6H3
4-HO-3-CH3OC6H3
H
H
H
4-Cl-C6H4
4-CH3OC6H4
4-CH3OC6H4
2
2
16
24
4
4
2
60
16
16
2.5
2.5
2
56
60
54
CH3 CH3 4-CH3OC6H4
60
H
H
H
H
H
H
H
H
4-CL-C6H4
4-CH3OC6H4
4-Cl-C6H4
25c
64
52
36c
42c
40c
56
4-CH3OC6H4
2
2
a
b
Yield of the major stereoisomer purified by column chromatography. Diastereomeric ratio measured by 300 MHz 1H-NMR on crude
reaction mixtures. c Yield of the major stereoisomer purified by crystallization. Yield of the exo isomer estimated from the mixtures
d
obtained from the column chromatography. This isomer couldn’t be isolated. e Yield of the endo isomer estimated from the mixtures
obtained from the column chromatography. This isomer could not be isolated.
Ta ble 2. Ster eoselectivity in th e [3 + 2] Cycloa d d ition
react with nitroalkenes 2a or 2b in the presence of AgOAc
betw een N-Meta la ted Azom eth in es Der ived fr om Im in e
and 2 equiv of triethylamine, the exo cycloadducts 4c and
4d were obtained with excellent stereocontrol (endo:exo
) 2:98). When the same reaction was carried out in the
presence of LiClO4, the exo cycloadducts were also the
major stereoisomers (see Table 1, entries 3 and 4). This
effect is significantly lower when the imine 1d , which
incorporates a 2-methoxyphenyl group was used instead
of 1b (see Table 1, entries 8 and 9). When azomethines
derived from imine 1f which has a 2-pyridyl group were
used, the “normal” sense of stereocontrol was observed,
although the stereoselectivity was lower (see Table 1,
entries 12 and 13). In all these systems, there is a Lewis
basic center located at the ortho position, which is
amenable to intramolecular coordination to the metallic
center. To test if this effect can also operate intermo-
lecularlly, imine 1g, derived from vanillin and the methyl
ester of glycine, was allowed to react with nitroalkenes
2a and 2b. The results are shown in entries 14 and 15
of Table 1. As can be seen, the ionized para-phenolic
moiety also promotes preferential formation of the exo
cycloadducts 4n and 4o in the presence of LiClO4,
although the effect is less pronounced than in the case
of entries 3 and 4 (vide supra).
With these results in mind, we reasoned that the
presence of additives incorporating highly coordinating
moieties such as nitro or phenoxy groups might modify
the stereochemical outcome of the reaction between
aromatic nitroalkenes and N-lithium azomethines de-
rived from 1a . We performed the experiments reported
in Table 2 to verify this hypothesis. In effect, the
presence of phenol (entry 4), 2,6-di-tert-butyl phenol
(entry 5) and nitrobenzene (entry 7) reverses the stereo-
chemical outcome observed in the reaction between imine
1a and nitroalkene 2a to form the cycloadducts endo- and
exo-4a . As expected, the effect induced by the hindered
2,6-di-tert-butylphenol is less pronounced than that
promoted by phenol itself (see Table 2, entries 4 and 5).
In the case of using AgOAc, these additives have negli-
1A a n d Alk en e 2A in th e P r esen ce of Sever a l Ad d itivesa
entry
MX
additiveb
time (h) endo-4a :exo-4a c
1
2
3
4
5
6
7
AgOAc none
LiClO4 none
3
3
5
5
2
5
5
27:73
76:24
30:70
40:60
51:49
27:73
36:64
AgOAc C6H5OHd
LiClO4 C6H5OHd
LiClO4 2,6-ditBu-PhOHd
AgOAc C6H5NO2
LiClO4 C6H5NO2
a
All experiments were conducted using acetonitrile as solvent
b
and triethylamine as base. One equivalent of the additive was
used. c Diastereomeric ratio determined by 300 MHz 1H-NMR on
d
crude reaction mixtures. Two equivalents of triethylamine were
used.
gible influence on the stereochemical outcome of the
1a +2a f 4a process (see Table 2, entries 1, 3, and 6).
According to our results azomethines derived from
substituted amino esters such as alanine (see compounds
1c,e in Chart 1) and (E)-1-methyl-2-(arylnitro)ethylenes
(see compound 2c in Chart 1) are much less reactive than
the preceding reactants. For example, the reaction
between 1c (R2 ) Me) and 2b to form 4e requires 48 h
at room temperature in the presence of AgOAc, whereas
the reaction between 1b (R2 d H) and 2b takes place in
only 2.5 h under the same conditions (see Table 1, entry
4). In any case, the use of highly substituted reactants
is of preparative value, since the cycloadducts are readily
purified and obtained in acceptable yields. In addition,
in these highly substituted substrates the effect of ortho-
phenolic groups is very attenuated with respect to their
less substituted analogues. Thus, in the reaction of imine
1c with nitroalkene 2c to yield cycloadducts endo- and
exo-4g (Table 1, entry 7) a loss of stereocontrol is observed
with respect to the nonmethylated substrates (see Table
1, entry 4). When LiClO4 is used, the proportion of exo-
4g is higher than that obtained in the case of the 1a +2b
f 4b process, although the endo cycloadduct is the major
one.

1798 J . Org. Chem., Vol. 63, No. 6, 1998
Ayerbe et al.
Sch em e 3
relative stereochemistry was secured by X-ray diffraction
analysis. The ORTEP diagrams of compounds exo-4c and
exo-4f are shown in Figures 1 and 2 of the Supporting
Information, respectively.16
In the case of compound exo-4p , its structure was
assigned by NOE experiments, by analogy with structur-
ally related compounds,17 and applying the endo-alkoxy
rule developed by Houk.18 In several cases, oily cycload-
ducts were transformed in their solid N-acetylated de-
rivatives19 in order to facilitate their purification. Thus,
cycloadducts endo-4h and exo-4h ,j were heated with
trifluoroacetic acid, and the corresponding salts were
acylated with acetic anhydride for 48 h to yield the
acetamides endo-5h and exo-5h ,j respectively (see Sup-
porting Information for additional details).
The next step in our study was to elucidate the origins
of the observed changes in the sense of stereocontrol.
Given that under kinetic conditions this stereocontrol is
determined by the differences in energy between the
corresponding diastereomeric transition structures,20 the
computational tools nowadays available to the organic
chemist are particularly convenient to address this
particular problem. Since it was presumed that solvent
and substituent effects should play a significant role, the
PM3 semiempirical Hamiltonian21 was used, in view of
its superior ability to model the particular features of the
nitro group.22
We located first the saddle point TS1 (Figure 1)
associated with the [3 + 2] cycloaddition between ethyl-
ene and the lithium azomethine ylide of N-(formylmeth-
yl)benzylideneamine. This transition structure corre-
sponds to a concerted supra-supra topology as would be
expected from the Woodward-Hoffmann rules for ther-
mal six-electron cycloadditions. However, inclusion of a
nitro group induces a drastic change in the potential
energy hypersurface. Thus, nitroalkene is highly π-de-
ficient and behaves as a Michael acceptor rather than a
dipolarophile. Therefore, the [3 + 2] cycloaddition actu-
ally consists of a tandem Michael-Henry reaction, as it
is shown in Scheme 4. The first step is a nucleophilic
attack of the lithium azomethine ylide on the nitroalkene
to form the zwitterionic species denoted in Scheme 4 as
INT, in which the C2-C3 bond is formed.23 The second
step is a nitroaldol process13 in which the nitronate
moiety adds to the iminic part of INT. This is in
agreement with the fact that several authors7,24 have
reported the formation of Michael adducts in the reaction
We next extended our study to the chiral nitroalkene
2d (see Chart 1). This compound was prepared via the
Henry reaction between (S)-2-(benzyloxy)propanal and
nitromethane in the presence of a catalytic amount of
triethylamine. The diastereomeric nitro alcohols13 thus
formed were transformed in the corresponding mesy-
lates14 and treated in situ with N,N-diisopropylethyl-
amine to yield the nitroalkene 2d in good yield.15 Reac-
tion of this dipolarophile with azomethine ylides de-
rived from 1a lead to the diastereomeric pair endo-4p
and endo-4′p (see Scheme 3). Interestingly, this alipha-
tic nitroalkene yields endo cycloadducts using both
lithium and silver azomethines, thus suggesting that the
aryl group in nitroalkenes 2a -c also plays a role in
the stereochemical outcome of the reaction (vide infra).
