Selectivity in 1,3-Dipolar cycloadditions of β-substituted captodative olefins - An experimental and DFT transition state study

Article abstract of DOI:10.1002/ejoc.200600988

Captodative olefins 1-acetylvinyl carboxylates substituted with alkyl groups at the β position, 12, strongly modified the regioselectivity of 1,3-dipolar cycloadditions with respect to the behavior observed for their unsubstituted analogs, 1. When the rea

Full text of DOI:10.1002/ejoc.200600988

FULL PAPER
DOI: 10.1002/ejoc.200600988
Selectivity in 1,3-Dipolar Cycloadditions of β-Substituted Captodative
Olefins – An Experimental and DFT Transition State Study
Rafael Herrera,[a,b,c] Jorge A. Mendoza,[a] Miguel A. Morales,[c] Francisco Méndez,[b,c]
Hugo A. Jiménez-Vázquez,*[a] Francisco Delgado,[a] and Joaquín Tamariz*[a]
Keywords: Cycloadditions / Nitrones / Regioselectivity / Density functional calculations / Transitions states
Captodative olefins 1-acetylvinyl carboxylates substituted
with alkyl groups at the β position, 12, strongly modified the
regioselectivity of 1,3-dipolar cycloadditions with respect to
the behavior observed for their unsubstituted analogs, 1.
When the reaction of 12 was carried out with diphenyl
nitrone (7a), the corresponding C-4 disubstituted isoxazolid-
ines were obtained as a mixture or as single stereoisomers,
in contrast to the isomeric C-5 disubstituted heterocycles
yielded by olefins 1. Nevertheless, olefins 12 reacted with
benzonitrile oxide (2a) to give the C-5 acetyl isoxazoles, as
observed with dipolarophiles 1. This intricate behavior of the
reactions between 1,3-dipoles and β-substituted captodative
olefins was rationalized on the basis of DFT calculations
[B3LYP/6-31(d)] of the transition states (TSs) for nitrone 7a
and olefin 12a. Thus, the observed C-4 and endo regio- and
stereoselectivities agreed with the most stable TSs, which are
mainly stabilized by dipolar and electrostatic interactions.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
Introduction
As a part of our ongoing research program concerning
the synthesis of new captodative olefins and their study in
cycloaddition reactions,[1] we have described the behavior
of 1-acetylvinyl carboxylates 1 with a diverse set of dipoles
such as nitrile oxides,[2] nitrile imines, diazo compounds,
and nitrones.[3] We found that the addition of aryl and alkyl
nitrile oxides, 2 and 3, C-aryl-N-phenylnitrile imines 4, and
diazo compounds 5 provided exclusively the corresponding
5-substituted aromatic isoxazoles and pyrazoles
6
(Scheme 1). The reactions with nitrones 7 were also highly
regioselective and yielded C-5 substituted heterocycles 8;
this process was stereoselective as well given that endo ad-
duct 8a was the major product.[3]
The regioselectivity observed in the 1,3-dipolar cycload-
ditions of monosubstituted alkenes can be explained in
terms of FMO theory.[4] However, many incorrect predic-
tions of this theory are reported for 1,3-dipolar reactions of
captodative and disubstituted olefins.[5] Diverse factors are
invoked to rationalize the preferred orientation of the cy-
Scheme 1. 1,3-Dipolar cycloadditions of captodative olefins 1 with
diverse dipoles.
cloaddends; among them we can cite steric effects,[6] closed-
shell repulsions,[7] and repulsive secondary orbital interac-
tions.[8] Calculations of transition states have also suggested
the presence of a diradicaloid or highly polarized species,[9]
and pointed out the importance of electrostatic interactions
and solvent effects in controlling the regiochemical out-
come.[10]
With respect to olefins 1, FMO analysis of the reaction
between benzonitrile oxide (2a) and nitrones 7 failed to pre-
dict the observed regioselectivity of the cycloaddition,
which favors the opposite orientation.[2,3b] A correct predic-
[a] Departamento de Química Orgánica, Escuela Nacional de Ci-
encias Biológicas, IPN
Prol. Carpio y Plan de Ayala, 11340 México, D.F., Mexico
Fax: +5255-5729-6300
E-mail: jtamariz@woodward.encb.ipn.mx
hjimenez@woodward.encb.ipn.mx
[b] Instituto de Investigaciones Quimicobiológicas, Universidad
Michoacana de San Nicolás de Hidalgo, Edif. B-1, Ciudad Uni-
versitaria,
Francisco J. Mújica S/N, 58066 Morelia, Mich., Mexico
[c] Departamento de Química, División de Ciencias Básicas e Inge-
niería, Universidad Autónoma Metropolitana-Iztapalapa,
A. P. 55-534, 09340 México, D.F., Mexico
2352
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 2352–2364

Selectivity in 1,3-Dipolar Cycloadditions of β-Substituted Captodative Olefins
FULL PAPER
tion of the regioselectivity was obtained through the estima-
tion of the interaction energy between the reactants in
terms of density functional theory (DFT) and the hard–soft
acid–base principle (HSAB).[11] The theoretical study of de-
rivatives 1 under the principles of the DFT/HSAB model
indicated that the regioselectivity depends mainly on the ef-
fect of the electron-donor group, which increases the nu-
cleophilicity of the unsubstituted terminus C-1 of the olefin.
A mutual electron donation between this carbon atom and
the nitrone C atom (Scheme 2), or the interaction with the
carbon atom of the nitrile oxide as the electrophilic center,
would predict the C-5 substituted adducts, which matches
the experimental results.[3b,11a]
Scheme 3.
ditions with benzonitrile oxide (2a) and nitrones 7. The ex-
perimental results and a theoretical study are described
herein.
Scheme 2. Mutual electron donation between olefins 1 and nitrones
7 (see text).
Results
Preparation of β-Substituted Captodative Alkenes 12
We have also assessed the perturbation effect of a third
substituent at the β position of the double bond of 1 on
Acyclic β-alkyl-substituted olefins 12a–12c were pre-
the reactivity, regio-, and stereoselectivity of these olefins in pared by a straightforward method through treatment of
Diels–Alder reactions.[12] We prepared β-functionalized 1- the base-generated enolate of corresponding α-diketones 13
acetylvinyl arenecarboxylates 9; however, they failed to re- with acid chlorides 14, under conditions similar to those
act with cyclopentadiene (10) and isoprene (11), except for used for the preparation of 1a (Scheme 4).[13] In a previous
β-brominated olefin 9a, which undergoes addition to these report, we described the synthesis of olefin 12a by a cross-
dienes to give the exo and para adducts, respectively, as the coupling reaction between olefin 9a and tetramethyltin in
major isomers (Scheme 3).[12a] The ratio of regioisomers ob- the presence of catalytic amounts of a PdII complex.[14] In
tained with diene 11, under thermal conditions, was similar contrast with that methodology, where 12a was obtained by
to that observed for unsubstituted olefin 1a. The low reac- a four-step synthesis in 38% overall yield, the direct con-
tivity of dienophiles 9b and 9c can be ascribed to efficient densation of 2,3-pentanedione (13a) with p-nitrobenzoyl
delocalization of the electron lone pair of the heteroatom chloride (14a) provided 12a in 94% yield. Similarly, by re-
of the β-substituent towards the double bond.
acting 2,3-hexanedione (13b) and 3,4-hexanedione (13c)
With the aim of evaluating the effect of β-substitution in with acid chlorides 14b and 14c, respectively, olefins 12b
captodative olefins on 1,3-dipolar cycloaddition reactions, and 12c were furnished in good yields. Under the thermo-
we decided to prepare cyclic and noncyclic captodative ole- dynamic conditions used in the reaction, neither geometric
fins substituted by groups with low polarizing strength, isomers 15 nor olefins 16 were obtained. NOE experiments
such as alkyl groups, and to study these alkenes in cycload- confirmed the configuration of the double bond as en-
Scheme 4.
Eur. J. Org. Chem. 2007, 2352–2364
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2353

H. A. Jiménez-Vázquez, J. Tamariz et al.
FULL PAPER
hancements of the signals attributed to the acyl methyl Table 1. 1,3-Dipolar additions of olefins 12a–12c and 12e with
nitrone 7a.[a]
group, and to the β-substituted alkyl group, were measured
upon irradiation of the vinylic 4-H. A procedure analogous
to that above was employed for the preparation of cyclic
enone 12e by using 1,2-cyclohexanedione (13d) and acid
chloride 14a.
Entry
Olefin
7a
[mol equiv.] [°C]
T
t
[h]
Products[b]
(ratio)
Yield[c]
[%]
1
2
3
4
12a
12b
12c
12e
1.9
3.6
3.7
2.0
110
110
80
12
12
12
24
17a/18a (70:30)
17b/18b (55:45)
17c/18c (66:44)
19
50[d]
30[e]
40[f]
63[g]
140
1,3-Dipolar Cycloadditions of Alkenes 12a–12c and 12e to
Nitrone 7a
[a] All reactions were performed under a N2 atmosphere in dry
benzene heated at reflux. [b] Determined by H NMR (300 MHz)
spectroscopy of the crude mixture. [c] Of the mixture after column
chromatography. [d] Olefin 12a (27%) was recovered. [e] Olefin 12b
(36%) was recovered. [f] Olefin 12c (30%) was recovered. [g] Olefin
12e (31%) was recovered.
1
The 1,3-dipolar cycloadditions between diphenyl nitrone
(7a) and olefins 12a–12c were carried out under thermal
conditions (Scheme 5) and provided the corresponding ste-
reoisomeric mixtures of 4-disubstituted isoxazolidines 17–
18 (Table 1). No evidence of the 5-disubstituted regioisomer
tron-withdrawing groups in the aroyloxy group of capto-
dative olefins 1 increase the reaction rate in concerted
Diels–Alder additions.[13a]
1
was found either by chromatography or by H NMR spec-
troscopy, even after the reaction was stopped before the di-
polarophile had disappeared. This regioselectivity strongly
contrasts that expected for both steric[6] and electronic
points of view,[15] and with the behavior of olefin 1a, which
adds to 7a to yield only the C-5 adduct (Scheme 1).[3] The
C-4 orientation would be the most crowded one because the
C-phenyl ring of the nitrone is vicinal to the geminal center
of the olefin. In principle, one might not expect that an
alkyl group at the β position would perturb the polarization
of the double bond as far as to reverse it.[16] However, the
1,3-dipolar additions of nitrones to 1,2-disubstituted olefins
bearing carbonyl and methyl groups with the (E) geometry,
such as olefins 12, yield only the C-4 regioisomer.[17] Other
captodative olefins,[17,18] and most dipolarophiles monosub-
stituted with either electron-donating or electron-with-
drawing groups add to nitrones[18f,19] to yield the C-5 orien-
tation, preferentially.
The endo stereoselectivity of these additions was not as
high as that shown by olefin 1a[3b] or by other captodative
alkenes,[17a,17d,18b,18d,18e] although lower stereoselectivities
were also reported.[18d,20] The endo approach is preferred,
in spite of the possible steric repulsions at the transition
state between the aroyloxy group and the C-phenyl group
of the nitrone (Figure 1). Besides, this endo preference
agrees with the steric interactions generated at the exo tran-
sition state between the nitrone N-phenyl group and both
the aroyloxy group and the substituent in the β position of
these dipolarophiles.[3b,19d] In addition to these interactions
during the exo approach, further steric interactions could
take place between the acyl group of olefins 12a–12c and
As expected, dipolarophile 12c, bearing the 3,5-dini-
trobenzoyloxy group, was more reactive than 12a and 12b,
although all these additions were slower than those of olefin
1a.[3b] The steric hindrance and the hyperconjugative effect
of the alkyl group in the double bond might be responsible
for the lower reactivity of the former dipolarophiles.[4f,6,16]
The faster addition of 12c towards 7a, in comparison with
Figure 1. Possible steric interactions at the exo and endo transition
states in the 1,3-dipolar cycloadditions of nitrone 7a and olefins
those of 12a and 12b, agrees with the observation that elec- 12a–12c.
Scheme 5.
2354
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 2352–2364

