Regio- and stereoselectivity of captodative olefins in 1,3-dipolar cycloadditions. A DFT/HSAB theory rationale for the observed regiochemistry of nitrones

Article abstract of DOI:10.1021/jo001393n

Captodative olefins 1-acetylvinyl carboxylates proved to be highly regioselective dipolarophiles in 1,3-dipolar cycloadditon to propionitrile oxide, arylphenylnitrile imines, diazoalkanes, and nitrones to yield the corresponding 5-substituted heterocycles

Full text of DOI:10.1021/jo001393n

1252
J . Org. Chem. 2001, 66, 1252-1263
Regio- a n d Ster eoselectivity of Ca p tod a tive Olefin s in 1,3-Dip ola r
Cycloa d d ition s. A DF T/HSAB Th eor y Ra tion a le for th e Obser ved
Regioch em istr y of Nitr on es
Rafael Herrera,† Arumugam Nagarajan,†,‡ Miguel A. Morales,§ Francisco Me´ndez,*,§
Hugo A. J ime´nez-Va´zquez,† L. Gerardo Zepeda,† and J oaqu´ın Tamariz*,†
Departamento de Quı´mica Orga´nica, Escuela Nacional de Ciencias Biolo´gicas, IPN, Prol. Carpio y Plan
de Ayala, 11340 Me´xico, D.F., and Departamento de Quı´mica, Divisio´n de Ciencias Ba´sicas e Ingenier´ıa,
Universidad Auto´noma Metropolitana-Iztapalapa, A.P. 55-534, 09340 Me´xico, D.F., Me´xico
jtamariz@woodward.encb.ipn.mx
Received September 19, 2000
Captodative olefins 1-acetylvinyl carboxylates proved to be highly regioselective dipolarophiles in
1,3-dipolar cycloadditon to propionitrile oxide, arylphenylnitrile imines, diazoalkanes, and nitrones
to yield the corresponding 5-substituted heterocycles. The addition of the latter was also
stereoselective, being slightly susceptible to steric demand of the carboxylate substituent in the
olefin. All atempts to cleave the isoxazolidine N-O bond under reductive conditions failed, providing
diverse products with side-group reduction. FMO theory was unsuccessful to explain the regio-
selectivity observed with nitrones, since the opposite orientation was predicted. The recently
formulated DFT/HSAB theoretical model was able to rationalize this regioselectivity, identifying
the nucleophilic and electrophilic atoms involved in the process via calculation of interaction
energies, suggesting the specific direction of the electronic process at each of the reaction sites.
In tr od u ction
variety of olefins, providing also the pyrazole substituted
on the C-5 position.4 Similar regioselectivity was ob-
served when diarylnitrile imines were used.5 In contrast,
nitrones react with olefins bearing strong electron-
withdrawing substituents, such as the nitro group, to give
the C-4 substituted isoxazolidines, while the C-5 regio-
isomer was preferred when electron-releasing and mod-
erate electron-withdrawing substituted olefins were
added.3a,6 Addition to 1,2-disubstituted olefins has proven
to be regioselective as well, but they were restricted to
have reverse electron-demand substituents.7 Although
some studies have been devoted to captodative olefins in
1,3-dipolar additions,8 for most of the reports, they have
been only treated as occasional examples.6h,9 However,
they have attracted especial interest in Diels-Alder
1,3-Dipolar cycloaddition is among the most important
pericyclic reactions, due to the versatile and straight-
forward way to generate a great number of hetero-
cycles, and to the relevance of the mechanistic aspects
involved in these processes.1 Multiple substituted olefins
with diverse electron-demand substituents have been
extensively studied in order to determine the factors
which control the regio- and stereoselectivity in these
cycloadditions.2 Thus, it has been established that in
the addition of nitrile oxides to monosubstituted olefins,
regardless of the electron-demand of the substituent,
the 5-susbtituted isoxazoles are preferentially obtained.3
Diazo compounds undergo 1,3-dipolar addition with a
* Corresponding authors. Dr. J oaqu´ın Tamariz: Phone and Fax:
(+525) 586-6621. Dr. Francisco Me´ndez: e-mail: fm@xanum.uam.mx.
† Escuela Nacional de Ciencias Biolo´gicas.
(4) Sustmann, R.; Sicking, W. Chem. Ber. 1987, 120, 1653, and
references therein.
(5) Bianchi, G.; Gandolfi, R.; De Micheli, C. J . Chem. Res. (S) 1981,
6.
‡ Present address: Department of Chemistry, Kangwon National
University, Chun-Cheon, 200-701, Republic of Korea.
§ Universidad Auto´noma Metropolitana-Iztapalapa.
(6) (a) Padwa, A.; Fisera, L.; Koehler, K. F.; Rodriguez, A.; Wong,
G. S. K. J . Org. Chem. 1984, 49, 276. (b) Huisgen, R. Angew. Chem.,
Int. Ed. Engl. 1963, 2, 633. (c) Tice, C. M.; Ganem, B. J . Org. Chem.
1983, 48, 5048. (d) David, D. M.; Bakavoli, M.; Pyne, S. G.; Skelton,
B. W.; White, A. H. Tetrahedron 1995, 51, 12393. (e) Gilbertson, S. R.;
Dawson, D. P.; Lopez, O. D.; Marshall, K. L. J . Am. Chem. Soc. 1995,
117, 4431. (f) DeShong, P.; Dicken, C. M.; Leginus, J . M.; Whittle, R.
R. J . Am. Chem. Soc. 1984, 106, 5598. (g) Belzecki, C.; Panfil, I. J .
Org. Chem. 1979, 44, 1212. (h) Keirs, D.; Moffat, D.; Overton, K.;
Tomanek, R. J . Chem. Soc., Perkin Trans. 1 1991, 1041.
(7) (a) Camiletti, C.; Dhavale, D. D.; Gentilucci, L.; Trombini, C. J .
Chem. Soc., Perkin Trans. 1 1993, 3157. (b) Fro¨hlich, J .; Fisera, L.;
Sauter, F.; Feng, Y.; Ertl, P. Monatsh. Chem. 1995, 126, 75. (c) Fisera,
L.; Sauter, F.; Fro¨hlich, J .; Feng, Y.; Ertl, P.; Mereiter, K. Monatsh.
Chem. 1994, 125, 553.
(8) (a) Do¨pp, D.; Henseleit, M. Chem. Ber. 1982, 115, 798. (b) Fisera,
L.; Konop´ıkova´, M.; Ertl, P.; Pro´nayova´, N. Monatsh. Chem. 1994, 125,
301. (c) Pyne, S. G.; Safaei-G., J .; Skelton, B. W.; White, A. H. Aust. J .
Chem. 1995, 48, 1511. (d) Do¨pp, D.; Walter, J . Heterocycles 1983, 20,
1055. (e) Kelly-Basetti, B. M.; Mackay, M. F.; Pereira, S. M.; Savage,
G. P.; Simpson, G. W. Heterocycles 1994, 37, 529. (f) Pereira, S. M.;
Savage, G. P.; Simpson, G. W.; Greenwood, R. J .; Mackay, M. F. Aust.
J . Chem. 1993, 46, 1401.
(1) (a) 1,3-Dipolar Cycloaddition Chemistry; Padwa, A., Ed.; Wiley-
Interscience: New York, 1984. (b) Carruthers, W. Cycloaddition
Reactions in Organic Synthesis; Pergamon Press: Oxford, 1990; pp
269-331. (c) Torssell, K. B. G. Nitrile Oxides, Nitrones, and Nitronates
in Organic Synthesis; VCH: New York, 1988. (d) Gru¨nanger, P.; Vita-
Finzi, P. Isoxazoles. In The Chemistry of Heterocyclic Compounds;
Taylor, E. C., Ed.; J ohn Wiley & Sons: New York, 1991; Vol. 49, part
1. (e) Coss´ıo, F. P.; Morao, I.; J iao, H.; Schleyer, P. v. R. J . Am. Chem.
Soc. 1999, 121, 6737.
(2) For some recent examples: (a) Borrachero, P.; Cabrera, F.;
Dia´nez, M. J .; Estrada, M. D.; Go´mez-Guille´n, M.; Lo´pez-Castro, A.;
Moreno, J . M.; de Paz, J . L.; Pe´rez-Garrido, S. Tetrahedron: Asymmetry
1999, 10, 77. (b) Bayo´n, P.; de March, P.; Figueredo, M.; Font, J .
Tetrahedron 1998, 54, 15691. (c) Taborski, W.; Bodura, A.; Baranski,
A. Chem. Heterocyclic Compd. 1998, 34, 339. (d) Sanchez-Blanco, A.
I.; Gothelf, K. V.; J ørgensen, K. A. Tetrahedron Lett. 1997, 38, 7923.
(e) Giera, H.; Huisgen, R. Liebigs Ann. 1997, 1685.
(3) (a) Houk, K. N.; Chang, Y.-M.; Strozier, R. W.; Caramella, P.
Heterocycles 1977, 7, 793, and references therein. (b) Martin, S. F.;
Dupre, B. Tetrahedron Lett. 1983, 24, 1337. (c) Boa, A. N.; Booth, S.
E.; Dawkins, D. A.; J enkins, P. R.; Fawcett, J .; Russell, D. R. J . Chem.
Soc., Perkin Trans. 1 1993, 1277.
10.1021/jo001393n CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/27/2001

