A DFT study of 1,3-dipolar cycloadditions of cyclic nitrones to unsaturated lactones. Part II

Article abstract of DOI:10.1016/j.tetasy.2008.09.003

The 1,3-dipolar cycloadditions of cyclic nitrones to five-membered unsaturated lactones are studied. Special attention was focused on a single and double asymmetric induction when one or both components were chiral. The energies of the cycloaddition reactions are investigated through application of molecular orbital calculations at the B3LYP/6-31+G(d) theory level. A study of the different reactants' approaches and their conformational aspects revealed the stereochemical preferences of these reactions. The results of these calculations correspond well with the previously reported experimental data.

Full text of DOI:10.1016/j.tetasy.2008.09.003

Tetrahedron: Asymmetry 19 (2008) 2140–2148
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
A DFT study of 1,3-dipolar cycloadditions of cyclic nitrones
to unsaturated lactones. Part II
a
a
b
b
c
a,
*
´
Sebastian Stecko , Konrad Pasniczek , Carine Michel , Anne Milet , Serge Perez , Marek Chmielewski
a Institute of Organic Chemistry of the Polish Academy of Sciences, Kasprzaka 44/52, 01-224 Warsaw, Poland
b Départament de Chimie Moléculaire, Université Joseph Fourier, BP53, 38041 Grenoble, France
c Centre de Recherches sur les Macromolécules Végétales, CNRS, BP53, 38041 Grenoble, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 July 2008
Accepted 4 September 2008
Available online 10 October 2008
The 1,3-dipolar cycloadditions of cyclic nitrones to five-membered unsaturated lactones are studied. Spe-
cial attention was focused on a single and double asymmetric induction when one or both components
were chiral. The energies of the cycloaddition reactions are investigated through application of molecular
orbital calculations at the B3LYP/6-31+G(d) theory level. A study of the different reactants’ approaches
and their conformational aspects revealed the stereochemical preferences of these reactions. The results
of these calculations correspond well with the previously reported experimental data.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
O
O
n( )
n( )
O
+
N
The 1,3-dipolar cycloadditions (1,3-DCs) between nitrones and
alkenes are a very useful tool for the construction of isoxazolidines.
The latter, after cleavage of the nitrogen–oxygen bond, can be
transformed into b-aminoalcohols.1 Owing to these attractive pos-
sibilities, such reactions are frequently used in the target-oriented
synthesis of nitrogen-containing products.2
O
N1
L1
E1
n=1
E2 n=2
n=1
L2 n=2
Ot-Bu
O
O
HO
+
N
As part of our research on the application of 1,3-DC reactions in
O
synthesis of polyhydroxy-alkaloids,3 we have recently reported
N2
L3
studies on the reactions of five-membered cyclic nitrones with c-
and d-lactones.4–6 In the case of d-lactones, the reactions usually
showed high diastereoselectivity and the exclusive formation of
exo-adducts.4 In marked contrast to d-lactones, the five-membered
lactones reacted with a lower diastereoselectivity and led to the
mixtures of products. This is a consequence of formation of endo-
products and the reversibility of the reaction.5,6 Such a course of
reaction has never been observed for d-lactones.4
Recently, we have presented a DFT study on the 1,3-DCs of nit-
rone N1 to the lactones L1/L2 and vinyl ethers E1/E2 (Chart 1).8
The calculated predictions were in good agreement with the exper-
imental results.7,9,10 For lactones, the meta-regioselectivity has
been predicted, whereas for ethers, which have an opposite elec-
tronic character, the ortho-regioselectivity was usually observed.
In all cases, the exo-adducts were found to have a lower energy
transition state than the endo ones. It has also been pointed out
that the high diastereoselectivity of the 1,3-DCs with six-mem-
bered dipolarophiles (which is manifested by the exclusive forma-
Chart 1.
tion of exo-adducts) is the result of a steric repulsion in their endo
transition state (TS).8
The DFT calculations and experimental observations8 prompted
us to conduct further studies on 1,3-DCs of cyclic nitrones to the
cyclic dipolarophiles, particularly to analyze the cases of chiral
reactants, the lactone or nitrone, or both. Herein, we report studies
on the reactions of nitrones N1 and N2 with 2-(5H)-furanones: L1,
L3, and ent-L3.
2. Computational methods
All calculations were carried out using the GAUSSIAN 03 suite of
software.11 The geometry optimization of critical points (reactants,
transition structures, and products) was carried out using DFT
method at the B3LYP/6-31+G(d) level theory.12 The stationary
points were characterized by the frequency calculations to verify
that both the minima and the transition structures (TSs) have zero
* Corresponding author. Tel./fax: +48 22 632 66 81.
E-mail address: chmiel@icho.edu.pl (M. Chmielewski).
0957-4166/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2008.09.003

S. Stecko et al. / Tetrahedron: Asymmetry 19 (2008) 2140–2148
2141
and one imaginary frequency, respectively. The optimizations were
carried out using the Berny analytical gradient optimization meth-
od.13 The electronic structures of the critical points were studied
by the natural bond orbital (NBO) method.14 The intrinsic reaction
coordinates (IRC)15 were also calculated to analyze the mechanism
in detail for all the transition structures. The molecular modeling
was carried out using HYPERCHEM 7.5 software (HyperCube, Inc.).
According to the experimental observations, for each presented
reaction only one regioisomeric approach of reactants8 was inves-
tigated, while both the exo and endo approaches of reactants were
analyzed.
(P1c1) has an almost flat A-ring with a coplanar arrangement of O1,
C1a, C4a and C4b atoms of B-ring (only the nitrogen atom is located
out-of-plane). For the two high-energy conformers P2c2 and P1c3,
the lactone ring A is twisted. In the latter case, atoms C1a, C4a, C4b
,
and N8 of the B-ring are almost coplanar while the O1 atom is lo-
cated out-of-plane. The geometries of P1c2 and P1c3 are very close,
differing mostly by the conformation of the C rings.
In contrast to P1, the P2 cycloadduct has only one set of con-
formers (type II). The P2c1 and P2c2 differ by the location of the
N8 atom in relation to the O1–C1a–C4a–C4b plane, which influences
the C-ring conformation. In further studies, for each investigated
reaction, the two most stable conformers of the product were ta-
ken into consideration. Such geometry of the reactants’ approach
gave a better idea of the cycloaddition, and we were able to deter-
mine that the type II structures correspond better to the related
transition state than the type I structures.