Selectivity in 1,3-Dipolar Cycloadditions of β-Substituted Captodative Olefins
FULL PAPER
the C-phenyl group of the nitrone (Figure 1). To investigate C4–C5 atoms in the plane of the envelope (dihedral angle
this hypothesis, we prepared olefin 12d (Scheme 4) and its = 3.2°), and the oxygen out of this plane and syn to the C3-
crystalline structure was determined by X-ray crystallogra- phenyl and C4-PNB groups. As a result of this arrange-
phy (Figure 2).[21] This structure confirmed that the config- ment, the latter two substituents are essentially eclipsed [di-
uration of the double bond was (Z), as anticipated by NOE hedral angle O10–C4–C3–C22 = –4.3°], as are the corre-
experiments. In addition, it showed that the aroyloxy group sponding substituents in the opposite face of the ring [dihe-
adopts a conformation out of the plane formed by the en- dral angle C7–C4–C3–H3 = 0.7°]. The methyl group at C5
one conjugated moiety and that the latter adopts the s-trans is also nearly eclipsed with the PNB group [dihedral angle
conformation. These structural features are in agreement C6–C5–C4–O10 = 22.1°], and it adopts the pseudoequato-
with those observed for analogous molecules,[12,14] and sup- rial position with respect to the conformation of the ring.
port the proposed transition states as long as the geometries Conversely, the configuration of the nitrogen atom is such
of the starting materials are maintained (Figure 1).
that the N-phenyl ring is anti with respect to the C3-phenyl.
Figure 3. X-ray structure of isoxazolidine 17a (ellipsoids with 30%
probability).
Figure 2. X-ray structure of 12d (ellipsoids with 30% probability).
The addition of nitrone 7a to cyclohexenone 12e was
slower than those to olefins 12a–12c because the former
reaction took place at a higher temperature and with a
longer reaction time (Table 1, Entry 4), to give adduct 19 as
a single regio- and stereoisomer (Scheme 5). Notably, the
regioselectivity was the same as that of dipolarophiles 12a–
12c; however, unlike these olefins, 12e showed a preference
for exo stereoselectivity. Therefore, the configuration of the
olefin, (Z) in 12a–12c or (E) in 12e, does not appear to have
a significant effect on the regiochemistry. Then again, the
endo approach of 12e would promote strong destabilizing
interactions between the N-phenyl ring of the nitrone and
the rest of the cyclohexenone σ skeleton. To gain further
insight on the conformation of this kind of cyclohexenone
derivatives, we undertook single-crystal X-ray analysis of a
related compound. Thus, conjugated cyclohexenone 12f was
prepared through the method described in Scheme 4, and
its structure was established by X-ray diffraction (Fig-
ure 4).[21] The ORTEP structure shows that the conforma-
tion of the cycle is a distorted half-chair. The crystal lattice
exhibited static disorder due to the random presence of the
It is well known that nitrones substituted with bulky
groups possess the (Z) configuration, with the N-alkyl or
N-phenyl and the C-aryl groups in a trans relationship,[5d]
as confirmed by X-ray crystallography and NOE experi-
ments.[3b] We assumed that this configuration is maintained
at the transition state, and that the reactants do not iso-
merize during the course of the cycloaddition.[22] The X-ray
structure of 7b (RЈ = phenyl and Ar = C6H4p-OMe) showed
both N-phenyl and C-anisyl groups in conformations
slightly out of the plane formed by the 1,3-dipole (vide in-
fra).[3b] In consequence, the steric interactions illustrated in
Figure 1 become more important if the conformation of the
aromatic rings of the nitrone is retained at such transition
states.
The structures of adducts 17 and 18 were established on
the basis of 2D NMR and NOE experiments. The multi-
plicity and coupling constants were consistent with the non-
vicinal 3-H and 5-H protons in the isoxazolidine ring
(Scheme 5). NOE enhancements of the signals of these pro-
tons were observed when the CH3CO group on C-4 was
irradiated in endo adducts 17a and 17b. In addition, for 17a,
an enhancement of the protons of the phenyl group at C-3
was recorded when the methyl group on C-5 was irradiated.
This assignment was further confirmed by single-crystal X-
ray diffraction of isoxazolidine 17a (Figure 3).[21] The con-
formation of the heterocycle is different to that shown by
the isoxazolidines arising from olefin 1a.[3b] In contrast with
the latter, where the p-nitrobenzoyloxy (PNB) group and
the N-phenyl group are pseudoaxial, in 17a only the N-
phenyl group exhibits this arrangement. The heterocyclic
Figure 4. X-ray structure of cyclohexenone 12f (ellipsoids with 30%
ring adopts the envelope conformation with the N2–C3– probability).
Eur. J. Org. Chem. 2007, 2352–2364
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2355

H. A. Jiménez-Vázquez, J. Tamariz et al.
FULL PAPER
two possible half-chair conformers of the cyclohexenone dition of the latter with 12e. Thus, under similar reaction
ring among different unit cells; the atoms involved in this conditions, 2a added to 12e to yield again a single product,
disorder were C3, C4, C5, C6, and O7. Occupancy factors whose structure corresponded to tetrahydrobenzoxazolone
were determined for both conformers; a value of 0.56 for 22 in good yield (86%) [Equation (2)]; the reaction showed
the half-chair in which C5 is pointing up with respect to the same regioselectivity as that found for 12a.
the plane of the cyclohexenone ring (see Figure 4) and 0.44
for the other conformer was found. According to the above
results, the exo preference found for enone 12e might be
attributed to repulsive interactions between the C4–C5–C6
cyclohexenone methylenes and the N-phenyl ring of the
nitrone, at the endo transition state.
(2)
1,3-Dipolar Cycloadditions of Alkenes 12a and 12e to
Benzonitrile Oxide (2a)
Nitrile oxides reacted with unsubstituted dipolarophile
1a to give directly 3-aryl- and 3-alkyl-5-acetylisoxazoles
(Scheme 1).[2] This regioselectivity parallels that obtained
with nitrones as only the C-5 substituted heterocycle was
observed.[3] From these results, one would expect that the
C-4 regioselectivity observed in the reactions of olefins 12
with nitrones would be preserved in the reactions with ni-
trile oxides. With the goal of confirming this hypothesis, we
carried out the reaction between benzonitrile oxide (2a) and
olefins 12a and 12e.
The structures of adducts 20 and 22 were established by
NOE experiments. Irradiation of the methyl group at C-4
of 20, and the C-4 methylene of 22, resulted in enhancement
of the signals corresponding to the ortho protons of the C-
3 phenyl ring in both molecules.
Discussion
Addition of 2a, prepared in situ from benzohydroxyimi-
noyl chloride and triethylamine in benzene, to olefin 12a
and heated at reflux for 12 h furnished a single isoxazole
[Equation (1)]. Expected 4-acetyl regioisomer 21 was not
isolated from the reaction mixture; instead 5-acetyl-4-
methyl-3-phenylisoxazole (20) was obtained pure in 33%
yield. The low yield of 20 was mainly due to the difficulty
of separating it from the dimer of 2a, but the actual conver-
The opposite regioselectivity observed in the 1,3-dipolar
cycloadditions of olefins 12 with nitrone 7a and benzoni-
trile oxide (2a), which gives rise to the C-4 and C-5 hetero-
cycle substitution of the captodative center, respectively,
suggests that electronic rather than steric control is involved
in the effect that the β substituent has on the polarization
of the double bond. The evaluation of such effects in these
reactions appears to be beyond the scope of FMO theory,
as we have formally demonstrated in the case of olefins 1.
Owing to the success of DFT/HSAB theory in accounting
for the regioselectivity of the 1,3-dipolar cycloadditions of
olefins 1,[3b,11a] we have also applied this theoretical ap-
proach to the rationalization of the behavior of olefins 12
in the additions described above, and these results will be
reported elsewhere. Herein we describe the calculation and
analysis of the geometries and energies of the possible tran-
sition states (TSs) derived from the reaction between
nitrone 7a and dipolarophile 12a.[25]
1
sion to the adduct was higher, as shown by H NMR spec-
troscopic analysis of the crude reaction mixture. The ob-
served regiochemistry agreed either with that observed in
the addition of nitrones to 1a or with the trend observed
for the reaction of diverse dipoles with 1a.[3] Unless the elec-
tron-deficient olefin has a β substituent such as a methyl
group, where the C-4 isomer will be preferred,[23] either cap-
todative olefins[18b,24] or both electron-rich and electron-de-
ficient alkenes,[19g,19h,23,24d,24f] will react with nitrile oxides
to provide preferentially the C-5 regioisomer.
The geometries of the reactants and transition states
were fully optimized at the HF/3-21G, HF/6-31G(d), and
B3LYP/6-31G(d) levels of theory[26] by using GAUSSIAN
94,[27] and by considering previous calculations of analo-
gous molecules.[3b,28] Although the orthogonal conforma-
tion of the aroyloxy group with respect to the enone moiety
for the calculated geometry of 12a was similar to that ob-
(1)
Although our previous results have suggested that steric served in the X-ray structures of 12a[14] and 12d (Figure 2),
interactions are not responsible for the regioselectivity ob- the conformation of the enone conjugated system in the
served in these 1,3-dipolar additions,[3] the relative stereo- most stable conformer calculated for 12d was not the s-trans
chemistry of the β substituent in the dipolarophile may have form (Figure 5) as expected from the X-ray structures.
an influence on the control of the regiochemistry observed However, the relative energy of the s-cis conformer was just
with dipole 2a. This prompted us to examine the cycload- 0.10 kcalmol–1 lower than that of the s-trans conformer.
2356
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 2352–2364