Captodative Olefins
J . Org. Chem., Vol. 66, No. 4, 2001 1253
Sch em e 1
reactions, because of the opposite electronic effect dis-
played by their geminally substituted functional groups.10
Regioselectivity in 1,3-dipolar cycloadditions has been
traditionally explained in terms of FMO theory, providing
a quite successful agreement with the experimental
result.1b,3a,11 Indeed, the prediction of monosubstituted
olefins with electron-withdrawing or electron-donating
groups appears to match very well, assuming a normal
and reverse electron-demand FMO interaction, respec-
tively.12 Several explanations have been furnished for the
unreliable predictions, including repulsive secondary
orbital interactions,13 and closed-shell repulsions.14 The
high level ab initio calculations of transition structures
and activation parameters of the regioisomeric ap-
proaches have satisfactorily accounted for the 1,3-dipolar
additions between dipoles and dipolarophiles with diverse
electron-demand.1e,15 A less predictable behavior may be
ascribed to captodative olefins, owing to the antagonic
electronic effect upon the double bond, and to the steric
hindrance generated around the disubstituted olefinic
carbon. Indeed, FMO theory fails to rationalize the
orientation observed with some captodative and disub-
stituted olefins.1d,4,8b,9a,16 Steric effects have been invoked
as controlling the cycloaddition.17 The presence of a
radicaloid or dipolaroid transition state generated by
these strongly polarized dipoles at the ground state has
also been suggested.10a,18
nitrones 6 (Scheme 1).20 In all cases, the reactions
were highly regioselective, yielding the C-5 substituted
heterocycle. For dipoles 2-5a , the aromatic products
were isolated as a consequence of the elimination of the
aroyloxy group, in contrast with nitrones for which the
adducts were stable, giving the endo stereoisomers
preferentially. Because FMO theory failed to account for
the regiochemistry observed in the additions with benzo-
nitrile oxide, an alternative novel theoretical model was
investigated, including density functional theory (DFT)
and the hard and soft acids and bases principle (HSAB),
proving to be in agreement with the experimental
results.21
With the aim to improve our knowledge of the factors
which govern the reactivity and selectivity of captodative
olefins in cycloadditon reactions, we herein describe an
extensive study of regioselectivity and stereoselectivity
of 1,3-dipolar additions of dienophiles 1 to diverse dipoles.
We also disclose full details about our calculations which
rationalizes the reactivity and regiochemistry found with
nitrones.
Recently, we reported the 1,3-dipolar cycloaddition of
captodative 1-acetylvinyl carboxylates 1 with arylnitrile
oxides 2,19 and preliminary results with propionitrile
oxide (3), diazoalkanes 4, diphenylnitrile imine (5a ), and
(9) (a) Padwa, A.; Kline, D. N.; Koehler, K. F.; Matzinger, M.;
Venkatramanan, M. K. J . Org. Chem. 1987, 52, 3909. (b) Yang, S.;
Hayden, W.; Griengl, H. Monatsh. Chem. 1994, 125, 469. (c) Bun˜uel,
E.; Cativiela, C.; Diaz-de-Villegas, M. D.; J imenez, A. I. Synlett 1992,
579. (d) Srivastava, V. P.; Roberts, M.; Holmes, T.; Stammer, C. H. J .
Org. Chem. 1989, 54, 5866. (e) Annunziata, R.; Benaglia, M.; Cinquini,
M.; Cozzi, F.; Raimondi, L. J . Org. Chem. 1995, 60, 4697. (f) DeShong,
P.; Dicken, C. M.; Staib, R. R.; Freyer, A. J .; Weinreb, S. M. J . Org.
Chem. 1982, 47, 4397. (g) Vasella, A. Helv. Chim. Acta 1977, 60, 1273.
(h) Aizikovich, A. Ya.; Korotaev, V. Yu.; Kodess, M. I.; Barkov, A. Yu.
Russ. J . Org. Chem. 1998, 34, 351. (i) Iwakura, Y.; Uno, K.; Shiraishi,
S.; Hongu, T. Bull. Chem. Soc. J pn. 1968, 41, 2954. (k) Wade, P. A.;
Hinney, H. R. Tetrahedron Lett. 1979, 139. (l) Palmer, J .; Rutledge, P.
S.; Woodgate, P. D. Heterocycles 1977, 7, 201.
(10) (a) Viehe, H. G.; J anousek, Z.; Mere´nyi, R.; Stella, L. Acc. Chem.
Res. 1985, 18, 148. (b) Odenkirk, W.; Rheingold, A. L.; Bosnich, B. J .
Am. Chem. Soc. 1992, 114, 6392. (c) Boucher, J .-L.; Stella, L.
Tetrahedron 1986, 42, 3871. (d) Seerden, J .-P. G.; Scheeren, H. W.
Tetrahedron Lett. 1993, 34, 2669. (e) Maruoka, K.; Imoto, H.; Yama-
moto, H. J . Am. Chem. Soc. 1994, 116, 12115. (f) Reyes, A.; Aguilar,
R.; Mun˜oz, A. H.; Zwick, J .-C.; Rubio, M.; Escobar, J .-L.; Soriano, M.;
Toscano, R.; Tamariz, J . J . Org. Chem. 1990, 55, 1024.
(11) (a) Huisgen, R. In 1,3-Dipolar Cycloaddition Chemistry; Padwa,
A., Ed.; Wiley-Interscience: New York, 1984; Vol. 1. (b) J aroskova´, L.;
Fisera, L.; Matejkova´, I.; Ertl, P.; Pro´nayova´, N. Monatsh. Chem. 1994,
125, 1413. (c) Houk, K. N.; Sims, J .; Duke, J r., R. E.; Strozier, R. W.;
George, J . K. J . Am. Chem. Soc. 1973, 95, 7287.
(12) Fleming, I. Frontier Orbitals and Organic Chemical Reactions;
J ohn Wiley & Sons: New York, 1976; pp 148-161.
(13) Houk, K. N. Theory of 1,3-Dipolar Cycloadditions. In 1,3-Dipolar
Cycloaddition Chemistry; Padwa, A., Ed.; Wiley: New York, 1984; Vol.
2.
(14) Sustmann, R.; Sicking, W.; Felderhoff, M. Tetrahedron 1990,
46, 783.
(15) Rastelli, A.; Gandolfi, R.; Amade´, M. S. J . Org. Chem. 1998,
63, 7425.
(16) Shanmugasundaram, M.; Raghunathan, R.; Padma Malar, E.
J . Heteroatom Chem. 1998, 9, 517.
(17) (a) Huisgen, R. Steric Course and Mechanism of 1,3-Dipolar
Cycloadditions. In Advances in Cycloaddition; Curran, D. P., Ed.; J AI
Press: Greenwich, 1988; Vol. 1, pp 1-31. (b) For a recent study, see:
Weidner-Wells, M. A.; Fraga-Spano, S. A.; Turchi, I. J . J . Org. Chem.
1998, 63, 6319.
Resu lts
The 1,3-dipolar cycloadditions between olefin 1a and
C-aryl-N-phenylnitrile imines 5a -d under thermal con-
ditions provided the corresponding 5-acetyl pyrazoles
9a -d in good yields (Scheme 2). The nitrile imines were
prepared in situ by treatment of arylaldehyde phenyl-
hydrazones 7a -d with chloramine-T in dry dioxane
under reflux. These aromatic heterocycles might have
been formed as a result of â-elimination of the p-nitro-
benzoyloxy group (PNB) from the first formed cyclo-
adducts 8a -d . No evidence of the latter was found either
by chromatography or by NMR, even after stopping the
reaction before the dipolarophile had disappeared. The
stabilization of the adducts by aromatization of the ring
could be the driving force to give the pyrazoles.
(19) J ime´nez, R.; Pe´rez, L.; Tamariz, J .; Salgado, H. Heterocycles
1993, 35, 591.
(20) Nagarajan, A.; Zepeda, G.; Tamariz, J . Tetrahedron Lett. 1996,
(18) (a) Baran, J .; Mayr, H. J . Org. Chem. 1989, 54, 5774. (b) Howe,
R. K.; Shelton, B. R. J . Org. Chem. 1990, 55, 4603. See also: Firestone,
R. A. Tetrahedron 1977, 33, 3009.
37, 6835.
(21) Me´ndez, F.; Tamariz, J .; Geerlings, P. J . Phys. Chem. A 1998,
102, 6292.

1254 J . Org. Chem., Vol. 66, No. 4, 2001
Herrera et al.
Sch em e 2
F igu r e 1. Steric interactions between the N-phenyl and
aroyloxy groups at the exo transition state.
Ta ble 1. 1,3-Dip ola r Ad d ition s of 1c w ith Dip oles
3 a n d 5a a
dipole
(mol equiv)
yield
(%)b
solvent
T (°C)
t (h)
product
3 (6.0)
5a (3.0)
C6H6
dioxane
80
90
2.0
2.0
12
9a
63
60
Sch em e 3
a
All under N2 atmosphere. b After column chromatography and/
or recrystallization.
It has been suggested that steric interactions are
responsible for the regioselectivity observed in many 1,3-
dipolar additions.16,17 This prompted us to examine the
steric effect of the PNB group on the control of the
regiochemistry observed with the aforementioned dipoles.
The aromatic ring could give rise to strong nonbonding
interactions, since the X-ray structure and the gas-phase
calculations of 1a have shown that the PNB group is out
of the plane formed by the enone π system.22 Therefore,
the sterically less hindered captodative olefin 1c was
evaluated, keeping the potential electron-donor effect of
the carboxylate group on the olefin. Table 1 summarizes
the results obtained in the cycloaddition reactions with
dipoles 3 and 5a . As observed with olefin 1a , the aromatic
heterocycles were isolated and they showed the same
regiochemistry.
On the other hand, no traces of the 4-acetyl regio-
isomers were detected in the crude mixtures by 1H
NMR. The structure of pyrazoles 9 was confirmed by
NOE experiments, since enhancements of the signals of
aromatic protons of the C-3 aryl group and acetyl protons
were observed when proton H-4 was irradiated.
The cycloaddition of 1a with diazoalkanes such as
diazomethane (4a ) took place efficiently at room temper-
ature to yield only pyrazole 10a (Scheme 3). In contrast,
with trimethylsilyl diazomethane (4b), a mixture of
stereoisomeric adducts 11a /11b (79:21), along with prod-
uct 10a in a ca. 1:1 ratio, were isolated. Interestingly,
the expected silylated pyrazole 10b was not detected.
Adducts 11a /11b seem to be precursors of pyrazole 10a ,
as suggested by the transformation into this pyrazole
(80%) by thermal (50 °C) treatment of 11a /11b. Ad-
ditional support for the above was obtained from the
reaction of dipole 4b with 1a at 50 °C for 1 h, to give the
expected single product 10a .
Thus, this reaction gives evidence that the cycloadduct
is the primary product formed in the reaction, and then
it is transformed into the more stable aromatic product.
A similar event might be taking place in the reaction with
the other dipoles which give rise to aromatic heterocycles,
including nitrile imines and nitrile oxides.
Similar results were obtained when olefins 1a and 1c
were added to N-phenylnitrones 6a -g: only the C-5
substituted isoxazolidines were observed as a mixture of
stereoisomers 13/14 and 15/16, respectively (Table 2).
However, comparatively, the endo stereoselectivity was
higher in the case of olefin 1a . Despite the predominance
of one or another diastereoisomer depending on a subtle
interplay of several steric and electronic factors,1d
this
endo preference of 1a could be rather associated with
stronger steric interactions leading to an exo transition
state between the N-phenyl group of the nitrone and the
PNB group in the dipolarophile (Figure 1). That steric
interactions would be controlling the endo preference in
the additions might be supported by the fact that no
traces of isomers exo 14h and 16h were detected when
the more crowded nitrone 6h was used with both olefins
1a and 1c (Table 2, entries 8 and 13).23 Therefore, the
presence of the bulky N-tert-butyl group influences the
distribution of diastereomers, although it does not affect
the regiochemistry of the cycloaddition.
Addition of the aliphatic nitrile oxide 3 (C6H6, reflux,
2 h) to olefin 1a furnished the single isoxazole 12 in 85%
yield (eq 1). The formation of the heterocycle and the
regiochemistry agreed with analogous results observed
for the reaction of arylnitrile oxides 2 with olefin 1a .19
Additions of olefin 1c to nitrones 6c and 6f were,
however, slower than those with olefin 1a (Table 2,
entries 11 and 12), correlating with their relative reac-
(22) J ime´nez-Va´zquez, H. A.; Ochoa, M. E.; Zepeda, G.; Modelli, A.;
J ones, D.; Mendoza, J . A.; Tamariz, J . J . Phys. Chem. A 1997, 101,
10082.
(23) The electrostatic interactions between the lone-pair electrons
on the oxygen of the electron-donating group, and the dipole may be
involved at the transition state, affecting the stereoselectivity, as
recently estimated for enol ethers: Liu, J .; Niwayama, S.; You, Y.;
Houk, K. N. J . Org. Chem. 1998, 63, 1064.