3. Results and discussion
3.1. Conformation analysis of tricyclic fused system
The conformational search using the MM+ force field for the two
simplest unsubstituted adducts P1 (exo) and P2 (endo) (Chart 2)
showed the presence of several conformers (Fig. 1). The extended
search (scanning range) also allowed us to find the energy of inver-
tomers P1c4 and P2c3 (Fig. 1, type III). Their large values of relative
energy suggest a high barrier of inversion of configuration at the
nitrogen atom. This result correlates well with the fact that stereo-
isomers related to inversion at the nitrogen atom of the 5–5–5
fused tricyclic systems have never been observed by us,5,6 or by
Font and de March.7b On the other hand, both invertomers can be
easily observed for systems having six-membered C-ring derived
from the nitrone and the five-, six-, or seven-membered lactone
ring.1,3e,7a,c All conformers of P1 and P2 are depicted in Figure 1.
3.2. The 1,3-DC reaction of nitrone N2 and lactone L 1—single
asymmetric induction system
The cycloaddition of the nitrone N2 to the unsubstituted lactone
L1 theoretically may afford four products: two exo- (P3, P4) and
two endo-adducts (P5, P6)—Scheme 1. The approach of reactants
can proceed with the diastereofacial selectivity anti to t-BuO sub-
stituent or syn to it. Our previous experiments have shown that un-
der the kinetic control a mixture of P3 and P5 can be formed in a
ratio of about 5.7:1.5 On the other hand, the mixture of P3 and
P4 was obtained when the experiment was carried out in boiling
toluene solution under thermodynamic conditions.6 Moreover, in
the latter case, the exo–syn adduct P4 has been found to be the ma-
jor one (P4:P3 ratio 3:2). Such a result was unexpected, since to the
best of our knowledge, the exo-approach in syn mode should be
disfavored. The formation of this product has been reported by
Brandi et al.,16 but the authors did not provide details of the reac-
tion conditions. Our careful studies and analysis, based on the NMR
and CD spectra as well as on the rentgenostructural analysis,
unambiguously confirmed the structure of compounds P3, P4, P5
as well as the conditions of their formation.5,6 The apparent stereo-
chemical pathway of the reaction encouraged us to undertake a
deeper analysis, involving computational methods.
O
A
O
O
O
2
4
4a
1a
4b
H
H
H
H
H
H
B
N
O
O1
N8
5
6
C
7
P1
P2
Chart 2.
Initially, the conformational studies for P3, P4, and P5 adducts
were performed. The same analysis was also carried out for the
hypothetical adduct P6, which has not been observed.5,6 In Figure
2, the two most stable conformers of each adduct are shown.
Two types of conformers were found for P1 adduct (Fig. 1, type I
and II). The most significant difference between these conformers
is the location of the oxygen atom of isoxazolidine in relation to
the C1a–C4a–C4b–N8 plane, Figure 1. The most stable P1 conformer
Figure 1. The conformers of adducts P1 and P2 with corresponding relative energies in kcal/mol (MM+ force field).

2142
S. Stecko et al. / Tetrahedron: Asymmetry 19 (2008) 2140–2148
is almost syn-periplanar with a small deviation of the correspond-
O
O
O
O
ing dihedral angle in range of 26° from 0° value (Scheme 2). Other
conformations of the t-BuO group are strongly disfavored resulting
from the apparent steric hindrance. For the P3 adduct, the con-
formers with an anti-periplanar arrangement of H5 atom and t-
Bu group have ca. 7.5–9.0 kcal/mol higher energy than conformers
with a lowest energy (Scheme 2). For other adducts, the corre-
sponding energy difference is even larger. These distortions are
well illustrated by the P4 adduct in which the t-BuO group and
isoxazolidine ring (fused to the lactone ring) are on the same side
of the pyrrolidine moiety. The accurate values of the H5–C5–O–t-Bu
dihedral angle for each adduct are presented in Table 1. The rela-
tive energies of these adducts are presented in Table 2. For each
of the adducts, only the two lower energy conformers (both type)
were taken into consideration.
H
H
H
H
H
H
O
O
O
t
-Bu
Ot-Bu
N
N
P3
P4
(
exo-anti
)
(exo-syn)
N2
+ L1
O
O
O
O
H
H
H
H
H
H
O
O
O
t
-Bu
Ot-Bu
N
N
P5
P6
(
endo-anti
)
(endo-syn)
Scheme 1.
N
N
N
t-Bu
Similarly, as in the case of the P1 cycloadduct, two sets of conform-
ers (type I and II) were also found for P3, P4, and P6. Adduct P5 be-
haves similarly as P2 showing only type II conformers. For every
type of conformer, several substructures were found resulting from
the relative mobility of the t-BuO group. The energy differences
between such substructures are small, and do not exceed
0.4 kcal/mol. Figure 3 presents the X-ray structures17 for the
cycloadducts P3, P4, and P5, and selected geometrical data are
presented in Table 1.
O
H
O
H
O
t-Bu
t-Bu
H
o
o
φ
10-26
φ
10-26
A
B
C
Scheme 2.
The main geometric parameters for all cycloadducts are col-
lected in Table 1. In most cases, the lengths of O1–C1a, C4a–C4b
and C4b–N8 bonds are slightly shorter for the more stable c2 con-
,
The obtained results revealed that only a narrow range of
H5–C5–O–t-Bu dihedral angles can be taken into consideration.
The most favorable arrangement of the H5 atom and t-Bu group
formers. An opposite trend is observed for the C1a–C4a and N8–O1
Figure 2. The conformers of adducts P3, P4, P5 and P6.
Figure 3. Molecular structures of the compounds P3–X, P4–X and P5–X with the crystallographic numbering scheme.17

S. Stecko et al. / Tetrahedron: Asymmetry 19 (2008) 2140–2148
2143
Table 1
Several geometrical parameters of cycloadducts P3, P4, P5 and P6
Bond lengths (Å)
Dihedral angles (°)
O1–C1a
C1a–C4a
C4a–C4b
C4b–N8
N8–O1
u1
22.1
u2
u3
u4
23.5
u5
P3c1
P3c2
P3–X
1.441
1.419
1.417
1.528
1.544
1.543
1.555
1.539
1.536
1.489
1.481
1.481
1.434
1.487
1.488
ꢀ3.7
16.9
ꢀ4.5
ꢀ0.7
13.1
ꢀ23.3
ꢀ20.7
ꢀ6.1
ꢀ22.9
ꢀ25.2
ꢀ1.2
ꢀ3.5
1.7
P4c1
P4c2
P4–X
1.442
1.419
1.418
1.526
1.541
1.538
1.545
1.534
1.529
1.496
1.486
1.484
1.435
1.484
1.486
ꢀ23.1
4.5
ꢀ0.1
5.8
22.7
26.4
ꢀ18.0
ꢀ23.1
0.5
ꢀ4.2
ꢀ15.9
ꢀ20.7
ꢀ21.6
2.3
ꢀ3.4
P5c1
P5c2
P5–X
1.444
1.416
1.417
1.531
1.536
1.476
1.564
1.554
1.492
1.471
1.506
1.499
1.441
1.478
1.419
ꢀ15.5
ꢀ26.0
ꢀ32.2
25.2
ꢀ2.7
3.5
ꢀ12.0
ꢀ17.0
ꢀ18.6
ꢀ17.0
ꢀ26.0
ꢀ25.6
15.9
14.9
11.6
P6c1
P6c2
1.451
1.423
1.532
1.541
1.555
1.543
1.462
1.486
1.448
1.484
25.2
ꢀ1.5
ꢀ34.6
ꢀ28.0
18.1
7.8
24.8
1.9
ꢀ27.3
27.8
u1—the values of O1–C1a–C4a–C4b dihedral angle.
