Synthesis of new captodative alkenes: Alkyl 2-aroyloxy acrylates - Structure, and reactivity in Diels - Alder cycloadditions

Article abstract of DOI:10.1002/1099-0690(200112)2001:24<4657::AID-EJOC4657>3.0.CO;2-3

Novel captodative alkenes, namely the methyl and ethyl esters of 2-aroyloxyacrylic acids, 2 and 3, have been prepared. Their reactivity and selectivity have been evaluated in Diels - Alder cycloadditions with unsymmetrical dienes, which generate the corresponding adducts with high regioselectivity. No significant stereoselectivity was observed in the reaction with cyclopentadiene (9), although Lewis acid catalysis improved the exo/endo isomeric ratio. Structural and electron spectroscopic studies of these alkenes are supported by MO calculations. FMO theory accounts for the regioselectivity observed with isoprene (7), and the reactivity seen in Diels - Alder additions correlates with the stabilization of the relevant vacant π^· MO in these alkenes, which is mainly due to the electron-withdrawing group.

Full text of DOI:10.1002/1099-0690(200112)2001:24<4657::AID-EJOC4657>3.0.CO;2-3

FULL PAPER
Synthesis of New Captodative Alkenes: Alkyl 2-Aroyloxy Acrylates ؊
Structure, and Reactivity in Diels؊Alder Cycloadditions
[a]
[a]
[b]
[c]
´
´
Rafael Herrera, Hugo A. Jimenez-Vazquez, Alberto Modelli, Derek Jones,
[d]
[a]
´
Björn C. Söderberg, and Joaquın Tamariz*
Keywords: Captodative alkenes / Cycloadditions / Photoelectron spectroscopy
Novel captodative alkenes, namely the methyl and ethyl es-
ters of 2-aroyloxyacrylic acids, 2 and 3, have been prepared.
Their reactivity and selectivity have been evaluated in
Diels−Alder cycloadditions with unsymmetrical dienes,
catalysis improved the exo/endo isomeric ratio. Structural
and electron spectroscopic studies of these alkenes are sup-
ported by MO calculations. FMO theory accounts for the re-
gioselectivity observed with isoprene (7), and the reactivity
which generate the corresponding adducts with high re- seen in Diels−Alder additions correlates with the stabilization
gioselectivity. No significant stereoselectivity was observed
of the relevant vacant π* MO in these alkenes, which is
in the reaction with cyclopentadiene (9), although Lewis acid
mainly due to the electron-withdrawing group.
Introduction
As an extension of our ongoing interest in understanding
the structural and electronic factors responsible for the high
reactivity and selectivity of captodative alkenes in cycload-
dition reactions, we present herein a new class of captod-
ative alkenes, namely the methyl and ethyl esters of 2-aroy-
loxyacrylic acids, 2 and 3,[10] as well as a study of their be-
havior in DielsϪAlder additions to unsymmetrical and car-
bocyclic dienes.
The DielsϪAlder cycloaddition reaction has been tradi-
tionally considered as a powerful method in the evaluation
of the structural and electronic factors that control chem-
ical reactivity and selectivity in pericyclic processes.[1] Al-
der’s rule has demonstrated its effectiveness in predicting
reactivity in DielsϪAlder additions.[2] It establishes that the
rate of cycloaddition increases when the diene bears elec-
tron-releasing groups and the dienophile bears electron-
withdrawing groups, whereas it is retarded with dienophiles
bearing electron-donating groups.[1a][2d] This behavior has
been rationalized in terms of frontier molecular orbital
(FMO) theory.[3]
Owing to the opposite electronic demands displayed by
their geminally substituted functional groups, captodative
alkenes have attracted particular interest in pericyclic cyclo-
additions.[4] The captodative alkenes 3-aroyloxy-3-buten-2-
ones 1 have been shown to be highly reactive,[5] stereoselec-
tive,[6] and regioselective[7] in DielsϪAlder and 1,3-dipolar
cycloadditions.[8] Furthermore, they have also proved to be
versatile synthons in natural product synthesis.[9]
In order to rationalize such behavior, a structural and
theoretical study of these molecules has been undertaken,
including gas-phase measurements by ultraviolet photoelec-
tron spectroscopy (UPS) and electron transmission spectro-
scopy (ETS). The latter two techniques measure, respect-
ively, the ionization energies (IEs) from the occupied MOЈs,
and the energies (vertical attachment energies, VAEs) of
electron attachment to the vacant MOЈs, in other words the
negative of the vertical electron affinities (VEAs). The two
complementary sets of data can be used to build a picture
of the frontier orbital structure. The major limitation is that
the formation of anion states stable with respect to the neut-
ral molecule cannot be observed by ETS, i.e. positive VEAs
cannot be measured. HartreeϪFock (HF) calculations us-
ing the 6Ϫ31G* basis set have been carried out to evaluate
the localization properties of the occupied and vacant MOЈs
involved in reactivity and the relative energies of the vacant
MOЈs associated with stable anion states that are not ac-
cessible experimentally.
[a]
Department of Organic Chemistry, Escuela Nacional de
´
´
Ciencias Biologicas, Instituto Politecnico Nacional,
´
Prol. Carpio y Plan de Ayala, 11340 Mexico, D.F., Mexico
Fax: (internat.) ϩ 52-5/396-3503
E-mail: jtamariz@woodward.encb.ipn.mx
Department of Chemistry ‘‘G. Ciamician’’, University of
Bologna,
[b]
[c]
[d]
via Selmi 2, 40126 Bologna, Italy
Istituto del Composti del Carbonio Contenenti Eteroatomi,
CNR,
via Gobetti 101, 40129 Bologna, Italy
Department of Chemistry, West Virginia University,
P. O. Box 6045, Morgantown, West Virginia 26506Ϫ6045, USA
Eur. J. Org. Chem. 2001, 4657Ϫ4669
WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ193X/01/1224Ϫ4657 $ 17.50ϩ.50/0
4657

J. Tamariz et al.
FULL PAPER
The thermal addition of dienophiles 2d and 2f to an ex-
cess of 7 gave mixtures of adducts 11 and 12 (Scheme 1).
Table 2 shows that these alkenes were less reactive than al-
kene 1a and methyl vinyl ketone (10) under comparable
conditions, since their reactions required higher temper-
atures and longer times. Probably as a consequence of this
lower reactivity, the regioselectivity was also lower, al-
though the para isomer was still the major adduct. This
behavior would seem to be readily rationalized, considering
that the methoxycarbonyl group of the captodative alkenes
2 has a lesser electron-withdrawing effect than the acetyl
group in alkene 1a. This would support the hypothesis that
the reactivity and regioselectivity of captodative alkenes of
this kind is mainly controlled by the electronic effect of the
electron-withdrawing group.[11] As expected,[4c,12] the re-
gioselectivity was found to be substantially improved by
adding aluminum chloride as a Lewis acid catalyst (Table 2,
entries 5 and 6).
Results
Preparation of the Captodative Alkenes, Alkyl 2-Aroyloxy
Acrylates 2 and 3
The preparation of the captodative alkenes 2aϪ2e and
3aϪ3d involved the use of methyl and ethyl esters of pyr-
uvic acid 4 and 5, respectively, as starting materials [Equa-
tion (1)]. O-Acylation of the base-promoted enolates of 4
and 5, under similar reaction conditions as those used for
the preparation of compounds 1,[5][9b] furnished the corres-
ponding alkenes 2 and 3 (Table 1). In contrast to the pro-
cedure for the preparation of 1, the pyruvic ester was added
to the mixture of acyl chloride 6 and triethylamine at a
lower temperature (Ϫ10 °C) to improve the yields. The
stabilities of the new alkenes are comparable to that of the
previously studied 1, although most of them were obtained
as colorless oils. Alkenes 2 and 3 have been fully character-
ized by spectroscopy and elemental analysis.
Table 1. Preparation of alkenes 2 and 3
Entry[a] Pyruvate6 (Ar)
Alkenem.p. (°C)Yield (%)[b]
1
2
3
4
5
6
7
8
9
4
4
4
4
4
4
5
5
5
5
6a (Ph)
2a
2b
2c
2d
oil
oil
65
55
6b (m-Me-C6H4)
6c (p-Cl-C6H4)
6d (p-NO2-C6H4)
42Ϫ43 71
80Ϫ81 60
91Ϫ92 60
oil
oil
oil
oil
oil
6e [3,5-(NO2)2-C6H3]2e
6f (p-OMe-C6H4)
6a (Ph)
6b (m-Me-C6H4)
6c (p-Cl-C6H4)
6d (p-NO2-C6H4)
2f
51
40
67
40
50
3a
3b
3c
3d
Scheme 1
10
[a]
Cycloaddition of diene 8 to alkenes 2d and 2f afforded
exclusively the ortho regioisomers (Scheme 1). Interestingly,
the process was also highly stereoselective since of two pos-
sible endo/exo isomers, 17/18, the latter was not detected in
the crude mixtures. These results are in agreement with
those obtained for the addition of alkene 1a,[7] in terms of
both regio- and stereoselectivities (Table 2, entries 8Ϫ10).
The structures of the major regioisomers 11 and 17 were
established by NMR spectroscopy, and by comparison with
the unambiguously established structures 13 and 19.[7]
Alkenes 2d and 2f were found to react with the cyclic
diene 9 under thermal conditions to give a mixture of the
exo/endo stereoisomers, 21/22, the former being the major
adduct (Scheme 2). The stereoisomeric mixtures were pro-
duced with smaller ratios as compared to those obtained
THF/HMPA (20:1) as solvent, containing Et3N; stirring the re-
action mixture at room temperature for 24 h. [b] After column chro-
matography and recrystallization.
(1)
Diels؊Alder Cycloadditions of Alkenes 2 to Unsymmetrical
and Carbocyclic 1,3-Dienes
The reactivities and selectivities of these alkenes in from 1a (Table 2, entries 11, 15, and 18). The presence of
DielsϪAlder reactions were assessed under thermal and AlCl3 as a catalyst[13] did not greatly increase the stereo-
catalytic conditions. In order to evaluate representative al- selectivity, but the excess of the exo isomer was maintained,
kenes, derivatives 2d and 2f were chosen, because of the in contrast to 1a, for which the ratio was inverted (entries
opposite electron demands of their aryl substituents. Re- 12, 16, and 20). Although high stereoselectivity was ob-
gioselectivity was studied with two dienes, isoprene (7) and tained with 1a using TiCl4 as the Lewis acid,[6] the reaction
1-acetoxybutadiene (8), which contain distinct substituents of alkene 2d with 9 in the presence of the same catalyst did
at different positions within the conjugated system. Diene not show any significant selectivity (entry 13). In contrast,
8 was also useful for probing the stereoselectivity of the under the same conditions, reaction of alkene 2f led to a
process, as was cyclopentadiene (9), a cyclic diene custom- marked improvement in the exo/endo ratio (entry 17). It is
arily employed for testing reactivity in these cycloadditions. noteworthy that captodative alkenes usually show a prefer-
4658
Eur. J. Org. Chem. 2001, 4657Ϫ4669