Selectivity in 1,3-Dipolar Cycloadditions of β-Substituted Captodative Olefins
FULL PAPER
Figure 6. Optimized geometry [B3LYP/6-31G(d)] of nitrone 7a
(left) and X-ray structure (right) of 7b (ellipsoids with 30% prob-
ability).
phase or including solvent effects.[5f] It is noteworthy that
both C4-endo-s-cis and C4-endo-s-trans TSs were the most
stable, and that they were also those which lead to experi-
mentally observed adduct 17a. Then again, the C4-exo TSs
were Ͼ3.4 kcalmol–1 more energetic than the C4-endo TSs.
This energy difference would be enough to provide the C4-
endo adduct 17a, exclusively. However, adduct 18a, which
Figure 5. Optimized geometries [B3LYP/6-31G(d)] of olefin 12a for
the s-cis and s-trans conformations. In parenthesis the relative zero-
point energies (kcalmol–1).
The calculated structure of nitrone 7a shows a nonplanar comes from the C4-exo TSs, was also observed in the prod-
conformation of the phenyl rings with respect to the plane uct mixture. In contrast, the C5-endo-s-cis TS was found to
of the dipole, with the N-phenyl ring having a larger twist be more stable than the C4-exo TSs but the corresponding
out of the plane (Figure 6). This structure is in agreement adduct was not detected in the reaction mixture.
with the X-ray structure of N-phenyl-C-(4-anisyl)nitrone
For the 1,3-dipolar cycloaddition of nitrones, it is likely
(7b) (Figure 6), whose synthesis was previously reported.[3b] that the reaction takes place through a concerted asynchro-
These and other nitrones prepared in precedent studies nous transition state.[26,29,30] Indeed, for the C4 TSs, one
show,[3b] either in the solid phase or in solution, the (Z) can observe a highly asynchronous formation of the C–C
configuration. There is evidence that this configuration and C–O bonds. The difference in distance between these
does not isomerize within the course of the reaction;[22] bonds is smaller in the C5 TSs (Figure 8). Interestingly, the
hence, we assumed that it is maintained at the transition shortest distance at the C4 TS corresponded to the forming
state.
C–O bond,[31] which is perhaps a result of the interaction
In terms of the possible modes of approach between the between the most polarized interacting sites: the oxygen of
reactants (7a + 12a) leading to the transition states, we as- the nitrone as the electron-rich center and the β-carbon
sumed that, in terms of the preferred conformation of the atom of the dipolarophile as the electron-deficient center.[32]
captodative olefin (Figure 5), the most favorable approach This strong dipolar (electrostatic) interaction could be also
of the 1,3-dipole would take place onto the face of the one of the main factors controlling the regioselectivity of
double bond anti to the carbonyl oxygen of the PNB group. the process.[10] Although the β-carbon atom of the olefin
Thus, we took into account only three variables: (1) The C- is expected to be less electrophilic because of the electron-
4/C-5 regioisomerism, (2) the endo/exo approaches, and (3) donating effect (+R) of the lone pairs of electrons on the
the s-cis/s-trans conformation of the enone fragment in the oxygen atom of the aroyloxy group, it was demonstrated
captodative olefin. In all, eight transition states were lo- that, because of the conformational preference, this effect is
cated (Figures 7 and 8). The electronic energies (Ee), the not significant in comparison with the electron-withdrawing
electronic energies corrected by the inclusion of zero-point effect (–R) of the acetyl group[28] and the (–I) inductive ef-
energies (E0), and the corresponding relative energies (∆Ee, fect of the aroyloxy group.[33] Another contribution to the
∆E0) for all the calculated TSs are summarized in Table 2. C-4 regioselectivity might be ascribed to CH···π stabilizing
Calculations were carried out in the gas phase; solvent ef- interactions. Visual inspection of the TSs (Figures 7 and 8)
fects were not included[10] because the experimental cyclo- shows a particularly close contact between the ortho pro-
additions took place in a nonpolar solvent (benzene). Usu- tons of the aroyloxy group and the C-phenyl ring of the
ally, a reliable account of regioselectivity can be obtained nitrone in both C4-endo geometries. The distance between
by means of theoretical calculations carried out in the gas the corresponding ortho proton of the aroyloxy group and
Eur. J. Org. Chem. 2007, 2352–2364
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2357

H. A. Jiménez-Vázquez, J. Tamariz et al.
FULL PAPER
Figure 7. Calculated [B3LYP/6-31G(d)] geometries of the endo and exo C4 TSs for the cycloaddition of the s-cis and s-trans conformations
of captodative olefin 12a and nitrone 7a.
Figure 8. Calculated [B3LYP/6-31G(d)] geometries of the endo and exo C5 TSs for the cycloaddition of the s-cis and s-trans conformations
of captodative olefin 12a and nitrone 7a.
2358
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 2352–2364

Selectivity in 1,3-Dipolar Cycloadditions of β-Substituted Captodative Olefins
FULL PAPER
Table 2. B3LYP/6-31G(d) electronic energies (Ee, HA), electronic energies including zero-point energy corrections (E0, HA), and the
corresponding relative energies (∆Ee and ∆E0 kcalmol–1) for the C4 and C5 TSs of the cycloaddition between 7a and 12a.[a]
Energy C4-endo-s-cis
C4-endo-s-trans
C4-exo-s-cis
C4-exo-s-trans
Ee
E0
∆Ee
∆E0
–1526.563685
–1526.137190
0.00
–1526.560558
–1526.134018
1.96
–1526.558143
–1526.131697
3.48
–1526.557640
–1526.130943
3.79
0.00
1.99
3.45
3.92
Energy C5-endo-s-cis
C5-endo-s-trans
C5-exo-s-cis
C5-exo-s-trans
Ee
E0
∆Ee
∆E0
–1526.559104
–1526.132638
2.87
–1526.554465
–1526.127881
5.79
–1526.556804
–1526.130482
4.32
–1526.553748
–1526.127319
6.24
2.86
5.84
4.21
6.19
[a] The relative energies (∆Ee and ∆E0) are given with respect to the most stable transition state.
the centroid of the phenyl ring is 2.747 Å for the C4-endo-
Unlike the stability of the TSs, where the C4-endo geome-
s-cis TS, and 2.704 Å for the C4-endo-s-trans TS. This inter- tries are the most stable, the stability of the adducts is re-
action may also be associated with the endo stereoselectivity versed given that the C5-exo-s-cis adduct is the most stable
of the cycloaddition (Table 1).
product (Table 3), which is in disagreement with the experi-
All the s-trans TSs are more energetic than the corre- mentally obtained C4 adducts. Also in contrast with the
sponding s-cis (Table 2). This can be attributed to a TSs, the rest of the adducts are highly destabilized relative
stronger steric repulsion between the acyl Me group of the to the most stable C5-exo-s-cis adduct. These results sup-
captodative olefin and either the C-phenyl or the N-phenyl ports the well-known hypothesis that the regio- and stereo-
ring of the nitrone in the s-trans TSs.
selectivity of the 1,3-dipolar cycloadditions are kinetically
In both C5-endo TSs, steric hindrance between the β- and not thermodynamically controlled.[4]
methyl group of the dipolarophile and the C-phenyl group
It can be argued that the captodative substitution of our
of the nitrone is progressively developed along the endo ap- alkenes would favor the formation of free radicals and thus
proach (Figure 8). In contrast, the less hindered TSs seem lead to a stepwise cycloaddition. However, the stereoselecti-
to be the C5-exo among all the series, although they are vity observed in these reactions strongly supports the idea
also among the most energetic. This supports the idea that of concerted processes. Alternatively, a more subtle issue
the stability of the regioisomeric TSs is mostly affected by would be whether the transition states for these concerted
electronic and electrostatic effects.[34] However, steric hin- processes have diradical character. One would expect that
drance might also contribute to the stereochemical outcome the highly polar nature of the 1,3-dipoles would lead to
of the cycloadditions.[19d,19i]
transition states with more polar than diradical character,
Notably, the forming C–O bond at the C5 TSs is much a fact that is supported by our current transition state cal-
longer than that of the C4 TSs. This is probably due to the culations; however, a transition state with diradical charac-
fact that, in the former, the captodative center is not as ter cannot be ruled out on the basis of these results. One
electrophilic as the β-carbon center of the double bond; way to rule out or confirm the diradical character of the
hence, the interaction would be less significant. Moreover, transition state would by means of TS calculations involv-
the methyl group at the β-carbon center will increase the π ing configuration interaction (CI) methods; nevertheless,
electron density of the captodative center by a hyperconju- the size of our system is very large for a study of this
gative effect, which in consequence reduces its electrophilic kind, at least with the computational resources we have at
character.[32]
hand.
Table 3. B3LYP/6-31G(d) electronic energies (Ee, HA), electronic energies including zero-point energy corrections (E0, HA), and the
corresponding relative energies (∆Ee and ∆E0, kcalmol–1) for the C4 and C5 adducts of the cycloaddition between 7a and 12a.[a]
Energy C4-endo-s-cis
C4-endo-s-trans
C4-exo-s-cis
C4-exo-s-trans
Ee
E0
∆Ee
∆E0
–1526.607394
–1526.177516
6.21
–1526.607921
–1526.177973
5.88
–1526.605490
–1526.175764
7.41
–1526.606231
–1526.176497
6.94
6.61
6.32
7.71
7.25
Energy C5-endo-s-cis
C5-endo-s-trans
C5-exo-s-cis
C5-exo-s-trans
Ee
E0
∆Ee
∆E0
–1526.612039
–1526.182588
3.30
–1526.606679
–1526.177031
6.66
–1526.617296
–1526.188049
0.00
–1526.607955
–1526.178146
5.86
3.43
6.91
0.00
6.21
[a] The relative energies (∆Ee and ∆E0) are given with respect to the most stable adduct.
Eur. J. Org. Chem. 2007, 2352–2364
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2359

