Remarkable influence of mild Lewis acid catalysts on cycloadditions leading to tetrasubstituted isoxazolidines: DFT analysis augmented by X-ray and NMR studies

Article abstract of DOI:10.1007/s00706-012-0828-z

Experimental and theoretical studies were carried out to highlight the influence of mild Lewis acid catalysts on the cycloadditions leading to stereochemically defined tetrasubstituted isoxazolidines. Considerable acceleration and increased exo stereoselectivity were observed for lithium triflate and magnesium bromide catalyzed processes. DFTbased reactivity indices reflected the differences in polar character of the catalyzed and uncatalyzed pathways. The regioselectivities were predicted from interaction energy calculations. Finally, the transition states were successfully located and the extent of bond formation analyzed. The catalytic acceleration was rationalized by density functional theory (DFT) calculated activation parameters. The rate constant ratios with Eyring's transition state theory (TST) rate equation and total partition functions were also theoretically evaluated to further rationalize the observed catalytic acceleration. Springer-Verlag 2012.

Full text of DOI:10.1007/s00706-012-0828-z

Monatsh Chem (2012) 143:1687–1703
DOI 10.1007/s00706-012-0828-z
ORIGINAL PAPER
Remarkable influence of mild Lewis acid catalysts
on cycloadditions leading to tetrasubstituted isoxazolidines:
DFT analysis augmented by X-ray and NMR studies
•
Nivedita Acharjee Avijit Banerji Thierry Prange
•
´
Received: 2 January 2012 / Accepted: 15 July 2012 / Published online: 8 September 2012
Ó Springer-Verlag 2012
Abstract Experimental and theoretical studies were car-
ried out to highlight the influence of mild Lewis acid catalysts
on the cycloadditions leading to stereochemically defined
tetrasubstituted isoxazolidines. Considerable acceleration
and increased exo stereoselectivity were observed for lithium
triflate and magnesium bromide catalyzed processes. DFT-
based reactivity indices reflected the differences in polar
character of the catalyzed and uncatalyzed pathways. The
regioselectivities were predicted from interaction energy
calculations. Finally, the transition states were successfully
located and the extent of bond formation analyzed. The cat-
alytic acceleration was rationalized by density functional
theory (DFT) calculated activation parameters. The rate
constant ratios with Eyring’s transition state theory (TST)
rate equation and total partition functions were also theoret-
ically evaluated to further rationalize the observed catalytic
acceleration.
Keywords Catalysis ꢀ Density functional theory ꢀ
Interaction energy ꢀ Transition states
Introduction
Catalytic studies of 1,3-dipolar cycloadditions (1,3DC)
have gained wide significance in the last decade [1–5]. The
need for regio- and stereospecific synthesis in natural
product studies has encouraged the growth of this particular
branch of 1,3DC chemistry. The reactions of C,N-disub-
stituted aldonitrones to a,b-unsaturated ketones generate
tetrasubstituted isoxazolidines. These reactions are accel-
erated by coordination of the enone moiety to appropriate
Lewis acid catalysts [6]. Consequently, the carbon–carbon
double bond of the alkene becomes highly polarized with
enhanced b-carbon electrophilicity. This induces a dramatic
rate enhancement and sometimes leads to remarkable
changes in the regio- and stereoselectivities. In the course of
our study, it was interesting to note that the reaction rates
and selectivities of the investigated cycloadditions changed
remarkably in the presence of mild Lewis acid catalysts,
e.g., metal triflates and magnesium bromide. Triflates were
selected because they are most promising green chemistry
catalysts, recoverable, and ensure better product separation.
It was further possible to change the catalytic behavior as
the well as the routine endo stereochemical pathway and
hence promote exo selectivity by simple structural varia-
tions at the non-reacting site of the dipolarophile.
Electronic supplementary material The online version of this
article (doi:10.1007/s00706-012-0828-z) contains supplementary
material, which is available to authorized users.