Synthesis of New Captodative Alkenes
FULL PAPER
Table 2. DielsϪAlder cycloadditions of alkenes 1a, 2d, 2f, and 10 with dienes 7, 8, 9, and 25
Entry[a]
Alkene
Diene (mol equiv.)
Solvent
Catalyst (mol equiv.)
T (°C)
t (h)
Products (ratio)[b]
Yield (%)[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
2d
2f
1a
10
2d
2f
1a
2d
2f
1a
2d
2d
2d
2d
2f
7 (15)
7 (35)
7 (7)
xylene
xylene
xylene
toluene
CH2Cl2
CH2Cl2
CH2Cl2
xylene
xylene
xylene
Ϫ
Ϫ
Ϫ
Ϫ
170
170
130
120
25
25
25
150
150
130
150
25
Ϫ50
55
150
25
Ϫ50
60
96
96
35
15
48
48
36
48
48
11
12
24
8
11a/12a (63:37)
11b/12b (67:33)
13/14 (75:25)
15/16 (71:29)
11a/12a (96:4)
11b/12b (88:12)
13/14 (94:6)
17a/18a (Ͼ98:Ͻ2)
17b/18b (Ͼ98:Ͻ2)
19/20 (Ͼ95:Ͻ5)
21a/22a (59:41)
21a/22a (75:25)
21a/22a (55:45)
21a/22a (61:39)
21b/22b (60:40)
21b/22b (57:43)
21b/22b (84:16)
23/24 (82:18)
23/24 (71:29)
23/24 (36:64)
26a
58
65
77[d]
[e]
7 (1.5)
7 (15)
7 (35)
7 (4)
AlCl3 (10)
AlCl3 (10)
ZnCl2 (5)
Ϫ
Ϫ
Ϫ
Ϫ
AlCl3 (10)
TiCl4 (5)
Ϫ
Ϫ
AlCl3 (10)
TiCl4 (5)
Ϫ
Ϫ
83
63
98[d]
52
8 (1)
8 (2)
64
8 (3)
79[d]
79
9 (19)
9 (19)
9 (19)
9 (19)
9 (24)
9 (24)
9 (24)
9 (5)
xylene
CH2Cl2
CH2Cl2
CH2Cl2
xylene
CH2Cl2
CH2Cl2
xylene
CH2Cl2
CH2Cl2
xylene
70
76
7
38[f]
76
12
48
8
8.5
7
0.3
48
48
12
2f
2f
73
70
1a
1a
1a
2d
2f
70[g]
70[g]
95[g]
92
9 (5)
9 (5)
25 (23)
25 (29)
25 (5)
60
25
150
150
110
AlCl3 (5)
Ϫ
Ϫ
Ϫ
xylene
toluene
26b
27
93
1a
91[h]
[a] All under N2 atmosphere. Thermal trials in the presence of 1Ϫ2% hydroquinone. [b] Proportions as determined by 1H NMR of the crude
reaction mixtures. [c] As a mixture of isomers after purification by column chromatography. [d] See ref.[7] [e] See ref.[14] [f] Corresponding to
[g]
the major isomer after column chromatography and recrystallization. See ref.[6] [h] See ref.[9f]
ence for exo stereoselectivity, which is increased by the pres- to the alkoxycarbonyl group, 3-Hx, is shifted downfield
ence of Lewis acid catalysts.[4]
with respect to that of the anti proton 3-Hn. This deshield-
ing effect of the carbonyl group on the syn 3-H proton is
characteristic of the skeleton of norbornene derivatives.[6,17]
This assignment was confirmed by X-ray crystallographic
analysis of the endo adduct 22a (Figure 1).[18]
Scheme 2
Unexpectedly, and in contrast to the results with 1a, for
which the reactivity and exo/endo selectivity were scarcely
modified by changing the polarity of the solvent (xylene,
CH2Cl2) (Table 2, entries 18 and 19),[6] the reactivity of 2d
toward diene 9 was found to be significantly enhanced in
the presence of dichloromethane, since the reaction took
place at 55 °C and in a shorter time. This can probably
be ascribed to an electrostatic effect.[15] An increase in the
Figure 1. ORTEP view of the X-ray diffraction structure of 22a;
thermal ellipsoids are drawn at the 30% probability level
In order to further evaluate the reactivity of alkenes 2
polarity of the solvent may enhance the solvophobic inter- with respect to 1a, cycloadditions of 2d and 2f to butadiene
actions with the reactants, facilitating their approach to the (25) were also carried out. The reactions were performed in
transition state,[16] without losing the concertedness of the a nonpolar solvent at high temperature (150 °C) for 72 h
mechanism. However, no significant solvent effect was (Table 2, entries 21 and 22), and led to the corresponding
found on the exo/endo ratios (Table 2, entries 11 and 14).
adducts 26a and 26b in good yields. Again, dienophiles 2
The exo configuration of adducts 21a and 21b was as- proved to be less reactive than 1a (entry 23). No significant
1
signed by means of H NMR spectroscopy. In the spectra difference in reactivity was found between alkenes 2d and
of both the compounds, the signal of the vicinal proton syn 2f. This analogous reactivity, along with the regio- and ste-
Eur. J. Org. Chem. 2001, 4657Ϫ4669
4659

J. Tamariz et al.
FULL PAPER
reoselectivities toward dienes 7؊9, suggests that the contri- pair orbital, associated with the shoulder, is thus stabilized
bution of the substituent on the benzene ring of the elec- by 0.6 eV with respect to 10 (9.60 eV). On going from 28 to
tron-donating group to the polarizability of the double 2b and 3d, all the corresponding MOЈs in the latter two
bond of the captodative alkene must be almost negligible.
compounds are inductively stabilized by the nitrobenzoate
group (Figure 3), so that the πCC MO, having mainly ethene
character, contributes to the second band of the UP spec-
trum, which shows a maximum at 11.2 eV.
Discussion
It is difficult to account for the slight exo stereoselectivity
displayed by alkenes 2 with cyclopentadiene (9) and for the
lack of significant changes on modification of the thermal
and catalytic reaction conditions, due to the great number
of possible stabilizing and destabilizing interactions at the
transition state between the diene and dienophile.[19] How-
ever, the regioselectivity shown by dienes 7 and 8 seems to
reflect electronic rather than steric control. This is much
more evident with diene 8, from which the most crowded
adduct is obtained exclusively.
FMO theory has proved useful in the study of reactivity
and regioselectivity of captodative alkenes 1.[7,11] It predicts
the correct orientation in the addition to dienes such as 7
and 8, showing that the interaction is controlled by the en-
ergetically more favorable HOMO-diene/LUMO-dienophile
gap. However, steric hindrance around the captodative
center may be invoked to account for the more than twofold
higher reactivity of alkene 10 with respect to 1a,[5] since the
LUMO of the latter is more stable than that of 10.[11] The
results of FMO calculations[20] were compared with the IEs
and VA E s measured for these molecules and were found to
reproduce the experimental energy trends.[11]
In order to correlate the structural and electronic factors
with the high reactivity and selectivity shown by captod-
ative alkenes in cycloaddition reactions, in particular the
novel alkenes 2 and 3, we undertook an extensive theoret-
ical and spectroscopic study of these molecules. Figure 2
displays the UP spectra of compounds 2d and 3d, along
Figure 2. UP spectra of 2d, 3d, and 28
The ET spectra of compounds 2d and 3d, along with that
with that of ethyl acrylate (28) for comparison purposes. A of the smaller reference molecule ethyl acrylate (28), are
pronounced difference is seen compared to the UP spec- presented in Figure 4. The spectra of 2d and 3d show one
trum of 1a,[11] due to the additional oxygen lone pair [nπ(en)
]
additional resonance, at about 3.4 eV, as compared to the
cross-conjugated with the enone (en) π system. The relative spectrum of 1a[11] (see Figure 3). Moreover, the width of the
intensity of the second band is appreciably higher and the first resonance is reduced from 1.0 to 0.65 eV. These find-
valley between the first and second bands is occupied, in ings fully confirm the previous assignment[11] of the first
agreement with the appearance of an additional signal from signal in the ET spectrum of 1a to unresolved contributions
the nπ(en) MO in this region. Moreover, replacement of the from the π*CO MO of the OCOAr group and the highest-
methyl group (bonded to the carbonyl carbon atom, as in lying [π*2(en)] of the two vacant π* MOЈs of the enone
alkene 1a) by an alkoxy group, as in alkenes 2 and 3, causes group. The latter is highly destabilized by mixing with the
a sizeable stabilization of the carbonyl oxygen lone pair methoxy or ethoxy oxygen lone pair, in line with literature
[no(en)], such that the first two occupied MOЈs of 2d and 3d ETS data,[22] indicating considerable localization at the car-
are likely to be the benzene πS and πA MOЈs (Figure 3). bonyl carbon atom. Most importantly from the reactivity
The stabilization of the carbonyl oxygen lone pair can be point of view, a similar destabilization should be experi-
evaluated by comparing the UP spectra of 10[11] and 28 enced by the lowest-lying vacant π* MO [π*1(en)] localized
(Figure 2). In the spectrum of the latter, the broad and in- at the enone group, i.e. the NLUMO (Figure 3). In this re-
tense first band seems to display a shoulder at about spect, it should be noted that the LUMO, with mainly NO2
10.2 eV and maxima at about 10.6 and 10.9 eV. In agree- character, is not involved in the reactivity of the ethene
ment, the (first) IE value determined by mass spectrometry double bond. The two enone vacant π* MOЈs arise from a
is 10.3 eV.[21] We thus assign the first band in the UP spec- strong interaction between the (energetically close) π*CO(en)
trum of 28 to the unresolved contributions from the no, and π*CC fragment orbitals. As a consequence, they possess
πCC, and nπ MOЈs, in order of increasing IE. The no σ lone similar localization properties, with nearly equal carbonyl
4660
Eur. J. Org. Chem. 2001, 4657Ϫ4669

Synthesis of New Captodative Alkenes
FULL PAPER
Figure 3. Correlation diagram for the FMO energies in 1a, 2d, 3d, 10, and 28, as deduced from the UP and ET spectra; the energy
positions of the vacant levels close to, or below, zero energy, as represented by dashed lines, are only tentative (see text); the UP and ET
data of 1a and 10 are taken from ref.[11]; no is the σ MO with mainly (carbonyl) oxygen lone pair character; nπ is the lone pair of the
benzoyl oxygen atom bonded to the enone moiety; πS and πA are the symmetric and antisymmetric components of the benzene e1g
HOMO; πS* and πA* are the corresponding components of the benzene e1u LUMO; πCO and πCC are the occupied π orbitals mainly
localized at the carbonyl and the ethene groups, respectively; π*CO is the vacant MO with mainly carbonyl (OCOAr) character; π*1(en)
and π*2(en) are the two vacant π* MOЈs of the enone group; σNO2 and πNO2 are the outermost occupied orbitals of the nitro group, and
π*NO2 is the vacant π* MO with mainly nitro group character; the MOЈs with mainly enone character are labeled ‘‘en’’
and alkene character, as confirmed by the wavefunction co-
The geometries of alkenes 2aϪ2f, 3a, 3d, and 28 were
efficients calculated for π*1(en) (see Table 3). In order to fully optimized at the ab initio 3Ϫ21G level, and then em-
confirm and to better evaluate the extent of the destabiliz- ployed as the starting points for ab initio 6Ϫ31G* optimiza-
ing effect caused by mixing with an adjacent oxygen lone tion (Table 4).[23] The nonplanar conformation of the aroyl-
pair on the enone vacant π* MOЈs, we obtained the ET oxy group (with respect to the plane formed by the acrylate
spectrum of 28 (Figure 4). Comparison with the ET spec- moiety) was found to be more stable than the planar form
trum of 10 (Figure 3) shows that the π*2(en) VAE increases for all the derivatives. In contrast, the s-cis/s-trans con-
by 0.85 eV, while the π*1(en) anion state, stable in 10, in- formers have almost the same energies, varying only over
creases to 0.34 eV above zero energy. According to the cal- very small energy ranges (0.15Ϫ0.50 kcal molϪ1); for ex-
culations (Table 3), the energy of the π*1(en) MO increases ample, in 2a, 2b, 2f, 3a, and 28, the s-cis conformer is more
by 0.4 eV on going from 1a to 2d and 3d. These results stable, while in 2d and 3d the s-trans conformer is more
are in qualitative agreement with experiment, but probably stable.
underestimate the destabilizing effect. On going from 28 to
Since these energy differences are too insignificant to af-
2d and 3d, the inductive electron-withdrawing effect of the firm that one of the two conformations is favored in the gas
nitrobenzoate group stabilizes the π*2(en) MO by 0.4 eV phase, and as we were unable to determine the equilibrium
(Figure 3). Exactly the same energy shift is predicted by the in solution by NMR, we carried out a crystallographic ana-
calculations (Table 3) for the π*1(en) MOЈs. Therefore, if the lysis of 2d in order to establish the conformation of the ester
VAE for 28 is 0.34 eV, the π*1(en) MOЈs for 2d and 3d should conjugated system in the solid state. The X-ray structure is
lie slightly below zero energy.
depicted in Figure 5.[18] The conformations of both conjug-
Eur. J. Org. Chem. 2001, 4657Ϫ4669
4661