H. A. Jiménez-Vázquez, J. Tamariz et al.
FULL PAPER
An additional comment should be made on the adducts. and the β carbon atom of the captodative olefin, as well as
Once these are formed, it does not really matter whether to a steric interaction between the CH3 group of the cap-
they arose from the s-cis or s-trans transition states, so that todative olefin and the C-Ph group of the dipole at the C5-
they can be described in terms of conformers of the C4/C5 endo TSs. Moreover, these calculations also accounted for
or endo/exo adducts. Owing to the very high conforma- the endo stereoselectivity shown in these cycloadditions, as
tional flexibility of saturated five-membered rings, and to a consequence of supplementary stabilizing CH···π interac-
all the rotamers that can be generated on these adducts, no tions at the C4-endo TSs that take place between the cy-
conformational analysis was carried out on any of them. cloaddends. Therefore, these calculations rationalize the ap-
The geometries whose energies are described in Table 3 parently subtle electronic effect of the β alkyl group in the
correspond to those coming from full optimization of geo- dipolarophiles, which reverses the regiochemical orientation
metries that mostly resembled the corresponding TSs, and with respect to their unsubstituted analogs, when they are
they might not correspond to the most stable conformers adding to nitrones.
of the adducts. In particular, the conformer of 17a observed
in the solid state (a C4-endo adduct) was not obtained
Experimental Section
through this methodology. An additional optimization at
the B3LYP/6-31G(d) level was carried out on the X-ray
structure, which led to a geometry not very different from
the experimental. Figure 9 shows a superposition of the ex-
perimental and theoretical geometries, which shows that
General: Melting points are uncorrected. NMR spectra were re-
corded with Varian Mercury (300 MHz) and Bruker DMX-500
(500 MHz) spectrometers by using TMS as an internal standard.
Mass spectra and high-resolution mass spectra (HRMS) were ob-
most of the differences originate in a small change in the tained in the electron impact (EI) mode at 70 eV and the fast-atom
bombardment (FAB) mode, respectively. Microanalyses were per-
formed by M-H-W Laboratories (Phoenix, AZ) and Centro de In-
vestigaciones Químicas, Universidad Autónoma del Estado de Hid-
algo. Analytical thin-layer chromatography was carried out with E.
Merck silica gel 60 F254 coated 0.25 plates, visualized by long- and
short-wavelength UV light. All air–moisture sensitive reactions
were carried out under a nitrogen atmosphere by using oven-dried
glassware. THF and benzene were freshly distilled from sodium,
and dichloromethane from calcium hydride, prior to use. Triethyl-
amine was freshly distilled from NaOH. All other reagents were
used without further purification.
conformation of the five-membered heterocyclic ring, as
well as in small rotations the bonds between the substitu-
ents and the heterocycle. The good agreement between the
structures shown in Figure 9, despite the fact that one was
obtained in the solid state and the other obtained through
theoretical calculations in the gas phase, gives us confidence
in the theoretical method employed.
General Method for the Preparation of Captodative Olefins 12a–12d:
To a solution of triethylamine in dry THF (19.0 mL) and HMPA
(1.0 mL), at –20 or –10 °C and under an N2 atmosphere, acid chlo-
ride 14 diluted in dry THF was slowly added. Then, a solution of
α-diketone 13 in dry THF was added dropwise. After being stirred
at room temperature for 24 to 36 h, the solvent was removed under
vacuum, and the residue was dissolved in CH2Cl2 (70 mL) and
washed successively with a cold 5% aqueous solution of HCl
Figure 9. Superposition of the X-ray (dark) and calculated [B3LYP/
6-31G(d)] (light) structures of 17a, obtained by superimposing the
N-2, C-4, and C-5 atoms of the heterocycle.
(2ϫ30 mL),
(3ϫ30 mL),
a
a
cold aqueous saturated solution of NH4Cl
cold 10% aqueous solution of NaHCO3
(2ϫ30 mL), and a cold saturated solution of NaCl (2ϫ30 mL).
The organic layer was dried (MgSO4), and the solvent was evapo-
rated under vacuum. The residue was successively purified by flash
column chromatography on silica gel treated with 10% triethyl-
amine (30 g/1.0 g of crude; hexane/EtOAc, 95:5), and by radial
chromatography (hexane/EtOAc, 90:10), to give corresponding ole-
fins 12a–12f.
Conclusions
The presence of a β alkyl substituent in captodative ole-
fins 12 modified their behavior substantially in 1,3-dipolar
cycloadditions with nitrone 7a, relative to unsubstituted di-
(Z)-3-(p-Nitrobenzoyloxy)-3-penten-2-one (12a): Following the ge-
polarophiles 1, and provided the products corresponding to neral procedure, Et3N (0.83 g, 8.26 mmol), 14a (1.0 g, 5.4 mmol) in
THF (25 mL), and 13a (0.413 g, 4.13 mmol) in THF (10 mL) were
mixed at –20 °C and stirred for 36 h to afford 12a (0.97 g, 94%) as
pale yellow crystals. Rf = 0.18 (hexane/EtOAc, 8:2). M.p. 104–
106 °C (ethyl ether/hexane, 8:2) [ref.[14] 104–106 °C]. 1H NMR
(500 MHz, CDCl3): δ = 1.88 (d, J = 7.1 Hz, 3 H, CH3C=), 2.39 (s,
3 H, CH3CO), 6.74 (q, J = 7.1 Hz, 1 H, HC=), 8.28–8.34 (m, 4 H,
H-Ar) ppm. 13C NMR (125 MHz, CDCl3): δ = 12.0 (CH3C=), 25.1
(CH3CO), 123.7, 129.0, 131.3, 134.2, 147.2, 150.9, 162.4, 190.3
C-4 regioselectivity. The formation of the opposite re-
gioisomer, the C-5 substituted aromatic heterocycle, with
benzonitrile oxide (2a) also suggested a significant pertur-
bation of the double bond by the third substituent.
The DFT calculations of TS energies and geometries for
the most stable approaches in both C-4 and C-5 regioiso-
mers, for the addition of nitrone 7a to olefin 12a, are in
agreement with the C-4 regioselectivity experimentally ob-
served. This preference might be associated with a stronger
(CH CO) ppm. IR (CH Cl ): ν = 1742, 1687, 1529, 1420, 1349,
˜
3
2
2
1256, 1245, 1096, 1012 cm–1. MS (70 eV): m/z (%) = 249 (0.6)
dipolar interaction between the oxygen atom of the nitrone [M]+, 150 (100), 134 (2), 120 (6), 104 (20), 92 (10), 76 (14).
2360
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 2352–2364