When lithium perchlorate was used instead of silver
acetate, the observed diastereomeric ratio was 95:5 in
favor of endo-4p (see Scheme 3). The major cycloadduct
was purified by crystallization in ethyl acetate-hexanes.
Despite the higher stereocontrol observed when the
lithium azomethine ylide of 1a was used, the chemical
yield of endo-4p was similar to that obtained using
AgOAc as the metalating reagent (see Scheme 3). There-
fore, the reaction was found to be of preparative value
and provides a useful access to the asymmetric synthesis
of compounds of this kind.
(16) Crystallographic data (excluding structure factors) for the
structures exo-4c and exo-4f have been deposited with the Cambridge
Crystallographic Data Centre. Copies of the data can be obtained free
of charge on application to the Director, CCDC, 12 Union Road,
Cambridge CB2 IEZ, UK. (fax: +44-(0)1223-336033; email: deposit@
chemcrys.cam.ac.uk).
(17) (a) Annunziata, R.; Cinquini, M.; Cozzi, F.; Raimondi, L.; Pilati,
T. Tetrahedron: Asymmetry 1991, 2, 1329. (b) Pa¨tzel, M.; Galley, G.;
J ones, P. G.; Chrapkowsky, A. Tetrahedron Lett. 1993, 34, 5707. (c)
Galley, G.; Liebscher, J .; Pa¨tzel, M. J . Org. Chem. 1995, 60, 5005.
(18) (a) Houk, K. N.; Moses, S. R.; Wu, Y.-D.; Rondan, N. G.; J ager,
V.; Schohe, R.; Fronczek, F. R. J . Am. Chem. Soc. 1984, 106, 3880. (b)
Houk, K. N.; Wu, Y.-D.; Duh, H.-Y.; Moses, S. R. J . Am. Chem. Soc.
1986, 108, 2754.
The structure of cycloadducts 4a -p was established
by means of their 1H NMR and 13C NMR spectra. The
most significant coupling constants are reported in Table
1 of the Supporting Information. In two cases, the
(13) (a) For
a
recent review see: Rosini, G. In Comprehensive
(19) Beausoleil, E.; Lubell, W. D. J . Am. Chem. Soc. 1996, 118,
12902.
Organic Synthesis; Trost, B. M., Ed.; Pergamon: New York, 1991; Vol.
2, pp 321-340. (b) For an study on the mechanism of this reaction,
see Lecea, B.; Arrieta, A.; Morao, I.; Coss´ıo, F. P. Eur. J . Chem. 1997,
3, 20.
(14) Melton, J .; McMurry, J . E. J . Org. Chem. 1975, 40, 2185.
(15) Pa¨tzel et al. have prepared structurally related chiral nitroalk-
enes using different reaction conditions and have observed a lower
stereocontrol. See Galley, G.; Hu¨bner, J .; Anklam, S.; J ones, P. G.;
Pa¨tzel, M. Tetrahedron Lett. 1996, 37, 6307.
(20) Seeman, J . I. Chem. Rev. (Washington, D.C.) 1983, 83, 83.
(21) (a) Stewart, J . J . P. J . Comput. Chem. 1989, 10, 209. (b) Anders,
E.; Koch, R.; Freunscht, P. J . Comput. Chem. 1993, 14, 1301.
(22) Stewart, J . J . P. J . Comput.-Aided Mol. Des. 1990, 4, 1.
(23) For a related reaction mechanism involving azomethine ylides
and R,â-unsaturated carbonyl compounds, see: Tatsukawa, A.; Ka-
watake, K.; Kanemasa, S.; Rudzinski, J . M. J . Chem. Soc., Perkin
Trans. 2 1994, 2525.

Synthesis of Highly Substituted Proline Esters
J . Org. Chem., Vol. 63, No. 6, 1998 1799
Sch em e 4
a
The possible substituents at the different positions are not
specified.
materials. On the other hand, we have found that the
reaction takes place more slowly when it is carried out
in the presence of only 0.15 equiv of LiClO4 (vide supra).
Therefore, we repeated the reaction between 1b and 2c
in the presence of 0.15 equiv of LiClO4 (Scheme 5). After
16 h reaction at room temperature, a 53:47 mixture of
endo/exo-4f and a mixture of Michael intermediates 7 was
observed. The cycloadducts 4f and the intermediates
were found to be in a ca. 2:1 ratio in the crude reaction
mixture, while ca. 25% of 2c remained un reacted. When
this mixture was allowed to react for additional 10 h, the
1H NMR signals of 7 disappeared and only a mixture of
cycloadducts 4f was observed. In this case a complex
mixture of pyrrolidines was obtained because of the
partial loss of the trans relationship between the nitro
and the 4-methoxyphenyl groups. All our attempts to
isolate and characterize these intermediates in a pure
form failed. However, hydrolysis of the imine moiety in
7 followed by in situ acylation with 4-bromobenzoyl
chloride allowed the complete characterization of amide
8, which was the major diastereomer observed in the
crude reaction mixture. The overall yield of 8 from 1b
was of 6%. Both the high δ value of the N-H signal and
molecular mechanics optimizations (vide infra) indicate
that in 8 there is a strong intramolecular hydrogen bond
between the amidic hydrogen and the nitro group, thus
yielding a pseudocyclic structure. Therefore, the relative
stereochemistry of this product was established on the
basis of the coupling constants between the contiguous
methinic protons (see Scheme 5). In summary, the
observation of intermediate 7, its conversion to the
corresponding cycloadducts and the complete character-
ization of its derivative 8 provide experimental support
for the stepwise mechanism predicted by our calculations.
Since no isomerization is observed either in the azo-
methine ylide or in the nitroalkene during the course of
the reaction, the stereochemistry of the cycloadducts is
determined by the approaching mode between both
partners. In Figure 1 we have included the two diaster-
eomeric endo- and exo-TS2 saddle points. In endo-TS2,
the nitro group is coordinated to the lithium atom, with
a Li‚‚‚ONO bond distance of 1.955 Å. In the latter saddle
F igu r e 1. Computer plots of the transition structures TS1,
endo-TS2, and exo-TS2 computed at the RHF/PM3 level. Bond
distances and angles are given in angstroms and deg, respec-
tively. In this and the following figures, which incorporate ball-
and-stick representations, unless otherwise noted, atoms are
represented by increased order of shadowing as follows: H,
C, O, N.
between π-deficient alkenes and azomethine ylides de-
rived from alkyl aldehydes or ketones.
We tried to isolate and characterize several of these
Michael intermediates in order to verify our computa-
tional results. However, in most cases the inherent
instability of the intermediates precluded their isolation,
1
although they were observed in the H NMR spectra of
several reaction mixtures. We reasoned that, since the
more substituted the nitronate is the lower its reactivity,13b
R-methyl nitroalkenes could be more convenient starting
(24) (a) Tsuge, O.; Kanemasa, S.; Ohe, M.; Yorozu, K.; Takenaka,
S.; Ueno, K. Chem. Lett. 1986, 1271. (b) Kanemasa, S.; Yoshioka, M.;
Tsuge, O. Bull. Chem. Soc. J pn. 1989, 62, 869.

1800 J . Org. Chem., Vol. 63, No. 6, 1998
Ayerbe et al.