Captodative Olefins
J . Org. Chem., Vol. 66, No. 4, 2001 1255
Ta ble 2. 1,3-Dip ola r Ad d ition s of Olefin s 1a a n d 1c w ith Nitr on es 6a -6ja
entry
olefinb
nitrone
R′
Ph
Ph
Ph
Ph
Ph
Ph
Ph
t-Bu
Bn
Bn
Ph
Ph
t-Bu
Y
t (h)
products (ratio)c
yield (%)d
1
2
3
4
5
6
7
8
9
10
11
12
13
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1c
1c
1c
6a
6b
6c
6d
6e
6f
6g
6h
6i
H
8
8
8
8
8
13a /14a (89:11)
13b/14b (90:10)
13c/14c (87:13)
13d /14d (85:15)
13e/14e (83:17)
13f/14f (95:5)
13 g/14g (89:11)
13h /14h (100:0)
13i/14i (85:15)
13j/14j (60:40)
15a /16a (73:27)
15b/16b (82:18)
15c/16c (100:0)
70
75
70
70
80
80
75
44
50
30
70
50
45
p-Cl
p-Br
p-NO2
p-MeO
p-Me
m-NO2
H
p-MeO
p-Me
p-Br
8
8
24
24
24
12
12
12
6j
6c
6f
6h
p-Me
H
a
b
All under N2 atmosphere, in dry benzene under reflux. 1.0 mol equiv of 1a , and 1.5 mol equiv of 1c. c Determined by 1H
d
NMR (300 MHz) from the crude. Of the major isomer after column chromatography and recrystallization.
tivities.24 An analogous trend is noticed with N-benzyl-
nitrones 6i and 6j in comparison with N-phenylnitrones,
since the former reacted with olefin 1a at longer reaction
times.
It has been well established that nitrones substituted
with bulky groups possess the Z configuration, with
N-alkyl or N-phenyl and C-aryl groups in a trans
relationship.9a Nevertheless, we have also examined the
stereochemistry of some of the nitrones used in this
study. X-ray crystal structures of 6e and 6k (N-benzyl-
C-phenylnitrone),25 and NOE experiments of the less
studied nitrones 6i-k , confirmed that in the solid phase
and in solution they show the Z configuration. For this
F igu r e 2. NOE observed upon irradiation of protons in 13b
and 13c.
reason, we assumed that the same configuration is
maintained at the transition state, and they do not
isomerize during the course of the cycloaddition.9f
A
signals attributed to the protons of the C-3 aromatic ring
and, in lesser extent, of the aromatic protons of the
N-phenyl group. This group appears to be oriented in the
R position with an anti relationship with respect to the
aryl group on C-3. This relative configuration also agrees
with the NOE effect observed in the signals of these
protons when the acetyl group was irradiated. A further
support for the above was the observable enhancement
of the protons of the p-nitrobenzoyloxy group by irradia-
tion of H-4â.
This assignment was confirmed by X-ray crystallog-
raphy of isoxazolines 13h and 14j,25 which were crystal-
lized from a mixture of hexane/EtOAc (7:3) (Figure 3).
In the adduct 13h , the heterocycle exhibits a conforma-
tion with the C-3 aryl ring and PNB group in pseudoaxial
positions, leaving the acetyl group in a pseudoequatorial
conformation. A similar conformation is displayed in the
crystal structure of 14j, the PNB group shows a quasi-
axial conformation, leaving the p-tolyl ring in the pseudo-
equatorial position. This preference of the PNB group for
the pseudoaxial conformation (as previously observed in
the half-chair conformation of cyclohexene adducts of
dienes with olefin 1a 10d,26) could be associated with the
further interesting feature to notice in the X-ray struc-
ture of 6e is that the N-phenyl and C-anisyl groups
present conformations slightly out of the plane formed
by the dipole. Accordingly, the steric interactions illus-
trated in Figure 1 become significant if the conformation
of the N-phenyl group of the nitrone is retained at the
transition state. An analogous argument could be invoked
for the low endo selectivity recorded in the reaction of
nitrone 6j (Table 2, entry 10), considering the motion of
the N-benzyl group around the C-N bond.
The structures of the major isomers were established
1
by H NMR spectroscopy. In the case of isomers 13 and
15, double irradiation, COSY, and HETCOR experiments
were determined in order to correlate the isoxazolidine
ring protons. The multiplicity and coupling constants
were consistent with vicinal CH and methylene groups.
The relative configuration on carbons C-3 and C-5 was
provided by NOE experiments. For example, in adducts
13b and 13c, an enhancement of the signals of H-4R and
the acetyl group on C-5 was observed when H-3 was
irradiated (Figure 2). At the same time, the latter
produced, as expected, a strong NOE enhancement of the
(24) Tamariz, J .; Vogel, P. Helv. Chim. Acta 1981, 64, 188.
(25) The authors have deposited the atomic coordinates for these
structures with the Cambridge Crystallographic Data Centre. The
coordinates can be obtained, on request, from the Director, Cambridge
Crystallographic Centre, 12 Union Road, Cambridge, CB2 1EZ, UK.
(26) Ochoa, M. E. Nuevos Equivalentes de Cetenas en Reacciones
de Diels-Alder. S´ıntesis de γ-Hidroxiciclohexenonas. M.S. Thesis,
Escuela Nacional de Ciencias Biolo´gicas, IPN: Me´xico, D. F., Mexico,
1997.

1256 J . Org. Chem., Vol. 66, No. 4, 2001
Herrera et al.
F igu r e 3. Structures of compounds 13h and 14j. They were constructed from their X-ray crystallographic data.
Sch em e 4
100 °C for 3 h, producing isoxazoline 20 as an oil in 60%
yield (eq 2). This compound has an interesting structure,
which could be considered as an endocyclic â-substituted
captodative olefin.28
Isoxazolidines have been useful synthons for the prep-
aration of, among others, substituted amines, natural
â-amino alcohols, and alkaloids by reductive cleavage of
the N-O bond.1a-d,29 With the aim to assess the synthetic
potential of isoxazolidines 13/14, we intended to cleave
the N-O bond by hydrogenation. Several catalysts were
used with derivatives 13e and 13j, and diverse products
were isolated and characterized, but the heterocycle was
not opened (Table 3). The hydrogenation of 13e with
Raney nickel produced alcohol 21 along with a small
amount of the nitro group reduction product 22; whereas
isoxazolidine 13j underwent reduction with the same
catalyst, but in the presence of a different solvent
(i-PrOH), to yield the product of reduction of the nitro
group 24 (Scheme 5). In contrast, compound 22 was
isolated as the major product by using PtO2 in both
solvents ethyl acetate and ethanol (entries 2 and 3).
Palladium on charcoal (10%) provided diol 23 as the main
product; however, it was unstable under purification
conditions, and it was not fully characterized. Other
catalysts were used such as Pd(OH)2 which has been
efficient for analogous substrates,30 but a mixture of
anomeric effect, as previously indicated in analogous C-5
oxygenated isoxazolidines.9f In both structures, 13h and
14j, the anti relationship between the N-substituent and
the aryl group on C-3 is preferred, as evidenced by the
NOE experiments.
To evaluate the effect of the electron-donating group
on the captodative center of olefins 1 on the control of
both regio- and stereoselectivity, we carried out the
cycloaddition of nitrone 6e with methyl vinyl ketone (17)
(Scheme 4). The reaction took place in benzene at reflux
for 24 h, to give a mixture of the four possible isomers
18a , 18b, 18c, and 18d in ca. 2:10:7:1 ratio.27 The
regioselectivity (acetyl on C-5/C-4, 60:40) was practically
lost in reference to olefins 1, even though the endo
selectivity remained in higher proportion. In accord with
these results, it seems that the aroyloxy and the acetoxy
groups of olefins 1 are playing a significant role in the
orientation of the dipole during the addition.
It is noteworthy that the PNB group in the nitrone
adducts 13/14 did not undergo â-elimination, as found
for the nitrile imines, nitrile oxides, and diazo com-
pounds. This suggests that the presence of a double bond
in the adducts induces the elimination of the carboxylate
group, giving rise to the most stable aromatic compound.
Nevertheless, the PNB group could be eliminated by
treatment of adduct 13e with 2,4,6-collidine (19) at
(28) â-Substituted captodative olefins have been prepared, and they
display interesting structural and reactivity features. See: (a) Peralta,
J .; Bullock, J . P.; Bates, R. W.; Bott, S.; Zepeda, G.; Tamariz, J .
Tetrahedron 1995, 51, 3979. (b) Villar, L.; Bullock, J . P.; Khan, M. M.;
Nagarajan, A.; Bates, R. W.; Bott, S. G.; Zepeda, G.; Delgado, F.;
Tamariz, J . J . Organomet. Chem. 1996, 517, 9.
(29) (a) Curran, D. P. J . Am. Chem. Soc. 1982, 104, 4024. (b) Keck,
G. E.; Fleming, S.; Nickell, D.; Weider, P. Synth. Commun. 1979, 9,
281. (c) Stork, G.; Danishefsky, S.; Ohashi, M. J . Am. Chem. Soc. 1967,
89, 5459. (d) Tufariello, J . J . Acc. Chem. Res. 1979, 12, 396. (e) Gothelf,
K. V.; J ørgensen, K. A. Chem. Rev. 1998, 98, 863. (f) Frederickson, M.
Tetrahedron 1997, 53, 403.
(27) Mixtures of regioisomers have been also obtained in the
presence of 6a with this and other olefins: (a) Asrof Ali, Sk.; Senaratne,
P. A.; Illig, C. R.; Meckler, H.; Tufariello, J . J . Tetrahedron Lett. 1979,
4167. (b) J oucla, M.; Hamelin, J . J . Chem. Res. (S) 1978, 276.

Captodative Olefins
J . Org. Chem., Vol. 66, No. 4, 2001 1257
Ta ble 3. Hyd r ogen a tion of Isoxa zolid in es 13e a n d 13ja
entry
isoxazolidine
catalyst
solvent
pressure (atm)
t (h)
products (ratio)b
yield (%)c
1
2
3
4
5
6
7
13e
13e
13e
13e
13e
13e
13j
Raney Ni
PtO2
PtO2
Pd/C, 10%
Pd(OH)2
Pd/CaCO3
Raney Ni
EtOAc
EtOAc
EtOH
EtOAc
EtOH
1.3
1.3
1.3
1.3
1.3
1.3
2.6
3.0
2.0
3.0
3.0
3.0
3.0
5.0
21/22 (92:8)
79
40
d
e
e
21/22/23 (9:86:5)
21/22/23 (23:69:8)
21/23 (5:95)
21/22/23 (16:8:76)
22/23 (25:75)
24
EtOH
i-PrOH
e
30
a
b
All at room temperature. Determined by 1H NMR (300 MHz). c After column chromatography and recrystallization of the major
product. The mixture of products was obtained in almost quantitative yield, but the major product was not separated. e 23 decomposes
d
under purification conditions.
Ta ble 4. Ab In itio 6-31G* En er gy Ga p s (eV) of F r on tier
Or bita lsa for Olefin 1a a n d Nitr on es 6
Sch em e 5
nitrone
HOMO-LUMO
LUMO-HOMO
diff
6a
6d
6e
6h
6k
10.4187
11.1034
10.0582
10.4781
10.5396
13.0406
11.8756
13.2401
13.6208
13.4788
2.6219
0.7722
3.1819
3.1427
2.9392
a
HOMO-nitrone/LUMO-olefin and LUMO-nitrone/HOMO-
olefin.
captodative effect,10a involving diradicaloid intermediates
or transition states,18 has also been considered in control-
ling the orientation of the cycloaddends.34 Both experi-
mental and theoretical investigations have revealed the
increased importance of electrostatic interactions in
controlling the approach to the transition state:35
A
dipolaroid character of the transition state may agree
with the 5-substituted adducts for the interaction with
nitrile oxides and nitrones as the dipoles, when the
transition states are simulated in diverse polarity
solvents.1e
It is generally agreed that 1,3-dipolar cycloadditions
are concerted,1,34,36 and the regiochemistry appears to
be controlled by frontier orbital interactions.11,29e,32 To
rationalize the exclusive C-5 regioselectivity displayed by
olefins 1 with nitrones 6, in contrast with other capto-
dative olefins,8a we performed an FMO analysis of the
calculated (RHF/3-21G, and 6-31G*) frontier orbitals.37
They show that the interaction between 1a and 6a , 6d ,
6e, 6h , and 6k is governed by the HOMO-dipole/LUMO-
dipolarophile (Table 4). According to FMO theory, regio-
selectivity is controlled by the magnitudes of the atomic
orbital coefficients at the terminal atoms of the dipole,
compounds 21/22/23 (entry 5) was isolated. Despite
hydrogenolysis of the isoxazolidine N-O bond seeming
to be slowed by steric encumbrance,1d the overstability
found in these isoxazolidines could be also due to the
presence of an oxygenated functionality on carbon C-5,
base of the endocyclic oxygen, as previously suggested.31
Discu ssion
The precedent results have shown that captodative
olefins 1 undergo highly regio- and stereoselective cy-
cloaddition reactions with a variety of 1,3-dipoles. No
significant effect on the regioselectivity could be detected
by changing the substituent in the electron-donating
group of the captodative olefin, suggesting a similar
electronic perturbation of this group on the double bond.
Moreover, the regioselectivity observed, furnishing the
C-5 heterocycle substitution in these cycloadditions with
all dipoles and olefins 1, suggests a steric control, keeping
far away the crowded centers of the olefin (the capto-
dative carbon C-3) and the nitrone (the substituted
carbon atom). However, neither the formation of the
less crowded C-5 substituted adducts from 1 could be
explained only by steric interactions, nor the loss of
regiochemistry when adding the noncaptodative olefin
17.32 Mostly electronic demand of the dipolarophile
substituents seems to account for these results.33 The
(33) (a) Sims, J .; Houk, K. N. J . Am. Chem. Soc. 1973, 95, 5798. (b)
Houk, K. N.; Sims, J .; Watts, C. R.; Luskus, L. J . J . Am. Chem. Soc.
1973, 95, 7301. (c) Houk, K. N.; Gonza´lez, J .; Li, Y. Acc. Chem. Res.
1995, 28, 81.
(34) However, NMR observations exclude the presence of diradical
intermediates: Houk, K. N.; Firestone, R. A.; Munchausen, L. L.;
Mueller, P. H.; Arison, B. H.; Garcia, L. A. J . Am. Chem. Soc. 1985,
107, 7227.
(35) (a) Kahn, S. D.; Pau, C. F.; Overman, L. E.; Hehre, W. J . J .
Am. Chem. Soc. 1986, 108, 7381. (b) Rastelli, A.; Bagatti, M.; Gandolfi,
R.; Burdisso, M. J . Chem. Soc., Faraday Trans. 1994, 90, 1077.
(36) (a) Huisgen, R. J . Org. Chem. 1976, 41, 403. (b) Bihlmaier, W.;
Geittner, J .; Huisgen, R.; Reissig, H.-U. Heterocycles 1978, 10, 147.
(c) Huisgen, R. Pure Appl. Chem. 1980, 52, 2283. (d) McDouall, J . J .
W.; Robb, M. A.; Niazi, U.; Bernardi, F.; Schlegel, H. B. J . Am. Chem.
Soc. 1987, 109, 4642.
(30) (a) DeShong, P.; Leginus, J . M. J . Am. Chem. Soc. 1983, 105,
1686.
(31) DeShong, P.; Lander, S. W., J r.; Leginus, J . M.; Dicken, C. M.
Dipolar Cycloadditions of Nitrones with Vinyl Ethers and Silane
Derivatives. In Advances in Cycloaddition; Curran, D. P., Ed.; J AI
Press: Greenwich, 1988; Vol. 1, pp 112-113.
(32) (a) Mullen, G. B.; Bennett, G. A.; Swift, P. A.; Marinyak, D.
M.; Dormer, P. G.; Georgiev, V. St. Liebigs Ann. Chem. 1990, 105. (b)
Confalone, P. N.; Huie, E. M. Org. React. 1988, 36, 1. (c) Bimanad, A.
Z.; Houk, K. N. Tetrahedron Lett. 1983, 24, 435.
(37) Calculated with GAUSSIAN94, Revision E.2: Frisch, M. J .;
Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; J ohnson, B. G.; Robb,
M. A.; Cheeseman, J . R.; Keith, T.; Petersson, G. A.; Montgomery, J .
A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J .
V.; Foresman, J . B.; Cioslowski, J .; Stefanov, B. B.; Nanayakkara, A.;
Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.;
Andres, J . L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J .;
Binkley, J . S.; Defrees, D. J .; Baker, J .; Stewart, J . P.; Head-Gordon,
M.; Gonzalez, C.; Pople, J . A. Gaussian, Inc., Pittsburgh, PA, 1995.
See the Supporting Information for details.