J. Tamariz et al.
FULL PAPER
(πCC) and the lowest unoccupied π* MO [π*1(en)]. The
atomic wavefunction coefficients were found to be largely
localized at the ethene group and mostly involved in the
DielsϪAlder reaction. The same parameters for dienes 7
and 8 are also included.[12]
Table 3. Ab initio 6Ϫ31G* calculations of energies (eV) of the fron-
tier molecular orbitals for alkenes 2aϪ2f, 3a, 3d, 1a, and 28, and
dienes 7 and 8
Compd.[a]
HOMO[b]
LUMO[c]
HF calculations neglect correlation and relaxation effects,
which tend to cancel out when IEs, but not EAs, are evalu-
ated. For this reason, π IEs are usually reproduced satisfact-
orily by this theoretical approach, whereas the VAEs are
overestimated by several eV.[24] In agreement, the energies
calculated for the occupied πCC MO’s in 1a, 2d, 3d
(Ϫ10.96 eV), and 28 (Ϫ10.6 eV) are close to the negatives
of the energies (11.2 and 10.9 eV, respectively, see Figure 3)
of the second bands in their UP spectra. For the vacant π*
2a
2b
2c
2d
2e
2f
3a
3d
1a
28
7[d]
8[e]
Ϫ10.6044
Ϫ10.5118
Ϫ10.7562
Ϫ10.9921
Ϫ11.2321
Ϫ10.5202
Ϫ10.5725
Ϫ10.9581
Ϫ11.0123
Ϫ10.6103
Ϫ8.6193
3.1770
3.2333
3.0322
2.8080
2.5261
3.2689
3.2140
2.8485
2.4588
3.1938
3.5337
3.2561
MO, the only experimental value available is the VAE
1(en)
Ϫ8.5695
(0.34 eV) in 28, the corresponding anion state in the other
compounds being stable (and thus not observed in the ET
spectra). As expected, the calculated orbital energy is much
higher (3.19 Ϫ 0.34 ϭ 2.85 eV). When the same shift is ap-
plied to the values calculated for 2d and 3d, the energy of
the π*1(en) MO is predicted to be slightly negative (Table 5),
in perfect agreement with the conclusions drawn above
from the VAE trends observed in the ET spectra.
[a]
Non-planar s-trans conformation for alkenes, and s-cis for di-
enes. [b] Energies of the corresponding πCC MO of alkenes 1a, 2, 3,
and 28. [c] Energies of the corresponding π*1(en) MO of alkenes 1a,
2, 3, and 28. See ref.[11] [e] Considered in its s-cis conformation.
[d]
The calculated FMO energies indicate that the
HOMO(πCC)-diene/NLUMO[π*1(en)]-alkene
interaction
[i.e. normal electron demand (NED)] is preferred (Table 6).
In this sense, considering a common diene, the reactivity of
the analogous series 2aϪ2f in DielsϪAlder additions would
be expected to be correlated with the stability of the vacant
π*1(en) MO. Thus, the lower the π*1(en) energy, the more
reactive the dienophile. This occurs when the alkene is sub-
stituted by electron-withdrawing groups on the aroyloxy
group. Alkenes 2d and 2e should be more reactive than 2a,
and even more than 2f, which bears an electron-donating
group on the aromatic ring, since the latter two have the
least energetically stable π*1(en) MO’s (Table 3). Even
though there is a lack of kinetic data on this series, one
could expect a correlation between π*1(en) MO energies and
reactivities, considering the antecedent of a similar reactiv-
ity trend found for the series of alkenes 1.[5,7] Therefore, the
reactivity of captodative alkenes of type 1, or 2 and 3, in
DielsϪAlder cycloaddition reactions, is most likely con-
trolled, to a large extent, by the electron-withdrawing group
(i.e. the acetyl and alkoxycarbonyl groups, respectively),
and by the inductive effect of the aroyloxy group.[25] This is
also supported by the fact that the π*1(en) MO in alkene 28
is higher in energy than those in alkenes 2 and 3 (Figure 3).
In principle, the high reactivity and regioselectivity
Figure 4. ET spectra of 2d, 3d, and 28; vertical lines mark the most
probable vertical AE values
ated moieties, the acrylate and p-nitrobenzoyl groups, show
a good agreement with the calculated gas-phase structure, shown by these alkenes suggests that there is an additional
being almost identical to that of alkene 1a.[11] Thus, the factor that counterbalances or inhibits the expected deactiv-
former is in a conjugated planar s-trans conformation, while ating effect of the electron-releasing group. As suggested
the latter lies out of the plane formed by the acrylate π- previously for alkenes 1a,[11] this inhibiting factor could be
system. Interestingly, the methyl group of the acrylate lies attributed to a conformational effect on the aroyloxy
in the plane of the carbonyl double bond, as predicted by group.[26] This group is oriented in an almost orthogonal
the calculations.
conformation (Figure 5), in which the lone pairs of the oxy-
The FMO energies for the s-trans conformations of the gen are restricted to be delocalized toward the double bond,
same alkenes were calculated at the HF/6-31G* level. favoring delocalization toward the carbonyl group of the
Table 3 lists the energies, of the highest occupied π MO aroyloxy ester.
4662
Eur. J. Org. Chem. 2001, 4657Ϫ4669

Synthesis of New Captodative Alkenes
Table 4. Ab initio (RHF/6-31G*) energies (au) of the minimum-energy conformations of alkenes 2aϪ2f, 3a, 3d, and 28
FULL PAPER
Compound
R
Conformation
E
Relative stability (kcal/mol)[a]
2a
2b
2c
2d
2e
2f
Me
Me
Me
Me
Me
Me
Et
s-cis
s-trans
s-cis
s-trans
s-cis
s-trans
s-cis
s-trans
s-cis
s-trans
s-cis
s-trans
s-cis
s-trans
s-cis
Ϫ721.836933
0.00
0.16
0.00
0.19
0.00
0.03
0.19
0.00
0.50
0.00
0.00
0.21
0.00
0.21
0.17
0.00
0.00
0.55
Ϫ721.836673
Ϫ760.873767
Ϫ760.874068
Ϫ1180.736084
Ϫ1180.736038
Ϫ925.304619
Ϫ925.304917
Ϫ1128.768845
Ϫ1128.769650
Ϫ835.719713
Ϫ835.719380
Ϫ760.876984
Ϫ760.876643
Ϫ964.344686
Ϫ964.344958
Ϫ343.720490
Ϫ343.719615
3a
3d
28
Et
s-trans
s-cis
s-trans
[a]
Considered for each compound in its two conformations.
Table 6. Ab initio HF/6-31G* energy gaps (eV) between the relev-
ant frontier orbitals for dienophiles 2 and 3, and diene 7
Alkene[a][b]
HOMO-π*1(en)
LUMOϪHOMO
Difference
2a
2b
2c
2d
2e
2f
11.7963
11.8526
11.6515
11.4273
11.1454
11.8882
11.8333
11.4678
14.1381
14.0455
14.2899
14.5258
14.7658
14.0539
14.1062
14.4918
2.3418
2.1929
2.6384
3.0985
3.6204
2.1657
2.2729
3.0240
Figure 5. ORTEP view of the X-ray diffraction structure of 2d;
thermal ellipsoids are drawn at the 30% probability level
3a
3d
[a]
HOMO-diene/π*1(en)-dienophile and LUMO-diene/πCC-dien-
Table 5. Calculated ab initio (RHF/6-31G*) energies (eV) and cor-
responding values deduced from the UP and ET spectra of the
highest occupied (πCC) and lowest unoccupied [π*1(en)] π MOЈs with
mainly enone character in alkenes 2d, 3d, and 28; the calculated π*
[b]
ophile. Of the nonplanar s-trans conformation.
value of an enol ester, such as vinyl acetate (IE ϭ
10.76 eV),[20b] is much closer to those of our alkenes.
On the other hand, it is noteworthy that a correlation is
found between the energy of the π*1(en) MO and experi-
mental reactivity when alkenes 2d and 1a are compared,
since the latter reacts faster than 2d with 7 (Table 2, entries
1 and 3), in agreement with a lower vacant π* MO energy
for 1a (Table 3).[11]
energies are shifted by 2.85 eV (see text)
1(en)
πCC
π*1(en)
Alkene
Exp. (ϪIE)
Calcd.
Exp. (VAE)
Calcd.
2d[a]
3d[a]
28[b]
Ϫ11.2
Ϫ11.2
Ϫ10.9
Ϫ10.99
Ϫ10.96
Ϫ10.61
Ϫ0.046
Ϫ0.005
0.34
0.34
[a]
[b]
Considered in its s-trans conformation.
cis conformation.
Considered in its s-
On the basis of the atomic coefficient differences for the
appropriate frontier orbital interaction [HOMO-diene/
πδ1(en)-alkene] between dienophiles 2 and 3 and dienes 7
From the viewpoint of the πCC(en) IE values (Figure 3), and 8 (Table 3), regioselectivity can be tentatively predict-
compounds 1؊3 should resemble an electron-deficient al- ed.[3c] The calculations show that the relative value of the
kene like 10 or 28 rather than an electron-rich alkene like coefficient on the unsubstituted carbon C-1 is greater than
methyl vinyl ether (IE ϭ 8.93 eV),[20b] even though the IE that on the captodative carbon C-2 for the LUMO of both
Eur. J. Org. Chem. 2001, 4657Ϫ4669
4663