Sch em e 5
a
Numbers in parentheses correspond to computed coupling
constants (see text). Nonrelevant hydrogen atoms have been
omitted for clarity.
point, exo-TS2 the lithium atom is dicoordinated and
therefore the stabilizing Li‚‚‚ONO interaction is not
present. In contrast, this transition structure is much
more flexible and the ∆G298 value for exo-TS2 is calcu-
lated to be 1.24 kcal/mol lower than that obtained for
endo-TS2 at the RHF/PM3 level (see Figure 1). These
results are not in agreement with our experimental data,
since we have observed that when azomethine ylides
derived from benzaldehyde are used, the endo cycload-
ducts are obtained with a relatively low stereocontrol
(vide supra). Therefore, we calculated additional transi-
tion structures taking into account as closely as possible
the effects operating under the experimental conditions,
namely the aryl substituent of the nitroalkene and the
solvent. In Figure 2 we present the endo- and exo-TS3
transition structures, which incorporate two acetonitrile
molecules. In endo-TS3, the lithium atom is pentacoor-
dinated, and its surrounding ligands adopt a bipyramidal
disposition, in which the nitrogens of the azomethine
moiety and of one solvent molecule occupy the axial
positions. In contrast, exo-TS3 shows a tetrahedrically
coordinated lithium atom, with a close proximity between
the methyl group of one acetonitrile molecule and the
phenyl group of the nitroalkene. These combined effects
result in a slightly higher stability of endo-TS3, whose
∆G298 calculated value is only 0.47 kcal/mol lower than
that found for its exo congener. This result is in fair
agreement with the diastereomeric excess experimentally
found for dipoles of this type.
F igu r e 2. Computer plots of the transition structures endo-
TS3 and exo-TS3 computed at the RHF/PM3 level. Bond
distances and angles are given in angstroms and deg, respec-
tively. See Figure 1 caption for additional details.
We next investigated computationally the effect of the
ortho-phenoxy group on the stereochemical outcome.
Figure 3 shows the three transition structures found for
the ortho-phenoxy analogues of TS2 saddle points in the
RHF/PM3 energy hypersurface. Our calculations show
that the phenoxy substituent has a remarkable effect.25
The first saddle point is endo′-TS4, whose chief geometric
features are similar to those of endo-TS3. However, in
this case there is a remarkable repulsive Coulombic
interaction between the nitro and the phenoxy groups
coordinated to the lithium atom. Thus, partitioning of
the total energy26 reveals that the bicentric term between
both groups is ∆E ) +0.84 kcal/mol, which corresponds
to an almost purely Coulombic repulsion. We have also
located another saddle point which is associated with the
formation of the corresponding endo cycloadduct. This
transition structure is denoted as endo-TS4 in Figure 3.
(25) For a related metal-imine complexes involving o-alkoxy groups,
see Gately, D. A.; Norton, J . R. J . Am. Chem. Soc. 1996, 118, 3479.
(26) (a) Dewar, M. J . S.; Lo, D. H. J . Am. Chem. Soc. 1971, 93, 7201.
(b) Olivella, S.; Vilarrasa, J . J . Heterocycl. Chem. 1981, 18, 1189.

Synthesis of Highly Substituted Proline Esters
J . Org. Chem., Vol. 63, No. 6, 1998 1801
F igu r e 4. Computer plots of the transition structures endo-
TS5 and exo-TS5 computed at the RHF/PM3 level. Bond
distances and angles are given in angstroms and deg, respec-
tively. See Figure 1 caption for additional details.
tures are also reported in Figure 3. This latter saddle
point does not exhibit an antiperiplanar arrangement
between the N1-C2 and C3-C4 and therefore has a heat
of formation ca. 0.2 kcal/mol higher than that calculated
for endo-TS4. However, the total entropy of exo-TS4 is
higher than that of its endo congener. Therefore, ac-
cording to our calculations the free energy at 298 K of
exo-TS4 is predicted to be 0.42 kcal/mol more stable than
endo-TS4 at the RHF/PM3 level (see Figure 3).
We have also calculated the transition structures
including two molecules of the solvent and an additional
phenyl ring for the nitroalkene moiety. The resulting
diastereomeric endo- and exo-TS5 saddle points are
depicted in Figure 4. Both transition structures are quite
similar to each other. Thus, in both stationary points
the lithium atom is pentacoordinated, the oxygen and
nitrogen atoms occupying the axial and equatorial posi-
tions, respectively. Once again, in endo-TS5 the ar-
rangement between the N1-C2 and C3-C4 bonds is
almost perfectly antiperiplanar, thus resulting in a lower
heat of formation for this transition structure. In con-
trast, the entropic term of exo-TS5 is higher than in its
endo congener. As a result, the free energy at 298 K of
exo-TS5 is calculated to be 0.82 kcal/mol lower than that
of endo-TS5 (see Figure 4). Therefore, the free energy
F igu r e 3. Computer plots of the transition structures endo′-
TS4 and endo-TS4 and exo-TS4 computed at the RHF/PM3
level. Bond distances and angles are given in angstroms and
deg, respectively. See Figure 1 caption for additional details.
Both RHF/PM3 calculated heats of formation and free
energies predict that this latter saddle point, despite
having a tricoordinated lithium cation, is significantly
more stable than endo′-TS4. In addition, endo-TS4
shows an almost perfect antiperiplanar arrangement
between the N1-C2 and C3-C4 bonds, and the calcu-
lated C2‚‚‚C3-C4 bond angle is found to be of 109.3°, a
value in good agreement with the Bu¨rgi-Dunitz trajec-
tory27 usually associated with nucleophilic attacks of
organometallic compounds to electrophilic double bonds.
Therefore, we can conclude that endo-TS4 resembles the
saddle point that one can expect for a Michael addition
of an organometallic carbon nucleophile on a π-deficient
olefin.
In addition, we have also located the transition struc-
ture exo-TS4, whose main geometric and energetic fea-
(27) (a) Bu¨rgi, H. B.; Lehn, J .-M.; Wipff, G. J . Am. Chem. Soc. 1974,
96, 1956. (b) Bu¨rgi, H. B.; Dunitz, J . D.; Lehn, J .-M. Tetrahedron 1974,
30, 1563.

1802 J . Org. Chem., Vol. 63, No. 6, 1998
Ayerbe et al.
balance favors formation of exo-TS5 with a stereocontrol
higher than that calculated for the non-ortho-substituted
analogues, in good agreement with our experimental
results.
of the substituents and on the competition between the
coordinating ability of the nitro group and that of the
solvent or ortho ionizable groups, if present.
(e) The nature of the metal, in particular, its hardness,
is closely related to the stereochemical outcome of the
reaction. Aside from other effects, the nitro group shows
more affinity for the hardest metal.
In summary, our computational studies provide a
useful information on the origins of the stereocontrol of
these cycloadditions. When nonionizable substituents
are present, entropic factors, the aryl group of the
dipolarophile (or Michael acceptor), and the solvent favor
the preferential formation of the endo cycloadduct by
coordination of the nitro group with the lithium atom.
In contrast, when groups other than the nitro group but
with high affinity for the lithium cation are present, the
same factors shift the sense of stereoselection toward the
preferential formation of the exo cycloadduct.
At this stage of our study, there was another aspect
which remained unexplored, namely the preferential
formation of exo cycloadducts when silver azomethine
ylides are used. Given that silver is not parametrized
in PM3, we used an indirect approach to address this
problem. As we have previously found, a possible source
of stereocontrol can arise from the competition between
the nitro and cyano groups. Therefore, the sign of the
energy associated with eq 1 can serve as an indicator of
the relative affinity of lithium and silver cations for the
Exp er im en ta l Section
Gen er a l. Commercially available compounds were used as
purchased without further purification. LiClO4 (ACS reagent)
was used as received (CAUTION: perchlorates are potentially
explosive materials. Under the experimental conditions de-
scribed below, this compound can be used safely). All solvents
were dried and distilled according to standard protocols.29
Nitroalkenes 2a -c were prepared following reported pro-
cedures.30(S)-2-(Benzyloxy)propanal was prepared as previ-
ously reported.31 Melting points are uncorrected. Chemical
1
shifts in the H NMR and 13C NMR spectra are reported as δ
values (ppm) with respect to TMS and deuterated chloroform.