N. Acharjee
Department of Chemistry, Jalpaiguri Government Engineering
College, Jalpaiguri 735102, West Bengal, India
e-mail: nivedita_acharjee@rediffmail.com
A. Banerji (&)
Centre of Advanced Studies on Natural Products Including
Organic Synthesis, Department of Chemistry, University of
Calcutta, 92, Acharya Prafulla Chandra Road,
Kolkata 700009, India
The aim of this work was to present a combined
experimental and density functional theory (DFT)-based
theoretical picture of some mild Lewis acid catalyzed
1,3DCs leading to tetrasubstituted isoxazolidines. The
present report covers the experimental recordings (reaction
conditions, total conversions, and product ratios) for the
e-mail: abnachem@rediffmail.com
´
T. Prange
Laboratoire de Cristallographie et RMN biologiques
´
(UMR 8015 CNRS), Universite Paris Descartes,
4, av. de l’Observatoire, 75006 Paris, France
123

1688
N. Acharjee et al.
cycloaddition reactions, structure elucidation of the gen-
erated adducts (particularly by NMR and X-ray studies),
analysis of the global and local (at the reactive sites)
properties [7–17] of the reactants in the presence and
absence of catalyst, interaction energy calculations from a
perturbative orbital independent theoretical model [18],
location of the stationary points for reactants, products, and
transition states, and finally the determination of the acti-
vation parameters for the catalyzed and uncatalyzed
processes in terms of DFT calculations. We then calculated
the rate constant ratios from Eyring’s transition state theory
(TST) rate equation [19] and total partition functions [20]
to underline the observed catalytic acceleration.
diastereomeric products 8 and 9 in the ratio 92:8 with a
total conversion of 74 % after 20 h as reported in our
recent communication [21]. In the present study, when we
carried out the reaction in the presence of lithium triflate,
we observed a remarkable acceleration with a total con-
version 91 % after 7 h. This can be attributed to the
preferential complexation of anisalacetophenone to
the lithium triflate catalyst. The diastereomeric excess of
the endo adduct 8 over the exo isomer 9 was almost the
same as in the case of the uncatalyzed reaction. In the
presence of magnesium bromide, the conversion was 81 %
after 3 h. However, a considerable decrease in the endo
selectivity of the process occurred. The product ratio 8:9
was noted to be 58:42. Probably, the greater bulk of the
magnesium bromide complexed unsaturated ketonic spe-
cies inhibits the endo approach. It was rather surprising to
note that the reaction was completely inhibited in the
presence of ytterbium triflate and copper triflate to the
Results and discussion
Experimental results
1
limits of 300 MHz H NMR detection.
The experimental data are collected in Table 1. The
cycloaddition reaction of C-(4-chlorophenyl)-N-phenyl-
nitrone (1) to anisalacetophenone (5) (Fig. 1) gave two
The uncatalyzed cycloaddition reaction of C-(4-nitro-
phenyl)-N-phenylnitrone (2) to anisalacetophenone (5)
afforded three cycloadducts 10, 11, and 12 in the ratio
Table 1 Experimental results
for Lewis acid catalyzed
Run no.
Reactants
Additive
Time/h
Product ratios
Conversion/%
cycloaddition reactions of C,N-
disubstituted aldonitrones 1–4
to a,b-unsaturated ketones 5–7
1
1 ? 5
1 ? 5
1 ? 5
1 ? 5
1 ? 5
2 ? 5
2 ? 5
2 ? 5
2 ? 5
2 ? 6
2 ? 6
2 ? 6
2 ? 6
3 ? 6
3 ? 6
3 ? 6
3 ? 6
1 ? 7
1 ? 7
1 ? 7
1 ? 7
4 ? 6
4 ? 6
4 ? 6
4 ? 6
4 ? 6
–
20
7
8:9 = 92:8
74
91
81
–
2
Li(OTf)
MgBr2ꢀEt2O
Cu(OTf)2
Yb(OTf)3
–
8:9 = 93:7
3