u2—the values of C1a–C4a–C4b–N8 dihedral angle.
u3—the values of C4–C4b–C4a–C2 dihedral angle.
u4—the values of H1a–C1a–C4a–H4a dihedral angle.
u5—the values of H5–C5–O–C(Me3) dihedral angle.
For atoms numbering see Scheme 1.
bonds; they are slightly longer for c2 conformers. Only a single
deviation from this trend was observed in the case of the C4b–N8
bond of endo-adducts, the length values of this bond are slightly
larger for c2 conformers than those for c1 ones.
vation for TS3 and TS5, the approximate product ratio of 3.7:1 was
assigned for the formation of P3 and P5. The obtained value re-
flects the experimental products’ ratio to be equal to about 5.7:1.5
Under thermodynamic control, the formation of P3/P4 mixture
was observed with predominance of the latter one.6 The theoretical
predictions for this process also correlate very well with the exper-
imental results. The equilibrium ratio of P3 and P4, based on their
free energy values, is close to the experimental assignment (1:1.6
and 2:3, respectively). As was mentioned earlier, the higher stabil-
ity of P4 compared to P3 was unexpected. To the best of our
knowledge, the formation of syn-adducts is less favorable and in-
deed the activation energy of TS4 supported this. The adduct P4,
however, has an additional stabilizing effect which decreases its
energy. The second order perturbation theory analysis of Fock ma-
trix in NBO basis18 shows the stabilizing effect between the lone
pair of the oxygen atom of t-BuO group and the relatively acidic
All cycloadditions are exothermic processes ranging from ꢀ11.9
to ꢀ20.5 kcal/mol. The energy difference (DDH) between both con-
formers of P3 is ca. 1.4 kcal/mol. The corresponding values for P4
and P5 are ca. 0.8 and 2.1 kcal/mol, respectively. The calculations
were also performed for the hypothetical adduct P6. In the latter
case, the conformers’ energy difference is ca. 4.2 kcal/mol.
Enthalpies and free energies of the adducts revealed that their
stability increases successively in series P6, P5, P3, and P4 (Table
1). The analysis of activation enthalpies for the TSs indicated that
the exo–anti approach (TS3) is the most favorable one. The endo–
anti (TS5) and the exo–syn (TS4) approaches have the activation
barrier higher than TS3 by ca. 0.9 and 2.0 kcal/mol, respectively.
The hypothetical endo–syn approach (TS6) is strongly disfavored
because of the steric interaction.
These results are in good agreement with our experimental
studies.5,6 Under the kinetic conditions, products with the lower
activation energy should be obtained and indeed the cycloaddition
affords a mixture of P3 and P5 adducts. The former one, having a
lower activation energy, is the major product (Scheme 1). Also,
the low temperature experiments (at ꢀ18 °C) demonstrated the
lowest activation energy for TS3. Based on the free energy of acti-
proton H4a located in
a-position to the electron-withdrawing
group (nO
!
rꢁC—H, 0.6 kcal/mol, Fig. 4). Such an extra stabilization
effect could account for P4 to be thermodynamically more stable
than P3. The interaction between H4a and the oxygen atom of t-
BuO group is also established by NMR spectra. The signal of H4a
proton is located downfield for P4 in comparison to its position
for P3, as well as for P5, which confirms an interaction between
both atoms.5,6
Table 2
Relative energies: free energies
(D
G, kcal/mol), enthalpies
(DH, kcal/mol), and
entropies (
DS, cal/mol K) at 25 °C, for TSs and products of reaction between N2 and
L1a
Direct reaction
Inverse reaction
D
H
D
S
D
G
D
H
D
S
DG
TS3
TS4
TS5
TS6
P3c1
P3c2
P4c1
P4c2
P5c1
P5c2
P6c1
P6c2
17.9
19.9
18.8
23.8
ꢀ18.8
ꢀ20.2
ꢀ19.7
ꢀ20.5
ꢀ17.3
ꢀ19.4
ꢀ11.9
ꢀ16.1
ꢀ17.7
ꢀ17.3
ꢀ17.0
ꢀ22.2
ꢀ49.5
ꢀ47.5
ꢀ49.1
ꢀ47.6
ꢀ48.0
ꢀ48.8
ꢀ48.3
ꢀ52.3
23.1
25.0
23.9
30.5
ꢀ4.1
ꢀ6.1
ꢀ5.1
ꢀ6.3
ꢀ3.0
ꢀ4.9
2.5
27.1
29.6
27.4
25.8
7.6
4.3
8.3
3.6
24.8
28.0
25.0
24.8
Figure 4. Extra stabilization of P4 adduct by hydrogen bonding.
Several geometrical parameters for TS3, TS4, TS5, and TS6 are
collected in Table 3. In all cases, the C1a–O1 distance is shorter than
C4a–C4b, which is typical for the cycloadditions with the electron-
poor dipolarophiles.8,21 Moreover, for both exo transition states
the C1a–O8 distance is longer than that for endo TSs. The opposite
trend was observed for the C4a–C4b distance. For all TSs, the O1–
C1a–C4a–C4b dihedral angle value is close to zero, indicating that
these four atoms are almost coplanar and only nitrogen atom is
located out-of-plane. These results correlate very well with the
ꢀ0.5
a
All TS energies refer to the energy value of [Nꢂ ꢂ ꢂL] van der Waals’ molecular
complex and energies of all adducts are referred to sum [N+L].

2144
S. Stecko et al. / Tetrahedron: Asymmetry 19 (2008) 2140–2148
Table 3
cycloaddition should be a slow process. The lowest value of activa-
tion energy of the reverse reaction was found for P5 adduct, and
the highest one was found for P4 adduct. These results are in
agreement with our experiments of isomerization of adducts dis-
cussed above.