Flash chromatography purifications were performed using
silica gel 70-230 mesh and AcOEt/hexanes mixtures varying
from 1:10 to 1:20. The organic layers were dried with Na2-
SO4 prior evaporation of the solvent under reduced pressure.
SCF -MO Ca lcu la tion s. All semiempirical calculations
were performed by means of the PM3 Hamiltonian.21a The
parameters for lithium developed by Anders et al.21b were used.
All transition structures were located using the eigenvector-
following algorithm32 as implemented in MOPAC.22 All saddle
points were characterized by harmonic analysis33 and showed
only one imaginary frequency in their diagonalized Hessian
matrixes, associated with nuclear motion along the reaction
coordinate under study. Thermodynamic quantities were
computed using the standard rigid rotor harmonic oscillator
approximation.34 The reported stationary points were refined
until their gradient norm was below 0.25 kcal/mol/Å, deg, mol.
In addition, the SCF convergence criteria were incremented
100 times, as suggested by Boyd et al.35
∆E
HNO2‚‚‚Ag+ + HCtN‚‚‚Li+
8
HNO2‚‚‚Li+ + HCtN‚‚‚Ag+ (1)
nitro and cyano groups. Given the number of electrons
of silver cation, the relativistic compact effective potential
developed by Stevens, Krauss, Basch and J asien (SKBJ )28
was used. At the MP2(FC)/SKBJ -6-31+G*+∆ZPVE level
(vide infra) the silver cation is calculated to be softer than
the lithium cation, the difference in hardness being of
∆η ) 1.01 au. In addition, the computed value for ∆E is
-2.74 kcal/mol, thus indicating a higher affinity of silver
for the cyano group. Therefore, in the presence of
acetonitrile as solvent, the cyano group can compete more
efficiently with the nitro group when silver azomethines
are used, thus resulting in the preferential formation of
exo cycloadducts.
Ab initio computations were performed using the GAUSS-
IAN 94 series of programs.36 For other elements than silver
the 6-31+G* basis set37 was used. Calculations involving
silver cation were performed using the relativistic compact
effective potential developed by Stevens, Krauss, Basch, and
J asien,28 and denoted as SKBJ . Electron correlation was
partially taken into account using the second-order Møller-
Plesset approximation, keeping the core electrons frozen.38
This level is denoted as MP2(FC). Zero point vibrational
energy (ZPVE) corrections were scaled by 0.96.39
Molecu la r Mech a n ics Com p u ta tion s. The minimum
energy conformation of compound 8 was computed using the
Con clu sion s
From the combined experimental and computational
studies reported in this paper, the following conclusions
can be drawn:
(29) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals; Pergamon: Oxford, 1993.
(30) Bourguinon, J .; LeNard, G.; Gueguiner, G. Can. J . Chem. 1985,
63, 2354.
(31) Zemribo, R.; Romo, D. Tetrahedron Lett. 1995, 36, 4159.
(32) Baker, J . Comput. Chem. 1986, 7, 385.
(33) McIver, J . W.; Komornicki, A. K. J . Am. Chem. Soc. 1972, 94,
2625.
(a) The reaction between substituted nitrostyrenes and
lithium azomethines having o-phenoxy groups results in
the preferential formation of the exo cycloadducts, both
in the case of lithium or silver azomethine ylides.
(b) The reaction between chiral nitroalkenes and
achiral lithium azomethines seems to be of preparative
value, although the use of silver azomethines results in
a lower stereocontrol.
(c) Computational studies suggest that the [3 + 2]
cycloaddition is actually a tandem Michael-Henry pro-
cess. In one case, a stable derivative of the reaction
intermediate has been isolated and characterized.
(d) According to the computational studies, the stereo-
chemical outcome of the reaction depends on the nature
(34) Dewar, M. J . S.; Ford, G. P. J . Am. Chem. Soc. 1977, 99, 7822.
(35) Boyd, D. B.; Smith, D. W.; Stewart, J . J . P.; Wimmer, E. J .
Comput. Chem. 1988, 9, 387.
(36) GAUSSIAN 94, Revision B.2: M. J . Frisch, G. W. Trucks, H.
B. Schlegel, P. M. W. Gill, B. G. J ohnson, M. A. Robb, J . R. Cheeseman,
T. Keith, G. A. Peterson, J . A. Montgomery, K. Raghavachari, M. A.
Al-Laham, V. G. Zakrzewski, J . V. Ortiz, J . B. Foresman, C. Y. Peng,
P. Y. Ayala, W. Chen, M. W. Wong, J . L. Andres, E. S. Replogle, R.
Gomperts, R. L. Martin, D. J . Fox, J . S. Binkley, D. J . Defrees, J . Baker,
J . P. Stewart, M. Head-Gordon, C. Gonzalez, and J . A. Pople, Gaussian
Inc.: Pittsburgh, PA, 1995.
(37) (a) Clark, T.; Chandrasekhar, J .; Spitznagel, G. W.; Schleyer,
P.v. R. J . Comput. Chem. 1983, 4, 294. (b) Frisch, M. J .; Pople, J . A.;
Binkley, J . S. J . Chem. Phys. 1994, 80, 3265.
(38) (a) Binkley, J . S.; Pople, J . A. Int. J . Quantum. Chem. 1975, 9,
229. (b) Pople, J . A.; Binkley, J . S.; Seeger, R. Int. J . Quantum Chem.
Symp. 1976, 10, 1.
(28) Stevens, W. J .; Krauss, M.; Basch, H.; J asien, P. G. Can. J .
Chem. 1992, 70, 612.

Synthesis of Highly Substituted Proline Esters
J . Org. Chem., Vol. 63, No. 6, 1998 1803
MM2* force field40 as implemented in the Macromodel 5.0 suite
of programs.41 The different possible diastereomers were
optimized and then Monte Carlo42 simulations were performed
on 1000 structures. The theoretical coupling constants in-
cluded in Scheme 5 were performed on the more stable
conformations using the Altona method.43
Gen er a l P r oced u r e for th e P r ep a r a tion of th e Im in es
1a -g. To a solution of the corresponding amino acid methyl
ester chlorhydrate (23 mmol) in CH2Cl2 (35 mL) were added
MgSO4 and of NEt3 (3.5 mL, 25 mmol) at room temperature.
The resulting mixture was stirred at the same temperature
for 1 h. Then the aldehyde was added, and the resulting
mixture was stirred for 16 h (except in the case of the imine
1f which only needed 7 h). Then the MgSO4 was filtered out
and the organic layer washed with water (2 × 25 mL), dried,
and evaporated. The imines showed satisfactory 1H NMR
spectra and were used in subsequent reactions immediately
without further purification.
The other reactions were performed according to the pro-
cedures described above. The reaction conditions as well as
the results are listed in Table 1.
(2S*,3R*,4S*,5S*)-3-(4-Ch lor op h en yl)-2-(m et h oxyca r -
bon yl)-4-n itr o-5-p h en ylp yr r olid in e (4a -en d o isom er ): mp
113-115 °C; IR 3360, 1733, 1547, 1362 cm-1; 1H NMR (δ ppm,
CDCl3) 7.38-7.18 (m, 9H), 5.23 (dd 1H, J ) 6.6 Hz, J ′ ) 3.9
Hz), 4.88 (tb, 1H), 4.18 (dd, 1H, J ) 7.6 Hz, J ′ ) 3.9 Hz), 4.08
(tb, 1H), 3.79 (s, 3H), 3.29 (tb, 1H); 13C NMR (δ ppm, CDCl3)
171.4, 136.8, 134.3, 134.0, 129.4, 128.9, 128.8, 128.7, 126.4,
96.7, 67.5, 67.2, 54.4, 52.7. Anal. Calcd for C18H17O4N2Cl: C,
59.91; H, 4.76; N, 7.77. Found: C, 60.83; H, 4.86; N, 7.02.