3
8:9 = 58:42
4
28
28
22
8
Not detected
5
Not detected
–
6
10:11:12 = 85:10:5
10:11:12 = 86:8:6
Not detected
79
86
–
7
Li(OTf)
Cu(OTf)2
Yb(OTf)3
–
8
28
28
22
7
9
Not detected
–
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
13:14:15 = 81:10:9
13:14:15 = 82:9:9
Not detected
79
85
–
Li(OTf)
Yb(OTf)3
Cu(OTf)2
–
28
28
21
7
Not detected
–
16:17:18 = 77:14:9
16:17:18 = 77:12:11
Not detected
80
89
–
Li(OTf)
Yb(OTf)3
Cu(OTf)2
–
28
28
18
7
Not detected
–
19:20 = 87:13
19:20 = 88:12
19:20 = 65:35
19:20 = 87:13
21:22:23 = 78:8:14
21:22:23 = 72:14:14
21:22:23 = 70:10:20
Not detected
81
94
71
57
65
71
66
–
Li(OTf)
Cu(OTf)2
Yb(OTf)3
–
3
3
17
7
Li(OTf)
MgBr2ꢀEt2O
Yb(OTf)3
Cu(OTf)2
7
28
28
Not detected
–
123

DFT analysis augmented by X-ray and NMR studies
1689
R
R2
O
1
Toluene, 110°C, N2, Lewis acid
4
R3
C
3
N
2
5
R1
H
O
1
1
R = 4-Cl; R = Ph ( )
R2 = OCH3; R3 = Ph ( )
5
R = 4-NO2; R1 = Ph (2)
R = 2-NO2; R = Ph ( )
R = 4-Cl; R1 = Me (4)
R2 = H, R3 = Ph (6)
1
3
R2 = H, R3 = 2-pyridyl ( )
7
R2
R
R
R
COR3
COR3
H
H
H
H
H
N
R
H
N
COR3
N
O
O
R1
R1
O
R1
R2
R2
R3
H
H
H
R1
Ph OCH3 Ph
R2
R3
R1
R2
R3
R
R1
R2
R
9
( )
4-Cl
Ph OCH3 Ph
(11) 4-NO2 Ph OCH3 Ph
12
15
(
(
)
)
4-NO2 Ph OCH3 Ph
(8)
10
4-Cl
4-NO2 Ph
H
H
H
Ph
Ph
Ph
(
)
4-NO2 Ph OCH3 Ph
14
(
(18) 2-NO2 Ph
(23) 4-Cl Me
)
4-NO2 Ph
(17) 2-NO2 Ph
H
H
H
H
Ph
Ph
(13) 4-NO2 Ph
16
H
H
H
H
Ph
(
)
2-NO2 Ph
Ph
2-pyridyl
Ph
(19) 4-Cl
22
Ph
2-pyridyl
Ph
(20) 4-Cl
(21) 4-Cl
Ph
(
)
4-Cl
Me
Me
Fig. 1 Lewis acid catalyzed cycloaddition reactions of C,N-disubstituted aldonitrones 1–4 to a,b-unsaturated ketones 5–7
85:10:5 with a total conversion 79 %. A 300 MHz 1H
NMR spectrum of the major endo/meta isomer 10 showed
signals at d = 5.51 ppm (d, J3,4 = 6.6 Hz), 4.44 ppm
Three products 16, 17, and 18 were obtained in the ratio
77:14:9. The ORTEP view and packing of the molecule 16
are depicted in Fig. 2. X-ray diffraction studies indicated
the all trans stereochemistry with a twist N2(O1) confor-
mation. The minor 3,4-cis diastereomeric adduct 17
showed signals at d = 5.89 (d, J3,4 = 9.3 Hz), 5.02 (t,
J = 9.3 Hz), and 5.80 ppm (d, J4,5 = 9.3 Hz) corre-
sponding to H3, H4, and H5 of the isoxazolidine ring,
respectively. The reaction systems 2/5, 2/6, and 3/6 were
completely inhibited in the presence of ytterbium triflate
and copper triflate and no product signals were detected in
(dd, J3,4 = 6.6 Hz, J4,5 = 9.0 Hz), and 5.21 ppm (d, J4,5
=
9.0 Hz) corresponding to H3, H4, and H5 of the isoxa-
zolidine ring, respectively. The 3,4-cis-4,5-trans stereo-
chemistry of cycloadduct 11 was determined by comparing
the chemical shift values and coupling constants with
those of similar cycloadducts isolated and analyzed in our
previous studies [21, 22] and also those available in
the literature [23] for reactions of benzylidene acetophe-
1
1
none. H NMR signals for H3 and H5 of 12 were shielded
the 300 MHz H NMR spectra of the crude reaction mix-
by 0.18 and 0.61 ppm compared to those of cycloadduct 10
as expected from its regioisomeric structure in accordance
with our previous studies [22]. The reaction between 2
and 5 gave 86 % yield after 7 h in the presence of lithium
triflate and was completely inhibited in the presence of
ytterbium triflate and copper triflate catalysts.
tures. The cycloaddition reactions of nitrones 1–6 were also
carried out in acetonitrile. The uncatalyzed reactions did
not occur in the presence of acetonitrile. We further
observed only 1–3 % total conversion for the catalyzed
reactions in acetonitrile and hence these were not pursued
further.