1258 J . Org. Chem., Vol. 66, No. 4, 2001
Herrera et al.
Sch em e 6. F MO Coefficien ts In ter a ction of th e
LUMO of 1a a n d HOMO (estim a ted a ver a ge
va lu es) of Nitr on es 6
chemical potential (µ), and is usually a manifestation of
the maximum hardness principle.44
In this perturbational treatment we can visualize the
various phenomena which occur during the process of
nitrone-olefin bonding, thus gaining more insight into
the factors that govern the reaction. We have recently
used such a method to study the chemical reactivity of
carbenes with olefins.45
For the study of the interaction from a global view-
point, which does not specify the interaction sites (atomic
sites) for nitrone A and olefin B, ∆Eint is given by eq 3.41b
and by those of the double bond of the dipolarophile.11c
Thus, the preferred orientation should be the result of
the interaction of the atoms with the largest coefficients.
For olefin 1a and calculated nitrones 6, the FMO argu-
ments predict the formation of the regioisomer C-4
substituted, in contrast to the observed regiochemistry
(Scheme 6). An analogous prediction is furnished from
the LUMO-HOMO interaction for olefin 17 and nitrones
6, respectively, minimizing the fact that the more abun-
dant C-5 regioisomer was obtained.
Therefore, this model did not provide a satisfactory
explanation of cycloadditions of our molecules with
nitrones, as anticipated from analogous results concern-
ing the use of FMO theory with arylnitrile oxides,19 and
in similar cycloadditions.15,35b,38 Owing to this conclusion,
we investigated a novel theoretical approach which is
based on the formulation of the interaction energy ∆Eint
of the reactants, in terms of density functional theory
(DFT)39 and the hard and soft acids and bases principle
(HSAB),40 for the 1,3-dipolar cycloaddition of benzonitrile
oxide and olefin 1a . The results we obtained provide
a successful rationalization of the experimental regio-
chemistry.21
∆Eint ) ∆Eν + ∆Eµ ≈
(µA - µB)2
1
1
λ
-
SASB -
(3)
2 SA + SB
2 SA + SB
where µA and µB are the electronic chemical potentials
(identified as the negative of the electronegativities øA
and øB)46 and SA and SB are the global softnesses of the
HSAB principle (identified as the inverse of the hard-
nesses ηA and ηB)47 for nitrone and olefin, respectively.
λ is a constant related to an effective number of valence
electrons.41b,44b
From a local viewpoint, nitrone A and olefin B interact
through their C, O, and C1 (unsubstituted carbon), C2
(captodative carbon) atoms, respectively. We assume that
only these atoms participate in the charge transfer and
the reshuffling of the charge distribution involved in the
bond formation processes. Therefore, ∆Eint is given by eq
4.41c,d
∆Eint ) ∆Eν + ∆Eµ ≈
(µA - µB)2
The essence of this treatment of chemical reactivity is
based, as already discussed in previous papers,41 in the
assumption that when molecule A and molecule B
interact with each other in order to form product AB, a
mutual perturbation of the molecular densities of both
reactants occurs. The resulting change in energy for
the reaction A + B f AB may be divided into two steps
1
1
λ
-
sAasBb
-
(4)
2 sAa + sBb
2 sAa + sBb
where sAa and sBb are the condensed softnesses and
characterize the softness of the C (or O) atom in the
nitrone and the C1 (or C2) atom in the olefin, respectively.
According to eq 3, when the first term (-1/2(µA
-
which take place in succession: A + B 9 9
i8 A* + B* ii8 AB,
µB)2SASB/(SA + SB)) becomes predominant, strong electron
transfer occurs between nucleophiles of low electronega-
tivity and electrophiles of high electronegativity, decreas-
ing the ionicity of the reagents. Chemical potential
differences (µA - µB)2 (electronegativity differences) drive
electron transfer, and electrons tend to flow from a region
of high chemical potential (low electronegativity) to a
region of low chemical potential (high electronegativity).
High values of the SASB/(SA + SB) term enhance the
magnitude of the ∆Eν term: high softness values of
nucleophile and electrophile lead to soft-soft interactions
and can be associated with covalent interaction.42 The
same reasoning may be extended to eq 4 at a local level.
Also in accord to eq 3, when the (-λ/(SA + SB) term
becomes predominant, very little electron transfer occurs
between nucleophile and electrophile. The ∆Eµ term is
enhanced by low softness values of the nucleophile and
electrophile, the magnitude of the (SA + SB)-1 term
(i) partial charge-transfer accompanied by soft-soft
interactions (covalent interactions),42 and (ii) electronic
reorganization accompanied by hard-hard interactions
(electrostatic interactions).42
The interaction energy ∆Eint ) ∆Eν + ∆Eµ represents
the energy involved in the two steps. The energy of the
first step, ∆Eν, corresponds to the charge-transfer process
between molecule A and molecule B at constant external
potential (ν(r ), which is generated by the nuclei), and it
is a consequence of the chemical potential equalization
principle.43 The energy of the second step, ∆Eµ, comes
from the reshuffling of charge distribution at constant
(38) Sustmann, R.; Sicking, W. Chem. Ber. 1987, 120, 1471.
(39) (a) Parr, R. G.; Yang, W. Density Functional Theory of Atoms
and Molecules; Oxford University Press: New York, 1989. (b) Parr, R.
G.; Yang, W. Annu. Rev. Phys. Chem. 1995, 46, 701-728.
(40) (a) Pearson, R. G. J . Am. Chem. Soc. 1963, 85, 3533. (b) Pearson,
R. G. J . Chem. Educ. 1987, 64, 561.
(41) (a) Chattaraj, P. K.; Lee, H.; Parr, R. G. J . Am. Chem. Soc. 1991,
113, 1855. (b) Ga´zquez, J . L. Chemical Hardness. In Structure and
Bonding; Sen, K. D., Ed., 1993; Vol. 80, p 27. (c) Me´ndez, F.; Ga´zquez,
J . L. J . Am. Chem. Soc. 1994, 116, 9298. (d) Me´ndez, F.; Ga´zquez, J .
L. Theoretical Models for Structure Properties and Dynamic in
Chemistry. Proc. Indian Acad. Sci., Chem. Sci. 1994, 106, 183.
(42) Klopman, G. J . Am. Chem. Soc. 1968, 90, 223.
(44) (a) Parr, R. G.; Chattaraj, P. K. J . Am. Chem. Soc. 1991, 113,
1854. (b) Parr, R. G.; Ga´zquez, J . L. J . Phys. Chem. 1993, 97, 3939. (c)
Ga´zquez, J . L.; Mart´ınez, A.; Me´ndez, F. J . Phys. Chem. 1993, 97, 4059.
(45) Me´ndez, F.; Garcia-Garibay, M. A. J . Org. Chem. 1999, 64, 7061.
(46) Parr, R. G.; Donnelly, R. A.; Levy, M.; Palke, W. E. J . Chem.
Phys. 1978, 68, 3801.
(43) Sanderson, R. T. Chemical Bond and Bond Energy; Academic
Press: New York, 1976.
(47) Yang, W.; Parr, R. G. Proc. Natl. Acad. Sci. U.S.A. 1985, 82,
6723.