Selectivity in 1,3-Dipolar Cycloadditions of β-Substituted Captodative Olefins
FULL PAPER
C12H11NO5 (249.13): calcd. C 57.83, H 4.45, N 5.62; found C
57.81, H 4.60, N 5.56.
(12), 76 (14). C13H11NO5 (261.14): calcd. C 59.77, H 4.24, N 5.36;
found C 59.62, H 4.25, N 5.30.
(Z)-3-(p-Bromobenzoyloxy)-3-hexen-2-one (12b): Following the ge-
neral procedure, Et3N (0.69 g, 6.84 mmol), 14b (0.977 g,
4.45 mmol) in THF (20 mL), and 13b (0.39 g, 3.42 mmol) in THF
(10 mL) were mixed at –10 °C and stirred for 24 h to afford 12b
(0.64 g, 63%) as white crystals. Rf = 0.38 (hexane/EtOAc, 8:2). M.p.
2-(3,5-Dinitrobenzoyloxy)-2-cyclohexen-1-one (12f): Following the
general procedure, Et3N (0.76 g, 7.5 mmol), 14a (1.09 g,
4.71 mmol) in THF (15 mL), and 13d (0.41 g, 3.62 mmol) in THF
(10 mL) were mixed at –10 °C and stirred for 24 h to afford 12f
(0.89 g, 82%) as pale brown needles. Rf = 0.26 (hexane/EtOAc, 8:2).
M.p. 153–154 °C (EtOAc/hexane, 8:2). 1H NMR (300 MHz,
CDCl3): δ = 2.16–2.22 (m, 2 H, 5-H), 2.61–2.68 (m, 4 H, 4-H, 6-
H), 6.90 (t, J = 4.2 Hz, 1 H, 3-H), 9.16–9.26 (m, 3 H, H-Ar) ppm.
13C NMR (75.4 MHz, CDCl3): δ = 22.3 (C-5), 24.7 (C-4), 37.7 (C-
6), 122.7, 129.7, 132.5, 137.0, 144.9, 148.6, 160.4, 190.6 (C-1) ppm.
1
55–56 °C (CH2Cl2/hexane, 7:3). H NMR (300 MHz, CDCl3): δ =
1.10 (t, J = 7.6 Hz, 3 H, CH3CH2C=), 2.28 (q, J = 7.6 Hz, 2 H,
CH3CH3C=), 2.36 (s, 3 H, CH3CO), 6.58 (t, J = 7.6 Hz, 1 H,
HC=), 7.61–7.64 (m, 2 H, H-Ar), 7.98–8.01 (m, 2 H, H-Ar) ppm.
13C NMR (75.4 MHz, CDCl3): δ = 12.6 (CH3CH2C=), 19.7
(CH3CH2C=), 25.1 (CH3CO), 127.6, 128.8, 131.60, 131.65, 134.6,
IR (CH Cl ): ν = 1748, 1696, 1630, 1541, 1349, 1142, 751 cm–1.
˜
2
2
3
2
2
MS (70 eV): m/z (%) = 306 (9) [M]+, 278 (2), 195 (100), 149 (18),
103 (7), 75 (24). C13H10N2O7 (306.14): calcd. C 50.99, H 3.29, N
9.15; found C 50.87, H 3.41, N 9.21.
145.7, 163.5, 191.0 (CH CO) ppm. IR (CH Cl ): ν = 1738, 1687,
˜
1589, 1484, 1399, 1366, 1300, 1282, 1241, 1141, 1109, 1093, 1067,
1013 cm–1. MS (70 eV): m/z (%) = 298 (1) [M + 2]+, 296 (1) [M]+,
185 (100), 183 (99), 157 (13), 155 (14), 131 (1), 129 (1), 104 (2), 77
(1). C13H13BrO3 (297.04): calcd. C 52.55, H 4.41; found C 52.67,
H 4.53.
General Procedure for the 1,3-Dipolar Addition of 12a–12d with
Nitrone 7a. (3R*,4S*,5S*)-4-Acetyl-5-methyl-4-(p-nitrobenzoyloxy)-
2,3-diphenylisoxazolidine (17a) and (3R*,4R*,5R*)-4-Acetyl-5-
(Z)-4-(3,5-Dinitrobenzoyloxy)-4-hexen-3-one (12c): Following the
general procedure, Et3N (0.69 g, 6.84 mmol), 14c (1.02 g,
4.45 mmol) in THF (20 mL), and 13c (0.39 g. 3.42 mmol) in THF
(10 mL) were mixed at –10 °C and stirred for 24 h to afford 12c
(0.88 g, 83%) as brown crystals. Rf = 0.48 (hexane/EtOAc, 8:2).
M.p. 73–75 °C (CH2Cl2/hexane, 7:3). 1H NMR (500 MHz, CDCl3):
δ = 1.15 (t, J = 7.3 Hz, 3 H, CH3CH2CO), 1.94 (d, J = 7.1 Hz, 3
H, CH3C=), 2.77 (q, J = 7.3 Hz, 2 H, CH3CH2CO), 6.81 (q, J =
7.1 Hz, 1 H, HC=), 9.23–9.25 (m, 2 H, H-Ar), 9.26–9.28 (m, 1 H,
H-Ar) ppm. 13C NMR (125 MHz, CDCl3): δ = 7.8 (CH3CH2CO),
11.9 (CH3C=), 30.1 (CH3CH2C=), 122.8, 128.5, 129.8, 132.5,
methyl-4-(p-nitrobenzoyloxy)-2,3-diphenylisoxazolidine (18a):
A
mixture of 7a (0.30 g, 1.52 mmol) and 12a (0.2 g, 0.8 mmol) in dry
benzene (10 mL), in a threaded ACE glass pressure tube with a
Teflon screw cap, was heated to 110 °C and stirred for 12 h under
an N2 atmosphere. The solvent was removed under vacuum to yield
a mixture of 17a/18a (70:30) as an oily residue. The mixture was
purified by flash column chromatography on silica gel (30 g, hex-
ane/EtOAc, 95:5) to give 17a and 18a (0.179 g, 50%) and recovered
12a (0.054 g, 27%). Additional purification by flash column
chromatography on silica gel (20 g, hexane/EtOAc, 10:2) allowed
the isolation of 17a (0.079 g, 22%) as pale green crystals, which
were recrystallized from hexane/EtOAc (10:2), and 18a (0.05 g,
14%) as a pale green oil. Data for 17a: Rf = 0.64 (hexane/EtOAc,
146.6, 148.6, 160.3, 192.7 (CH CH CO) ppm. IR (CH Cl ): ν =
˜
3
2
2
2
1751, 1688, 1627, 1547, 1345, 1282, 1225, 1152 cm–1. MS (70 eV):
m/z (%) = 308 (4) [M]+, 279 (4), 195 (100), 179 (3), 149 (24), 103
(5), 75 (19). C13H12N2O7 (308.14): calcd. C 50.66, H 3.92, N 9.09;
found C 50.88, H 4.05, N 9.09.
1
7:3). M.p. 146–147 °C. H NMR (300 MHz, CDCl3): δ = 1.63 (d,
J = 6.3 Hz, 3 H, CH3), 2.24 (s, 3 H, COMe), 4.70 (q, J = 6.3 Hz,
1 H, 5-H), 5.39 (s, 1 H, 3-H), 6.97–7.03 (m, 3 H, H-Ph), 7.19–7.29
(m, 5 H, H-Ph), 7.46–7.54 (m, 2 H, H-Ph), 7.70–7.76 (m, 2 H, H-
Ar), 8.13–8.20 (m, 2 H, H-Ar) ppm. 13C NMR (75.4 MHz, CDCl3):
δ = 14.4 (CH3), 25.3 (COCH3), 74.5 (C-3), 78.2 (C-5), 96.4 (C-4),
115.1, 122.5, 123.4, 128.2, 128.3, 128.9, 129.3, 130.7, 133.7, 136.1,
(Z)-4-(p-Nitrobenzoyloxy)-4-hexen-3-one (12d): Following the gene-
ral procedure, Et3N (0.69 g, 6.84 mmol), 14a (0.82 g, 4.45 mmol) in
THF (20 mL), and 13c (0.39 g, 3.42 mmol) in THF (10 mL) were
mixed at –10 °C and stirred for 24 h to afford 12d (0.66 g, 73%)
as yellow crystals. Rf = 0.38 (hexane/EtOAc, 8:2). M.p. 98–100 °C
149.3, 150.7, 163.8 (CO Ar), 199.4 (COMe) ppm. IR (KBr): ν =
˜
2
1730, 1598, 1489, 1349, 1279, 1099 cm–1. HRMS (FAB): calcd. for
1
(CH2Cl2/hexane, 7:3). H NMR (500 MHz, CDCl3): δ = 1.15 (t, J
C25H22N2O6 [M]+ 446.1478; found 446.1495. Data for 18a: Rf =
= 7.3 Hz, 3 H, CH3CH2CO), 1.88 (d, J = 7.1 Hz, 3 H, CH3C=),
2.74 (q, J = 7.3 Hz, 2 H, CH3CH2CO), 6.73 (q, J = 7.1 Hz, 1 H,
HC=), 8.31–8.35 (m, 4 H, H-Ar) ppm. 13C NMR (125 MHz,
CDCl3): δ = 7.9 (CH3CH2CO), 11.8 (CH3C=), 30.3 (CH3CH2C=),
123.6, 127.7, 131.2, 134.2, 146.7, 150.8, 162.4, 193.4 (CH3CH2CO)
1
0.72 (hexane/EtOAc, 7:3). H NMR (300 MHz, CDCl3): δ = 1.34
(d, J = 6.4 Hz, 3 H, CH3), 1.82 (s, 3 H, COMe), 5.14 (s, 1 H, 3-
H), 5.30 (q, J = 6.4 Hz, 1 H, 5-H), 6.93–7.03 (m, 3 H, H-Ph), 7.21–
7.30 (m, 2 H, H-Ph), 7.34–7.45 (m, 3 H, H-Ph), 7.55–7.62 (m, 2 H,
H-Ph), 8.04–8.12 (m, 2 H, H-Ar), 8.26–8.33 (m, 2 H, H-Ar) ppm.
13C NMR (75.4 MHz, CDCl3): δ = 14.9 (CH3), 26.9 (COCH3), 76.7
(C-3), 77.6 (C-5), 98.1 (C-4), 115.3, 122.4, 123.7, 127.9, 128.8, 129.1
(2 C), 131.1, 133.9, 134.8, 150.8, 150.9, 163.7 (CO2Ar), 199.2
ppm. IR (CH Cl ): ν = 1741, 1728, 1530, 1348, 1282, 1099 cm–1.
˜
2
2
MS (70 eV): m/z (%) = 263 (1) [M]+, 234 (1), 150 (100), 120 (22),
104 (21), 76 (19). C13H13NO5 (263.14): calcd. C 59.31, H 4.98, N
5.32; found C 59.29, H 5.07, N 5.32.
(COMe) ppm. IR (film): ν = 1730, 1599, 1528, 1490, 1349, 1279,
˜
2-(p-Nitrobenzoyloxy)-2-cyclohexen-1-one (12e): Following the ge-
neral procedure, Et3N (0.76 g, 7.5 mmol), 14a (0.87 g, 4.71 mmol)
in (15 mL) THF, and 13d (0.41 g, 3.62 mmol) in THF (10 mL) were
mixed at –10 °C and stirred for 24 h to afford 12e (0.85 g, 90%) as
pale brown needles. Rf = 0.32 (hexane/EtOAc, 8:2). M.p. 130–
131 °C (EtOAc/hexane, 8:2). 1H NMR (500 MHz, CDCl3): δ =
1101 cm–1. HRMS (FAB): calcd. for C25H22N2O6 [M]+ 446.1478;
found 446.1475.
(3R*,4S*,5S*)-4-Acetyl-4-(p-bromobenzoyloxy)-5-methyl-2,3-di-
phenylisoxazolidine (17b) and (3R*,4R*,5R*)-4-Acetyl-4-(p-bromo-
benzoyloxy)-5-methyl-2,3-diphenylisoxazolidine (18b): A mixture of
2.12–2.17 (m, 2 H, 5-H), 2.58–2.64 (m, 4 H, 4-H, 6-H), 6.78 (t, J 7a (0.5 g, 2.5 mmol) and 12b (0.2 g, 0.7 mmol) in dry benzene
= 4.3 Hz, 1 H, 3-H), 8.26–8.29 (m, 4 H, H-Ar) ppm. 13C NMR (15 mL), in a threaded ACE glass pressure tube with a Teflon screw
(125 MHz, CDCl3): δ = 22.5 (C-5), 24.9 (C-4), 38.0 (C-6), 123.6,
cap, was stirred and heated to 110 °C for 12 h under an N2 atmo-
131.3, 134.3, 136.7, 145.2, 150.8, 162.6, 191.2 (C-1) ppm. IR
sphere. The solvent was removed under vacuum to yield an oily
(CH Cl ): ν = 1743, 1690, 1530, 1350, 1239, 1110 cm–1. MS residue of a mixture of 17b/18b (55:45). The mixture was purified
˜
2
2
(70 eV): m/z (%) = 261 (2) [M]+, 150 (100), 120 (8), 104 (24), 92
by flash column chromatography on silica gel (30 g, hexane/EtOAc,
Eur. J. Org. Chem. 2007, 2352–2364
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2361