J. Tamariz et al.
FULL PAPER
Varian Gemini-300 instrument, with samples in CDCl3 solution
containing TMS as an internal standard. Mass spectra (MS) and
high-resolution mass spectra (HRMS) were obtained, in electron
impact (EI) (70 eV) and fast-atom bombardment (FAB) modes, on
a HewlettϪPackard 5971A or a Jeol JMS-AX 505 HA spectro-
meter. X-ray data were collected on a Siemens P-4 diffractometer.
The UP spectra were obtained with a PerkinϪElmer PS18 photoe-
lectron spectrometer. The ET spectroscopy apparatus was in the
format devised by Sanche and Schulz,[30] as has been described pre-
viously.[31] Microanalyses were performed by M-H-W Laboratories
2 and 3. Therefore, the predicted main interaction is that
between carbon C-1 of alkenes 2 and 3 and the terminal
carbon C-1 of diene 7 (Scheme 1),[11] since the largest
HOMO coefficient of the latter is located on this carbon.
Although the reaction of dienophile 3d was not examined
experimentally, according to this analysis the regioselectiv-
ity can be expected to be comparable to that shown by its
homologue 2d. For diene 8, several possible starting geo-
metries, with s-cis and s-trans conformations, were ana-
´
lyzed. However, only a very small difference in the coeffi- (Phoenix, AZ) and at the Centro de Investigaciones Quımicas, Un-
´
iversidad Autonoma de Hidalgo (Pachuca, Hgo., Mexico). Analyt-
cients of carbons C-1 and C-4 of the HOMO was observed.
Therefore, the reaction would not be expected to be regiose-
lective. This is in contrast to the high ortho regioselectivity
observed experimentally, hence there might be other factors
involved in controlling the outcome of DielsϪAlder reac-
tions with 8.
This regioselectivity could find a better rationalization
than that given by FMO theory by applying the novel DFT/
HSAB theory.[8a,27] This has proved to be a reliable model
ical thin-layer chromatography was performed using E. Merck sil-
ica gel 60 F254 coated plates (0.25 mm), and spots were visualized
using long- and short-wavelength UV lamps. Radial chromato-
graphy was performed on a ‘‘Chromatotron’’ from Harrison Re-
search Instruments. All air- and moisture-sensitive reactions were
carried out under nitrogen using oven-dried glassware. Dioxane,
diethyl ether, THF, toluene, and xylene were freshly distilled from
sodium, and dichloromethane from calcium hydride, prior to use.
Li2CO3 was dried overnight at 120 °C prior to use. All other re-
in correlating the values of the condensed local softness for agents were used without further purification.
the terminal carbons of cycloaddends in DielsϪAlder reac-
General Method for the Preparation of Alkyl 2-Aroyloxyacrylates
2aϪ2f and 3aϪ3d: To solution of triethylamine (6.0 mL,
tions with the regiochemistry observed for a wide range of
a
dienes and dienophiles under NED conditions.[28]
4.3 mmol) in dry THF (20 mL) and HMPA (1.0 mL) at Ϫ10 °C
under N2 atmosphere, a solution of pyruvate 4 or 5 in dry THF
(10 mL) was added dropwise. Then, at the same temperature, a so-
lution of the acid chloride 6 in dry THF (10 mL) was slowly added.
After stirring the mixture at room temperature for 24 h, the solvent
was removed in vacuo, and the residue was redissolved in CH2Cl2
(50 mL). A saturated aqueous solution of NH4Cl (100 mL) was
added and the aqueous layer was extracted with CH2Cl2
(3 ϫ 50 mL). The combined organic extracts were dried (MgSO4)
and the solvent was evaporated in vacuo. The residue was success-
ively purified by flash column chromatography on silica gel (20 g;
hexane/EtOAc, 95:5) and by radial chromatography (hexane/
EtOAc, 90:10), to give the corresponding alkenes 2aϪ2f and
3aϪ3d.
Conclusion
Representatives of a new class of captodative alkenes, 2
and 3, have been prepared from alkyl pyruvates 4 and 5
and the corresponding aroyl chlorides. Highly regio- and
stereoselective cycloadditions are observed when alkenes 2
react with unsymmetrical and carbocyclic dienes. The elec-
tronic structures of these alkenes, as determined from ex-
perimental UPS and ETS data, as well as by FMO calcula-
tions, are consistent with the relative reactivity displayed
by analogous captodative alkenes 1. These results suggest a
greater influence of the electron-withdrawing group than of
its geminally substituted electron-donating group in con-
trolling reactivity and selectivity in DielsϪAlder cycloaddi-
tions.
Methyl 2-(Benzoyloxy)-2-propenoate (2a): Following the general
procedure, reaction of 4 (1.0 g, 9.8 mmol) with 6a (1.4 g, 10 mmol)
afforded 1.31 g (65%) of 2a as a colorless oil.[10] Rf ϭ 0.7 (hexane/
˜
EtOAc, 8:2). IR (film): ν ϭ 1736, 1649, 1452, 1439, 1321, 1317,
1
1267, 1196, 1152, 1088, 1067 cmϪ1. H NMR (300 MHz, CDCl3):
The regioselectivity shown by the new alkenes 2 and di-
ene 7 has been rationalized by considering the wavefunction δ ϭ 3.73 (s, 3 H, CO2CH3), 5.59 (d, J ϭ 1.7 Hz, 1 H, 3a-H), 6.12
(d, J ϭ 1.7 Hz, 1 H, 3b-H), 7.44Ϫ7.52 (m, 2 H, 7-H), 7.57Ϫ7.65
(m, 1 H, 8-H), 8.09Ϫ8.16 (m, 2 H, 6-H). 13C NMR (75.4 MHz,
CDCl3): δ ϭ 51.9 (CO2CH3), 114.0 (C-3), 127.8 (C-6), 128.5 (C-4),
129.8 (C-5), 134.2 (C-7), 144.3 (C-2), 161.6 (PhCO2), 164.0
(CO2CH3). GC-MS (70 eV): m/z (%) ϭ 206 (1) [Mϩ], 175 (3), 105
(100), 77 (38). C11H10O4 (206.2): calcd. C 64.08, H 4.89; found C
64.11, H 4.82.
coefficients of the appropriate (occupied π-diene and π*-
alkene) FMOs of the cycloaddends. An unsatisfactory pre-
diction was obtained by this theoretical approach in the
case of diene 8.
Interestingly, the structural features of these alkenes, such
as the presence of hydroxyl and carboxylic groups, make
them attractive potential synthons for the preparation of
important synthetic targets such as functionalized α-hy-
droxy acids or α-hydroxy-β-amino acids.[29]
Methyl 2-(m-Methylbenzoyloxy)-2-propenoate (2b): Following the
general procedure, reaction of 4 (1.0 g, 9.8 mol) with 6b (1.54 g,
9.97 mmol) afforded 1.19 g (55%) of 2b as a colorless oil. Rf ϭ 0.8
˜
(hexane/EtOAc, 8:2). IR (film): ν ϭ 1736, 1650, 1314, 1275, 1189,
1
1151, 1093, 1072 cmϪ1. H NMR (300 MHz, CDCl3): δ ϭ 2.42 (s,
Experimental Section
3 H, ArMe), 3.81 (s, 3 H, CO2CH3), 5.61 (d, J ϭ 1.8 Hz, 1 H, 3a-
H), 6.16 (d, J ϭ 1.8 Hz, 1 H, 3b-H), 7.33Ϫ7.46 (m, 2 H, ArH),
7.90Ϫ7.97 (m, 2 H, ArH). 13C NMR (75.4 MHz, CDCl3): δ ϭ 21.1
(ArMe), 52.6 (CO2CH3), 114.2 (C-3), 127.4 (C-9), 128.3 (C-4),
128.4 (C-8), 130.6 (C-5), 134.6 (C-7), 138.3 (C-6), 144.7 (C-2), 162.5
General Remarks: Melting points (uncorrected) were determined
on an Electrothermal capillary melting point apparatus. IR spectra
were recorded on a PerkinϪElmer 1600 spectrophotometer. 1H
(300 MHz) and 13C (75.4 MHz) NMR spectra were recorded on a
4664
Eur. J. Org. Chem. 2001, 4657Ϫ4669