(2S*,3S*,4R*, 5S*)-3-(4-Ch lor op h en yl)-2-(m eth oxyca r -
bon yl)-4-n itr o-5-p h en ylp yr r olid in e (4a -exo isom er ): mp
115-117 °C; IR 3358, 1734, 1548, 1379 cm-1; 1H NMR (δ ppm,
CDCl3) 5.10 (t, 1H, J ) 7.7 Hz), 4.74 (t, 1H, J ) 7.7 Hz), 4.49
(dd, 1H, J ) 8.8 Hz, J ′ ) 5.4 Hz), 4.32 (dd, 1H, J ) 8.8 Hz, J ′
) 7.7 Hz), 3.34 (s, 3H), 2.70 (tb, 1H); 13C NMR (δ ppm, CDCl3)
171.5, 137.5, 134.7, 134.0, 129.2, 129.0, 128.9, 126.8, 95.1, 67.2,
64.0, 52.7, 51.9. Anal. Calcd for C18H17O4N2Cl: C, 59.91; H,
4.76; N, 7.77. Found: C, 56.79; H, 5.03; N, 7.12.
Syn th esis of (E)-(3S)-3-(Ben zyloxy)-1-n itr o-1-bu ten e
(2d ). A mixture of (S)-2-(benzyloxy)propanal (1.64 g, 10
mmol), nitromethane (2.7 mL, 50 mmol), and NEt3 (0.2 mL)
was stirred at room temperature for 3 h. The excess of
nitromethane was removed by evaporation under reduced
pressure. The nitroaldol thus obtained as a mixture of
diastereomers was dissolved in CH2Cl2 (20 mL) and cooled to
-70 °C. MsCl (0.9 mL, 12 mmol) was added dropwise followed
by a solution of N,N-diisopropylethylamine (4.25 mL, 25 mmol)
in CH2Cl2 (5 mL), keeping the reaction mixture below -60 °C.
The mixture was stirred at -70 °C for 2.5 h and then allowed
to reach room temperature. The solution was washed with
water (5 mL), HCl 1 N (4 × 5 mL), and brine (5 mL), dried,
and evaporated. The product was purified by flash column
chromatography, yielding 75% of a pale yellow oil: bp 92-94
(2S*,3R*,4S*,5S*)-2-(Met h oxyca r b on yl-3-(4-m et h oxy-
p h en yl)-4-n itr o-5-p h en ylp yr r olid in e (4b-en d o isom er ):
1
mp 99-100 °C; IR 3362, 1739, 1533, 1364 cm-1; H NMR (δ
ppm, CDCl3) 5.23 (dd 1H, J ) 4.4 Hz, J ′ ) 2.3 Hz), 4.89 (sb,
1H), 4.17-4.08 (m 2H), 3.81 (s, 3H), 3.80 (s, 3H), 3.34 (sb, 1H);
13C NMR(δ ppm, CDCl3) 171.8, 159.2, 134.4, 130.4, 128.6,
128.5, 126.4, 114.5, 97.1, 67.5, 67.4, 55.2, 54.7, 52.6. Anal.
Calcd for C19H20O5N2: C, 64.02; H, 5.67; N, 7.86. Found: C,
63.91; H, 5.48; N, 7.92.
(2S*,3S*,4R*,5S*)-3-(4-Meth oxyp h en yl)-2-(m eth oxyca r -
bon yl)-4-n itr o-5-p h en ylp yr r olid in e (4b-exo isom er ): mp
140-142 °C; IR 3328, 1730, 1542, 1359 cm-1; 1H NMR (δ ppm,
CDCl3) 5.16 (t, 1H J ) 8.0 Hz), 4.74 (db 1H J ) 7.9 Hz), 4.46
(d, 1H, J ) 9.0 Hz), 4.32 t, 1H, J ) 8.0 Hz), 3.77 (s, 3H), 3.34
(s, 3H), 2.72 (sb, 1H); 13C NMR (δ ppm, CDCl3) 172.0, 159.3,
137.7, 129.1, 129.0, 128.7, 127.7, 126.8, 114.1, 95.3, 67.4, 64.2,
55.2, 53.2, 51.9. Anal. Calcd for C19H20O5N2: C, 64.02; H,
5.67; N, 7.86. Found: C, 64.96; H, 5.75; N, 7.98.
1
°C (0.04 mmHg); IR (film) 1523, 1351 cm-1; H NMR (δ ppm,
CDCl3) 7.19-7.17 (m, 2H), 4.61, (s, 2H), 4.32 (qd, 1H, J ) 6.6
Hz, J ′ ) 3.3 Hz), 1.39 (d, 3H, J ) 6.6 Hz); 13C NMR (δ ppm,
CDCl3) 142.8, 139.3, 137.3, 128.4, 127.8, 127.4, 70.9, 19.9; [R]25
) -39.11 (c ) 1.23, CH2Cl2).
D
Gen er a l P r oced u r e for th e Rea ction of th e Im in es 1
a n d th e Nitr oa lk en es 2. Meth od A. As a typical example
the reaction between imine 1a and nitroalkene 2a is described
below: The imine 1a (0.89 g, 5 mmol) was solved in CH3CN
(50 mL), and then NEt3 (0.7 mL, 5 mmol), the nitroalkene 2a
(0.92 g, 5 mmol), and AgOAc (0.13 g, 0.75 mmol) were added.
The progress of the reaction was monitored by TLC. After
completion of the reaction, the mixture was filtered through
a Celite pad and washed with NH4Cl saturated water solution
(2 × 10 mL) and water (2 × 10 mL). After drying, the solution
was evaporated, and the crude mixture separated by flash
chromatography to give both endo and exo isomers which were
crystallized separately in EtOH.
Meth od B. As a typical example the reaction between
imine 1c and nitroalkene 2b is described below: The imine
1c (0.97 g, 5 mmol) was solved in CH3CN (25 mL), and then
NEt3 (1.4 mL, 10 mmol), the nitroalkene 2b (0.90 g, 5 mmol),
and LiClO4 (2.66 g, 25 mmol) were added. The progress of
the reaction was monitored by TLC. After completion of the
reaction, the mixture was washed with NH4Cl-saturated water
solution (2 × 10 mL) and water (4 × 5 mL). The residue was
dried and evaporated. The crude mixture was triturated in
Et2O to give a white powder as a mixture of endo and exo
isomers of 4d which were separated by fractional crystalliza-
tion in EtOH.
(2S *,3S *,4R *,5S *)-3-(4-Ch lor op h e n yl)-5-(2-h yd r oxy-
p h en yl)-2-(m eth oxyca r bon yl)-4-n itr op yr r olid in e (4c-exo
isom er ): mp 177-179 °C; IR 3270, 1737, 1545, 1355 cm-1
;
1H NMR (δ ppm, CDCl3) 10.07 (sb, 1H), 5.45 (t, 1H, J ) 9.3
Hz), 4.93 (db, 1H, J ) 9.0 Hz), 4.56 (db, 1H, J ) 10.4 Hz), 4.41
(t, 1H, J ) 10.1 Hz, 3.29 (s, 3H), 3.00 (sb, 1H); 13C NMR (δ
ppm, CDCl3) 171.6, 156.8, 134.6, 132.3, 130.7, 129.4, 129.1,
129.0, 119.8, 119.7, 118.0, 90.6, 66.6, 61.9, 52.3, 51.6. Anal.
Calcd for C18H17N2O5Cl: C,57.34; H, 4.56; N, 7.44. Found: C,
56.98; H, 4.67; N, 7.39.