The reaction of nitrone 2 and benzylidene acetophenone
(6) afforded three cycloadducts 13, 14, and 15 in the ratio
81:10:9 (total conversion 79 %). X-ray studies indicated
the 3,4-trans-4,5-trans structure of 13 including the lone
pair of nitrogen. The compound showed a pure O1 enve-
lope (E) conformation. The ORTEP view and packing of
13 are shown in Fig. 2. We then carried out the reaction of
C-(2-nitrophenyl)-N-phenylnitrone (3) and dipolarophile 6.
The cycloaddition reaction of nitrone 1 to the benzyli-
dene derivative of 2-acetylpyridine (7) was exo stereo-
selective, in contrast to the reactions of 5 and 6, and
afforded two diastereomeric cycloadducts 19 and 20 in
the ratio 87:13 (total conversion 81 %) after 18 h. The
3,4-cis-4,5-trans stereochemistry of adduct 19 was deter-
mined from X-ray crystallographic analysis (Fig. 2).
The compound showed an intermediate envelope/twist
123

1690
N. Acharjee et al.
Fig. 2 ORTEP view and
packing of cycloadducts 13, 16,
and 19
13
16
19
conformation with the O1 atom out of the mean plane
defined by the four others.
presence of these catalysts. The uncatalyzed reaction (total
conversion 65 %) afforded three adducts 21, 22, and 23
in the ratio 78:8:14 after 17 h [21]. The reaction time
was reduced to 7 h in the presence of Li(OTf) and
MgBr2ꢀEt2O with a total conversion 71 and 66 %,
respectively. No products were detected for ytterbium and
H3, H4, and H5 of the minor endo adduct 20 appeared
at d = 5.40, 5.20, and 6.00 ppm, respectively. Compara-
tive chemical shift values of the isoxazolidine ring protons
implied a diastereomeric relationship between 19 and 20.
In the presence of lithium triflate, the total conversion was
94 % after 7 h with 76 % de of the exo adduct 19 over the
endo adduct 20. When the reaction was carried out in the
presence of ytterbium triflate, the total conversion was
57 % after 3 h with the product ratio 19:20 = 87:13.
Copper triflate resulted in considerable reduction of the
exo approach and the diastereomeric excess of exo adduct
19 over 20 was reduced to 30 % after 3 h. We then car-
ried out the reaction between C-(4-chlorophenyl)-N-
methylnitrone (4) and benzylideneacetophenone (6) in the
1
copper triflate catalyzed reactions on the H NMR spectra
of crude reaction mixtures.
Irrespective of the para C-aryl substituents with varying
electronic demand, the effective nitrone–Lewis acid com-
plexation [6] probably blocked the cycloaddition pathway
for nitrones 1 and 2 to a,b-unsaturated ketones 5 and 6.
However, dipolarophile 7 reacted with nitrone 1 in the
presence of ytterbium and copper triflates. Presumably, this
is due to preferential complexation of the Lewis acid cat-
alysts to the latter.
123

DFT analysis augmented by X-ray and NMR studies
1691
Table 2 DFT/B3LYP/6-31G(d) calculated HOMO and LUMO energies and global properties of the reactants
Optimized energy/au
HOMO/au
LUMO/au
l/au
g/au
S/au
x/eV
1
-1,091.512317
-768.558418
-0.207
-0.215
-0.230
-0.223
-0.218
-0.068
-0.065
-0.098
-0.095
-0.100
-0.139
-0.140
-0.166
-0.168
-0.169
0.257
0.266
0.244
0.246
0.251
1.946
1.880
2.049
2.033
1.992
1.023
1.002
1.536
1.561
1.548
5
5-I
5-II
1-II
-1,737.641582
-1,236.409799
-1,559.359611
Analysis based on electronic chemical potentials,
global hardness, electrophilicity, and softness matching
indices
of anisalacetophenone increases the electrophilicity of the
corresponding complex. 5-I has an electrophilicity value of
1.536 eV. 5-II has an x value of 1.561 eV. x values of 5-I
and 5-II allowed us to assert the use of the LiCl–anisalace-
tophenone complex 5-II as a reduced model of 5-I and hence
required less computational effort in the transition state
calculations. Dx (0.5 eV) for the coordinated reaction sys-
tems 1/5-I and 1/5-II indicates a greater polar character
compared to 1/5. 1-II represents LiCl-coordinated dipol-
arophile 1. The HOMOdipole–LUMOalkene energy gaps for
1/5, 1/5-II, and 1-II/5 systems are 372.8, 294.1, and
401.7 kJ mol-1 and HOMOalkene–LUMOdipole gaps corre-
spond to 385.9, 406.9, and 301.9 kJ mol-1, respectively.