Selected geometric parameters for transition structures TS3, TS4, TS5, and TS6
Bond lengths (Å)
C1a–O1 C4a–C4b
Dihedral angles
O1–C1a–C4a–C4b
O1–N8–C4b–C4a
TS3
TS4
TS5
TS6
1.909
1.935
1.861
1.871
2.230
2.200
2.500
2.222
6.9
2.2
49.3
47.2
3.3. The 1,3-DC reaction of nitrone N1 and lactone L3—single
asymmetric induction system
ꢀ0.4
ꢀ43.3
7.5
46.8
For atoms numbering see Chart 2.
We focused our attention on the reaction of nitrone N1 with the
chiral lactone L3. The four possible products are depicted in
Scheme 3. The hypothetical endo–syn adduct P10 was omitted
from further discussion for simplification.
previous observation for P1 and P2,9 and show that the introduc-
tion of a t-BuO substituent to the nitrone molecule does not influ-
ence the geometry of the transition state significantly.
Previously, we have shown that adduct P3 can be partially
isomerized to adduct P4 as a result of the increase of heating dura-
tion.6 The same phenomenon was observed for P5 which after pro-
longed heating gave the equilibrium mixture of P4 and P3.6 Figure
5 shows the changes of the adducts’ ratio during isomerization of
P3, and Figure 6 presents the changes of adducts’ ratio during
the heating of P3/P5 mixture obtained under kinetic conditions.
Both experiments reveal that the isomerization process is slow
and that the equilibrium can usually be reached after 8 days with
a significant decrease of overall yield, due to the low stability of
nitrone.
O
N
O
O
O
HO
HO
H
H
H
H
H
H
O
O
N
P7
exo-anti
P8
exo-syn)
(
)
(
N1
+ L3
O
O
O
O
HO
HO
H
H
H
H
The values of the energy barrier for the retro-1,3-cycloaddition
reactions are shown in Table 2. For adducts, P3, P4 and P5, the acti-
vation energy values of the reverse process are ca. 8.6–9.7 kcal/mol
higher than for the synthesis, which indeed suggests that the retro-
H
H
O
N
O
N
P9
endo-anti
P10
endo-syn)
(
)
(
Scheme 3.
The result of conformational analysis of adducts P7, P8, and P9
is analogous to previous examples. Only the discussion of rotation
of the hydroxymethyl substituent is presented here.
For adducts P7, P8, and P9 (both type I and II) there are three
rotamers of the CH2OH group relative to the lactone moiety. These
three conformations of this substituent are depicted in Scheme 4.
The conformers A and B have similar energies, whereas the rot-
amer C is ca. 0.02–0.2 kcal/mol higher in energy than the former
ones. The C conformation, however, was found to be particularly
preferred in a solid state, resulting in a better fit within an elemen-
tary cell (see X-ray structures in Ref. 5).
O
O
O
Figure 5. Changes of adducts’ ratio during isomerization of adduct P3 ( ) to
adduct P4 ( ).
O
H
O
H
O
HO
H
H
H
OH
H
H
H
H
OH
A
B
C
Scheme 4.
Analogously to the reaction of nitrone N2 with L1, in the case of
the N1–L3 pair all possible directions of cycloaddition are exother-
mic processes in the range from ꢀ12.2 to ꢀ14.2 kcal/mol. The sta-
bility of the adducts increases in order P9, P8 and P7 (Table 4).
The calculated activation barriers reveal that the exo–anti ap-
proach is the most favorable (TS7). The endo approach (TS9) is
ca. 0.6 kcal/mol (DDHà) higher in energy. According to the compu-
tational results, the second exo-approach (syn addition to hydroxy-
methyl group of the lactone) is strongly disfavored. The activation
enthalpy for TS8 is ca. 7.8 kcal/mol higher than for the activation
barrier of TS7.
Figure 6. Changes of adducts’ ratio during heating of P3/P5 mixture obtained under
kinetic control. Legend: P3 ( ), P4 ( ), and P5 ( ).

S. Stecko et al. / Tetrahedron: Asymmetry 19 (2008) 2140–2148
2145
Table 4
3.4. The 1,3-DC reaction of nitrone N2 and lactone L3—double
asymmetric induction system
Relative energies: free energies (DG, kcal/mol), enthalpies (DH, kcal/mol), and
entropies (D
S, cal/mol K) at 25 °C, for TSs and products of reaction between N1 and L3a
Direct reaction
Reverse reaction
The 1,3-DC reaction of nitrone N2 to chiral lactone L3 is shown
in Scheme 6. According to our previous studies, the three adducts
P11, P12 and P13 are formed in a ratio ca. 21:27:52. The formation
of the endo–syn P14 cycloadduct has never been observed, and
therefore it was omitted from further analysis. The preference for
the endo product P13 formation is attributable to the mismatched
effect of substituents at both reactants in the exo-approach.
D
H
D
S
D
G
D
H
D
S
DG
TS7
TS8
TS9
P7c1
P7c2
P8c1
P8c2
P9c1
P9c2
17.2
25.0
17.8
ꢀ12.5
ꢀ14.2
ꢀ12.6
ꢀ13.8
ꢀ11.9
ꢀ12.2
ꢀ19.9
ꢀ11.5
ꢀ18.5
ꢀ48.4
ꢀ47.6
ꢀ49.1
ꢀ48.2
ꢀ48.2
ꢀ48.2
23.1
28.5
23.3
1.9
0.0
2.0
0.6
2.4
2.2
27.8
28.8
26.2
2.0
2.2
0.7
27.2
28.1
26.0
O
O
O
O
HO
HO
H
H
H
H
H
H
a
All TS energies refer to the energy value of [Nꢂ ꢂ ꢂL] van der Waals’ molecular
O
O
O
t
-Bu
Ot-Bu
N
N
complex and energies of all adducts are referred to sum [N+L].
P11
P12
exo-anti
(
exo-syn
)
(
)
The above results show that under kinetic control, the forma-
tion of P7/P9 mixture should be expected. Moreover, such an
expectation appeared justifiable when cycloaddition of nitrone
N1 to the lactone L1 is taken into consideration.7b,9 These predic-
tions only partially agree with our experimental findings. Accord-
ing to our preliminary studies, cycloaddition of nitrone N1 to
lactone L3 gave mixture of two adducts P7 and P8 in a ratio of
about 7:3.5 In case of minor product, we were unable to unambig-
uously establish its configuration by 1H NMR spectra. Moreover,
rapid decomposition of P8 was observed, which additionally com-
plicated its analysis. Previously, we showed that CD spectroscopy
is a convenient, sensitive, and rapid technique for the stereochem-
ical assignment of products between nitrones N1, N2 and six-
membered lactones. These assignments are based on the ‘ring-chi-
rality rule’ which correlates the positive/negative sign of the n–p*
CD band (Cotton effect) with the positive/negative sign of the O–
N2
+ L3
O
O
O
O
HO
HO
H
H
H
H
H
H
O
O
O
t
-Bu
Ot-Bu
N
N
P13
endo-anti
P14
(endo-syn)
(
)
Scheme 6.