Captodative Olefins
J . Org. Chem., Vol. 66, No. 4, 2001 1259
Ta ble 5. Globa l P r op er ty Va lu es (eV) for Nitr on es 6a ,
increases. All these properties correspond to those as-
sociated with a hard-hard interaction, and the electro-
static effect can be identified with it. Such an effect
reflects an ionic type of interaction and will be predomi-
nant between highly charged species. It leads to a low
covalent interaction.42 Again, the same reasoning may
be extended to eq 4 at a local level.
Once the necessary DFT and HSAB parameters for
nitrones and olefins are known, we can characterize the
global and local interactions using eqs 3 and 4. The
electronic chemical potential (µA and µB) and the global
softness (SA and SB) can be calculated in terms of the
vertical ionization potential (I) and the vertical electron
affinity (A) by the expressions39 µ ) - (I + A)/2, S ) η-1
and η ) (I - A)/2, where η is the global hardness of the
HSAB principle.48
6d -f, a n d 6h -k , a n d Olefin s 1a , 1c, a n d 17
chemical potential
global softness
compd
(µ)a
(S)b
6a
6d
6e
6f
6h
6i
-2.59
-3.12
-2.45
-2.52
-2.25
-2.23
-2.31
-2.36
-4.56
-3.63
-3.17
0.153
0.178
0.154
0.155
0.142
0.143
0.144
0.143
0.107
0.095
0.099
6j
6k
1a c
1c
17
a
b
Calculated as -(I + A)/2. Calculated as S ) η-1 and η )
(I + A)/2, S values in eV-1 c See ref 21.
.
Ta ble 6. Globa l In ter a ction En er gy (∆Ein t) Va lu es
(k J /m ol) for th e P a ir Nitr on e-Olefin , Ca lcu la ted
w ith Eq 3
The condensed softness sAa for the C (or O) atom in
the nitrone and sBb for the C1 (or C2) atom in the olefin,
can be obtained for nucleophilic s-Aa ) SAf A-a and elec-
trophilic s+Aa ) SAf + attacks.39 The f -Aa and f A+a are the
∆Eint
Aa
S a
olefin 1c
(S ) 0.095 eV-1
olefin 17
(S ) 0.099 eV-1
olefin 1a
(S ) 0.107 eV-1
)
condensed fukui functions for nucleophilic and electro-
nitrone (eV-1
)
)
)
philic attacks respectively; they can be evaluated by f A+a
) qAa(NA + 1) - qAa(NA) and f - ) qAa(NA) - qAa(NA
-
6d
6f
6e
6a
6j
6i
6k
6h
0.178
-177.614
-196.761
-198.062
-197.749
-206.717
-208.340
-206.999
-209.195
-173.919
-190.993
-192.081
-192.102
-200.172
-201.575
-200.551
-202.432
-175.966
-196.964
-198.629
-197.396
-207.127
-209.086
-207.042
-209.733
Aa
0.155
0.154
0.153
0.144
0.143
0.143
0.142
1),49 where qAa(NA - 1), qAa(NA), and qAa(NA + 1) are
the charges of the reactive atom in the cation, neutral,
and anion of the nitrone, respectively. The superscript
(+) in the f +Aa term indicates the electrophilic tendency
of the C (or O) atom in the nitrone to increase its charge
(qAa(NA) f qAa(NA + 1)) when the olefin carries out a
nucleophilic attack with its C1 (or C2) atom. The super-
script (-) in the f -Aa term indicates the nucleophilic
tendency of the C (or O) atom in the nitrone to decrease
its charge (qAa(NA) f qAa(NA - 1)) when the olefin carries
out an electrophilic attack with its C1 (or C2) atom.
The statement hard likes hard and soft likes soft at
global and local levels, the DFT perturbational ap-
proximation, and the orbital molecular calculation will
be used to provide a rationalization of the experimental
regioselectivity in the addition of nitrones and olefins.
The electronic structures of the neutral and ionic species
of the eight nitrones 6a , 6d -f, and 6h -k and the two
olefins 1c and 17 were calculated at the equilibrium
geometry of the neutral species at the RHF and UHF/6-
31G** level with GAUSSIAN98.50 The electronic struc-
ture for 1a has been previously calculated.21
a
Calculated as in Table 5.
energy for each nitrone-olefin pair, using the global
properties of the eight nitrones and three olefins, and
with a λ value of one.21,45 Table 6 shows that the
interaction energy becomes, in general, predominant
when the global softnesses of the nitrones decrease. The
same trend is observed for olefins 1c and 17 except for
olefin 1a . This is probably due to a small contribution of
the covalent interaction in the latter (vide infra). Accord-
ing to eq 3 the hard-hard interactions will be very
important, and the noncovalent contribution will domi-
nate the interaction (electrostatic interaction). This would
imply that the ∆Eν term (energy involved in the charge-
transfer process between nitrone and olefin) should be
not as important as the ∆Eµ term (the reshuffling of the
charge distribution). The calculations of the ∆Eµ and ∆Eν
terms confirm this suggestion, since for olefins 1c and
17, the 99% of the interaction energy is coming from the
∆Eµ term. However, for olefin 1a , 93% of the interaction
energy is coming from the ∆Eµ term, and a small but
significant contribution of covalent interactions (7%) is
provided.
Our results supports the idea suggested by Sustman
and Sicking4,38 that noncovalent interactions are impor-
tant in some 1,3-dipolar cycloadditions. Recently, Coss´ıo
et al. investigated, via transition state calculations, the
regiochemistry of several 1,3-dipolar cycloadditions, and
they concluded that electrostatic interactions and solvent
effects can modify the regiochemical outcome.1e
To determine the regiochemistry of each nitrone-olefin
pair, it is necessary to study the interaction at a local
level. In accord to eq 4, only the C and O atoms of the
nitrone and the C1 and C2 atoms of the olefin will be
involved in the bonding process. Table 7 shows the
condensed fukui functions for nucleophilic and electro-
philic attacks for the C and O atoms for the eight
Table 5 summarizes the chemical potential and the
global softness values for the nitrones and olefins; they
were calculated in terms of the vertical ionization poten-
tial and electron affinity. The global parameter values
for olefin 1a were obtained from ref 21.
To study the process of nitrone-olefin bonding at a
global level, we calculated from eq 3 the interaction
(48) Parr, R. G.; Pearson, R. G. J . Am. Chem. Soc. 1983, 105, 7512.
(49) Yang, W.; Mortier, W. J . J . Am. Chem. Soc. 1986, 108, 5708.
(50) GAUSSIAN98, Revision A.7: Frisch, M. J .; Trucks, G. W.;
Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J . R.;
Zakrzewski, V. G.; Montgomery, J r., J . A.; Stratmann, R. E.; Burant,
J . C.; Dapprich, S.; Millam, J . M.; Daniels, A. D.; Kudin, K. N.; Strain,
M. C.; Farkas, O.; Tomasi, J .; Barone, V.; Cossi, M.; Cammi, R.;
Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J .;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J . B.; Cioslowski, J .; Ortiz,
J . V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J .; Keith, T.;
Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill,
P. M. W.; J ohnson, B.; Chen, W.; Wong, M. W.; Andres, J . L.; Gonzalez,
C.; Head-Gordon, M.; Replogle, E. S.; Pople, J . A. Gaussian, Inc.,
Pittsburgh, PA, 1998.

1260 J . Org. Chem., Vol. 66, No. 4, 2001
Herrera et al.
Ta ble 7. Va lu es of th e Con d en sed F u k u i F u n ction s for
th e C a n d O Atom s in th e Nitr on e a n d th e C1 a n d C2
Atom s in th e Olefin , Resp ectively, Va lu es in eV-1
f C+
f C-
f O+
f O-
a
b
a
b
nitrone
6d
6k
6h
6j
0.283
0.182
0.180
0.183
0.202
0.194
0.195
0.211
0.089
0.110
0.112
0.111
0.117
0.114
0.114
0.121
-0.079
0.152
0.145
0.149
0.148
0.156
0.152
0.149
0.442
0.438
0.450
0.438
0.438
0.443
0.443
0.443
6i
6a
6f
6e
f C+
f C-
f C+
f C-
c
d
c
d
olefin
1
1
2
2
F igu r e 4. Histogram of the relative interaction energy ∆Eint
between olefins an nitrones. Values relative to the value of
the interaction energy of the pair olefin 1c - nitrone 6d . This
energy is calculated using eq 4, where s-C of nitrone and sC- of
1c
1a e
7
0.316
0.034
0.335
-0.085
-0.067
0.002
0.128
-0.059
0.052
-0.008
-0.035
0.098
+
a
b
olefin are used. The values of s-C and sC- were calculated fr1om
sC- ) SNitrone f - and s-C ) SOlefin f - . They1increase from nitrone
6d to 6e andCfrom ol1efin 1c to C117, respectively.
Calculated as f Aa ) qAa(NA + 1) - qAa(NA). Calculated as
f A-a ) qAa(NA) - qAa(NA - 1). c Calculated as f +Bb ) qBb(NB + 1) -
-
d
q
Bb(NB). Calculated as f Bb ) qBb(NB) - qBb(NB - 1). e See ref 21.
nitrones, and C1 and C2 for the three olefins. The charges
of the reactive atoms were obtained from the Mu¨lliken
population analysis of the neutral and ionic species.
In the cycloaddition of a nitrone and an olefin
there are four pairs of possible atomic interactions
(C-C1, C-C2, O-C1, and O-C2), and two kinds of
condensed softness for each atom (nucleophilic and
electrophilic attacks). Therefore, by statistics, there
will be 16 probable interactions involved in the nitrone-
Figure 4 depicts a histogram of the relative calculated
s-Cs-C values for the interaction energy term for each
pair1 nitrone-olefin versus the nucleophilic condensed
softness s-C and s-C , in relation to the value of the pair
1
s-C (olefin 1c)-s-C (nitrone 6d ). In general, the interac-
1
tion energies increase when the nucleophilic condensed
softness of both atoms decreases. At a local level, these
results suggest that the hard-hard interactions will be
very relevant in the nitrone-olefin bond process, also
supporting our previous analysis obtained at a global
level. In addition, noncovalent interactions will be present
in the most important local interaction: nitrone C atom-
olefin C1 atom (eq 4), since 99% of the interaction energy
is provided by the ∆Eµ interaction term.
olefin bond process: (s+C sC- , sC- sC+ , sC+ s+C , sC- s-C , ...). The
1
s+C s-C term means that the nitro1ne C a1tom is1 the elec-
1
tron-acceptor (electrophilic center) and the olefinic
C1 atom is the electron-donor (nucleophilic center), while
the s-Cs+C term means that the nitrone C atom is the
1
electron-donor (nucleophilic center) and the olefinic C1
atom is the electron-acceptor (electrophilic center). We
The experimental results would suggest that the
control of regioselectivity depends mainly on the presence
of the electron-donating group, and not only from the
effect of the electron-withdrawing group on the olefins
1. The theoretical results confirm this idea: Figure 4
shows that the interaction energy increases in the order
of olefins 1c > 1a > 17, (1:0.9:0.5). The electron-donating
groups OCOCH3 and OCOPhNO2, in the captodative
olefins 1c and 1a , respectively, increase the interaction
energy s-Cs-C term (approximately at the same extent),
enhancing t1he electron-donor capacity and the nucleo-
philicity of the olefinic C1 atom, and consequently a high
regioselectivity should be observed. While the electron-
withdrawing group OCOCH3 in olefin 17 cuts by half
consider the sC+sC+ and sC-s-C terms in agreement with
1
Klopman et al., who suggest1ed that an electron transfer
occurs between the 1,3-dipole and the dipolarophile, but
the specific direction of the electronic transfer process is
not defined.51 The sC+ sC+ term indicates that the nitrone
1
C atom and the olefinic C1 atom are electron-acceptors
(electrophilic centers), while the s-CsC- term indicates
1
that the nitrone C atom and the olefinic C1 atom are
electron-donors (nucleophilic centers).
To predict the main interaction term in the bond
process nitrone-olefin, we calculated the interaction
energy considering the 16 probable interactions for each
pair nitrone-olefin. The results show that the most
important interaction corresponds to the s-CsC- term,
the energy of the interaction s-CsC- term, decreasing
1
1
and suggests that the interaction will occur between the
nitrone C atom and the olefinic C1 atom with a mutual
electron donation between these atoms in the addition
process. These results are very interesting, because a
mutual donation has been supported by experimental and
theoretical evidence. For example, Fujitake and Hirota52
corroborated by millimeter and submillimeter wave
spectrum of dichlorocarbene (:CCl2) the calculations made
by Carter and Goddard,53 who found a pπ donation from
Cl to C (πc ) 0.25) and a σ electron donation from C to
Cl (iσ ) 0.20).
the electron-donor capacity and the nucleophilicity of
the olefinic C1 atom, hence the reaction would not be
regioselective, and a mixture of regiosomers should be
obtained, as observed experimentally (Scheme 4).
Con clu sion s
Captodative olefins 1-acetylvinyl carboxylates 1 under-
went 1,3-dipolar cycloadditions with diverse dipoles in
highly regio- and stereoselectivities. The formation of the
5-substituted aromatic heterocycle with propionitrile
oxide, nitrile imines, and diazo compounds and olefin 1a
was the result of the succesive dipolar addition and
elimination of the PNB group, as confirmed by the
reaction with trimethylsilyl diazomethane (4b). The
cycloaddition with nitrones 6 also proceeded with the
(51) Klopman, G. In Chemical Reactivity and Reaction Paths;
Klopman, G., Ed.; Wiley: New York, 1974; p 141.
(52) Fujitake, M.; Hirota, E. J . Chem. Phys. 1989, 91, 3426.
(53) Carter, E. A.; Goddard, W. A., III. J . Chem. Phys. 1988, 88,
1752.