The coordination of alkene to Lewis acid decreases the
LUMOalkene energy and the HOMOdipole–LUMOalkene gap.
On the other hand, coordination of the dipole lowers the
HOMOalkene–LUMOdipole energy gap compared to the
uncatalyzed reaction. Both possibilities suggest the accel-
eration of the reaction in the presence of the Lewis acid.
The condensed Fukui functions [15–17] (electrophilic
and nucleophilic) of the reactants were calculated for the
atomic sites in order to rationalize the observed regiose-
lectivity in respect of the cycloadditions. Chandra and
Nguyen [27] pointed out that the evaluation of condensed
Fukui functions requires atomic electron population in a
molecule, which is not an unambiguously defined quantity.
They further reported that electrostatic potential driven
atomic charges [28] and natural orbital-based charges [29]
are more consistent for the calculation of Fukui functions.
For the present study, we calculated the condensed Fukui
functions on the basis of natural population analysis (NPA)
[29] and Merz-Kollman (MK) [28] analysis. These are
listed in Table 3. Cb has a larger xk than Ca in both the
analyses. Therefore Cb will be the preferred position for the
nucleophilic attack by the dipole. In the case of nitrone 1,
the f-k value from NPA and MK calculations of the oxygen
atom O1 is greater than that of the carbon atom C3. This
implies that nucleophilic attack of O1 of 1 at Cb of the
dipolarophile is preferred in both 1/5 and 1/5-II reaction
systems. If we consider the 1-II/5 system, C3 of 1-II has
greater xk than O1 in both the analyses indicating nucle-
ophilic attack by Ca (fk- value of Ca is greater than of Cb) of
5 at C3. Both these predictions lead to meta regioisomers in
conformity with the experiments.
Lithium triflate proved to be the best catalyst in terms of
reaction time and total conversion of the investigated reac-
tions. The lithium catalyzed reaction pathway is cost
effective and appropriate to be investigated in terms of
B3LYP/6-31G(d) level of theory unlike the other catalytic
systems. Therefore, we have attempted to present a DFT
interpretation for the uncatalyzed and lithium catalyzed
reactions of nitrone 1 and dipolarophile 5. The global
properties are listed in Table 2. The HOMOdipole
–
LUMOalkene and HOMOalkene–LUMOdipole energy gaps for
the 1/5 system are 372.8 and 385.9 kJ mol-1, respectively.
The similarity of the HOMOdipole–LUMOalkene and
HOMOalkene–LUMOdipole gaps for the reaction indicates
that this reaction is of the intermediate type in Sustmann’s
classification (type II). Hence, the coordination of a Lewis
acid to the dipole should also accelerate the reaction (at least
from the molecular orbital theory viewpoint). We therefore
attempted to consider both cases when lithium is coordi-
nated to the dipole as well as to the dipolarophile molecule.
The electronic chemical potential l is an index pointing
to the direction of the electronic flux during the cycloaddi-
tion, i.e., the charge transfer (CT) within the system in the
ground state. Definitions of electronic chemical potential
(l), hardness (g), softness (S), electrophilicity (x), and
Fukui functions (f?k , fk-) are available in many well-known
reviews [24] and textbooks and need no further elaboration
at this stage, except for their use and applications. The
global indexes [7–12] are indicative of the polar character of
a cycloaddition reaction especially in terms of the global
electrophilicity difference Dx between the reagents. Anal-
ysis of global electrophilicities can be used to choose the
more appropriate computational model for Lewis acid cat-
alyzed cycloadditions involving large molecules [25, 26].