(2S*,3S*,4R*,5S*)-5-(2-Hyd r oxyp h en yl)-2-(m eth oxyca r -
bon yl)-3-(4-m et h oxyp h en yl)-4-n it r op yr r olid in e (4d -exo
isom er ): mp 210 °C (dec); IR 2902, 1722, 1545, 1350 cm-1
;
1H NMR (δ ppm, CDCl3) 10.18 (sb, 1H), 5.48 (t, 1H, J ) 9.3
Hz), 4.93 (d, 1H, J ) 9.2 Hz), 4.53 (d, 1H, J ) 10.5 Hz), 4.40
(t, 1H, J ) 10.2 Hz), 3.77 (s, 3H), 3.31 (s, 3H), 2.97 (sb, 1H);
13C NMR (δ ppm, CDCl3) 171.9, 159.6, 156.9, 130.6, 129.2,
125.4, 128.9, 119.8, 119.7, 118.0, 90.9, 66.7, 62.1, 55.3, 52.2,
51.8. Anal. Calcd for C19H2N2O6: C, 61.27; H, 5.42; N, 7.52.
Found: C, 61.96; H, 5.39; N, 7.57.
(2S *,3R *,4S *,5S *)-5-(2-H yd r oxyp h e n yl)-2-m e t h yl-2-
(m eth oxyca r bon yl)-3-(4-m eth oxyp h en yl)-4-n itr op yr r oli-
d in e (4e-en d o isom er ): mp 161-163 °C; IR 3312, 1728, 1512,
1342 cm-1; 1H NMR (δ ppm, CDCl3) 10.63 (sb 1H), 5.36 (t, 1H,
J ) 8.8 Hz), 4.82 (db, 1H, J ) 8.8 Hz), 4.32 (d, 1H, J ) 8.8
Hz), 3.91 (s, 3H), 3.80 (s, 3H), 3.77 (sb, 1H), 1.09 (s, 3H); 13C
NMR (δ ppm, CDCl3) 176.0, 159.6, 157.1, 130.3, 130.1, 128.9,
126.6, 119.9, 117.6, 114.3, 93.9, 67.1, 65.7, 55.2, 54.4, 53.5, 24.3.
Anal. Calcd for C20H22N2O6: C, 62.16; H, 5.75; N, 7.25.
Found: C, 63.05; H, 5.82; N, 7.19.
(2S*,3S*,4R*, 5S*)-5-(2-H yd r oxyp h en yl)-2-(m et h oxy-
c a r b o n y l)-2-m e t h y l-3-(4-m e t h o x y p h e n y l)-4-n i t r o -
p yr r olid in e(4e-exo isom er ): mp 164-166 °C; IR 3340, 1723,
1553, 1368 cm-1; 1H NMR (δ ppm, CDCl3) 10.23 (sb, 1H), 5.63
(dd, 1H, J ) 11.2 Hz, J ′ ) 9.4 Hz), 5.00 (dd, 1H, J ) 9.4 Hz,
J ′ ) 5.3 Hz), 3.92 (d, 1H, J ) 11.2 Hz), 3.76 (s, 3H), 3.35 (s,
(39) Pople, J . A.; Schlegel, B.; Krishnan, R.; DeFrees, D. J .; Binkley,
J . S.; Frisch, H.; Whiteside, R.; Hout, R. F., J r.; Hehre, W. J . Int. J .
Quantum Chem. Symp. 1981, 15, 269.
(40) Allinger, N. L. J . Am. Chem. Soc. 1977, 99, 8127.
(41) MacroModel v5.0: Mohamadi, F.; Richards, N. G. S.; Guida,
W. C.; Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson,
T.; Still, W. C. J . Comput. Chem. 1990, 11, 440.
(42) (a) Chang, G.; Guida, W. C.; Still, W. C. J . Am. Chem. Soc. 1989,
111, 4379. (b) Saunders: M.; Houk, K. N.; Wu, Y.-D.; Still, W. C.; Lipton,
M.; Chang, G.; Guida, W. J . Am. Chem. Soc. 1990, 112, 1419.
(43) Haasnoot, C. A. G.; de Leevw, F. A. A. M.; Altona, C. Tetrahe-
dron 1980, 36, 2783.

1804 J . Org. Chem., Vol. 63, No. 6, 1998
Ayerbe et al.
3H), 2.44 (db, 1H), 1.76 (s, 3H); 13C NMR (δ ppm, CDCl3) 173.6,
159.8, 157.0, 130.4, 128.9, 90.4, 67.9, 65.3, 59.9, 55.2, 52.5, 25.1.
Anal. Calcd for C20H22N2O6: C, 62.16; H, 5.75; N, 7.25.
Found: C, 62.45; H, 5.70; N, 7.30.
(2S*,3S*,4R*,5S*)-5-(2-Hyd r oxyp h en yl)-2-(m eth oxyca r -
b on yl)-4-m e t h yl-3-(4-m e t h oxyp h e n yl)-4-n it r op yr r oli-
d in e (4f-en d o isom er ): mp 178-180 °C; IR 3350, 1732, 1529,
1386 cm-1; 1H NMR (δ ppm, CDCl3) 11.05 (sb 1H), 5.41 (d, 1H,
J ) 6.2 Hz), 4.82 (dd, 1H, J ) 10.8 Hz, J ′ ) 3.8 Hz), 3.77 (s,
3H), 3.64 (s, 3H), 3.59 (s, 1H), 3.27 (sb, 1H), 1.28 (s, 3H); 13C
NMR (δ ppm, CDCl3) 171.7, 160.0, 157.9, 130.2, 129.6, 129.3,
123.2, 121.4, 119.7, 118.2, 114.3, 99.4, 70.0, 61.3, 59.7, 55.2,
52.7, 20.6. Anal. Calcd for C20H22N2O6: C, 62.16;H, 5.75; N,
7.25. Found: C, 61.85; H, 5.68; N, 7.31.
(2S*,3S*,4R*,5S*)-5-(2-Hyd r oxyp h en yl)-2-(m eth oxyca r -
b on yl)-4-m e t h yl-3-(4-m e t h oxyp h e n yl)-4-n it r op yr r oli-
d in e (4f-exo isom er ): mp 181-183 °C; IR 3348, 1735, 1530,
1387 cm-1; 1H NMR (δ ppm, CDCl3) 10.69 (sb 1H), 5.21 (d, 1H,
J ) 4.4 Hz), 4.62 (dd, 1H, J ) 9.9 Hz, J ′ ) 4.4 Hz), 4.55 (d,
1H, J ) 9.9 Hz), 3.78 (s, 3H), 3.49 (s, 3H), 3.10 (sb 1H), 1.22
(s, 3H); 13C NMR (δ ppm, CDCl3) 171.4, 159.4, 158.0, 130.7,
130.2, 129.3, 125.9, 118.7, 117.6, 113.9, 98.2, 70.6, 61.9, 55.2,
55.0, 52.2, 18.0. Anal. Calcd for C20H22N2O6: C, 62.16;H, 5.75;
N, 7.25. Found: C, 61.40; H, 5.82; N, 7.14.
(2S*,3S*,4R*,5S*)-2,4-Dim eth yl-2-(m eth oxyca r bon yl)-
3-(4-m eth oxyp h en yl)-5-(2-m eth oxyp h en yl)-4-n itr op yr r o-
lid in e (4k -exo isom er ): mp 160-162 °C; IR 3344, 1725, 1536,
1334 cm-1; 1H NMR (δ ppm, CDCl3) 5.51 (s, 1H), 4.50 (s, 1H),
3.82 (s, 3H), 3.79 (s, 3H), 3.50 (s, 3H), 2.75 (sb, 1H), 1.72 (s,
3H), 1.56 (s, 3H); 13C NMR (δ ppm, CDCl3) 175.9, 159.1, 157.3,
131.7, 129.3, 127.4, 125.0, 120.6, 113.6, 110.2, 103.0, 69.6, 66.3,
63.6, 55.2, 51.9, 25.9, 21.6. Anal. Calcd for C22H26N2O6: C,
63.74; H, 6.34; N, 6.76. Found: C, 63.95; H, 6.16; N, 6.88.