Captodative Olefins
J . Org. Chem., Vol. 66, No. 4, 2001 1261
188.6 (CH3CO); MS (70 eV) 262 (M+, 100), 247 (82), 233 (4),
219 (12), 171 (8), 116 (27), 89 (21), 77 (33). Anal. Calcd for
exclusive formation of the 5-substituted isoxazolidines,
and yielding the endo adduct as the major stereoisomer.
No significant effect on the regiochemistry was detected
by modifying the electron-donating group of the capto-
dative olefin, in contrast with olefin 17, which bears a
single electron-withdrawing group, where the reaction
was not regioselective. The structural characterization
of the adducts was supported by NMR experiments and
X-ray crystallography. Thus, the preference of the PNB
group in the isoxazolidines for the pseudoaxial conforma-
tion was determined, which may be associated to the
anomeric effect. Cleavage of the isoxazolidine N-O bond
was not successfully made, despite the diverse catalytic
hydrogenation conditions applied. This may be explained
as an overstability of this bond due to the presence of
the C-5 oxygenated functionality.
FMO theory has been used to rationalize the regio-
selectivity observed in 1,3-dipolar cycloadditions with
nitrones. However, the calculations furnished a predic-
tion of the regiochemistry opposite to the experimental
results. In contrast, the chemical reactivity of nitrones 6
and olefins 1 and 17, and the various factors which affect
the process of nitrone-olefin bonding, were estimated in
terms of interaction energy values obtained from the
HSAB principle and DFT, predicting a regiochemistry in
agreement with the experimental results.
C
17H14N2O: C, 77.84; H, 5.38; N, 10.68. Found: C, 77.97; H,
5.19; N, 10.85.
5-Acetyl-1-p h en yl-3-(p-tolyl)p yr a zole (9b). Following the
general procedure with 7b (0.63 g) afforded 0.2 g (72%) of 9b
as colorless crystals: Rf 0.23 (hexane/EtOAc, 8:2); mp 88-89
°C; IR (KBr) 1690, 1614, 1504, 1424, 1175 cm-1; 1H NMR (500
MHz, CDCl3) δ 2.36 (s, 3H, CH3Ar), 2.51 (s, 3H, CH3CO), 7.20-
7.23 (m, 2H, Ar-H), 7.22 (s, 1H, H-4), 7.39-7.47 (m, 5H, Ph-
H), 7.73-7.75 (m, 2H, Ar-H); 13C NMR (125 MHz, CDCl3) δ
21.3 (CH3Ar), 28.8 (CH3CO), 109.5 (C-4), 125.6, 125.9, 128.6,
128.7, 129.3, 129.4, 138.4, 140.6, 140.7, 151.4, 187.8 (CH3CO).
Anal. Calcd for C18H16N2O: C, 78.24; H, 5.84; N, 10.14.
Found: C, 78.39; H, 6.04; N, 10.02.
5-Acetylp yr a zole (10a ). Meth od A. A solution of 0.24 g
(1.0 mmol) of 1a in dry ether (15 mL) and 0.21 g (5.0 mmol) of
4a (2.5 mL solution in dry ether) was stirred in dark at room
temperature, under an N2 atmosphere, for 5 h. The solvent
was removed under vacuum, and the residue was purified by
column chromatography on silica gel (30 g, hexane/EtOAc, 7:3)
to give 0.075 g (68%) of 10a as an yellow oil. Meth od B. A
solution of 0.24 g (1.0 mmol) of 1a in dry ether (15 mL) and
0.34 g (5.0 mmol) of 4b (2.5 mL, 2 M solution in hexane) was
stirred in dark at room temperature, under an N2 atmosphere,
for 24 h. The solvent was removed under vacuum, affording a
mixture of 11a /11b (71:29) and 10a in 1:1 ratio. The mixture
was heated in dry ether (10 mL) to 50 °C for 1 h, to yield an
oily residue, which was purified by column chromatography
on silica gel (30 g, hexane/EtOAc, 7:3) to give 0.088 g (80%) of
10a as an yellow oil: Rf 0.65 (hexane/EtOAc, 7:3); IR (film)
To the best of our knowledge, this seems to be the first
theoretical study of captodative olefins that provides an
insight into the relevance of the electron-donor group of
these molecules, as the main controlling factor of the
regiochemistry in 1,3-dipolar additions with nitrones, and
probably also with other dipoles.
1
3224, 1656, 1467, 1348, 1299, 792 cm-1; H NMR (500 MHz,
CDCl3) δ 2.61 (s, 3H, CH3CO), 6.79 (d, J ) 2.3 Hz, 1H), 7.67
(d, J ) 2.3 Hz, 1H), 10.90 (br s, 1H, NH); 13C NMR (75.4 MHz,
CDCl3) δ 26.9 (CH3CO), 107.4, 134.2, 147.4, 191.9 (CH3CO).
Anal. Calcd for C5H6N2O: C, 54.54; H, 5.49; N, 25.44. Found:
C, 54.39; H, 5.31; N, 25.21.
5-Acetyl-3-eth ylisoxa zole (12). Meth od A. A solution of
1.08 g (12.0 mmol) of 3 and 1.20 g (12.0 mmol) of triethylamine
in dry benzene (15 mL) was stirred at room temperature,
under an N2 atmosphere, for 30 min. At 0 °C, 1.44 g (12.0
mmol) of phenyl isocyanate and 0.47 g (2.0 mmol) of 1a were
added, and the mixture was stirred for 2 h at room tempera-
ture and then heated to reflux for 2 h. The suspension was
filtered, and the solvent was removed under vacuum. The
residue was purified by column chromatography on silica gel
(40 g, hexane/EtOAc, 7:3) to give 0.236 g (85%) of 12 as an
yellow oil. Meth od B. Following the same conditions as
Method A, but using 0.26 g (2.0 mmol) of 1c, to give 0.175 g
(63%) of 12: Rf 0.65 (hexane/EtOAc, 7:3); IR (CH2Cl2) 2982,
Exp er im en ta l Section
Gen er a l. Melting points are uncorrected. NMR spectra
were recorded at 300 and 500 MHz for 1H, and at 75.4 and
125.0 MHz for 13C, using TMS as internal standard. Mass
spectra (EI) were obtained at 70 eV. X-ray analyses were
collected using Mo KR radiation (graphite crystal mono-
chromator, l ) 71073 Å). Microanalyses were performed by
M-H-W Laboratories (Phoenix, AZ). Analytical thin-layer
chromatography was carried out using E. Merck silica gel
60 F254 coated 0.25 plates, visualizing by long- and short-
wavelength UV lamp. All air moisture sensitive reactions
were carried out under nitrogen using oven-dried glassware.
Dioxane, ethyl ether, THF, toluene, and xylene were freshly
distilled from sodium, and methylene chloride from calcium
hydride, prior to use. K2CO3 was dried overnight at 120 °C
before using. Triethylamine was freshly distilled from NaOH.
All other reagents were used without further purification.
Gen er a l Meth od for th e P r ep a r a tion of 5-Acetyl-3-
a r yl-1-p h en ylp yr a zoles 9a -d . To a solution of arylaldehyde
phenylhydrazones 7a -d (3.0 mmol) and 0.24 g (1.0 mmol) of
olefin 1a in dry dioxane (15 mL) was added 0.68 g (3.0 mmol)
of chloramine-T‚H2O at room temperature, under an N2
atmosphere. The mixture was heated to reflux for 2 h, and
the solvent was removed under vacuum. The residue was
purified by column chromatography on silica gel (30 g, hexane/
EtOAc, 7:3) to give the corresponding pyrazoles 9a -d .
1
1603, 1474, 1453, 1431, 1036, 954 cm-1; H NMR (300 MHz,
CDCl3) δ 1.30 (t, J ) 7.6 Hz, 3H, CH3CH2), 2.60 (s, 3H,
CH3CO), 2.77 (q, J ) 7.6 Hz, 2H, CH3CH2), 6.79 (s, 1H, H-4);
13C NMR (75.4 MHz, CDCl3) δ 12.6 (CH3CH2), 19.6 (CH3CH2),
27.2 (CH3CO), 106.6, 166.0, 166.5, 187.0 (CH3CO); MS (70 eV)
139 (M+, 61), 124 (7), 96 (33), 68 (100). HRMS (EI) calcd for
[M+] C7H9NO2: 139.0633. Found: 139.0630.
Gen er a l P r oced u r e for th e 1,3-Dip ola r Ad d ition of 3-p-
Nit r ob en zoyloxy-3-b u t en -2-on e (1a ) w it h Nit r on es 6a ,
6d , 6e, a n d 6f. (3R*,5R*)-5-Acetyl-2,3-d ip h en yl-5-(p-n itr o-
ben zoyloxy)isoxa zolid in e (13a ) a n d (3R*,5S*)-5-Acetyl-
2,3-d ip h en yl-5-(p-n itr oben zoyloxy)isoxa zolid in e (14a ). A
mixture of 0.50 g (2.54 mmol) of 6a , and 0.60 g (2.5 mmol) of
1a in dry benzene (20 mL) was stirred and heated to 80 °C
for 8 h, under an N2 atmosphere. The solvent was removed
under vacuum, to yield an oily residue of a mixture of 13a /
14a (89:11). The major isomer 13a was isolated by crystal-
lization from EtOH/CH2Cl2, 5:1, to yield 0.77 g (70%) as
brownish yellow crystals: Rf 0.5 (hexane/EtOAc, 8:2); mp
100-101 °C; IR (CH2Cl2): 1706, 1601, 1529, 1492, 1348, 1263,
5-Acetyl-1,3-d ip h en ylp yr a zole (9a ). Meth od A. Follow-
ing the general procedure with 7a (0.59 g) afforded 0.21 g (80%)
of 9a as colorless prisms. Meth od B. Following the same
conditions as method A, but using 0.13 g (1.0 mmol) of 1c to
give 0.157 g (60%) of 9a : Rf 0.37 (hexane/EtOAc, 8:2); mp 112-
113 °C; IR (CCl4) 1694, 1498, 1424, 1356, 1276, 1214 cm-1; 1H
NMR (300 MHz, CDCl3) δ 2.58 (s, 3H, CH3CO), 7.31 (s, 1H,
756, 696 cm-1 1H NMR (300 MHz, CDCl3) δ 2.55 (s, 3H,
;
COMe), 3.12 (dd, J ) 14.1, 6.4 Hz, 1H, H-4â), 3.27 (dd, J )
14.1, 9.1 Hz, 1H, H-4R), 4.68 (dd, J ) 9.1, 6.4 Hz, 1H, H-3),
7.01-7.08 (m, 3H, H-9, Ar-H), 7.20-7.27 (m, 2H, Ar-H),
7.33-7.42 (m, 3H, Ar-H), 7.50-7.55 (m, 2H, Ar-H), 8.03-
H-4), 7.36-7.58 (m, 8H, Ph-H), 7.90-7.98 (m, 2H, Ph-H); 13
C
NMR (75.4 MHz, CDCl3) δ 29.0 (CH3CO), 110.9 (C-4), 126.5,
126.8, 129.1, 129.3, 129.4, 129.7, 133.5, 142.1, 142.4, 152.1,