Due to the influence of the three independent factors on the
conformational properties (tricyclic fused system, t-BuO group in
the pyrrolidine ring, and hydroxymethyl group in the lactone),
the conformational analysis of the cycloaddition reaction between
the nitrone N2 and the lactone L3 is more complicated than that for
the single asymmetric induction case (see Sections 3.1–3.3). The
general trends, however, found in previous cases, are still observed.
For adducts P11 and P12, the most stable type I conformer and two
separate type II conformers were found. Analogously, as it was ob-
served that in the case of the adducts P2, P5, and P9, only the type
II conformers were found for the adduct P13. For each type of the
tricyclic fused system, several substructures were found due to the
different arrangement of both substituents. The calculations pre-
dicted the preference of the anti-periplanar arrangement of the
OH group and the lactones’ oxygen atom (conformer A, Scheme
4) and similar in energy gauche conformation B. The t-BuO group
and the H5 atom are in a syn-periplanar conformation. The more
complex character of conformational interactions makes the en-
ergy differences between particular types of conformers larger
than those found in previous cases. Although for the P11 c2–c1,
the enthalpy difference is only 1.8 kcal/mol, the corresponding val-
ues for two other products P12 and P13 are higher and amount to
5.5 and 10.5 kcal/mol (Table 5).
C(@O)–C –Cb torsional angle in lactone moiety.19 Recently, we
a
proved that the ‘ring-chirality rule’ can be applied also in case of
the cycloaddition reaction involving five-membered lactones as
well (L1 and L3).20 However, the assignment of P8’s configuration
by CD spectra failed due to deviation from generally observed
trends. One could suggest that partial isomerization of P8 into
more stable P80 may occur (Scheme 5). Such behavior, however,
was not observed for the structurally related compound P12.
O
O
O
2
HO
O
H
H
HO
1a
5a
5b
N
H
H
H
H
O
O
N
P8
exo-syn
P8'
exo-anti)
(
)
(
Scheme 5.
Table 5
Relative energies: free energies
(
D
G, kcal/mol), enthalpies
(DH, kcal/mol), and
The discrepancy between the calculations and experiments
forced us to carry out a deeper analysis of the investigated cycload-
dition reaction. Its reinvestigation revealed the presence of endo-
adduct P9. Due to its unexpected high polarity and a weak visual-
ization on the TLC plates, it was lost in our preliminary experi-
ments during the standard column chromatography. In repeated
experiments, the mixture of adducts P7/P8/P9 in a ratio ca.
78:7:15 was obtained. In a other experiment, where the silylated
lactone (protected with TBDPS) was used instead of the unpro-
tected lactone L3, the mixture of silylated cycloadducts P7 and
P9 (Si–P7, Si–P9) was obtained in the ratio 94:6, exclusively. The
silylated derivative of P8 was not observed. Hence, after the above
revision, the theoretical predictions about cycloaddition of nitrone
N1 to lactone L3 fit well with the experimental findings.
entropies (
D
S, cal/mol K) at 25 °C, for TSs and products of reaction between N2 and L3a
Direct reaction Reverse reaction
D
H
D
S
D
G
D
H
D
S
DG
TS11
TS12
TS13
P11c1
P11c2
P12c1
P12c2
P13c1
P13c2
19.1
19.1
18.7
ꢀ16.8
ꢀ17.5
ꢀ19.4
ꢀ48.5
ꢀ47.5
ꢀ48.2
ꢀ47.4
ꢀ49.8
ꢀ48.9
24.1
24.3
24.5
ꢀ4.8
ꢀ6.9
0.1
28.4
25.1
23.8
2.5
4.2
1.8
27.6
23.9
23.3
ꢀ19.3
ꢀ21.1
ꢀ14.2
ꢀ19.7
ꢀ8.6
ꢀ5.6
6.2
ꢀ19.4
ꢀ4.9
a
All TS energies refer to the energy value of [Nꢂ ꢂ ꢂL] van der Waals’ molecular
complex and energies of all adducts are referred to sum [N+L].

2146
S. Stecko et al. / Tetrahedron: Asymmetry 19 (2008) 2140–2148
Table 6
The geometry optimization of adducts revealed that the syn-ad-
Relative energies: free energies
(
D
G, kcal/mol), enthalpies
(DH, kcal/mol), and
duct P11 is ca. 1.4 kcal/mol lower in energy than the anti-adduct
P12 (Table 5), analogously to the P3/P4 pair (Table 2). On the other
hand, the P12 and P13 have an analogous stability, and the energy
difference between them is only 0.3 kcal/mol which is smaller than
that for the P3/P5 pair, which was found to be 0.8 kcal/mol.
The activation enthalpy values confirm the experimental obser-
vations that the endo-adduct is a kinetic product of the reaction
(Table 5). The activation barrier for both exo-adducts is the same,
and is ca. 0.4 kcal/mol higher than that for TS13. On the other hand,
the activation enthalpies for the reverse reaction are higher and fall
within the range of 5.1–9.3 kcal/mol (Table 5). The retro-process is
the easiest for adduct P13, and is more difficult for adduct P11.
As we pointed out,5,6 the reversibility of this cycloaddition is not
straightforward. The long heating time of the reaction mixture
leads to the slow disappearance of the endo-adduct P13. It does
not lead, however, to the most stable P11 adduct or P11/P12 mix-
ture as observed for the P3/P4 pair. In this case, the formation of
adduct P15 (Scheme 7) arises because of the thermal racemization
of the lactone via a hydroxyfuran stage, as well as of the slow
decomposition of the nitrone being observed, hence the thermody-
namic equilibrium cannot be reached.
The increase in the amount of P15 in the reaction mixture is
attributable to its very low activation enthalpy (TS15, Table 6). This
value is ca. 4.5 kcal/mol lower than that for TS13. Moreover, the
barrier for the retro-process is almost twice as high than that for
the synthesis which makes the reversibility of this process difficult.
Additionally, the lower TS15 energy value upsets the L3/ent-L3
equilibrium by consuming the ent-L3. These factors appear to con-
tribute to the experimental observation that the P15 adduct is the
major product of the reaction under thermodynamic conditions.