(2S*,3R*,4S*,5S*)-3-(4-Ch lor op h en yl)-2-(m et h oxyca r -
b on yl)-4-n it r o-5-(2-p yr id yl)p yr r olid in e (4l-en d o iso-
1
m er ): mp 144-147 °C; IR 3277, 1742, 1548, 1362 cm-1; H
NMR (δ ppm, CDCl3) 5.38 (dd, 1H, J ) 6.7 Hz, J ′ ) 4.5 Hz),
4.96 (tb, 1H), 4.23 (dd, 1H, J ) 7.4 Hz, J ′ ) 4.5 Hz), 4.10 (tb,
1H), 3.92 (tb, 1H), 3.79 (s, 3H); 13C NMR (δ ppm, CDCl3) 171.4,
154.2, 149.4, 136.8, 134.0, 129.4, 128.9, 123.4, 122.0, 95.9, 67.9,
67.6, 54.5, 52.6. Anal. Calcd for C17H16N3O4Cl: C, 56.43; H,
4.47; N, 11.62. Found: C, 57.25; H, 4.52; N, 11.41.
(2S*,3R*,4S*,5S*)-3-(4-Ch lor op h en yl)-2-(m et h oxyca r -
bon yl)-4-n itr o-5-(2-p yr id yl)p yr r olid in e (4l-exo isom er ):
1
mp 154-156 °C; IR 3290, 1741, 1549, 1364 cm-1; H NMR (δ
ppm, CDCl3) 8.70 (db, 1H), 5.32 (dd, 1H, J ) 7.2 Hz, J ′ ) 6.1
Hz), 4.72 (sb, 1H), 4.52 (sb, 1H), 4.38 (dd, 1H, J ) 8.7 Hz, J ′ )
6.1 Hz), 3.82 (sb, 1H), 3.33 (s, 3H); 13C NMR (δ ppm, CDCl3)
170.2, 155.4, 150.0, 137.2, 135.4, 134.0, 129.6, 128.8, 124.0,
123.2, 95.8, 69.2, 65.6, 54.2, 52.0. Anal. Calcd for C17H16N3O4-
Cl: C, 56.43; H, 4.47; N, 11.62. Found: C, 56.98; H, 4.39; N,
11.28.
(2S*,3R*,4S*,5S*)-2,4-Dim eth yl-5-(2-h yd r oxyp h en yl)-2-
(m eth oxyca r bon yl)-3-(4-m eth oxyp h en yl)-4-n itr op yr r oli-
d in e (4g-en d o isom er ): mp 123-125 °C; IR 3278, 1726, 1588,
1
1377 cm-1; H NMR (δ ppm, CDCl3) 11.08 (sb, 1H), 5.22 (db,
(2S *,3R *,4S *,5S *)-2-(Me t h oxyca r b on yl)-3-(4-m e t h -
oxyp h en yl)-4-n itr o-5-(2-p yr id yl)p yr r olid in e (4m -exo- iso-
1H, J ) 5.9 Hz), 4.54 (s, 1H), 3.91 (db, 1H, J ) 5.9 Hz), 3.82
(s, 3H), 3.80, (s, 3H), 1.36 (s, 3H), 1.26 (s, 3H); 13C NMR (δ
ppm, CDCl3) 176.4, 159.4, 158.1, 132.6, 129.9, 129.2, 125.6,
119.7, 118.7, 117.4, 113.8, 99.5, 69.5, 66.9, 58.7, 55.2, 53.4, 24.2,
18.8. Anal. Calcd for C21H24N2O6: C, 62.98; H, 6.05; N, 7.00.
Found: C, 63.47; H, 6.02; N, 6.91.
1
m er ): mp 112-114 °C; IR 3290, 1742, 1550, 1368 cm-1; H
NMR (δ ppm, CDCl3) 5.32 (t, 1H, J ) 6.5 Hz), 4.72 (d, 1H, J
) 7.5 Hz), 4.50 (d, 1H, J ) 8.7 Hz), 4.37 (dd, 1H, J ) 8.7 Hz,
J ′ ) 6.5 Hz), 3.77 (s, 3H), 3.73 (sb, 1H), 3.32 (s, 3H); 13C NMR
(δ ppm, CDCl3) 170.6, 159.2, 155.6, 150.0, 137.1, 129.3, 128.6,
123.8, 123.1, 114.0, 96.0, 69.1, 65.6, 55.2, 54.4, 51.9. Anal.
Calcd for C18H19N3O5: C, 60.49; H, 5.37; N, 11.76. Found: C,
60.05; H, 5.29; N, 11.95.
(2S*,3S*,4R*,5S*)-2,4-Dim eth yl-5-(2-h yd r oxyp h en yl)-2-
(m eth oxyca r bon yl)-3-(4-m eth oxyp h en yl)-4-n itr op yr r oli-
d in e (4g-exo isom er ): mp 179-180 °C; IR 3316, 1716, 1534,
1
1375 cm-1; H NMR (δ ppm, CDCl3) 11.10 (sb, 1H), 5.38 (sb,
(2S*,3R*,4S*,5S*)-3-(4-Ch lor op h en yl)-5-(4-h yd r oxy-3-
m et h oxyp h en yl)-2-(m et h oxyca r b on yl)-4-n it r op yr r oli-
d in e (4n -exo isom er ): mp 167-169 °C; IR 3310, 1726, 1549,
1363 cm-1; 1H NMR (δ ppm, CDCl3) 5.71 (sb, 1H), 5.09 (t, 1H,
J ) 7.8 Hz), 4.67 (t, 1H, J ) 7.3 Hz), 4.46 (dd, 1H, J ) 5.2 Hz,
J ’ ) 9.1 Hz), 4.33 (t, 1H, J ) 8.3 Hz), 3.91 (s, 3H), 3.34 (s,
3H), 2.64 (tb, 1H); 13C NMR (δ ppm, CDCl3) 171.9, 147.0, 146.2,
134.3, 134.1, 129.4, 129.2, 129.0, 119.7, 114.8, 109.3, 94.8, 67.4,
63.8, 55.9, 52.7, 52.0. Anal. Calcd for C19H19N2O6Cl: C, 56.09;
H, 4.72; N, 6.89. Found: C, 55.82; H, 4.57; N, 6.71.
(2S*,3R*,4S*,5S*)-5-(4-H yd r oxy-3-m et h oxyp h en yl)-2-
(m eth oxyca r bon yl)-3-(4-m eth oxyp h en yl)-4-n itr op yr r oli-
d in e (4o-exo isom er ): mp 163-164 °C; IR 3312, 1725, 1547,
1364 cm-1; 1H NMR (δ ppm, CDCl3) 5.71 (sb, 1H), 5.15 (t, 1H,
J ) 8.2 Hz), 4.67 (db, 1H), 4.43 (d, 1H, J ) 9.0 Hz), 4.32 (t,
1H, J ) 8.8 Hz), 3.92 (s, 3H), 3.77 (s, 3H), 3.33 (s, 3H), 2.63
(sb, 1H); 13C NMR (δ ppm, CDCl3) 172.4, 159.4, 147.0, 146.1,
129.6, 128.9, 127.3, 119.7, 114.8, 114.1, 109.4, 95.0, 67.5, 64.0,
55.9, 55.2, 53.1, 51.9. Anal. Calcd for C20H22N2O7: C, 59.69;
H, 5.52; N, 6.96. Found: C, 59.22; H, 5.39; N, 6.96.