1262 J . Org. Chem., Vol. 66, No. 4, 2001
Herrera et al.
8.07 (m, 2H, Ar-H), 8.22-8.26 (m, 2H, Ar-H). Further signals
attributed to isomer 14a : 2.59 (s, COMe), 4.99 (dd, J ) 9.9,
7.1 Hz, H-3); 13C NMR (75.4 MHz, CDCl3) δ 25.9 (COCH3),
47.7 (C-4), 69.5 (C-3), 105.9 (C-5), 117.2, 123.5, 124.1, 126.9,
128.2, 128.8, 129.1, 131.1, 134.5, 139.3, 148.9, 150.9, 163.3
(CO2Ar), 198.5 (COMe); MS (70 eV) 432 (M+, 2), 352 (12), 275
(10), 256 (11), 222 (76), 198 (72), 167 (100), 154 (45), 121 (43),
77 (34), 65 (53). Anal. Calcd for C24H20N2O6: C, 66.66; H, 4.66;
N, 6.48. Found: C, 63.70; H, 5.01; N, 6.04.
Gen er a l P r oced u r e for th e 1,3-Dip ola r Ad d ition of
3-Acetoxy-3-bu ten -2-on e (1c) w ith Nitr on es 6c, 6f, a n d
6h . (3R*,5R*)-5-Acet oxy-5-a cet yl-3-(p -b r om op h en yl)-2-
p h en ylisoxa zolid in e (15a ) a n d (3R*,5S*)-5-Acetoxy-5-
a cetyl-3-(p-br om op h en yl)-2-p h en ylisoxa zolid in e (16a ). A
mixture of 0.28 g (1.0 mmol) of nitrone 6c and 0.13 g (1.0
mmol) of 1c in dry benzene (15 mL) was stirred and heated to
80 °C for 12 h, under an N2 atmosphere. The solvent was
removed under vacuum, to yield an oily residue of a mixture
of 15a /16a (73:27). This residue was purified by column
chromatography on silica gel (15 g, hexane/EtOAc, 7:3) to give
a mixture 15a /16a (73:27). The major isomer was isolated by
recrystallization from hexane/CH2Cl2 (95:5), to afford 0.283 g
(70%) of 15a as white needles: Rf 0.6 (hexane/EtOAc, 7:3); mp
122-123 °C; IR (KBr) 1735, 1595, 1481, 1360, 1233, 1067, 1000
cm-1; 1H NMR (300 MHz, CDCl3) δ 2.12 (s, 3H, CH3CO2), 2.41
(s, 3H, CH3CO), 2.70 (dd, J ) 14.1, 7.4 Hz, 1H, H-4â), 3.19
(dd, J ) 14.1, 9.0 Hz, 1H, H-4R), 4.43 (dd, J ) 9.0, 7.4 Hz, 1H,
H-3), 6.93-7.00 (m, 2H, Ph-H), 7.01-7.07 (m, 1H, Ph-H),
7.17-7.25 (m, 2H, Ph-H), 7.32-7.38 (m, 2H, Ar-H), 7.47-
7.53 (m, 2H, Ar-H). Signals attributed to isomer 16a : 2.22
(s, CH3CO2), 2.34 (s, CH3CO), 3.03 (dd, J ) 12.7, 6.7 Hz, H-4),
4.84 (dd, J ) 9.9, 6.7 Hz, H-3); 13C NMR (75.4 MHz, CDCl3) δ
20.8 (CH3CO2), 25.6 (CH3CO), 48.0 (C-4), 69.3 (C-3), 104.6
(C-5), 117.8, 122.1, 124.4, 128.7, 128.9, 132.1, 132.3, 138.1,
(3R*,5R*)-5-Acet yl-3-(p -ch lor op h en yl)-5-(p -n it r ob en -
zoyloxy)-2-p h en ylisoxa zolid in e (13b) a n d (3R*,5S*)-5-
Acetyl-3-(p-ch lor oph en yl)-5-(p-n itr oben zoyloxy)-2-ph en yl-
isoxa zolid in e (14b). The same procedure as for 13a /14a was
used, with 0.23 g (1.0 mmol) of 6b , and 0.24 g (1.0 mmol)
of 1a in dry benzene (15 mL), to give a mixture of 13b/14b
(90:10). Purification by column chromatography on silica gel
(30 g, hexane/EtOAc, 7:3), yielded 0.35 g (75%) of 13b, which
was recrystallized from hexane/CH2Cl2, 95:5, to give pale
yellow crystals: Rf 0.48 (hex/EtOAc, 7:3); mp 57-58 °C; IR
(KBr) 1730, 1663, 1600, 1521, 1346, 1263, 1097 cm-1; 1H NMR
(300 MHz, CDCl3) δ 2.52 (s, 3H, COMe), 2.94 (dd, J ) 14.1,
6.4 Hz, 1H, H-4â), 3.30 (dd, J ) 14.1, 9.0 Hz, 1H, H-4R), 4.65
(dd, J ) 9.0, 6.4 Hz, 1H, H-3), 6.98-7.02 (m, 2H, Ar-H), 7.03-
7.10 (m, 1H, Ar-H), 7.22-7.29 (m, 2H, Ar-H), 7.35-7.40 (m,
2H, Ar-H), 7.43-7.48 (m, 2H, Ar-H), 8.04-8.09 (m, 2H, Ar-
H), 8.27-8.31 (m, 2H, Ar-H). Further signals attributed to
isomer 14b: 2.59 (s, COMe), 4.98 (dd, J ) 9.6, 6.8 Hz, H-3);
13C NMR (75.4 MHz, CDCl3) δ 25.8 (COCH3), 47.7 (C-4), 69.0
(C-3), 106.1 (C-5), 117.5, 123.6, 124.5, 128.2, 128.7, 129.3,
131.1, 134.2, 134.3, 137.8, 148.7, 151.1, 163.4 (CO2Ar), 198.4
(COMe). Anal. Calcd for C24H19N3O8: C, 60.38; H, 4.01; N, 8.80.
Found: C, 60.40; H, 4.09; N, 9.00.
(3R*,5R*)-5-Acetyl-2-(ter t-bu tyl)-5-(p-n itr oben zoyloxy)-
3-p h en ylisoxa zolid in e (13h ). The same procedure as for
13a /14a was used, with 0.10 g (0.56 mmol) of 6e, and 0.13 g
(0.56 mmol) of 1a in dry benzene (15 mL), and heating to reflux
for 24 h, to give 13h as a pale yellow oil. Purification by column
chromatography on silica gel (15 g, hexane/EtOAc, 7:3), yielded
0.1 g (44%) of 13h , which was recrystallized from EtOH/
CH2Cl2, 4:1, to give pale yellow needles: Rf 0.43 (hex/EtOAc
8:2); mp 129-130 °C; IR (KBr) 1728, 1529, 1354, 1284, 1095,
846, 719 cm-1; 1H NMR (300 MHz, CDCl3) δ 1.12 (s, 9H, t-Bu),
2.44 (s, 3H, COMe), 2.82 (dd, J ) 14.1, 5.2 Hz, 1H, H-4â), 3.07
(dd, J ) 14.1, 9.7 Hz, 1H, H-4R), 4.46 (dd, J ) 9.7, 5.2 Hz, 1H,
H-3), 7.23-7.37 (m, 3H, Ph-H), 7.50-7.55 (m, 2H, Ph-H),
7.95-8.00 (m, 2H, Ar-H), 8.20-8.26 (m, 2H, Ar-H); 13C NMR
(75.4 MHz, CDCl3) δ 25.7 (COCH3), 26.1 (t-Bu), 48.7 (C-4), 59.5
(CMe3), 61.3 (C-3), 106.3 (C-5), 123.4, 127.3, 127.9, 128.5, 131.1,
134.6, 142.4, 150.7, 163.4 (CO2Ar), 199.4 (COMe). Anal. Calcd
for C22H24N2O6: C, 64.07; H, 5.87; N, 6.79. Found: C, 60.10;
H, 5.76; N, 6.85.
148.6, 169.9 (CH3CO2), 199.1 (CH3CO). Anal. Calcd for C19H18
-
BrNO4: C, 56.45; H, 4.49; N, 3.46. Found: C, 56.32; H, 4.68;
N, 3.41.
(3R*,5R*)-5-Acetoxy-5-a cetyl-2-p h en yl-3-(p-tolyl)isox-
a zolid in e (15b) a n d (3R*,5S*)-5-Acetoxy-5-a cetyl-2-p h en -
yl-3-(p-tolyl)isoxa zolid in e (16b). The same procedure as for
15a /16a was used, with 0.10 g (0.47 mmol) of 6f, and 0.10 g
(0.76 mmol) of 1c, to give a mixture of 15b/16b (82:18). This
residue was purified by column chromatography on silica gel
(30 g, hexane/EtOAc, 7:3) to give a mixture 15b/16b (82:18)
as pale green oil. The major isomer was isolated by crystal-
lization from hexane/CH2Cl2, 8:2, to afford 0.08 g (50%) of 15b
as colorless crystals: Rf 0.37 (hexane/EtOAc, 8:2); mp 126-
127 °C; IR (CH2Cl2) 1732, 1720, 1597, 1488, 1359, 1230, 1184,
1074 cm-1; 1H NMR (300 MHz, CDCl3) δ 2.15 (s, 3H, CH3CO2),
2.36 (s, 3H, CH3Ar), 2.42 (s, 3H, CH3CO), 2.76 (dd, J ) 14.1,
8.0 Hz, 1H, H-4â), 3.19 (dd, J ) 14.1, 8.8 Hz, 1H, H-4R), 4.40
(dd, J ) 8.8, 8.0 Hz, 1H, H-3), 6.97-7.06 (m, 3H, Ar-H), 7.17-
7.23 (m, 4H, Ar-H), 7.32-7.37 (m, 2H, Ar-H). Signals
attributed to isomer 16b: 1.87 (s, CH3CO2), 2.35 (s, CH3Ar),
2.50 (s, CH3CO), 2.99 (dd, J ) 13.0, 6.9 Hz), 4.81 (dd, J ) 10.2,
6.9 Hz, H-3); 13C NMR (75.4 MHz, CDCl3) δ 20.4 (CH3Ar), 20.8
(CH3CO2), 26.1 (CH3CO), 48.7 (C-4), 69.3 (C-3), 104.9 (C-5),
118.0, 124.0, 127.3, 128.5, 129.6, 129.9, 135.8, 137.9, 149.0,
170.0 (CH3CO2), 199.7 (CH3CO). Anal. Calcd for C20H21NO4:
C, 70.78; H, 6.24; N, 4.13. Found: C, 70.69; H, 6.12; N, 4.18.
(3R*,5S*)-5-Acet yl-3-(p -a n isyl)-2-p h en ylisoxa zolid in e
(18a ), (3R*,5R*)-5-Acetyl-3-(p-a n isyl)-2-p h en ylisoxa zoli-
d in e (18b), (3R*,4S*)-4-Acetyl-3-(p-a n isyl)-2-p h en ylisox-
a zolid in e (18c), a n d (3R*,4R*)-4-Acetyl-3-(p-a n isyl)-2-
p h en ylisoxa zolid in e (18d ). A mixture of 0.50 g (2.2 mmol)
of nitrone 6e and 0.15 g (2.2 mmol) of 17 in dry benzene (20
mL) was stirred and heated to 80 °C for 24 h, under an N2
atmosphere. The solvent was removed under vacuum, to yield
an oily residue of a mixture of 18a /18b/18c/18d (2:10:7:1). This
residue was purified by column chromatography on silica gel
(30 g, hexane/EtOAc, 9:1) to give 0.52 g (80%) of a mixture of
four stereoisomers and pure oily small fractions of each one:
0.026 g (4%) of 18a , Rf 0.57 (hexane/EtOAc, 8:2); 0.052 g (8%)
of 18b, Rf 0.48 (hexane/EtOAc, 8:2); 0.039 g (6%) of 18c, Rf
0.40 (hexane/EtOAc, 8:2); 0.02 g (3%) of 18d , Rf 0.31 (hexane/
EtOAc, 8:2); Data of 18a : IR (CCl4) 1703, 1604, 1578, 1511,
(3R*,5R*)-5-Acet yl-2-b en zyl-5-(p -n it r ob en zoyloxy)-3-
(p -t olyl)isoxa zolid in e (13j) a n d (3R*,5S*)-5-Acet yl-2-
ben zyl-5-(p-n itr oben zoyloxy)-3-(p-tolyl)isoxazolidin e (14j).
The same procedure as for 13a /14a was used, with 0.103 g
(0.458 mmol) of 6j, and 0.107 g (0.458 mmol) of 1a , to give a
mixture of 13j/14j (60:40). The major isomer 13j was isolated
by radial chromatography (hexane/EtOAc, 8:2), yielding 0.063
g (30%) of 13j, which was recrystallized from hexane/CH2Cl2,
3:1, to give pale yellow needles: Rf 0.3, hex/EtOAc, 7:3;
1
131-132 °C; IR (KBr) 1726, 1527, 1350, 1286, 1064 cm-1. H
NMR (300 MHz, CDCl3) δ 2.37 (s, 3H, Me-Ar), 2.49 (s, 3H,
COMe), 2.66 (dd, J ) 13.2, 11.3 Hz, 1H, H-4â), 2.86 (dd, J )
13.2, 5.4 Hz, 1H, H-4R), 3.90 (d, J ) 14.7 Hz, 1H, PhCH2),
4.28 (d, J ) 14.7 Hz, 1H, PhCH2), 4.30 (dd, J ) 11.3, 5.4 Hz,
1H, H-3), 6.98-7.14 (m, 3H, Ar-H), 7.18-7.26 (m, 4H,
Ar-H), 7.37-7.43 (m, 2H, Ar-H), 8.04-8.09 (m, 2H, Ar-H),
8.24-8.29 (m, 2H, Ar-H). Further signals attributed to isomer
14j: 3.20 (dd, J ) 14.1, 7.8 Hz, H-4), 3.94 (d, J ) 14.8 Hz,
PhCH2), 4.16 (d, J ) 14.8 Hz, PhCH2); 13C NMR (75.4 MHz,
CDCl3) δ 21.2 (Ar-CH3), 25.5 (COCH3), 47.7 (C-4), 59.8
(PhCH2), 65.3 (C-3), 106.1 (C-5), 123.4, 127.4, 127.6, 128.0,
129.7, 129.9, 131.0, 133.1, 134.5, 135.1, 138.5, 150.8, 163.1
(CO2Ar), 201.6 (COMe). Anal. Calcd for C26H24N2O6: C, 67.82;
H, 5.25; N, 6.08. Found: C, 67.62; H, 5.23; N, 6.03.
1485, 1258, 1161, 1036 cm-1 1H NMR (300 MHz, CDCl3) δ
;
2.38 (s, 3H, CH3CO), 2.62 (ddd, J ) 12.7, 6.4, 5.0 Hz, 1H, H-4â),
2.96 (ddd, J ) 12.7, 9.0, 8.4 Hz, 1H, H-4R), 3.80 (s, 3H, CH3O),
4.44 (dd, J ) 8.4, 6.4 Hz, 1H, H-3), 4.58 (dd, J ) 9.0, 5.0 Hz,
1H, H-5), 6.58-7.35 (m, 9H, Ar-H); 13C NMR (75.4 MHz,
CDCl3) δ 25.8 (CH3CO), 41.9 (C-4), 55.2 (MeO), 69.1 (C-3), 81.5
(C-5), 114.2, 116.6, 123.1, 128.0, 128.7, 132.1, 149.9, 159.1,
200.5 (CH3CO). Anal. Calcd for C18H19NO3: C, 72.71; H, 6.44;