H. A. Jiménez-Vázquez, J. Tamariz et al.
FULL PAPER
9:1) to give a mixture of 17b and 18b (0.11 g, 30%) as a green oil,
and recovered 12b (0.072 g, 36%). The mixture was separated by
radial chromatography (hexane/EtOAc, 95:5) to isolate 17b (0.07 g,
21%) as a colorless oil and 18b (0.02 g, 6%) as a colorless oil.
161.7 (CO Ar), 201.4 (COEt) ppm. IR (film): ν = 1738, 1720, 1545,
˜
2
1344, 1278, 1159, 1074 cm–1. HRMS (FAB): calcd. for C26H23N3O8
[M]+ 505.1485; found 505.1493.
(3R*,4R*,5S*)-3a-(p-Nitrobenzoyloxy)-4-oxo-2,3-diphenyl-
2,3,3a,4,5,6,7,7a-octahydrobenzo-4H-isoxazole (19): A mixture of
7a (0.30 g, 1.52 mmol) and 12e (0.20 g, 0.77 mmol) in dry benzene
(10 mL), in a threaded ACE glass pressure tube with a Teflon screw
cap, was stirred and heated to 140 °C for 24 h under an N2 atmo-
sphere. The solvent was removed under vacuum to yield a green
oily residue, which was purified by flash column chromatography
on silica gel (20 g, hexane/EtOAc, 100:6) to give 19 (0.22 g, 63%)
as yellow crystals and recovered 12e (0.062 g, 31%). Data for 19: Rf
= 0.4 (hexane/EtOAc, 7:3). M.p. 157–158 °C. 1H NMR (300 MHz,
CDCl3): δ = 1.18–1.37 (m, 1 H, 6-H), 1.62–2.00 (m, 3 H, 6-H, 7-
H), 2.23–2.43 (m, 2 H, 5-H), 4.86 (dd, J = 6.9, 6.5 Hz, 1 H, 7a-H),
5.56 (s, 1 H, 3-H), 7.02–7.08 (m, 1 H, H-Ph), 7.10–7.18 (m, 2 H,
H-Ph), 7.26–7.44 (m, 7 H, H-Ph, H-Ar), 7.54–7.60 (m, 2 H, H-Ph),
8.02–8.10 (m, 2 H, H-Ar) ppm. 13C NMR (75.4 MHz, CDCl3): δ
= 18.1 (C-6), 28.3 (C-7), 39.8 (C-5), 77.7 (C-3), 85.1 (C-7a), 96.7
(C-3a), 114.3, 122.1, 123.3, 128.4, 128.6, 128.7, 129.2, 130.8, 134.0,
1
Data for 17b: Rf = 0.7 (hexane/EtOAc, 7:3). H NMR (300 MHz,
CDCl3): δ = 1.23 (t, J = 7.4 Hz, 3 H, CH3CH2), 1.64–1.80 (m, 1
H, CH3CH2), 2.08–2.22 (m, 1 H, CH3CH2), 2.21 (s, 3 H, COMe),
4.44 (dd, J = 10.5, 2.8 Hz, 1 H, 5-H), 5.27 (s, 1 H, 3-H), 6.93–7.04
(m, 3 H, H-Ph), 7.17–7.30 (m, 5 H, H-Ph), 7.38–7.54 (m, 6 H, H-
Ph, H-Ar) ppm. 13C NMR (75.4 MHz, CDCl3 ): δ = 10.9
(CH3CH2), 22.0 (CH3CH2), 25.3 (COCH3), 74.3 (C-3), 83.2 (C-5),
95.6 (C-4), 115.2, 122.3, 127.3 (2 C), 128.1, 128.2, 128.8, 129.3,
131.1, 131.6, 136.2, 144.0, 162.8 (CO2Ar), 200.7 (COMe) ppm. IR
(film): ν = 1720, 1591, 1486, 1285, 1087, 1011, 750 cm–1
.
˜
C26H24BrNO4 (494.19): calcd. C 63.17, H 4.89, N 2.83; found C
63.23, H 5.03, N 2.86. Data for 18b: Rf = 0.6 (hexane/EtOAc, 7:3).
1H NMR (300 MHz, CDCl3): δ = 1.12 (t, J = 7.3 Hz, 3 H,
CH3CH2), 1.48–1.68 (m, 2 H, CH3CH2), 1.79 (s, 3 H, COMe), 5.10
(dd, J = 10.0, 3.1 Hz, 1 H, 5-H), 5.10 (s, 1 H, 3-H), 6.94–7.06 (m,
3 H, H-Ph), 7.20–7.32 (m, 3 H, H-Ph), 7.34–7.64 (m, 6 H, H-Ph,
H-Ar), 7.77–7.83 (m, 2 H, H-Ar) ppm. 13C NMR (75.4 MHz,
CDCl3): δ = 10.9 (CH3CH2), 22.2 (CH3CH2), 26.8 (COCH3), 76.5
(C-3), 83.0 (C-5), 97.5 (C-4), 114.6, 122.0, 127.9, 128.7, 128.8,
128.9, 129.0, 129.2, 131.4, 132.0, 135.0, 150.8, 162.5 (CO2Ar), 199.9
134.1, 150.6, 151.1, 162.3 (CO Ar), 202.5 (C-4) ppm. IR (KBr): ν
˜
2
= 1717, 1599, 1527, 1489, 1347, 1277, 1102 cm–1. MS (70 eV): m/z
(%) = 458 (14) [M]+, 291 (35), 262 (30), 234 (10), 180 (15), 150
(100), 104 (33), 76 (20). C26H22N2O6 (458.29): calcd. C 68.11, H
4.84, N 6.11; found C 68.29, H 5.07, N 5.95.
(COMe) ppm. IR (film): ν = 1686, 1497, 1421, 1360, 1277, 1218,
˜
955, 769, 693 cm–1. HRMS (FAB): calcd. for C26H24BrNO4 [M]+
493.0889; found 493.0889.
5-Acetyl-4-methyl-3-phenylisoxazole (20): To a solution of 12a
(0.15 g, 0.60 mmol) in dry benzene (10 mL), benzohydroxyiminoyl
chloride (2.0 g, 12.8 mmol) and Et3N (0.36 g, 3.59 mmol) were
added at room temperature. After the mixture was stirred and
heated to 80 °C for 12 h, the solvent was removed under vacuum,
and the residue was dissolved in CH2Cl2 (20 mL), washed with a
5% aqueous solution of HCl (2ϫ30 mL), and water (2ϫ20 mL).
The organic layer was dried (MgSO4), and the solvent was evapo-
rated under vacuum to yield a brown residue, which was purified
by flash column chromatography on silica gel (15 g, hexane) to give
20 (0.04 g, 33%) as a pale brown oil. Rf = 0.66 (hexane/EtOAc,
(3R*,4S*,5S*)-5-Methyl-4-(3,5-dinitrobenzoyloxy)-2,3-diphenyl-4-
propanoylisoxazolidine (17c) and (3R*,4R*,5R*)-5-Methyl-4-(3,5-
dinitrobenzoyloxy)-2,3-diphenyl-4-propanoylisoxazolidine (18c): A
mixture of 7a (0.48 g, 2.43 mmol) and 12c (0.20 g, 0.65 mmol) in
dry benzene (15 mL), in a threaded ACE glass pressure tube with
a Teflon screw cap, was stirred and heated to 80 °C for 12 h under
an N2 atmosphere. The solvent was removed under vacuum to yield
a green oily residue of a mixture of 17c/18c (66:44). The mixture
was purified by flash column chromatography on silica gel (30 g,
hexane/EtOAc, 9:1) to give a mixture of 17c and 18c (0.2 g, 40%)
and recovered 12c (0.06 g, 30 %). By crystallization (hexane/
CH2Cl2, 8:2) of the mixture, 17c (0.128 g, 25%) was isolated as pale
brown needles and 18c (0.041, 8%) as a pale brown oil. Data for
17c: Rf = 0.5 (hexane/EtOAc, 7:3). M.p. 152–153 °C. 1H NMR
(300 MHz, CDCl3): δ = 1.14 (t, J = 7.1 Hz, 3 H, CH3CH2CO),
1.61 (d, J = 6.3 Hz, 3 H, CH3), 2.59–2.63 (m, 2 H, CH3CH2CO),
4.82 (q, J = 6.3 Hz, 1 H, 5-H), 5.14 (s, 1 H, 3-H), 6.98–7.01 (m, 2
H, H-Ph), 7.21–7.50 (m, 8 H, H-Ph), 8.60 (br. s, 2 H, H-Ar), 9.14
(br. s, 1 H, H-Ar) ppm. 13C NMR (75.4 MHz, CDCl3): δ = 7.9
(CH3CH2CO), 14.9 (CH3), 30.7 (CH3CH2CO), 74.6 (C-3), 77.9 (C-
5), 99.4 (C-4), 115.6, 122.7, 122.8, 128.6, 128.8, 128.9, 129.0, 129.2,
132.9, 134.9, 148.8, 150.8, 161.5 (CO2Ar), 201.3 (COEt) ppm. IR
1
8:2). H NMR (300 MHz, CDCl3): δ = 2.42 (s, 3 H, CH3), 2.66 (s,
3 H, COMe), 7.47–7.55 (m, 3 H, H-Ph), 7.62–7.67 (m, 2 H, H-
Ph) ppm. 13C NMR (75.4 MHz, [D6]acetone): δ = 8.1 (CH3), 27.3
(COCH3), 117.9, 128.3, 128.6, 129.0, 130.0, 161.8, 163.8, 188.5
(COMe) ppm. IR (film): ν = 1696, 1604, 1452, 1359, 1286, 1204,
˜
924, 772, 701 cm–1. HRMS (FAB): calcd. for C12H12NO2 [M+H]+
202.0868; found 202.0866.
7-Oxo-3-phenyl-4,5,6,7-tetrahydrobenzo-7H-isoxazole (22): The
same procedure used for 20 was followed: 12e (0.1 g, 0.4 mmol),
benzohydroxyiminoyl chloride (0.51 g 3.3 mmol), and Et3N (0.73 g
7.2 mmol) reacted to afford a residue that was purified by column
chromatography on silica gel (20 g, hexane/EtOAc, 8:2) to give 22
(0.07 g, 86%) as colorless crystals (hexane/EtOAc, 8:2). Rf = 0.5
(hexane/EtOAc, 7:3). M.p. 114–15 °C. 1H NMR (300 MHz,
CDCl3): δ = 2.23–2.36 (m, 2 H, 5-H), 2.69–2.74 (m, 2 H, 6-H), 2.97
(t, J = 6.0 Hz, 2 H, 4-H), 7.49–7.55 (m, 3 H, H-Ph), 7.72–7.79 (m,
2 H, H-Ph) ppm. 13C NMR (75.4 MHz, CDCl3): δ = 21.6 (C-5),
24.4 (C-4), 38.6 (C-6), 127.5, 127.9, 128.2, 129.1, 130.3, 160.8,
160.9, 186.3 (C-7) ppm. IR (KBr): 1690, 1600, 1521, 1444, 1327,
1302, 928, 884 cm–1. MS (70 eV): m/z (%) = 213 (100) [M]+, 185
(47), 184 (95), 156 (64), 129 (65), 103 (25), 77 (32). HRMS (FAB):
calcd. for C13H11NO2 [M]+ 213.0790; found 213.0805.
(KBr): ν = 1736, 1597, 1546, 1489, 1455, 1344, 1282, 1161,
˜
1077 cm–1. MS (70 eV): m/z (%) = 505 (79) [M]+, 448 (1), 461 (4),
294 (10), 278 (20), 238 (10), 196 (48), 195 (77), 180 (50), 149 (28),
105 (25), 91 (100), 77 (37). HRMS (FAB): calcd. for C26H23N3O8
[M]+ 505.1485; found 505.1486. Data for 18c: Rf = 0.6 (hexane/
EtOAc, 7:3). 1H NMR (300 MHz, CDCl3): δ = 0.70 (t, J = 7.1 Hz,
3 H, CH3CH2CO), 1.39 (d, J = 6.5 Hz, 3 H, CH3), 2.31 (q, J =
7.1 Hz, 2 H, CH3CH2CO), 5.25 (s, 1 H, 3-H), 5.28 (q, J = 6.5 Hz,
1 H, 5-H), 6.94–7.30 (m, 3 H, H-Ph), 7.21–7.29 (m, 2 H, H-Ph),
7.34–7.47 (m, 3 H, H-Ph), 7.58–7.62 (m, 2 H, H-Ph), 9.06 (d, J =
2.1 Hz, 2 H, H-Ar), 9.26 (t, J = 2.1 Hz, 1 H, H-Ar) ppm. 13C
Single-Crystal X-ray Crystallography: Enones 12d and 12f and isox-
NMR (75.4 MHz, CDCl3): δ = 7.9 (CH3CH2CO), 14.8 (CH3), 30.7 azolidine 17a were obtained as pale brownish, yellow, and green
(CH3CH2CO), 74.5 (C-3), 77.9 (C-5), 99.1 (C-4), 114.8, 122.4, crystals, respectively. These were mounted on glass fibers. Crystal-
122.9, 127.5, 128.9, 129.1, 129.2, 129.6, 132.7, 134.8, 148.7, 150.8,
lographic measurements were performed with a Siemens P4 dif-
2362
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 2352–2364