(2S,3R,4S,5S)-3-(1(S)-Ben zyloxyeth yl)-2-(m eth oxyca r -
bon yl)-4-n itr o-5-p h en ylp yr r olid in e (4p -en d o isom er ): mp
106-108 °C; IR 3309, 1736, 1545, 1378 cm-1; 1H NMR (δ ppm,
CDCl3) 5.39 (dd, 1H, J ) 6.0 Hz, J ’ ) 2.6 Hz), 4.72 (d, 1H, J
) 11.6 Hz), 4.54 (tb, 1H), 4.43, (d, 1H, J ) 11.6 Hz), 3.96 (tb,
1H), 3.88 (dd, 1H, J ) 6.2 Hz, J ’ ) 2.7 Hz), 3.81 (s, 3H), 3.32
(tb, 1H), 2.92 (dt, 1H, J ) 7.5 Hz, J ’)2.7 Hz), 1.35 (d, 3H, J )
6.3 Hz); 13C NMR (δ ppm, CDCl3) 172.0, 137.7, 134.3, 128.6,
128.4, 128.3, 127.9, 126.2, 91.3, 73.2, 70.9, 67.9, 62.5, 57.7, 52.6,
18.2. Anal. Calcd for C21H24N2O5: C, 65.60; H, 6.30; N, 7.29.
Found: C, 66.33; H, 6.52; N, 7.13.
1H), 4.27 (s, 1H), 3.78 (s, 3H), 3.54 (s, 3H), 2.53 (sb, 1H), 1.75
(s, 3H), 1.49 (s, 3H); 13C NMR (δ ppm, CDCl3) 174.2, 159.7,
158.4, 131.3, 130.2, 129.2, 124.5, 119.4, 117.7, 114.1, 113.8,
98.6, 69.0, 67.0, 63.7, 55.2, 52.5, 27.3, 16.2. Anal. Calcd for
C
21H24N2O6: C, 62.98; H, 6.05; N, 7.00. Found: C, 61.35; H,
6.13; N, 6.89.
(2S *,3S *,4R *,5S *)-3-(4-Me t h o x yp h e n y l)-5-(2-m e t h -
oxyp h en yl)-4-n itr op yr r olid in e (4i-exo isom er ): mp 94-
97 °C; IR 3326, 1733, 1548, 1354 cm-1; 1H NMR (δ ppm, CDCl3)
5.21 (dd, 1H, J ) 7.6 Hz, J ′ ) 6.2 Hz), 4.93 (db, 1H, J ) 7.6
Hz), 4.53 (db, 1H, J ) 8.7 Hz), 4.17 (dd, 1H, J ) 8.7 Hz, J ′ )
6.2 Hz), 3.84 (s, 3H), 3.76 (s, 3H), 3.32 (s, 3H), 3.30 (sb, 1H);
13C NMR (δ ppm, CDCl3) 171.2, 159.2, 157.2, 129.8, 129.2,
128.6, 128.4, 125.1, 121.0, 113.9, 110.7, 95.3, 64.5, 55.2, 54.3,
51.7. Anal. Calcd for C20H22N2O6: C, 62.16; H, 5.75; N, 7.25.
Found: C, 61.89; H, 5.70; N, 7.36.
(2S *,3R *,4S *,5S *)-2-(Me t h oxyca r b on yl)-3-(4-m e t h -
oxyp h en yl)-5-(2-m eth oxyp h en yl)-2-m eth yl-4-n itr op yr r o-
lid in e (4j-en d o isom er ): mp 122-123 °C; IR 3342, 1724,
1
1541, 1372 cm-1; H NMR (δ ppm, CDCl3) 5.25 (db, 1H, J )
6.8 Hz), 4.96 (dd, 1H, J ) 6.8 Hz, J ’ ) 4.5 Hz), 4.17 (d, 1H, J
) 4.5 Hz), 3.86 (s, 3H), 3.77 (s, 3H), 3.70 (s, 3H), 2.94 (sb, 1H),
1.01 (s, 3H); 13C NMR (δ ppm, CDCl3) 176.9, 159.0, 157.0,
130.0, 128.9, 127.8, 126.4, 120.6, 113.8, 110.2, 97.0, 68.6, 60.1,
56.8, 55.2, 54.7, 52.9, 23.2. Anal. Calcd for C21H24N2O6: C,
62.98; H, 6.05; N, 7.00. Found: C, 62.19; H, 5.89; N, 7.56.
(2S*,3R*,4S*,5S*)-2,4-Dim eth yl-2-(m eth oxyca r bon yl)-
3-(4-m eth oxyp h en yl)-5-(2-m eth oxyp h en yl)-4-n itr op yr r o-
lid in e (4k -en d o isom er ): mp 132-133 °C; IR 3332, 1731,
1
1510, 1348 cm-1; H NMR (δ ppm, CDCl3) 5.48 (sb 1H), 4.17
(s, 1H), 3.85 (s, 3H), 3.63 (s, 3H), 2.89 (sb, 1H), 1.18 (s, 3H),
0.75 (s, 3H); 13C NMR (δ ppm, CDCl3) 177.7, 158.9, 156.9,
132.6, 128.6, 128.0, 127.1, 126.7, 120.4, 113.2, 109.6, 98.3, 67.7,
63.4, 61.5, 55.1, 54.7, 52.7, 23.7, 20.6. Anal. Calcd for
Isola tion a n d Ch a r a cter iza tion of Com p ou n d 8. Reac-
tion between 1b and 2c was performed in the presence of 0.15
equiv of LiClO4 and following the general procedure. After
16 h of reaction at room temperature, a mixture of cyloadducts
4f and imines 7 was obtained. Purification of the crude
reaction mixture by flash chromatography gave compounds 7
C
22H26N2O6: C, 63.74; H, 6.34; N, 6.76. Found: C, 63.93; H,
6.52; N, 6.98.

Synthesis of Highly Substituted Proline Esters
J . Org. Chem., Vol. 63, No. 6, 1998 1805
as a syrup. To a solution of 7 (0.300 g, 0.8 mmol) in THF (2.4
mL) was added 2 N HCl (0.24 mL), and the resulting mixture
was stirred for 2 h at rt. Usual basic workup and evaporation
of the solvent gave a residue which was diluted with CH2Cl2
(2 mL). To the resulting solution triethylamine (0.06 mL, 0.43
mmol) and 4-bromobenzoyl chloride (0.114 g, 0.52 mmol) were
added, and the resulting mixture was stirred at rt for 7 h.
Usual workup gave a solid residue which was purified by flash
chromatography and crystallization in AcOEt/hexane. Yield:
Ack n ow led gm en t. This work has been supported
by the Gobierno Vasco/Eusko J aurlaritza (Project GV
170.215-EX97/11) and by the Universidad del Pa´ıs
Vasco/Euskal Herriko Unibertsitatea (Project UPV
170.215-EA126/96).
Su p p or tin g In for m a tion Ava ila ble: Table reporting the
main coupling constants in cycloadducts endo- and exo-4. 1H
NMR data of imines 1a -g. Complete characterization data
and experimental procedures for compounds 4h and 5a ,b.
ORTEP plots of the structures of compounds exo-4c and exo-
4f. Cartesian coordinates of all the stationary points men-
tioned in this work (14 pages). This material is contained in
libraries on microfiche, immediately follows this article in the
microfilm version of the journal, and can be ordered from the
ACS; see any current masthead page for ordering information.
6%; mp ) 161-163 °C; IR: 3380, 173, 1674, 1549 cm-1
.
1H
NMR (δ ppm, CDCl3): 6.42 (db, J ) 8.2 Hz); 5.21 (dd, J ) 3
Hz, J ′ ) 8.2 Hz, 1H); 5.02 (dq, 1H, J ) 6.8 Hz, J ′ ) 11 Hz);
4.07 (dd, 1H, J ) 3 Hz, J ′ ) 11 Hz); 3.80 (s, 3H); 3.77 (s, 3H);
1.38 (d, 3H, J ) 6.8 Hz). 13C NMR (δ ppm, CDCl3): 170.7,
166.5, 159.6, 131.3, 129.7, 128.7, 125.5, 114.6, 113.9, 82.9, 55.2,
53.5, 52.9, 50.6, 19.4. Anal. Calcd for C20H21N2O6Br: C, 51.62;
H, 4.56; N, 6.02. Found: C, 52.45; H, 4.54; N, 5.92.
J O971212Q