Generally, the cycloaddition of nitrone N2 to ent-L3 can afford
up to four cycloadducts as presented in Scheme 7. Owing to the
matching of the reactants, however, the cycloaddition proceeds
with a high diastereoselectivity and leads to the exo–anti adduct
P15 exclusively.
entropies (
DS, cal/mol K) at 25 °C, for TSs and products of reaction between N2 and
ent-L3a
Direct reaction
Reverse reaction
D
H
D
S
D
G
D
H
D
S
DG
TS15
TS16
TS17
TS18
P15c1
P15c2
P16c1
P16c2
P17c1
P17c2
P18c1
P18c2
14.2
28.2
28.1
30.0
ꢀ15.3
ꢀ20.3
ꢀ16.8
ꢀ16.4
ꢀ14.6
ꢀ14.2
ꢀ12.1
ꢀ13.7
ꢀ10.8
ꢀ12.8
ꢀ14.0
ꢀ17.2
ꢀ47.0
ꢀ47.6
ꢀ48.3
ꢀ47.9
ꢀ45.2
ꢀ47.4
ꢀ47.5
ꢀ50.6
17.5
32.0
32.2
35.1
ꢀ1.3
ꢀ6.1
ꢀ2.4
ꢀ2.2
ꢀ1.1
ꢀ0.1
2.0
27.4
29.8
26.1
27.7
2.4
3.1
26.7
28.8
26.5
25.9
ꢀ1.5
5.8
1.4
a
All TS energies refer to the energy value of [Nꢂ ꢂ ꢂL] van der Waals’ molecular
complex and energies of all adducts are referred to sum [N+L].
attractive way for the synthesis of adducts derived from the ent-L3,
predominantly due to its relative poor availability compared to the
commercially available (S)-enantiomer.
According to the computational predictions, the difference of
activation enthalpy between TS15 and TS13 is ca. 4.5 kcal/mol,
which indicates that the adduct P15 should be formed exclusively.
The experimentally obtained P15/P13 ratio (4:1) does not agree
with the predictions. Due to arguments presented above, however,
we are unable to prove that the thermodynamic equilibrium has
been reached.
4. Conclusions
The presented study demonstrates that the DFT calculations at
B3LYP/6-31+G(d) level of theory can be used for the detailed
description of the cycloaddition reaction between the five-mem-
bered cyclic nitrones and five-membered lactones. The above
methodology is applicable, not only to the simplest model systems
such as N1/L2 providing simple tricyclic fused ring systems, but
also to the analysis of cycloadditions involving substituted reac-
tants. The quantum-mechanic calculations in combination with
the molecular modeling are in the agreement with the experimen-
tally observed trends in the 1,3-dipolar cycloaddition reactions of
O
O
O
O
HO
HO
H
H
H
H
H
H
O
O
Ot
-Bu
Ot-Bu
N
N
P15
exo-anti
P16
(
)
(
exo-sy
n
)
N2 + ent-L3
nitrones N1 and N2 with unsaturated
c-lactones. Computational
predictions successfully reproduced the exo/endo selectivity as
well as the diastereofacial selectivity connected with the inductive
effect of substituents.5,6 As was pointed out for the cycloaddition of
N1 to lactone L3, the computational methods in combination with
the spectroscopy data can be a most useful analytical tool for the
study of organic reactions, and can provide help in resolving some
stereochemical controversies.
O
O
O
O
HO
HO
H
H
H
H
H
H
O
O
Ot
-Bu
Ot-Bu
N
N
P17
endo-anti
P18
n
(endo-sy )
(
)
Scheme 7.
5. Experimental
The activation enthalpy for adduct P15 (TS15) is only 14.2 kcal/
mol, and is much lower than that for other adducts, hence, it ex-
plains the high stereoselectivity of the cycloaddition (Table 6).
The computational predictions correspond very well with the
experimental findings. Moreover, they explain some other obser-
vations. For example, when the cycloaddition of N1 to L3 was car-
ried out in triethylamine in the presence of trace amounts of water,
the mixture of adducts P15 and P13 was obtained in a 4:1 ratio.
Under these reaction conditions, the base-mediated racemiza-
tion of L3 occurs faster and we hoped that the dynamic kinetic res-
olution of the lactone can be achieved. This approach would be an
Proton and carbon NMR spectra were recorded on a Bruker DRX
500 Avance Spectrometer at 500 MHz and 125 MHz, respectively,
using deuterated solvents and TMS as an internal standard. Chem-
ical shifts are reported as d values in ppm and coupling constants
are in hertz. Infrared spectra were recorded on an FT-IR-1600 Per-
kin–Elmer spectrophotometer. The optical rotations were mea-
sured with a JASCO J-2000 digital polarimeter. High resolution
mass spectra were recorded on ESI-TOF Mariner spectrometer
(Perspective Biosystem). The HPLC analysis was carried out on Hit-
achi chromatograph with L-2130 pump and L-2450 DAD detector
equipped with LiChrospherÒ Si60 analytical column.

S. Stecko et al. / Tetrahedron: Asymmetry 19 (2008) 2140–2148
2147
Thin layer chromatography (TLC) was performed on aluminum
sheets Silica Gel 60 F254 (20 ꢃ 20 ꢃ 0.2) from Merck. Column chro-
matography was carried out using Merck silica gel 230–400 mesh.
The TLC spots were visualized in UV (254 nm) and by treatment
with alcoholic solution of ninhydrine, aqueous solution of KMnO4,
or with ceric sulphate/phosphomolybdenic acid solution.
Toluene was purified and dried by applying standard tech-
niques.22 The TBDPS protected lactone L3 (Si–L3) was prepared
according to literature procedure from commercially available lac-
tone L3 (Fluka).23 Nitrone N1 was obtained from 1,4-
dibromobutane.24
(1 H, dd, J 2.8, 2.3 Hz, H-2), 3.69 (1H, m, H-4b), 3.58 (1H, dd, J
11.4, 2.8 Hz, CHHOSi), 3.32–3.28 (2H, m, J 11.4, 7.2, 2.3 Hz, H-4a,
CHHOSi), 3.22 (1H, ddd, J 14.0, 8.0, 3.3 Hz, H-7), 2.66 (1H, m, H-
70), 1.73 (1H, m, H-6), 1.45 (1H, m, H-5), 1.27–1.15 (2H, m, H-50,
H-60), 1.10 (9H, s, t-Bu); 13C NMR (125 MHz, C6D6): d 175.8,
136.0, 135.8, 133.3, 132.6, 130.3, 130.2, 128.5, 128.3, 85.3, 78.8,
70.9, 64.4, 56.04, 55.98, 30.0, 26.9, 24.4, 19.3; IR (film, CH2Cl2) m:
1769 cmꢀ1; HR MS (ESI): m/z [M+Na+], calcd for C25H31NO4NaSi
460.1914. Found 460.1892.
5.6. (1aS,2R,4aR,4bS)-2-(tert-Butyldiphenylsilyloxymethyl)-
hexahydrofuro[3,4-d]pyrrolo[1,2-b]isoxazol-4(3H)-one (Si–P9)
5.1. Isomerization of P3
Colorless oil; [
a
]
D = ꢀ37.7 (c 1.06, CH2Cl2); 1H NMR (500 MHz,
A sample of P3 (75 mg)5 in dry toluene (10 mL) was refluxed
under an argon atmosphere. The progress of isomerization was
monitored by TLC, and the P3/P4 ratio was assigned by 1H NMR
spectra in benzene-d6. After ca. 8 days, equilibrium mixture was
obtained (P4/P3 ratio 3:2). For spectral data of P3 and P4, see Refs.