Synthesis of New Captodative Alkenes
FULL PAPER
(ArCO2), 164.7 (CO2CH3). C12H12O4 (220.2): calcd. C 65.45, H
5.49; found C 65.55, H 5.49.
(C-4), 130.7 (C-5), 134.5 (C-7), 145.0 (C-2), 161.5 (PhCO2), 164.9
(CO2Et). GC-MS (70 eV): m/z (%) ϭ 220 (1) [Mϩ], 191 (1), 175
(5), 105 (100), 77 (75). C12H12O4 (220.2): calcd. C 65.45, H 5.49;
found C 65.52, H 5.70.
Methyl 2-(p-Chlorobenzoyloxy)-2-propenoate (2c): Following the
general procedure, reaction of 4 (1.0 g, 9.8 mmol) with 6c (1.75 g,
10.0 mmol) afforded 1.67 g (71%) of 2c as colorless crystals. Rf ϭ
Ethyl 2-(m-Methylbenzoyloxy)-2-propenoate (3b): Following the
˜
0.8 (hexane/EtOAc, 8:2). IR (KBr): ν ϭ 1729, 1649, 1590, 1442, general procedure, reaction of 5 (1.50 g, 12.9 mol) with 6b (2.0 g,
1401, 1320, 1273, 1197, 1150, 1092, 1013, 935 cmϪ1 1H NMR 13 mmol) afforded 2.0 g (67%) of 3b as a colorless oil. Rf ϭ 0.6
.
(300 MHz, CDCl3): δ ϭ 3.82 (s, 3 H, CO2CH3), 5.63 (d, J ϭ (hexane/EtOAc, 8:2). IR (film): ν˜ ϭ 1734, 1650, 1274, 1192, 1150,
1
1.9 Hz, 1 H, 3a-H), 6.18 (d, J ϭ 1.9 Hz, 1 H, 3b-H), 7.45Ϫ7.52 (m, 1093, 1071 cmϪ1. H NMR (300 MHz, CDCl3): δ ϭ 1.27 (t, J ϭ
2 H, ArH), 8.04Ϫ8.11 (m, 2 H, ArH). 13C NMR (75.4 MHz,
7.1 Hz, 3 H, CO2CH2CH3), 2.40 (s, 3 H, ArMe), 4.26 (q, J ϭ
CDCl3): δ ϭ 52.8 (CO2CH3), 114.6 (C-3), 127.0 (C-4), 128.8 (C-6), 7.1 Hz, 2 H, CO2CH2CH3), 5.58 (d, J ϭ 1.7 Hz, 1 H, 3a-H), 6.13
131.8 (C-5), 140.4 (C-7), 144.6 (C-2), 161.9 (ArCO2), 163.8 (d, J ϭ 1.7 Hz, 1 H, 3b-H), 7.34Ϫ7.45 (m, 2 H, 8-H, 7-H),
(CO2CH3). GC-MS (70 eV): m/z (%) ϭ 240 (1) [Mϩ], 139 (52), 111
7.89Ϫ7.96 (m, 2 H, 5-H, 9-H). 13C NMR (75.4 MHz, CDCl3): δ ϭ
(44), 75 (100). C11H9ClO4 (240.6): calcd. C 54.90, H 3.77, Cl 14.73; 14.2 (CO2CH2CH3), 21.6 (ArMe), 61.4 (CO2CH2CH3), 114.5 (C-
found C 55.14, H 4.00, Cl 15.00.
3), 127.3 (C-9), 128.5 (C-8), 128.8 (C-4), 130.6 (C-5), 134.5 (C-7),
138.3 (C-6), 144.9 (C-2), 161.8 (ArCO2), 164.6 (CO2Et). GC-MS
(70 eV): m/z (%) ϭ 234 (1) [Mϩ], 136 (75), 135 (100), 107 (20), 92
(30), 77 (50). C13H14O4 (234.2): calcd. C 66.66, H 6.02; found C
66.82, H 5.83.
Methyl 2-(p-Nitrobenzoyloxy)-2-propenoate (2d): Following the gen-
eral procedure, reaction of 4 (1.0 g, 9.8 mmol) with 6d (1.85 g,
9.97 mmol) afforded 1.47 g (60%) of 2d as pale-yellow crystals.
˜
Rf ϭ 0.6 (hexane/EtOAc, 8:2); m.p. 80Ϫ81 °C. IR (KBr): ν ϭ 1744,
1727, 1648, 1607, 1526, 1439, 1346, 1324, 1272, 1197, 1152,
Ethyl 2-(p-Chlorobenzoyloxy)-2-propenoate (3c): Following the gen-
1094 cmϪ1 1H NMR (300 MHz, CDCl3): δ ϭ 3.84 (s, 3 H, eral procedure, reaction of 5 (2.0 g, 17 mmol) with 6c (3.01 g,
.
CO2CH3), 5.70 (d, J ϭ 2.0 Hz, 1 H, 3a-H), 6.23 (d, J ϭ 2.0 Hz, 1 17.2 mmol) afforded 1.75 g (40%) of 3c as a colorless oil. Rf ϭ 0.8
H, 3b-H), 8.28Ϫ8.37 (m, 4 H, ArH). 13C NMR (75.4 MHz, (hexane/EtOAc, 8:2). IR (film): ν ϭ 1736, 1650, 1594, 1401, 1315,
CDCl3): δ ϭ 52.8 (CO2CH3), 115.0 (C-3), 123.6 (C-6), 131.4 (C-5), 1265, 1152, 1089, 1015, 754 cmϪ1 1H NMR (300 MHz, CDCl3):
133.9 (C-4), 144.3 (C-2), 150.9 (C-7), 161.5 (ArCO2), 162.8 δ ϭ 1.28 (t, J ϭ 7.1 Hz, 3 H, CO2CH2CH3), 4.28 (q, J ϭ 7.1 Hz,
(CO2CH3). C11H9NO6 (251.2): calcd. C 52.60, H 3.61, N 5.58; 2 H, CO2CH2CH3), 5.62 (d, J ϭ 1.8 Hz, 1 H, 3a-H), 6.17 (d, J ϭ
˜
.
found C 52.69, H 3.75, N 5.48.
1.8 Hz, 1 H, 3b-H), 7.45Ϫ7.48 (m, 2 H, ArH), 8.05Ϫ8.08 (m, 2 H,
ArH). 13C NMR (75.4 MHz, CDCl3): δ ϭ 14.0 (CO2CH2CH3),
61.8 (CO2CH2CH3), 114.2 (C-3), 127.0 (C-4), 128.9 (C-6), 131.6
(C-5), 140.3 (C-7), 144.8 (C-2), 161.3 (ArCO2), 163.9 (CO2Et). GC-
MS (70 eV): m/z (%) ϭ 254 (1) [Mϩ], 209 (1), 141 (21), 139 (67),
111 (41), 75 (100). C12H11ClO4 (254.7): calcd. C 56.60, H 4.35;
found C 56.54, H 4.46.
Methyl 2-(3,5-Dinitrobenzoyloxy)-2-propenoate (2e): Following the
general procedure, reaction of 4 (1.0 g, 9.8 mol) with 6e (2.19 g,
9.50 mmol) afforded 1.69 g (60%) of 2e as colorless crystals. Rf ϭ
0.5 (hexane/EtOAc, 8:2); m.p. 91Ϫ92 °C. IR (KBr): ν˜ ϭ 1749, 1651,
1631, 1552, 1349, 1318, 1284, 1199, 1160 cmϪ1
.
1H NMR
(300 MHz, CDCl3): δ ϭ 3.86 (s, 3 H, CO2CH3), 5.77 (d, J ϭ
2.2 Hz, 1 H, 3a-H), 6.29 (d, J ϭ 2.2 Hz, 1 H, 3b-H), 9.25Ϫ9.28 (m, Ethyl 2-(p-Nitrobenzoyloxy)-2-propenoate (3d): Following the gen-
2 H, ArH), 9.29Ϫ9.32 (m, 1 H, ArH). 13C NMR (75.4 MHz,
eral procedure, reaction of 5 (2.0 g, 17 mmol) with 6d (3.19 g,
CDCl3): δ ϭ 53.0 (CO2CH3), 115.5 (C-3), 123.1 (C-7), 130.0 (C-5), 17.2 mmol) afforded 2.28 g (50%) of 3d as a colorless oil. Rf ϭ 0.7
132.2 (C-4), 144.0 (C-2), 148.7 (C-6), 160.7 (ArCO2), 161.0 (hexane/EtOAc, 8:2). IR (film): ν˜ ϭ 1735, 1650, 1530, 1348, 1317,
(CO2CH3). C11H8N2O8 (296.2): calcd. C 44.61, H 2.72, N 9.46; 1267, 1151, 1088, 1015 cmϪ1 1H NMR (300 MHz, CDCl3): δ ϭ
.
found C 44.74, H 3.00, N 9.36.
1.24 (t, J ϭ 7.1 Hz, 3 H, CO2CH2CH3), 4.23 (q, J ϭ 7.1 Hz, 2 H,
CO2CH2CH3), 5.67 (d, J ϭ 2.0 Hz, 1 H, 3a-H), 6.15 (d, J ϭ 2.0 Hz,
1 H, 3b-H), 8.32 (m, 4 H, ArH). 13C NMR (75.4 MHz, CDCl3):
δ ϭ 14.0 (CO2CH2CH3), 62.0 (CO2CH2CH3), 114.5 (C-3), 123.6
(C-6), 131.3 (C-5), 134.0 (C-4), 144.6 (C-2), 149.9 (C-7), 161.0
(ArCO2), 162.8 (CO2Et). GC-MS (70 eV): m/z (%) ϭ 265 (1) [Mϩ],
222 (3), 150 (100), 120 (47), 104 (26), 76 (17). C12H11NO6 (265.2):
calcd. C 54.34, H 4.18, N 5.28; found C 54.15, H 4.32, N 5.23.
Methyl 2-(p-Methoxybenzoyloxy)-2-propenoate (2f): Following the
general procedure, reaction of 4 (1.0 g, 9.8 mol) with 6f (1.7 g,
10 mmol) afforded 1.18 g (51%) of 2f as a colorless oil. Rf ϭ 0.26
˜
(hexane/EtOAc, 8:2). IR (film): ν ϭ 1731, 1649, 1606, 1581, 1511,
1440, 1311, 1260, 1151, 1080, 1027, 846, 765 cmϪ1 1H NMR
.
(300 MHz, CDCl3): δ ϭ 3.81 (s, 3 H, CO2CH3 or ArOMe), 3.88
(s, 3 H, ArOMe or CO2CH3), 5.59 (d, J ϭ 1.7 Hz, 1 H, 3a-H), 6.14
(d, J ϭ 1.7 Hz, 1 H, 3b-H), 6.93Ϫ6.98 (m, 2 H, 6-H), 8.05Ϫ8.10 1-Methoxycarbonyl-4-methyl-3-cyclohexen-1-yl
(m, 2 H, 5-H). 13C NMR (75.4 MHz, CDCl3): δ ϭ 52.6 (CO2CH3),
(11a) and 1-Methoxy-3-methyl-3-cyclohexen-1-yl p-Nitrobenzoate
55.5 (ArOCH3), 113.8 (C-6), 114.1 (C-3), 120.7 (C-4), 132.4 (C-5), (12a) ؊ Method A: In a screw-capped ACE pressure tube under
144.7 (C-2), 162.2 (ArCO2), 164.1 (C-8 or CO2CH3), 164.4 N2 atmosphere, a stirred mixture of 2d (0.1 g, 0.4 mmol) and 7
(CO2CH3 or C-8). C12H12O5 (236.2): calcd. C 61.02, H 5.12; found (0.40 g, 5.9 mmol) in dry xylene (5 mL) was heated to 170 °C for 4
p-Nitrobenzoate
C 61.34, H 5.16.
days. The solvent was then removed in vacuo, to leave 0.075 g
(58%) of a mixture of 11a and 12a (63:37) as a greenish oily residue.
Ethyl 2-(Benzoyloxy)-2-propenoate (3a): Following the general pro-
cedure, reaction of 5 (1.0 g, 8.6 mmol) with 6a (1.24 g, 8.80 mmol)
Method B: In a screw-capped ACE pressure tube under N2 atmo-
afforded 0.76 g (40%) of 3a as a colorless oil. Rf ϭ 0.8 (hexane/ sphere, a mixture of 2d (0.1 g, 0.4 mmol), AlCl3 (0.53 g, 4.0 mmol),
˜
EtOAc, 8:2). IR (film): ν ϭ 1735, 1650, 1452, 1373, 1302, 1266, and 7 (0.40 g, 5.9 mmol) in dry CH2Cl2 (5 mL) was stirred at room
1151, 1087, 1067, 1024 cmϪ1 1H NMR (300 MHz, CDCl3): δ ϭ
temperature for 48 h. The mixture was then diluted with CH2Cl2
1.29 (t, 3 H, CO2CH2CH3), 4.29 (q, 2 H, CO2CH2CH3), 5.62 (d, (50 mL) and washed with saturated aqueous NaHCO3 solution
.
J ϭ 1.7 Hz, 1 H, 3a-H), 6.17 (d, J ϭ 1.7 Hz, 1 H, 3b-H), 7.46Ϫ8.15
(2 ϫ 30 mL). The organic phase was dried (Na2SO4) and the solv-
(m, 5 H, PhH). 13C NMR (75.4 MHz, CDCl3): δ ϭ 14.0 ent was removed in vacuo to leave 0.42 g (83%) of a mixture of 11a
(CO2CH2CH3), 61.8 (CO2CH2CH3), 113.8 (C-3), 128.1 (C-6), 128.4 and 12a (96:4) as a greenish oily residue. This was purified by flash
Eur. J. Org. Chem. 2001, 4657Ϫ4669
4665