Captodative Olefins
J . Org. Chem., Vol. 66, No. 4, 2001 1263
N, 4.71. Found: C, 72.90; H, 6.59; N, 4.75. Data of 18b: IR
Crystallographic measurements were performed on a Siemens
P4 diffractometer with Mo KR radiation (λ ) 0.7107 Å;
graphite monochromator) at room temperature. Two standard
reflections were monitored periodically; they showed no change
during data collection. Unit cell parameters were obtained
from least-squares refinement of 26 reflections in the range
2 < 2Θ < 20°. Intensities were corrected for Lorentz and
polarization effects. No absorption correction was applied.
Anisotropic temperature factors were introduced for all non-
hydrogen atoms. Hydrogen atoms were placed in idealized
positions and their atomic coordinates refined. Unit weights
were used in the refinement. Structures were solved using
SHELXTL54 on a personal computer. Data for 13h : Formula:
(CHCl3) 1718, 1676, 1512, 1492, 1251, 1176, 1029 cm-1 1H
;
NMR (300 MHz, CDCl3) δ 2.31 (s, 3H, CH3CO), 2.55 (ddd, J )
12.5, 8.2, 5.8 Hz, 1H, H-4â), 2.89 (ddd, J ) 12.5, 8.0, 6.3 Hz,
1H, H-4R), 3.76 (s, 3H, CH3O), 4.54 (dd, J ) 8.0, 5.8 Hz, 1H,
H-3), 4.66 (dd, J ) 8.2, 6.3 Hz, 1H, H-5), 6.85-7.03 (m, 5H,
Ar-H), 7.14-7.22 (m, 2H, Ar-H), 7.29-7.35 (m, 2H, Ar-H);
13C NMR (75.4 MHz, CDCl3) δ 26.7 (CH3CO), 41.4 (C-4), 55.1
(MeO), 67.6 (C-3), 81.4 (C-5), 114.1, 116.2, 122.3, 128.0, 128.4,
132.1, 149.4, 159.0, 207.6 (CH3CO). Anal. Calcd for C18H19
-
NO3: C, 72.71; H, 6.44; N, 4.71. Found: C, 72.80; H, 6.18; N,
4.94. Data of 18c: IR (CCl4) 1720, 1612, 1598, 1512, 1490,
1
1249, 1173, 1040 cm-1; H NMR (300 MHz, CDCl3) δ 2.12 (s,
3H, CH3CO), 3.61-3.70 (m, 1H, H-4), 3.80 (s, 3H, CH3O), 4.21
(dd, J ) 8.2, 6.9 Hz, 1H, H-5â), 4.41 (dd, J ) 8.2, 8.0 Hz, 1H,
H-5R), 4.65 (d, J ) 6.1, 1H, H-3), 6.89-6.99 (m, 5H, Ar-H),
7.16-7.24 (m, 2H, Ar-H), 7.38-7.44 (m, 2H, Ar-H); 13C NMR
(75.4 MHz, CDCl3) δ 29.7 (CH3CO), 55.3 (MeO), 66.8 (C-4),
68.2 (C-5), 71.5 (C-3), 114.4, 115.5, 122.3, 127.9, 128.7, 133.1,
150.4, 159.3, 204.2 (CH3CO). Anal. Calcd for C18H19NO3: C,
72.71; H, 6.44; N, 4.71. Found: C, 72.92; H, 6.33; N, 4.75. Data
of 18d : IR (CCl4) 1719, 1617, 1599, 1513, 1500, 1250, 1183,
1042 cm-1; 1H NMR (300 MHz, CDCl3) δ 1.81 (s, 3H, CH3CO),
3.80 (s, 3H, CH3O), 3.80-3.92 (m, 1H, H-4), 4.26 (t, J ) 8.4
Hz, 1H, H-5â), 4.61 (dd, J ) 8.4, 6.7 Hz, 1H, H-5R), 5.02 (d, J
) 8.8, 1H, H-3), 6.85-7.09 (m, 5H, Ar-H), 7.23-7.31 (m, 2H,
Ar-H), 7.32-7.38 (m, 2H, Ar-H); 13C NMR (75.4 MHz, CDCl3)
δ 30.4 (CH3CO), 55.2 (MeO), 60.5 (C-4), 66.9 (C-5), 71.5 (C-3),
114.1, 115.2, 122.1, 128.9, 129.1, 129.5, 150.4, 159.4, 204.2
(CH3CO). Anal. Calcd for C18H19NO3: C, 72.71; H, 6.44; N,
4.71. Found: C, 72.94; H, 6.56; N, 4.70.
C22H24N2O6; molecular weight: 412.43; cryst syst: triclinic;
space group: P1; unit cell parameters: a, 9.1316 (6), b, 11.266
(2), c, 11.7911 (8) (Å); R, 82.150 (7), â, 68.905 (4), γ, 75.488 (6)
(deg); temp (K): 293 (2); Z: 2; R: 0.0421; GOF: 1.060. Data
for 14j: Formula: C26H24N2O6; molecular weight: 460.47; cryst
syst: triclinic; space group: P1; unit cell parameters: a, 6.778,
b, 10.1390 (10), c, 18.013 (2) (Å); R, 79.610 (10), â, 79.330 (10),
γ, 82.320 (10) (deg); temp (K): 293 (2); Z: 2; R: 0.0391; GOF:
1.042.
Ack n ow led gm en t. We thank Fernando Labarrios
for his help in spectrometric measurements. J .T. would
like to acknowledge DEPI/IPN (Grant 921769) and
CONACYT (Grant 1570P-E9507) for financial support.
F.M. acknowledges a grant from CONACYT (400200-
5-29299E). H.A.J .-V. thanks CONACYT (Grant 3251P)
for financial support. R.H. is grateful to CONACYT
for a graduate fellowship, and to the Ludwig K.
Hellweg Foundation for a partial scholarship. A.N.
wishes to thank CONACYT for a postdoctoral fellow-
ship (Ca´tedra Patrimonial Nivel II, No 920347).
M.A.M. thanks CONACYT for a Ph.D. scholarship.
∆4-5-Acetyl-3-(p-an isyl)-2-ph en ylisoxazolin e (20). A mix-
ture of 0.10 g (0.22 mmol) of 13e, and 19 (10 mL) was stirred
and heated to 100 °C for 3 h, under an N2 atmosphere. The
mixture was diluted with CH2Cl2 (50 mL) and washed with a
20% aqueous solution of HCl (3 × 30 mL), a saturated aqueous
solution of NaHCO3 (3 × 30 mL), and brine (3 × 30 mL). The
organic phase was dried (Na2SO4), and the solvent was
removed under vacuum, to give an oily residue, which was
purified by column chromatography on silica gel (30 g, hexane/
EtOAc, 9:1) to yield 0.038 g (60%) of 20 as a pale yellow oil:
Rf 0.2 (hexane/EtOAc, 8:2); IR (CCl4) 1706, 1600, 1535, 1510,
Su p p or tin g In for m a tion Ava ila ble: X-ray structures of
nitrones 6e and 6k , tables of crystal data, structure solution
and refinement, bond angles, lengths, and anisotropic thermal
parameters for the structures 13h and 14j. Preparation
methods and spectroscopic data for nitrones 6, and compounds
9c, 9d , 13c/14c, 13d /14d , 13e/14e, 13f/14f, 13g/14g, 13i/14i,
15c, 21, 22, 23, and 24. Table of ab initio (RHF/3-21G, and
6-31G*) calculated energies and coefficients of the frontier
molecular orbitals of olefins 1a and 17, and nitrones 6a , 6d ,
6e, 6h , and 6k . This material is available free of charge via
the Internet at http://pubs.acs.org.
1260, 1249, 1124, 1086 cm-1 1H NMR (300 MHz, CDCl3) δ
;
2.43 (s, 3H, CH3CO), 3.84 (s, 3H, MeO), 5.85 (d, J ) 10.0 Hz,
1H, H-3), 6.90-7.45 (m, 9H, Ar-H), 8.37 (br d, J ) 10.0 Hz,
1H, H-4); 13C NMR (75.4 MHz, CDCl3) δ 24.0 (CH3CO), 55.5
(MeO), 96.5 (C-3), 114.8, 115.2, 122.1, 125.6, 128.2, 129.6,
129.8, 144.5, 151.4, 159.5, 185.5 (CH3CO); MS (70 eV) 295
(M+, 5), 252 (100), 209 (25), 180 (5), 77 (6). Anal. Calcd for
C
18H17NO3: C, 73.20; H, 5.80; N, 4.74. Found: C, 73.37; H,
J O001393N
5.83; N, 4.93.
Sin gle-Cr ysta l X-r a y Cr ysta llogr a p h y. Isoxazolidines
13h and 14j were obtained as pale yellow and colorless
crystals, respectively. These were mounted in glass fibers.
(54) SHELXTL, v. 5.03, Siemens Energy & Automation, Germany,
1995.