Selectivity in 1,3-Dipolar Cycloadditions of β-Substituted Captodative Olefins
FULL PAPER
fractometer with Mo (12d and 12f) and Cu (17a) Kα radiation (λ
= 0.7107 Å and 1.54178 Å, respectively; graphite monochromator)
at room temperature. Three standard reflections were monitored
periodically; they showed no appreciable change during data collec-
tion. Unit cell parameters were obtained from least-squares refine-
ment of 50 (12d), 39 (12f), and 41 (17a) reflections in the ranges
10.2Ͻ2θϽ24.9°, 9.8Ͻ2θϽ25.7°, and 21.4Ͻ2θϽ56.1°, respectively.
Intensities were corrected for Lorentz and polarization effects. No
absorption corrections were applied for 12d and 12f, whereas for
17a an empirical absorption correction was employed. 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 refine-
ment. Structures were solved by using SHELXTL[35] on a personal
computer. Data of 12d: Formula: C13H13NO5; molecular weight:
263.24 gmol–1; cryst. syst.: monoclinic; space group: P21/n; unit cell
parameters: a, 8.1617 (9), b, 14.0618 (9), c, 11.7048 (12) (Å); α, 90,
β, 104.131 (12), γ, 90 (°); V = 1302.7(2) Å3; temp.: 293 (2)°K; Z: 4;
no. of reflections collected: 2333; no. of observed reflections: 1700;
R: 0.0370; wR = 0.0949; GOF: 1.043. Data for 12f: Formula:
C13H10N2O7; molecular weight: 306.23 gmol–1; cryst. syst.: mono-
clinic; space group: P21/n; unit cell parameters: a, 10.7549 (11), b,
5.4943 (4), c, 22.919 (4) (Å); α, 90, β, 91.214 (9), γ, 90 (°); V =
1354.0(3) Å3; temp.: 293 (2)°K; Z: 4; no. of reflections collected:
4321; no. of observed reflections: 3036; R: 0.0543; wR = 0.1304;
GOF: 1.013. Data for 17a: Formula: C25H22N2O6; molecular
921769, 200410, 20050151, and 20060583) and CONACYT (Grants
1570P, 32273-E, and 43508Q) for financial support. F. M. acknowl-
edges 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 Lud-
wig K. Hellweg Foundation for a partial scholarship. M. A. M.
thanks CONACYT for a Ph.D. scholarship. J. T., F. D., and
H. A. J.-V. are fellows of EDI/IPN and COFAA/IPN.
[1] a) R. Aguilar, A. Reyes, J. Tamariz, J.-L. Birbaum, Tetrahedron
Lett. 1987, 28, 865–868; b) A. Reyes, R. Aguilar, H. Muñoz,
J.-C. Zwick, M. Rubio, J.-L. Escobar, M. Soriano, R. Toscano,
J. Tamariz, J. Org. Chem. 1990, 55, 1024–1034; c) O.
García de Alba, J. Chanona, F. Delgado, G. Zepeda, F. Labar-
rios, R. W. Bates, S. Bott, E. Juaristi, J. Tamariz, Anal. Chem.
Int. Ed. 1996, 92, 108–117; d) R. Herrera, H. A. Jiménez-
Vázquez, A. Modelli, D. Jones, B. C. Söderberg, J. Tamariz,
Eur. J. Org. Chem. 2001, 4657–4669. Even though the term
captodative was originally coined for radical species that are
overstabilized by the presence of two substituent groups with
opposite electronic demand (captodative effect), it was also
used to describe those alkenes that are geminally substituted
by an electron-withdrawing group and an electron-donating
group;[1a–1d,9a] for some more recent examples, see: e) A. Y. Ru-
lev, Russ. Chem. Rev. 2002, 71, 195–221; f) A. Liard, T.-H.
Nguyen, A. I. D. Smir, M. Vaultier, A. Derdour, J. Mortier,
Chem. Eur. J. 2003, 9, 1000–1007; g) H. R. Memarian, M. Deh-
ghani, G. Henkel, D. Döpp, Monatsh. Chem. 2004, 135, 425–
433.
–1
¯
weight: 446.45 gmol ; cryst. syst.: triclinic; space group: P1; unit
cell parameters: a, 8.0780 (5), b, 12.2879 (8), c, 12.4725 (6) (Å); α,
67.057 (5), β, 87.978 (4), γ, 78.508 (6) (°); V = 1116.02(11) Å3;
temp.: 293 (2)°K; Z: 2; no. of reflections collected: 3685; no. of
observed reflections: 2926; R: 0.0881; wR = 0.2691; GOF: 1.103.
[2] R. Jiménez, L. Pérez, J. Tamariz, H. Salgado, Heterocycles
1993, 35, 591–598.
[3] a) A. Nagarajan, G. Zepeda, J. Tamariz, Tetrahedron Lett.
1996, 38, 6835–6838; b) R. Herrera, A. Nagarajan, M. A. Mo-
rales, F. Méndez, H. A. Jiménez-Vázquez, L. G. Zepeda, J. Ta-
mariz, J. Org. Chem. 2001, 66, 1252–1263.
Calculation Methods: All the calculations described in this work
were carried out with the Gaussian 94 program package,[27] on per-
sonal computers running under the Linux operating system. All
optimizations were first carried out at the at the HF/3-21G level of
theory, and the resulting geometries were employed as starting
points for further optimizations at the HF/6-31G(d) and B3LYP/
6-31G(d) levels. In all optimizations (minima and transition states)
the OPT = TIGHT keyword was employed; in addition, all the
DFT calculations were carried out with the INT(GRID = 99590)
keyword, to obtain better energies and vibrational frequencies by
means of a finer integration grid. The initial geometries of the reac-
tants were derived either from previous studies or from X-ray struc-
tures. For each one of the approaches of the reactants described
in the main text, geometries were generated for the corresponding
products (in this case a single conformer was optimized, no attempt
was made to carry out a conformational search). From these geo-
metries, and those of the reactants, the transition states were ob-
tained employing the QST2 (QST3 at the highest levels) option of
the OPT keyword. For all stationary points, vibrational analyses
were carried out at each level of ab initio theory; each point was
characterized by the appropriate number of imaginary vibrational
frequencies. Each transition state was further characterized by vis-
ual inspection of the normal mode corresponding to its single
imaginary frequency. Relative energies were obtained by sub-
tracting the energy of the lowest-energy structures (TSs or minima)
from the energies of all the other geometries and converting these
differences into kcalmol–1.
[4] a) R. Huisgen in 1,3-Dipolar Cycloaddition Chemistry (Ed.: A.
Padwa), Wiley, New York, 1984, vol. 1; b) K. N. Houk, J. Sims,
R. E. Duke Jr, R. W. Strozier, J. K. George, J. Am. Chem. Soc.
1973, 95, 7287–7301; c) W. Carruthers, Cycloaddition Reactions
in Organic Synthesis, Pergamon Press, Oxford, 1990, pp. 269–
331; d) R. Huisgen, J. Org. Chem. 1976, 41, 403–419; e) K. N.
Houk, J. Sims, C. R. Watts, L. J. Luskus, J. Am. Chem. Soc.
1973, 95, 7301–7315; f) K. N. Houk, Acc. Chem. Res. 1975,
8, 361–369; g) R. Jasinski, A. Ciezkowska, A. Lyubimtsev, A.
Baranski, Chem. Heterocycl. Compd. 2004, 40, 206–210.
[5] a) P. Grünanger, P. Vita-Finzi, “Isoxazoles” in The Chemistry
of Heterocyclic Compounds (Ed.: E. C. Taylor), John Wiley &
Sons, New York, 1991, vol. 49, part 1 ; b) R. Sustmann, W.
Sicking, Chem. Ber. 1987, 120, 1653–1658; c) L. Fisera, M.
Konopíková, P. Ertl, N. Prónayová, Monatsh. Chem. 1994, 125,
301–312; d) A. Padwa, D. N. Kline, K. F. Koehler, M. Matz-
inger, M. K. Venkatramanan, J. Org. Chem. 1987, 52, 3909–
3917; e) M. Shanmugasundaram, R. Raghunathan, E. J.
Padma Malar, Heteroat. Chem. 1998, 9, 517–522; f) A. Rastelli,
R. Gandolfi, M. S. Amadé, J. Org. Chem. 1998, 63, 7425–7436.
[6] a) R. Huisgen, “Steric Course and Mechanism of 1,3-Dipolar
Cycloadditions” in Advances in Cycloaddition (Ed.: D. P. Cur-
ran), JAI Press, Greenwich, 1988, vol. 1, pp. 1–31; b) M. A.
Weidner-Wells, S. A. Fraga-Spano, I. J. Turchi, J. Org. Chem.
1998, 63, 6319–6328.
[7] R. Sustmann, W. Sicking, M. Felderhoff, Tetrahedron 1990, 46,
783–792.
[8] K. N. Houk, “Theory of 1,3-Dipolar Cycloadditions” in 1,3-
Dipolar Cycloaddition Chemistry (Ed.: A. Padwa), Wiley, New
York, 1984, vol. 2.
[9] a) H. G. Viehe, Z. Janousek, R. Merényi, L. Stella, Acc. Chem.
Res. 1985, 18, 148–154; b) J. Baran, H. Mayr, J. Org. Chem.
1989, 54, 5774–5783; c) R. K. Howe, B. R. Shelton, J. Org.
Chem. 1990, 55, 4603–4607; d) R. A. Firestone, Tetrahedron
Acknowledgments
We thank Fernando Labarrios for his help in spectrometric mea-
surements. J. T. would like to acknowledge DEPI/IPN (Grants
Eur. J. Org. Chem. 2007, 2352–2364
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2363