J. Tamariz et al.
FULL PAPER
column chromatography on silica gel (15 g; hexane/EtOAc, 100:2).
8:2). IR (film): ν
˜ ϭ 1750, 1718, 1606, 1286, 1259, 1228, 1168, 1100,
1
˜
1024 cm
Ϫ1. H NMR (300 MHz, CDCl3): δ ϭ 1.96Ϫ2.12 (br. s, 1
11a: Rf ϭ 0.57 (hexane/EtOAc, 8:2). IR (film): ν ϭ 1747, 1728,
1529, 1349, 1292, 1251, 1103, 1078 cmϪ1 1H NMR (300 MHz, H, 5-H), 2.04 (s, 3 H, AcO), 2.16Ϫ2.32 (m, 2 H, 5-H, 6-H),
CDCl3): δ ϭ 1.71 (br. s, 3 H, 4-CH3), 2.00Ϫ2.21 (m, 3 H, 5-H, 6- 2.63Ϫ2.78 (m, 1 H, 6-H), 3.74 (s, 3 H, CO2CH3), 3.86 (s, 3 H,
H), 2.40Ϫ2.56 (m, 1 H, 6-H), 2.63 (br. d, J ϭ 18.1 Hz, 2-H), 2.77 ArOCH3), 5.42 (d, J ϭ 4.9 Hz, 1 H, 2-H), 5.87Ϫ5.93 (m, 1 H, 3-
.
(br. d, J ϭ 18.1 Hz, 1 H, 2-H), 3.77 (s, 3 H, CO2CH3), 5.34 (br. s, H), 6.10Ϫ6.15 (m, 1 H, 4-H), 6.89Ϫ6.94 (m, 2 H, ArH), 7.27Ϫ7.95
1 H, 3-H), 8.14Ϫ8.32 (m, 4 H, ArH). Signals attributed to 12a: δ ϭ (m, 2 H, ArH). 13C NMR (75.4 MHz, CDCl3): δ ϭ 20.8 (CH3CO2),
3.79 (s, CO2CH3), 5.50 (br. s, 4-H). 13C NMR (75.4 MHz, CDCl3): 21.8 (C-5), 22.4 (C-6), 52.3 (CO2CH3), 55.4 (ArOCH3), 68.4 (C-2),
δ ϭ 23.1 (4-CH3), 26.3 (C-5), 28.5 (C-6), 32.8 (C-2), 52.6
79.1 (C-1), 133.7 (C-11), 121.4 (C-9), 121.7 (C-3), 131.9 (C-10),
(CO2CH3), 79.9 (C-1), 116.2 (C-3), 123.5 (C-11), 130.8 (C-10), 133.5 (C-4), 163.8 (C-8), 164.5 (C-12), 169.5 (MeCO2), 170.3
133.2 (C-4), 135.3 (C-9), 150.6 (C-12), 163.6 (C-8), 172.3 (CO2CH3). C18H20O7 (348.4): calcd. C 62.06, H 5.79; found C
(CO2CH3). Signals attributed to 12a: δ ϭ 21.5, 23.2, 28.1, 37.0,
80.5, 119.8. C16H17NO6 (319.3): calcd. C 60.18, H 5.37, N 4.39;
found C 60.43, H 5.23, N 4.24.
62.19, H 5.82.
1-Methoxycarbonyl-3-cyclohexen-1-yl p-Nitrobenzoate (26a): Ac-
cording to the method used to prepare 17a, reaction of 2d (0.2 g,
0.8 mmol) with 25 (1.00 g, 18.5 mmol) for 72 h gave 0.22 g (92%)
of 26a as a colorless oil. Rf ϭ 0.71 (hexane/EtOAc, 8:2). IR (film):
ν˜ ϭ 1747, 1728, 1529, 1349, 1296, 1256, 1113, 1104 cmϪ1. 1H NMR
(300 MHz, CDCl3): δ ϭ 2.04Ϫ2.16 (m, 1 H, 6-H), 2.17Ϫ2.25 (m,
2 H, 5-H), 2.44Ϫ2.53 (m, 1 H, 6-H), 2.64 (dm, J ϭ 18.3 Hz, 1 H,
2-H), 2.80 (dm, J ϭ 18.3 Hz, 1 H, 2-H), 3.76 (s, 3 H, CO2CH3),
5.62Ϫ5.70 (m, 1 H, 3-H), 5.75Ϫ5.84 (m, 1 H, 4-H), 8.17Ϫ8.35 (m,
4 H, ArH). 13C NMR (75.4 MHz, CDCl3): δ ϭ 21.6 (C-5), 28.1
(C-6), 32.5 (C-2), 52.6 (CO2CH3), 79.9 (C-1), 122.3 (C-3), 123.5 (C-
11), 126.0 (C-4), 130.8 (C-10), 135.3 (C-9), 150 (C-12), 163.6 (C-8),
172.9 (CO2CH3). C15H15NO6 (305.3): calcd. C 59.02, H 4.95, N
4.59; found C 59.42, H 4.99, N 4.47.
1-Methoxycarbonyl-4-methyl-3-cyclohexen-1-yl p-Methoxybenzoate
(11b) and 1-Methoxycarbonyl-3-methyl-3-cyclohexen-1-yl p-Me-
thoxybenzoate (12b): According to Method A for the preparation
of 11a/12a, a stirred mixture of 2f (0.04 g, 0.17 mmol) and 7 (0.40 g,
5.9 mmol) was heated to 170 °C for 96 h. Workup gave 0.033 g
(65%) of a mixture of 11b and 12b (67:33) as a greenish oily residue.
According to Method B for the preparation of 11a/12a, a mixture
of 2f (0.04 g, 0.17 mmol), AlCl3 (0.53 g, 4.0 mmol), and 7 (0.40 g,
5.9 mmol) was stirred at room temperature for 48 h. Workup gave
0.032 g (63%) of a mixture of 11b and 12b (88:12) as a greenish
˜
oily residue. 11b: Rf ϭ 0.65 (hexane/EtOAc, 8:2). IR (CH2Cl2): ν ϭ
1746, 1712, 1606, 1291, 1257, 1167, 1101 cmϪ1 1H NMR
.
1-Methoxycarbonyl-3-cyclohexen-1-yl p-Methoxybenzoate (26b):
According to the method used to prepare 26a, reaction of 2f
(0.15 g, 0.63 mmol) with 25 (1.00 g, 18.5 mmol) yielded 0.17 g
(300 MHz, CDCl3): δ ϭ 1.70 (br. s, 3 H, 4-CH3), 1.92Ϫ2.20 (m, 3
H, 5-H, 6-H), 2.36Ϫ2.64 (m, 2 H, 2-H, 6-H), 2.72 (br. d, J ϭ
18.1 Hz, 1 H, 2-H), 3.75 (s, 3 H, CO2CH3), 3.86 (s, 3 H, ArOCH3),
5.31 (br. s, 1 H, 3-H), 6.88Ϫ6.94 (m, 2 H, ArH), 7.92Ϫ7.98 (m, 2
H, ArH). Signals attributed to 12b: δ ϭ 3.76 (s, CO2CH3), 5.46 (br.
s, 4-H). 13C NMR (75.4 MHz, CDCl3): δ ϭ 23.2 (4-CH3), 26.5 (C-
5), 28.6 (C-6), 33.0 (C-2), 52.4 (CO2CH3), 55.4 (ArOCH3), 78.3 (C-
1), 113.6 (C-11), 116.5 (C-3), 122.4 (C-9), 131.8 (C-10), 133.1 (C-
4), 163.5 (C-8), 165.2 (C-12), 172.9 (CO2CH3). Signals attributed
to 12b: δ ϭ 21.6, 23.3, 28.3, 37.1, 55.5, 79.1, 119.7, 122.3, 132.5,
173.1. FAB HRMS (mNBA): calcd. 305.1389 [M ϩ H]ϩ; found
305.1399.
(93%) of 26b as a colorless oil. Rf ϭ 0.63 (hexane/EtOAc, 8:2). IR
Ϫ1
˜
(film): ν ϭ 1746, 1712, 1606, 1295, 1254, 1168, 1102, 1069 cm
.
1H NMR (300 MHz, CDCl3): δ ϭ 2.02Ϫ2.10 (m, 1 H, 6-H),
2.17Ϫ2.22 (m, 2 H, 5-H), 2.44 (dm, J ϭ 13.2 Hz, 1 H, 6-H), 2.63
(dm, J ϭ 18.3 Hz, 1 H, 2-H), 2.75 (dm, J ϭ 18.3 Hz, 1 H, 2-H),
3.76 (s, 3 H, CO2CH3), 3.85 (s, 3 H, ArOCH3), 5.59Ϫ5.68 (m, 1 H,
3-H), 5.73Ϫ5.81 (m, 1 H, 4-H), 6.89Ϫ6.93 (m, 2 H, ArH),
7.95Ϫ7.98 (m, 2 H, ArH). 13C NMR (75.4 MHz, CDCl3): δ ϭ 21.6
(C-5), 28.2 (C-6), 32.6 (C-2), 52.3 (CO2CH3), 55.3 (ArOCH3), 78.3
(C-1), 113.5 (C-11), 122.2 (C-9), 122.5 (C-3), 125.9 (C-4), 131.8
(C-10), 163.5 (C-8), 165.1 (C-12), 172.9 (CO2CH3). FAB HRMS
(mNBA): calcd. 291.1232 [M ϩ H]ϩ; found 291.1230.
(1R*,2R*)-2-Acetoxy-1-methoxycarbonyl-3-cyclohexen-1-yl p-Nitro-
benzoate (17a): In a screw-capped ACE pressure tube under N2
atmosphere, a stirred mixture of 2d (0.2 g, 0.8 mmol) and 8
(0.094 g, 0.84 mmol) in dry xylene (5 mL) was heated to 150 °C for
48 h. The solvent was then removed in vacuo to leave a greenish
oily residue. This was purified by flash column chromatography on
silica gel (20 g; hexane/EtOAc, 90:10) to yield 0.15 g (52%) of 17a
as colorless crystals. Rf ϭ 0.34 (hexane/EtOAc, 8:2). IR (KBr): ν˜ ϭ
1751, 1724, 1607, 1534, 1438, 1347, 1296, 1232, 1104, 1026, 844,
(1R*,2R*,4R*)-2-Methoxycarbonyl-5-norbornen-2-yl p-Nitrobenzo-
ate (21a) and (1R*,2S*,4R*)-2-Methoxycarbonyl-5-norbornen-2-yl
p-Nitrobenzoate (22a) ؊ Method A: According to Method A for
the preparation of 17a, a mixture of 2d (0.2 g, 0.8 mmol) and 9
(1.0 g, 15 mmol) in dry xylene (2 mL) was stirred for 12 h. Column
chromatographic workup on silica gel (15 g; hexane/EtOAc, 9:1)
yielded 0.20 g (79%) of a mixture of 21a and 22a (59:41) as a color-
less oil.
1
719 cmϪ1. H NMR (300 MHz, CDCl3): δ ϭ 1.90Ϫ2.30 (m, 1 H,
5-H), 2.06 (s, 3 H, AcO), 2.23Ϫ2.41 (m, 2 H, 5-H, 6-H), 2.63Ϫ2.72
(m, 1 H, 6-H), 3.76 (s, 3 H, CO2CH3), 5.46 (d, J ϭ 4.9 Hz, 1 H, 2-
H), 5.88Ϫ5.96 (m, 1 H, 3-H), 6.11Ϫ6.19 (m, 1 H, 4-H), 8.13Ϫ8.17
(m, 2 H, ArH), 8.27Ϫ8.32 (m, 2 H, ArH). 13C NMR (75.4 MHz,
Method B: Under an N2 atmosphere at Ϫ50 °C, a mixture of 2d
(0.2 g, 0.8 mmol), 9 (1.0 g, 15 mmol), and TiCl4 (0.76 g, 4.0 mmol)
in dry CH2Cl2 (5 mL) was stirred for 8 h. The mixture was then
diluted with CH2Cl2 (50 mL) and washed with saturated aqueous
NaHCO3 solution (2 ϫ 30 mL). The organic phase was dried
(Na2SO4) and the solvent was removed in vacuo to leave a greenish
oily residue. This was purified by flash column chromatography on
silica gel (15 g; hexane/EtOAc, 9:1), giving 0.19 g (76%) of a mix-
ture of 21a and 22a (55:45) as a colorless oil.
CDCl3):
δ ϭ 20.8 (CH3CO2), 21.7 (C-5), 22.3 (C-6), 52.7
(CO2CH3), 68.2 (C-2), 80.5 (C-1), 121.6 (C-3), 123.7 (C-11), 131.0
(C-10), 133.4 (C-4), 134.4 (C-9), 150.8 (C-12), 162.9 (C-8), 169.4
(MeCO2), 169.6 (CO2CH3). C17H17NO8 (363.3): calcd. C 56.20, H
4.72, N 3.85; found C 56.50, H 4.52, N 3.71.
(1R*,2R*)-2-Acetoxy-1-methoxycarbonyl-3-cyclohexen-1-yl p-Me-
thoxybenzoate (17b): According to the method used to prepare 17a,
reaction of 2f (0.20 g, 0.85 mmol) with 8 (0.140 g, 1.25 mmol) gave
Method C: According to Method B for the preparation of 11a/12a,
a mixture of 2d (0.2 g, 0.8 mmol), 9 (1.0 g, 15 mmol), and AlCl3
0.19 g (64%) of 17b as a colorless oil. Rf ϭ 0.40 (hexane/EtOAc, (1.08 g, 8.01 mmol) was stirred for 24 h. The mixture was then
4666 Eur. J. Org. Chem. 2001, 4657Ϫ4669