5 and 6, respectively.
C6D6 with 5% CD3OD): d 7.70–7.18 (10H, 2 ꢃ Ph), 4.66 (1H, d, J
6.4 Hz, H-1a), 4.14 (1H, dd, J 2.6, 1.9 Hz, H-2), 4.02 (1H, m, H-4b),
3.73 (1H, d, J 9.5, 6.4 Hz, H-4a), 3.53 (1H, dd, J 11.6, 2.6 Hz, CHHO-
Si), 3.21 (1H, dd, J 11.6, 1.9 Hz, CHHOSi), 3.17 (1H, ddd, J 13.7, 7.5,
4.2 Hz, H-7), 2.61(1H, m, H-70), 2.05 (1H, m, H-5), 1.75 (1H, m, H-6),
1.58 (1H, m, H-50), 1.32 (1H, m, H-60), 1.05 (9H, s, t-Bu); 13C NMR
(125 MHz, C6D6 with 5% CD3OD): d 175.6, 135.9, 135.8, 132.9,
132.3, 130.4, 130.3, 128.5, 128.3, 81.7, 81.5, 68.2, 64.3, 55.7, 55.2,
5.2. Cycloaddition of nitrone N1 to lactone L3
26.8, 26.7, 24.6, 19.2; IR (film, CH2Cl2)
m
: 1771 cmꢀ1; HR MS
The lactone L3 (100 mg, 0.88 mmol) and nitrone N1 (96 mg,
1.14 mmol) were dissolved in dry toluene (10 mL) and stirred at
room temperature under an argon atmosphere. The progress of
reaction was monitored by TLC (ethyl acetate–hexane 4:1). After
disappearance of the lactone, the solvent was removed and the res-
idue was chromatographed on silica gel (ethyl acetate–hexane 2:1
with the addition of 1% triethylamine) affording adducts P7
(95 mg) and P8 (8 mg). The adduct P9 was eluted with ethyl ace-
tate (18 mg). Overall yield was 70%. For analytical data of P7 and
P8, see Ref. 5.
(ESI): m/z [M+Na+], calcd for C25H31NO4NaSi 460.1914. Found
460.1909.
Acknowledgments
Authors are most indebted to the Department de Chimie Molec-
ulaire (UJF, Grenoble), CIMENT/CECIC (Grenoble), and CERMAV
(CNRS, Grenoble) and to Mr. Grzegorz Lipner (IOC PAS, Warsaw)
for providing us with computer capabilities. We are also indebted
to Marie Curie European Program for financial support for one of us
(S.S.). Financial support for this work was provided through the
Polish Ministry of Science and Informatics (Grant # 3 T09A 025 28).
5.3. (1aS,2R,4aR,4bS)-2-Hydroxymethyl-hexahydrofuro-
[3,4-d]pyrrolo[1,2-b]isoxazol-4(3H)-one P9
References
Colorless oil; [
a]
D = ꢀ67.4 (c 0.6, CH2Cl2); 1H NMR (500 MHz,
C6D6 with 5% CD3OD): d 4.60 (1H, d, J 6.3 Hz, H-1a), 4.19 (1H, dd,
J 2.7, 2.2 Hz, H-2), 3.59 (1H, dd, J 9.6, 6.3 Hz, H-4a), 3.54–3.48
(2H, m, CHHOH, H-4b), 3.26 (1H, dd, J 12.3, 2.2 Hz, CHHOH), 3.12
(1H, ddd, J 13.6, 7.3, 4.1 Hz, H-7), 2.55 (1H, ddd, J 13.6, 8.3,
7.8 Hz, H-70), 2.08–2.00 (1H, m, H-5), 1.75–1.64 (1H, m, H-6),
1.56–1.47 (1H, m, H-50), 1.30–1.22 (1H, m, H-60); 13C NMR
(125 MHz, C6D6 with 5% CD3OD): d 176.2, 82.1, 81.8, 68.3, 62.1,
1. (a) Padwa, A. 1,3-Dipolar Cycloaddition Chemistry; Wiley: New York, 1984; (b)
Gothelf, K. V.; Jørgensen, K. A. Chem. Rev. 1998, 98, 863–909; (c) Karlsson, S.;
Högberg, H.-E. Org. Prep. Proced. Int. 2001, 33, 103–172; (d) Kobayashi, S.;
Jørgensen, K. A. Cycloaddition Reactions in Organic Synthesis; Wiley-VCH:
Weinheim, 2002; (e) Pellissier, H. Tetrahedron 2007, 63, 3235–3285; (f)
Padwa, A. Synthetic Applications of 1,3-Dipolar Cycloaddition Chemistry Toward
Heterocycles and Natural Products; Padwa, A., Pearson, W. H., Eds.; Wiley & Sons:
Hoboken, NJ, 2003; (g) Merino, P. Science of Synthesis; Padwa, A., Ed.; George
Thieme: New York, NY, 2004, Vol. 27, p 511.
55.8, 53.7, 26.6, 24.5; IR (film, CH2Cl2) m
: 3360, 1767 cmꢀ1; HR
2. Frederickson, M. Tetrahedron 1997, 53, 403–425.
MS (ESI): m/z [M+H+], calcd for C9H14NO4 200.0917. Found
200.0921.
3. (a) Socha, D.; Jurczak, M.; Chmielewski, M. Carbohydr. Res. 2001, 336, 315–318;
(b) Socha, D.; Pas´niczek, K.; Jurczak, M.; Solecka, J.; Chmielewski, M. Carbohydr.
Res. 2006, 341, 2005–2011; (c) Pas´niczek, K.; Socha, D.; Solecka, J.; Jurczak, M.;
Chmielewski, M. Can. J. Chem. 2006, 84, 534–539; (d) Panfil, I.; Solecka, J.;
Chmielewski, M. J. Carbohydr. Chem. 2006, 25, 673–684; (e) Pas´niczek, K.;
Solecka, J.; Chmielewski, M. J. Carbohydr. Chem. 2007, 26, 195–211; (f) Stecko,
S.; Jurczak, M.; Urban´ czyk-Lipkowska, Z.; Chmielewski, M. Carbohydr. Res.