H. A. Jiménez-Vázquez, J. Tamariz et al.
FULL PAPER
1977, 33, 3009–3039; e) R. Arnaud, P. Juvin, Y. Vallée, J. Org.
Chem. 1999, 64, 8880–8886.
[10] F. P. Cossío, I. Morao, H. Jiao, P. v. R. Schleyer, J. Am. Chem.
paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.
ac.uk/data_request/cif.
Soc. 1999, 121, 6737–6746. A semiquantitative method was de- [22] P. DeShong, C. M. Dicken, R. R. Staib, A. J. Freyer, S. M.
veloped for predicting the outcome of reactions of nitrones
with mono- and disubstituted alkenes: M. A. Silva, J. M.
Goodman, Tetrahedron 2002, 58, 3667–3671.
Weinreb, J. Org. Chem. 1982, 47, 4397–4403.
[23] P. A. Wade, H. R. Hinney, Tetrahedron Lett. 1979, 20, 139–142.
[24] a) R. E. Sammelson, C. D. Gurusinghe, J. M. Kurth, M. M.
Olmstead, M. J. Kurth, J. Org. Chem. 2002, 67, 876–882; b)
B. M. Kelly-Basetti, M. F. Mackay, S. M. Pereira, G. P. Savage,
G. W. Simpson, Heterocycles 1994, 37, 529–539; c) S. M. Per-
eira, G. P. Savage, G. W. Simpson, R. J. Greenwood, M. F.
Mackay, Aust. J. Chem. 1993, 46, 1401–1412; d) S. Yang, W.
Hayden, H. Griengl, Monatsh. Chem. 1994, 125, 469–477; e)
R. Annunziata, M. Benaglia, M. Cinquini, F. Cozzi, L. Rai-
mondi, J. Org. Chem. 1995, 60, 4697–4706; f) Y. Iwakura, K.
Uno, S. Shiraishi, T. Hongu, Bull. Chem. Soc. Jpn. 1968, 41,
2954; g) P. Micúch, L. Fisera, M. K. Cyranski, T. M. Krygow-
ski, J. Krajcík, Tetrahedron 2000, 56, 5465–5472.
[11] a) F. Méndez, J. Tamariz, P. Geerlings, J. Phys. Chem. A 1998,
102, 6292, and references cited therein. A review about HSAB
was published: b) S. Woodward, Tetrahedron 2002, 58, 1017–
1050.
[12] a) J. Peralta, J. P. Bullock, R. W. Bates, S. Bott, G. Zepeda, J.
Tamariz, Tetrahedron 1995, 51, 3979–3996; b) J. A. Mendoza,
H. A. Jiménez-Vázquez, R. Herrera, J. Liu, J. Tamariz, Rev.
Soc. Quím. Méx. 2003, 47, 108–116.
[13] a) J. Tamariz, P. Vogel, Helv. Chim. Acta 1981, 64, 188–197; b)
A. Orduña, L. G. Zepeda, J. Tamariz, Synthesis 1993, 375–377.
[14] L. Villar, J. P. Bullock, M. M. Khan, A. Nagarajan, R. W.
Bates, S. G. Bott, G. Zepeda, F. Delgado, J. Tamariz, J. Or-
ganomet. Chem. 1996, 517, 9–17.
[15] a) A. Padwa (Ed.), 1,3-Dipolar Cycloaddition Chemistry, Wiley,
New York, 1984; b) R. Sustmann, W. Sicking, R. Huisgen, J.
Am. Chem. Soc. 1995, 117, 9679–9685; c) J. J. W. McDouall,
M. A. Robb, U. Niazi, F. Bernardi, H. B. Schlegel, J. Am.
Chem. Soc. 1987, 109, 4642–4648.
[16] R. Sustmann, Tetrahedron Lett. 1971, 12, 2717–2720.
[17] a) F. Viton, G. Bernardinelli, E. P. Kündig, J. Am. Chem. Soc.
2002, 124, 4968–4969; b) S. Kanemasa, N. Ueno, M. Shirahase,
Tetrahedron Lett. 2002, 43, 657–660; c) D. M. David, M. Baka-
voli, S. G. Pyne, B. W. Skelton, A. H. White, Tetrahedron 1995,
51, 12393–12402; d) S. R. Gilbertson, D. P. Dawson, O. D. Lo-
pez, K. L. Marshall, J. Am. Chem. Soc. 1995, 117, 4431–4432;
e) K. Hori, H. Kodama, T. Ohta, I. Furukawa, J. Org. Chem.
1999, 64, 5017–5023.
[18] a) D. Döpp, M. Henseleit, Chem. Ber. 1982, 115, 798–800; b)
S. G. Pyne, J. Safaei-G., B. W. Skelton, A. H. White, Aust. J.
Chem. 1995, 48, 1511–1533; c) A. Vasella, Helv. Chim. Acta
1977, 60, 1273–1295; d) A. Benavides, R. Martínez, H. A. Jimé-
nez-Vázquez, F. Delgado, J. Tamariz, Heterocycles 2001, 55,
469–485; e) U. Chiacchio, A. Corsaro, V. Pistarà, A. Rescifina,
D. Iannazzo, A. Piperno, G. Romeo, R. Romeo, G. Grassi,
Eur. J. Org. Chem. 2002, 1206–1212; f) D. Keirs, D. Moffat, K.
Overton, R. Tomanek, J. Chem. Soc., Perkin Trans. 1 1991,
1041–1051.
[25] For recent computational studies about the regio- and stereose-
lectivity in 1,3-dipolar cycloaddition of substituted nitrones
and dipolarophiles, see: a) P. Merino, J. Revuelta, T. Tejero, U.
Chiacchio, A. Rescifina, G. Romeo, Tetrahedron 2003, 59,
3581–3592; b) K. Marakchi, O. Kabbaj, N. Komiha, R. Jalal,
M. Esseffar, J. Mol. Struct. (Theochem) 2003, 620, 271–281; c)
A. Milet, Y. Gimbert, A. E. Greene, J. Comput. Chem. 2006,
27, 157–162.
[26] The B3LYP/6-31G(d) level of theory was considered adequate
for modeling the TSs of the 1,3-dipolar cycloaddition of nitro-
nes: C. Di Valentin, M. Freccero, R. Gandolfi, A. Rastelli J.
Org. Chem. 2000, 65, 6112–6120; and other dipoles: M.-D. Su,
H.-Y. Liao, W.-S. Chung, S.-Y. Chu, J. Org. Chem. 1999, 64,
6710–6716.
[27] M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G.
Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Pe-
tersson, J. A. Montgomery, K. Raghavachari, M. A. Al-La-
ham, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslow-
ski, B. B. Stefanov, A. Nanayakkara, M. Challacombe, 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. Gon-
zalez, J. A. Pople, Gaussian 94, Gaussian, Inc., Pittsburgh, PA,
1995, revision E.2.
[28] H. A. Jiménez-Vázquez, M. E. Ochoa, G. Zepeda, A. Modelli,
D. Jones, J. A. Mendoza, J. Tamariz, J. Phys. Chem. A 1997,
101, 10082–10089.
[19] a) D. S. C. Black, R. F. Crozier, V. C. Davis, Synthesis 1975,
205–221; b) C. Belzecki, I. Panfil, J. Org. Chem. 1979, 44, 1212– [29] L. R. Domingo, Eur. J. Org. Chem. 2000, 2265–2272.
1214; c) C. M. Tice, B. Ganem, J. Org. Chem. 1983, 48, 5048–
5050; d) A. Padwa, L. Fisera, K. F. Koehler, A. Rodriguez,
G. S. K. Wong, J. Org. Chem. 1984, 49, 276–281; e) P. De-
Shong, C. M. Dicken, J. M. Leginus, R. R. Whittle, J. Am.
Chem. Soc. 1984, 106, 5598–5602; f) C. Camiletti, D. D. Dhav-
ale, L. Gentilucci, C. Trombini, J. Chem. Soc., Perkin Trans. 1
1993, 3157–3171; g) J. Fröhlich, L. Fisera, F. Sauter, Y. Feng,
P. Ertl, Monatsh. Chem. 1995, 126, 75–83; h) L. Fisera, F. Sau-
ter, J. Fröhlich, Y. Feng, P. Ertl, K. Mereiter, Monatsh. Chem.
1994, 125, 553–563; i) M. Carda, R. Portolés, J. Murga, S. Ur-
iel, J. A. Marco, L. R. Domingo, R. J. Zaragozá, H. Röper, J.
Org. Chem. 2000, 65, 7000–7009.
[30] a) J. J. W. McDouall, M. A. Robb, U. Niazi, F. Bernardi, H. B.
Schlegel, J. Am. Chem. Soc. 1987, 109, 4642–4648; b) K. N.
Houk, J. González, Y. Li, Acc. Chem. Res. 1995, 28, 81–90.
[31] K. Tanaka, T. Imase, S. Iwata, Bull. Chem. Soc. Jpn. 1996, 69,
2243–2248.
[32] a) J. Liu, S. Niwayama, Y. You, K. N. Houk, J. Org. Chem.
1998, 63, 1064–1073; b) M. J. Aurell, L. R. Domingo, P. Pérez,
R. Contreras, Tetrahedron 2004, 60, 11503–11509.
[33] J. Mendoza, E. Pérez, H. A. Jiménez-Vázquez, J. Tamariz, J.
Mex. Chem. Soc. 2006, 50, 47–56.
[34] A. Rastelli, M. Bagatti, R. Gandolfi, M. Burdisso, J. Chem.
Soc., Faraday Trans. 1994, 90, 1077–1082.
[35] SHELXTL, v. 5.03, Siemens Energy & Automation, Germany,
1995.
[20] D. Iannazzo, A. Piperno, V. Pistarà, A. Rescifina, R. Romeo,
Tetrahedron 2002, 58, 581–587.
[21] CCDC-290852 (for 12d), -290851 (for 12f), and -290853 (for
17a) contain the supplementary crystallographic data for this
Received: November 15, 2006
Published Online: March 16, 2007
2364
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 2352–2364