Synthesis of New Captodative Alkenes
FULL PAPER
worked-up by flash column chromatography on silica gel (15 g;
(dm, J ϭ 8.7 Hz, 1 H, 7-Hs), 1.83 (dd, J ϭ 13.2, 3.3 Hz, 1 H, 3-
hexane/EtOAc, 9:1), giving 0.177 g (70%) of a mixture of 21a and Hx), 1.99 (br. d, J ϭ 8.7 Hz, 1 H, 7-Ha), 2.44 (dd, J ϭ 13.2, 3.0 Hz,
22a (75:25) as a colorless oil. The isomers were separated by radial 1 H, 3-Hn), 2.98 (br. s, 1 H, 4-H), 3.25 (br. s, 1 H, 1-H), 3.68 (s, 3
chromatography (hexane/EtOAc, 100:2), giving 0.06 g (24%) of 21a H, CO2CH3), 3.87 (s, 3 H, ArOMe), 5.91 (dd, J ϭ 5.8, 3.0 Hz, 1
as a colorless oil and 0.025 g (10%) of 22a as colorless crystals. 21a: H, 6-H), 6.43 (dd, J ϭ 5.8, 3.0 Hz, 1 H, 5-H), 6.91Ϫ6.95 (m, 2 H,
Rf ϭ 0.46 (hexane/EtOAc, 8:2). IR (film): ν ϭ 1742, 1728, 1530, ArH), 7.99Ϫ8.03 (m, 2 H, ArH). 13C NMR (75.4 MHz, CDCl3):
1349, 1290, 1250, 1164, 1117, 1102 cmϪ1 1H NMR (300 MHz, δ ϭ 39.1 (C-3), 41.9 (C-4), 49.0 (C-7), 51.7 (C-1), 55.2 (CO2CH3),
˜
.
CDCl3): δ ϭ 1.47 (dd, J ϭ 13.2, 3.6 Hz, 1 H, 3-Hn), 1.60 (dm, J ϭ 55.4 (ArOCH3), 86.3 (C-2), 113.6 (C-12), 122.1 (C-10), 130.2 (C-
9.1 Hz, 1 H, 7-Hs), 1.90 (br. d, J ϭ 9.1 Hz, 1 H, 7-Ha), 2.82 (dd, 6), 131.9 (C-11), 141.8 (C-5), 163.6 (C-9), 165.9 (C-13), 171.6 (C-8).
J ϭ 13.2, 3.6 Hz, 1 H, 3-Hx), 3.03 (br. s, 1 H, 4-H), 3.45 (br. s, 1 C17H18O5 (302.3): calcd. C 67.54, H 6.00; found C 67.85, H 6.14.
H, 1-H), 3.77 (s, 3 H, CO2CH3), 6.20 (dd, J ϭ 5.6, 3.0 Hz, 1 H, 6-
X-ray Crystallographic Study: Alkene 2d (hexane/CH2Cl2, 30:5) and
H), 6.46 (dd, J ϭ 5.6, 3.0 Hz, 1 H, 5-H), 8.11Ϫ8.18 (m, 2 H, ArH),
adduct 22a (hexane/EtOAc, 100:2) were obtained as pale-yellow
8.26Ϫ8.32 (m, 2 H, ArH). 13C NMR (75.4 MHz, CDCl3): δ ϭ 40.1
and colorless crystals, respectively. The selected specimens were
(C-3), 42.1 (C-4), 47.5 (C-7), 51.0 (C-1), 52.8 (CO2CH3), 88.0 (C-
mounted on glass fibers. Crystallographic measurements were per-
2), 123.5 (C-12), 130.8 (C-11), 132.6 (C-6), 135.1 (C-10), 140.5 (C-
formed at room temperature on a Siemens P4 diffractometer using
5), 150.6 (C-13), 164.2 (C-9), 172.5 (C-8). C16H15NO6 (317.3):
Mo-Kα radiation (graphite crystal monochromator, λ ϭ 0.71073
calcd. C 60.57, H 4.77, N 4.41; found C 60.63, H 4.80, N 4.11.
˚
A). Two standard reflections were monitored periodically; they
Data for 22a: Rf ϭ 0.43 (hexane/EtOAc, 8:2); m.p. 139Ϫ140 °C.
showed no change during data collection. Unit cell parameters
were obtained from a least-squares refinement of 26 reflections in
the range 2 Ͻ 2Θ Ͻ 20°. Intensities were corrected for Lorentz and
polarization effects. No absorption correction was applied. Aniso-
tropic temperature factors were introduced for all non-hydrogen
atoms. Hydrogen atoms were placed in idealized positions and their
IR (KBr): ν˜ ϭ 1749, 1727, 1528, 1288, 1102, 1053 cmϪ1. 1H NMR
(300 MHz, CDCl3): δ ϭ 1.77 (dm, J ϭ 8.8 Hz, 1 H, 7-Hs), 1.86
(dd, J ϭ 13.3, 3.8 Hz, 1 H, 3-Hx), 1.98 (br. d, J ϭ 8.8 Hz, 1 H, 7-
Ha), 2.48 (dd, J ϭ 13.3, 2.9 Hz, 1 H, 3-Hn), 3.03 (br. s, 1 H, 4-H),
3.30 (br. s, 1 H, 1-H), 3.69 (s, 3 H, CO2CH3), 5.92 (dd, J ϭ 5.6,
3.2 Hz, 1 H, 6-H), 6.46 (dd, J ϭ 5.6, 2.9 Hz, 1 H, 5-H), 8.21Ϫ8.33
atomic coordinates were refined. Unit weights were used in the re-
(m, 4 H, ArH). 13C NMR (75.4 MHz, CDCl3): δ ϭ 39.2 (C-3), 42.0
finement. The structures were solved using SHELXTL[32] on a per-
(C-4), 49.1 (C-7), 51.8 (C-1), 52.5 (CO2CH3), 87.7 (C-2), 123.6 (C-
12), 130.1 (C-6), 130.9 (C-11), 135.1 (C-10), 142.0 (C-5), 150.6 (C-
13), 164.3 (C-9), 170.9 (C-8). C16H15NO6 (317.3): calcd. C 60.57,
H 4.77, N 4.41; found C 60.29, H 5.01, N 4.68.
sonal computer. Data for 2d: Formula: C11H9NO6; molecular
weight: 251.19; crystal system: monoclinic; space group: C2/c; unit
˚
cell parameters: a ϭ 24.087(4), b ϭ 7.0305(5), c ϭ 13.6270(9) A;
3
˚
α ϭ 90°, β ϭ 100.533(12)°, γ ϭ 90°; V ϭ 2268.7(5) A ; T ϭ
293(2) K; Z ϭ 8; θ scan range: 1.72Ϫ27.00°; no. of reflections col-
lected: 3056; no. of observed reflections: 2443; R ϭ 0.0597; GoF ϭ
1.025. Data for 22a: Formula: C16H15NO6; molecular weight:
317.29; crystal system: monoclinic; space group: C2/c; unit cell
(1R*,2R*,4R*)-2-Methoxycarbonyl-5-norbornen-2-yl p-Methoxyb-
enzoate (21b) and (1R*,2S*,4R*)-2-Methoxycarbonyl-5-norbornen-
2-yl p-Methoxybenzoate (22b) ؊ Method A: According to Method
A for the preparation of 21a/22a, a mixture of 2f (0.15 g,
0.63 mmol) and 9 (1.0 g, 15 mmol) in dry xylene (2 mL) was al-
lowed to react to give 0.14 g (76%) of a mixture of 21b and 22b
(60:40) as a colorless oil.
˚
parameters: a ϭ 23.926(3), b ϭ 7.9688(9), c ϭ 16.659(2) A; α ϭ
3
˚
90°, β ϭ 109.541(13)°, γ ϭ 90°; V ϭ 2993.4(6) A ; T ϭ 293(2) K;
Z ϭ 8; θ scan range: 1.81Ϫ26.99°; no. of reflections collected: 4028;
no. of observed reflections: 3273; R ϭ 0.0775; GoF ϭ 1.052.
Method B: According to Method B for the preparation of 11a/11b,
a mixture of 2f (0.15 g, 0.63 mmol), 9 (1.0 g, 15 mmol), and AlCl3
(0.84 g, 6.3 mmol) was allowed to react to give 0.14 g (73%) of a
mixture of 21b and 22b (57:43) as a colorless oil.
UP and ET Spectroscopy: The UP spectra were obtained with a
HeI source in conjunction with a Datalab DL400 signal analysis
system. The bands, calibrated against rare-gas lines, were located
using the positions of their maxima, which were taken as corres-
ponding to the vertical ionization energies. The accuracy of the IE
values was estimated to be better than Ϯ0.05 eV. The ET spectra
were obtained by operating the instrument in such a mode as to
obtain a signal related to the nearly total scattering cross section.
The energy scales were calibrated using the (1s1Ϫ2s2) 2S anion state
of helium and the estimated accuracy of the measured VAE values
was Ϯ0.05 eV.
Method C: According to Method B for the preparation of 21a/22a,
a mixture of 2f (0.15 g, 0.63 mmol), 9 (1.0 g, 15 mmol), and TiCl4
(0.57 g, 3.0 mmol) was allowed to react to give 0.134 g (70%) of
21b and 22b (84:16). The mixture was purified by flash column
chromatography on silica gel (15 g; hexane/EtOAc, 9:1). The iso-
mers were separated by radial chromatography (hexane/EtOAc,
100:2), giving 0.04 g (21%) of 21b and 0.021 g (11%) of 22b as col-
orless oils. Data for 21b: Rf ϭ 0.43 (hexane/EtOAc, 8:2). IR (film):
ν˜ ϭ 1742, 1712, 1607, 1511, 1291, 1257, 1162, 1102 cmϪ1. 1H NMR
(300 MHz, CDCl3): δ ϭ 1.43 (dd, J ϭ 13.0, 3.7 Hz, 1 H, 3-Hn),
1.60 (dm, J ϭ 9.1 Hz, 1 H, 7-Hs), 1.86 (br. d, J ϭ 9.1 Hz, 1 H, 7-
Ha), 2.79 (dd, J ϭ 13.0, 3.6 Hz, 1 H, 3-Hx), 2.98 (br. s, 1 H, 4-H),
3.41 (br. s, 1 H, 1-H), 3.75 (s, 3 H, CO2CH3), 3.85 (s, 3 H, ArOMe),
6.19 (dd, J ϭ 5.7, 3.1 Hz, 1 H, 6-H), 6.41 (dd, J ϭ 5.7, 3.0 Hz, 1
H, 5-H), 6.89Ϫ6.92 (m, 2 H, ArH), 7.90Ϫ7.93 (m, 2 H, ArH). 13C
NMR (75.4 MHz, CDCl3): δ ϭ 39.9 (C-3), 42.1 (C-4), 47.3 (C-7),
51.0 (C-1), 52.6 (CO2CH3), 55.4 (ArOCH3), 86.7 (C-2), 133.5 (C-
Theoretical Calculations: The ab initio SCF/RHF calculations were
carried out with the 3Ϫ21G and 6Ϫ31G* basis sets using
GAUSSIAN-94[23a] (PC-Linux) and GAMESS[23b] for Windows 95/
98/NT (v. 6.0). Geometries were optimized using the 3Ϫ21G level
basis set, and these were employed as starting points for optimiza-
tion at the 6Ϫ31G* level.
Acknowledgments
´
12), 122.0 (C-10), 131.8 (C-11), 132.8 (C-6), 140.1 (C-5), 163.5 (C- We are grateful to Dr. Francisco Mendez for helpful comments. We
9), 165.8 (C-13), 173.2 (C-8). C17H18O5 (302.3): calcd. C 67.54, H thank Fernando Labarrios for his help in performing the spectro-
6.00; found C 67.28, H 6.07. Ϫ Data for 22b: Rf ϭ 0.40 (hexane/ metric measurements. J.T. gratefully acknowledges the DEPI/IPN
EtOAc, 8:2). IR (film): ν˜ ϭ 1745, 1712, 1606, 1511, 1288, 1259, (grant no. 200410) and CONACYT (grants nos. 1570P-E9507 and
1167, 1101, 1032 cmϪ1
.
1H NMR (300 MHz, CDCl3): δ ϭ 1.73
32273-E) for financial support; he also thanks the OAS (F54111),
Eur. J. Org. Chem. 2001, 4657Ϫ4669
4667