2008, 343, 2215–2220.
5.4. Cycloaddition of nitrone N1 to Si–L3
The lactone Si–L3 (154 mg, 0.44 mmol) and nitrone N1 (48 mg,
0.57 mmol) were dissolved in dry toluene (10 mL) and stirred at
room temperature under an argon atmosphere. The progress of
the reaction was monitored by TLC (ethyl acetate–hexane 1:4).
After disappearance of the lactone, the solvent was removed and
the residue was chromatographed on silica gel (hexane–ethyl ace-
tate 4:1) to afford adducts Si–P7 and Si–P9 in a 94:6 ratio (HPLC,
hexane–2-propanol 97:3, 1 mL/min, tSi–P7 4.7 min, tSi–P9 10.3 min)
with an overall yield of 73%.
4. (a) Pas´niczek, K.; Socha, D.; Jurczak, M.; Frelek, J.; Suszczyn´ ska, A.; Urban´ czyk-
Lipkowska, Z.; Chmielewski, M. J. Carbohydr. Chem. 2003, 22, 613–629; (b)
´
Socha, D.; Jurczak, M.; Frelek, J.; Klimek, A.; Rabiczko, J.; Urbanczyk-Lipkowska,
Z.; Suwin´ ska, K.; Chmielewski, M.; Cardona, F.; Goti, A.; Brandi, A. Tetrahedron:
Asymmetry 2001, 12, 3163–3172; (c) Jurczak, M.; Rabiczko, J.; Socha, D.;
Chmielewski, M.; Cardona, F.; Goti, A.; Brandi, A. Tetrahedron: Asymmetry 2000,
11, 2015–2022.
5. Stecko, S.; Pas´niczek, K.; Jurczak, M.; Urban´ czyk-Lipkowska, Z.; Chmielewski,
M. Tetrahedron: Asymmetry 2006, 17, 68–79.
6. Stecko, S.; Pas´niczek, K.; Jurczak, M.; Urban´ czyk-Lipkowska, Z.; Chmielewski,
M. Tetrahedron: Asymmetry 2007, 18, 1085–1093.
7. (a) Cid, P.; de March, P.; Figueredo, M.; Font, J.; Milan, S. Tetrahedron Lett. 1992,
33, 667–670; (b) Alonso-Perarnau, D.; de March, P.; Figueredo, M.; Font, J.;
Soria, A. Tetrahedron 1993, 49, 4267–4274; (c) Cid, P.; de March, P.; Figueredo,
M.; Font, J.; Milan, S.; Soria, A.; Virgili, A. Tetrahedron 1993, 49, 3857–3870.
8. Stecko, S.; Pas´niczek, K.; Michel, C.; Milet, A.; Perez, S.; Chmielewski, M.
Tetrahedron: Asymmetry 2008. doi:10.1016/j.tetasy.2008.07.008.
9. Tufariello, J.; Tette, J. J. Org. Chem. 1975, 40, 3866–3869.
5.5. (1aS,2R,4aR,4bR)-2-(tert-Butyldiphenylsilyloxymethyl)-
hexahydrofuro[3,4-d]pyrrolo[1,2-b]isoxazol-4(3H)-one Si–P7
Colorless oil; [
C6D6): d 7.80–7.18 (10H, 2 ꢃ Ph), 4.60 (1H, d, J 7.2 Hz, H-1a), 4.43
a]
D = +37.5 (c 1.2, CH2Cl2); 1H NMR (500 MHz,

2148
S. Stecko et al. / Tetrahedron: Asymmetry 19 (2008) 2140–2148
10. Kakisawa, H.; Iwashita, T.; Kusumi, T. J. Org. Chem. 1982, 47, 230–233.
11. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe,
M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
GAUSSIAN 03, Revision B.04; Gaussian: Pittsburgh, PA, 2003.
15. (a) Fukui, K. Acc. Chem. Res. 1981, 14, 363–368; (b) Head-Gordon, M.; Pople, J. A.
J. Chem. Phys. 1988, 89, 5777–5786.
16. Goti, A.; Cicchi, S.; Cordero, F. M.; Fedi, V.; Brandi, A. Molecules 1999, 4, 1–12.
17. Crystallographic data for the structures reported in this paper have been
previously deposited with the Cambridge Crystallographic Data Center,
Cambridge, UK, as a Supplementary publications: P3–X (CCDC 282892), P4–
X (CCDC 634391), and P5–X (CCDC 634282).
18. Weinhold, F.. In Encyclopedia of Computational Chemistry; Schleyer, P., Allinger,
N. L., Clark, T., Gasteiger, J., Kollman, P. A., Schaefer, H. F., Eds.; Wiley:
Chichester, UK, 1998; Vol. 3, pp 1792–1811.
19. Legrand, M.; Bucourt, R. Bull. Soc. Chim. Fr. 1967, 2241–2242.
20. Stecko, S., Frelek, J., Chmielewski, M., in preparation.
21. (a) Carda, M.; Portoles, P.; Murga, J.; Uriel, S.; Marco, J. A.; Domingo, L. R.;
Zaragoza, R.; Röper, H. J. Org. Chem. 2000, 65, 7000–7009; (b) Merino, P.; Mates,
J. A.; Revuelta, J.; Tejero, T.; Chiacchio, U.; Romeo, G.; Iannazze, D.; Romeo, R.
Tetrahedron: Asymmetry 2002, 13, 173–190; (c) Merino, P.; Revualta, J.; Tejero,
T.; Chiacchio, U.; Rescifina, A.; Romeo, G. Tetrahedron 2003, 59, 3581–3592; (d)
Merino, P.; Tejero, T.; Chiaccio, U.; Romeo, G.; Rescifina, A. Tetrahedron 2007,
63, 1448–1458.
12. (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652; (b) Becke, A. D. Phys. Rev. A
1988, 38, 3098–3100; (c) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37,
785–789.
13. Schlegel, H. B. Geometry Optimization on Potential Energy Surface. In Modern
Electronic Structure Theory; Yarkony, D. R., Ed.; World Scientific Publishing:
Singapore, 1994.
14. Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735–
746.
22. Armarego, W.; Chai, C. Purification of Laboratory Chemicals, 5th ed.;
Butterworth-Heinemann, 2003.
23. Mann, J.; Weymouth-Wilson, A. Org. Synth. 1998, 75, 139–142.
24. Cordero, F.; Machetti, F.; de Sarlo, F.; Brandi, A. Gazz. Chim. Ital. 1997, 127, 25–
29.