J. Tamariz et al.
FULL PAPER
[10]
[11]
Alkene 2a has been reported previously and has been used in
polymerization reactions: H. Tanaka, S. Yoshida, Macromolec-
ules 1995, 28, 8117Ϫ8121.
CONACYT (250099), and COFAA-IPN (9711220214) for support
during sabbatical leave in the U.S.A. H.A.J.-V. thanks CONACYT
(grant no. 3251P) for financial support. R.H. is grateful to CON-
ACYT for a graduate scholarship, and to the Ludwig K. Hellweg
Foundation for a partial scholarship. The Italian Ministero dellЈU-
´
´
H. A. Jimenez-Vazquez, M. E. Ochoa, G. Zepeda, A. Modelli,
D. Jones, J. A. Mendoza, J. Tamariz, J. Phys. Chem. A 1997,
101, 10082Ϫ10089.
`
niversita e della Ricerca Scientifica e Tecnologica is also acknow-
[12] [12a]
K. N. Houk, R. W. Strozier, J. Am. Chem. Soc. 1973, 95,
ledged for financial support.
[12b]
4094Ϫ4096;
26, 129Ϫ158.
T. Oh, M. Reilly, Org. Prep. Proc. Int. 1994,
[13]
[14]
[15]
˜
A. Sbai, V. Branchadell, R. M. Ortuno, A. Oliva, J. Org. Chem.
1997, 62, 3049Ϫ3054.
[1] [1a]
J. Sauer, R. Sustmann, Angew. Chem. Int. Ed. Engl. 1980,
[1b]
´
K. N. Houk, J. Gonzalez, Y. Li, Acc. Chem.
19, 779Ϫ807;
Res. 1995, 28, 81Ϫ90;
B. Thomas IV, J. Am. Chem. Soc. 1995, 117, 8594Ϫ8599;
F. Bernardi, P. Celani, M. Olivucci, M. A. Robb, G. Suzzi-Valli,
E. F. Lutz, G. M. Bailey, J. Am. Chem. Soc. 1964, 86,
3899Ϫ3901.
J. A. Berson, Z. Hamlet, W. A. Mueller, J. Am. Chem. Soc.
1962, 84, 297Ϫ304.
[1c]
O. Wiest, K. N. Houk, K. A. Black,
[1d]
[1e]
J. Am. Chem. Soc. 1995, 117, 10531Ϫ10536;
E. Goldstein,
[16] [16a]
R. Breslow, T. Guo, J. Am. Chem. Soc. 1988, 110,
B. Beno, K. N. Houk, J. Am. Chem. Soc. 1996, 118,
[16b]
5613Ϫ5617;
Engberts, J. Am. Chem. Soc. 1991, 113, 4241Ϫ4246;
Grieco, Aldrichimica Acta 1991, 24, 59Ϫ66.
W. Blokzijl, M. J. Blandamer, J. B. F. N.
[1f]
6036Ϫ6043;
P. Binger, P. Wedemann, R. Goddard, U. H.
[16c]
P. A.
Brinker, J. Org. Chem. 1996, 61, 6462Ϫ6464.
[2] [2a]
[2b]
K. Alder, Experientia 1955, Suppl. II, 86Ϫ118;
O. Eis-
[17]
[18]
´
´
´
A. Benavides, R. Martınez, H. A. Jimenez-Vazquez, F.
Delgado, J. Tamariz, Heterocycles 2001, 55, 469Ϫ485.
Crystallographic data (excluding structural factors) for the
structures reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary
publication nos. CCDC-162378 (2d) and CCDC-162379
(22a). Copies of the data can be obtained free of charge
on application to the CCDC, 12 Union Road, Cambridge
CB2 1EZ, U.K. [Fax: (internat.) ϩ44 (0)1223/336023; E-mail:
deposit@ccdc.cam.ac.uk].
[2c]
enstein, N. T. Anh, Bull. Soc. Chim. Fr. 1973, 2721Ϫ2723;
O. Eisenstein, N. T. Anh, Bull. Soc. Chim. Fr. 1973, 2723Ϫ2726;
[2d]
N. T. Anh, E. Canadell, O. Eisenstein, Tetrahedron 1978,
[2e]
34, 2283Ϫ2288;
R. Sustmann, R. Schubert, Angew. Chem.
Int. Ed. Engl. 1972, 11, 840.
[3] [3a]
[3b]
R. Sustmann, Tetrahedron Lett. 1971, 2721Ϫ2724;
I.
Fleming, Frontier Orbitals and Organic Chemical Reactions,
[3c]
John Wiley & Sons, Chichester, 1976;
Chem. Res. 1975, 8, 361Ϫ369;
K. N. Houk, Acc.
O. Eisenstein, J. M. Lefour,
[3d]
[3e]
N. T. Anh, R. F. Hudson, Tetrahedron 1977, 33, 523Ϫ531;
[19] [19a]
´
J. I. Garcıa, J. A. Mayoral, L. Salvatella, Acc. Chem. Res.
´
C. Spıno, M. Pesant, Y. Dory, Angew. Chem. Int. Ed. 1998,
2000, 33, 658Ϫ664, and references included therein; [19b] S. Ya-
37, 3262Ϫ3265.
[19c]
mabe, T. Minato, J. Org. Chem. 2000, 65, 1830Ϫ1841;
J. I.
[4] [4a]
´
H. G. Viehe, Z. Janousek, R. Merenyi, L. Stella, Acc. Chem.
´
´
Garcıa, V. Martınez-Merino, J. A. Mayoral, L. Salvatella, J.
Am. Chem. Soc. 1998, 120, 2415Ϫ2420.
[4b]
Res. 1985, 18, 148Ϫ154;
W. Odenkirk, A. L. Rheingold, B.
[4c]
Bosnich, J. Am. Chem. Soc. 1992, 114, 6392Ϫ6398;
Boucher, L. Stella, Tetrahedron 1986, 42, 3871Ϫ3885;
J.-L.
[20]
Assuming Koopmans’ theorem, (see ref. [20a]) the energies of
the occupied and vacant MOЈs can be correlated with the nega-
tive of the corresponding IEs and VEA values: (see refs.
[2c,3c,20b]). [20a] T. Koopmans, Physica 1933, 1, 104Ϫ113; [20b] K.
N. Houk, J. Sims, R. E. Duke, Jr., R. W. Strozier, J. K. George,
J. Am. Chem. Soc. 1973, 95, 7287Ϫ7301.
J. P. Morizur, J. Mercier, M. Sarraf, Org. Mass Spectrom. 1982,
17, 327Ϫ330.
A. Modelli, Trends Chem. Phys. 1997, 6, 57Ϫ95.
RHF/6-31G* calculations were performed using GAUSSIAN-
94[23a] and GAMESS:[23b] [23a] 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. Petersson, J. A. Montgomery, K.
Raghavachari, M. A. Al-Laham, V. G. Zakrzewski, J. V. Ortiz,
J. B. Foresman, J. Cioslowski, 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,
[4d]
K.
Maruoka, H. Imoto, H. Yamamoto, J. Am. Chem. Soc. 1994,
[4e]
116, 12115Ϫ12116;
1983, 24, 885Ϫ888;
D. Döpp, H. Libera, Tetrahedron Lett.
V. K. Aggarwal, D. E. Jones, A. M.
[4f]
[4g]
Martin-Castro, Eur. J. Org. Chem. 2000, 2939Ϫ2945;
Kouklovsky, O. Dirat, T. Berranger, Y. Langlois, M. E. Tran-
C.
[21]
[4h]
Hun-Dau, C. Riche, J. Org. Chem. 1998, 63, 5123Ϫ5128;
E. Murray, A. Alvarez-Larena, J. F. Piniella, V. Branchadell,
[22]
[23]
R. M. Ortuno, J. Org. Chem. 2000, 65, 388Ϫ396; [4i] E. J. Corey,
˜
A. Guzman-Perez, Angew. Chem. Int. Ed. 1998, 37, 388Ϫ401;
[4j]
P. Vogel, J. Cossy, J. Plumet, O. Arjona, Tetrahedron 1999,
[4k]
55, 13521Ϫ13642;
K. Ishihara, H. Kurihara, M. Matsum-
oto, H. Yamamoto, J. Am. Chem. Soc. 1998, 120, 6920Ϫ6930;
[4l]
M. L. Quan, C. D. Ellis, A. Y. Liauw, R. S. Alexander, R.
M. Knabb, G. Lam, M. R. Wright, P. C. Wong, R. R. Wester,
J. Med. Chem. 1999, 42, 2760Ϫ2773.
[5]
[6]
J. Tamariz, P. Vogel, Helv. Chim. Acta 1981, 64, 188Ϫ197.
´
O. Garcıa de Alba, J. Chanona, F. Delgado, G. Zepeda, F. Lab-
M. Head-Gordon, C. Gonzalez, J. A. Pople, Gaussian, Inc.,
´
arrios, R. W. Bates, S. Bott, E. Juaristi, J. Tamariz, Anal. Quım.,
[23b]
Pittsburgh, PA, 1995;
M. W. Schmidt, K. K. Baldridge, J.
Int. Ed. 1996, 92, 108Ϫ117.
[7]
A. Boatz, S. T. Elbert, M. S. Gordon, J. H. Jensen, S. Koseki,
N. Matsunaga, K. A. Nguyen, S. J. Su, T. L. Windus, M. Du-
pui, J. A. Montgomery, J. Comput. Chem. 1993, 14,
1347Ϫ1363.
A. Modelli, M. Venuti, F. Scagnolari, M. Contento, D. Jones,
J. Phys. Chem. A 2001, 105, 219Ϫ226.
Analysis of captodative alkenes 1 in 1,3-dipolar cycloadditions
with nitrones according to the DFT/HSAB theory shows that
the hardϪhard interactions between these cycloaddends con-
trol the process and that the main contribution is furnished by
the electron-releasing group.[8a]
The role of the conformation of the electron-donor group may
be related to the calculated preferential s-trans conformation
of enol ethers at the transition state in DielsϪAlder cycloaddi-
tions: J. Liu, S. Niwayama, Y. You, K. N. Houk, J. Org. Chem.
1998, 63, 1064Ϫ1073.
˜
A. Reyes, R. Aguilar, A. H. Munoz, J.-C. Zwick, M. Rubio, J.-
L. Escobar, M. Soriano, R. Toscano, J. Tamariz, J. Org. Chem.
1990, 55, 1024Ϫ1034.
[8] [8a]
´
R. Herrera, A. Nagarajan, M. A. Morales, F. Mendez, H.
[24]
[25]
´
´
A. Jimenez-Vazquez, L. G. Zepeda, J. Tamariz, J. Org. Chem.
2001, 66, 1252Ϫ1263; [8b] A. Nagarajan, G. Zepeda, J. Tamariz,
[8c]
´
R. Jimenez, L.
Tetrahedron Lett. 1996, 37, 6835Ϫ6838;
´
Perez, J. Tamariz, H. Salgado, Heterocycles 1993, 35, 591Ϫ598.
[9] [9a]
˜
R. M. Andrade, A. H. Munoz, J. Tamariz, Synth. Commun.
1992, 22, 1603Ϫ1609; [9b] A. Orduna, L. G. Zepeda, J. Tamariz,
˜
Synthesis 1993, 375Ϫ377; [9c] R. Aguilar, A. Reyes, A. Orduna,
˜
[26]
´
´
G. Zepeda, R. W. Bates, J. Tamariz, Rev. Soc. Quım. Mex. 2000,
[9d]
44, 91Ϫ96;
Z. Dienes, P. Vogel, J. Org. Chem. 1996, 61,
6958Ϫ6970; [9e] J. D. Dudones, P. Sampson, J. Org. Chem. 1997,
[9f]
62, 7508Ϫ7511;
M. E. Ochoa, M. S. Arias, R. Aguilar, F.
Delgado, J. Tamariz, Tetrahedron 1999, 55, 14535Ϫ14546.
4668
Eur. J. Org. Chem. 2001, 4657Ϫ4669

Synthesis of New Captodative Alkenes
FULL PAPER
[27] [27a]
[29f]
´
´
´
C. Cativiela, M. D. Dıaz-
J. L. Gazquez, F. Mendez, J. Phys. Chem. 1994, 98,
Asymmetry 2001, 12, 347Ϫ353;
[27b]
[29g]
´
4591Ϫ4593;
F. Mendez, J. Tamariz, P. Geerlings, J. Phys.
de-Villegas, Tetrahedron: Asymmetry 1998, 9, 3517Ϫ3599;
P. G. M. Wuts, R. L. Gu, J. M. Northuis, Tetrahedron: Asym-
metry 2000, 11, 2117Ϫ2123.
Chem. A 1998, 102, 6292Ϫ6296.
[28]
[29]
´
S. Damoun, G. Van de Woude, F. Mendez, P. Geerlings, J.
[30] [30a]
Phys. Chem. A 1997, 101, 886Ϫ893.
J. Sanche, G. J. Schulz, Phys. Rev. A 1972, 5, 1672Ϫ1683;
[29a]
[30b]
[30c]
For some recent examples, see:
Enantioselective Synthesis
G. J. Schulz, Rev. Mod. Phys. 1973, 45, 378Ϫ422;
G.
of β-Amino Acids (Ed.: E. Juaristi), Wiley-VCH, New York,
J. Schulz, Rev. Mod. Phys. 1973, 45, 423Ϫ486.
1997; [29b] P. Camps, M. Font-Bardia, S. Gimenez, F. Perez, X.
A. Modelli, D. Jones, G. Distefano, Chem. Phys. Lett. 1982,
[31]
[32]
´
´
Solans, N. Soldevilla, Tetrahedron: Asymmetry 1999, 10,
86, 434Ϫ437.
[29c]
3123Ϫ3138;
J.-W. Chang, D.-P. Jang, B.-J. Uang, F.-L.
SHELXTL, v. 5.03, Siemens Energy & Automation, Ger-
Liao, S.-L. Wang, Org. Lett. 1999, 1, 2061Ϫ2063; [29d] C. Gaul,
many, 1995.
[29e]
´
D. Seebach, Org. Lett. 2000, 2, 1501Ϫ1504;
J. M. Andres,
Received June 8, 2001
[O01275]
´
´
M. A. Martınez, R. Pedrosa, A. Peres-Encabo, Tetrahedron:
Eur. J. Org. Chem. 2001, 4657Ϫ4669
4669