5-I and 5-II represent Li(OTf)- and LiCl-coordinated
dipolarophile 5, respectively. The electronic chemical
potential l of nitrone 1 (l = -0.139 au) is higher than 5, 5-
I, and 5-II, thereby unanimously indicating net charge
transfer from dipole to dipolarophile. The low Dx value
(0.021 eV) for the 1/5 reaction system indicates a low polar
character. Coordination of lithium metal to the oxygen atom
123

1692
N. Acharjee et al.
model [18]. We recently applied the model to rationalize
the regioselectivity of the cycloaddition reaction between
C,N-diphenylnitrone and methyl crotonate [30]. In the
present study, we considered two approaches to analyze the
factor k. The exact value of k is difficult to determine with
a simple model. The positive factor k may be different for
different 1,3-dipole–dipolarophile couples. Gazquez
reported [31–33] that the value of k was obtained by factors
yielding at least an order of magnitude 1. Let this be coined
as approach A. With approach A under consideration, the
more important term is DEl0 as can be anticipated from the
small difference in chemical potentials, especially for the
uncatalyzed reaction.
Table 3 DFT/B3LYP/6-31G(d) calculated local properties at the
reactive site (k)
k
fk?
s?/au
x?k /eV
fk-
s-/au
NPA
1
O1
C3
0.099
0.193
0.101
0.126
0.217
0.422
0.123
0.239
0.125
0.243
MK
1
O1
C3
0.118
0.206
0.230
0.401
0.121
0.211
0.187
0.118
0.364
0.230
NPA
5
Ca
Cb
0.028
0.126
0.053
0.237
0.028
0.126
0.126
0.008
0.237
0.015
MK
Pal and Chandrakumar [34, 35] related the parameter k to
the change in the electron densities at the interacting site
before and after the interaction process. This change will give
the effective number of valence electrons that have partici-
pated in the interaction process. This can be termed as
approach B. k values calculated from Eq. 16 by NPA for the
1/5, 1/5-II, and 1-II/5 reaction systems are 0.021, 0.015, and
0.008, respectively. Approach B leads to comparable mag-
nitudes of DEt and DEl for the catalyzed reaction channels.
The calculated interaction energy values (Tables 5, 6) sug-
gest that both the possibilities of electrophilic and
nucleophilic attack at the dipole by the dipolarophile and vice
versa need to be considered for the investigated reactions.
Let us analyze the preferred interactions between the
electrophilic and nucleophilic sites of the reactants. The
5
Ca
Cb
-0.006
-0.011
-0.006
0.156
0.004
0.293
0.008
0.165
0.310
0.165
NPA
1-II
O1
C3
0.095
0.152
0.189
0.303
0.147
0.235
0.058
0.025
0.116
0.050
MK
1-II
O1
C3
0.175
0.292
0.345
0.581
0.271
0.452
0.028
0.034
0.056
0.068
NPA
5-II
Ca
Cb
-0.017
-0.035
-0.027
0.051
0.104
0.144
0.293
0.225
-0.020
-0.041
MK
5-II
Ca
Cb
0.016
0.168
0.033
0.342
0.025
0.262
0.039
0.017
0.079
0.035
k
interaction energy ðDEintÞdipole will be dominated by the local
properties of the carbon and oxygen atoms of the dipoles.
When we consider the local viewpoint for dipolarophiles, the
l
Table 4 DFT/B3LYP/6-31G(d) calculated softness matching index
for the catalyzed and uncatalyzed reactions
interaction energy ðDEintÞdipolarophile will be dominated by the
local properties of Ca and Cb atoms. Both approach A and
approach B predict interaction between C3 (electrophilic
attack) of dipoles (1, 1-II) and Ca (nucleophilic attack) of the
dipolarophiles (5, 5-II) by NPA and MK analyses. Consid-
ering the other way round, the interaction is prevalent
between O1 (nucleophilic attack) of the dipoles and Cb
(electrophilic attack) of the dipolarophiles. Both approach A
and B imply the generation of meta regioisomers and hence
are in agreement with the experiments.
1 ? 5
1 ? 5-II
1-II ? 5
NPA
Dortho
Dmeta
MK
0.136
0.070
0.211
0.060
0.085
0.035
Dortho
Dmeta
0.147
0.061
0.122
0.039
0.331
0.197
Meta regioselectivity is also predicted from the lower
value of softness matching index [14, 15, 21] Dmeta com-
pared to Dortho for both the catalyzed and uncatalyzed
reactions (Table 4).
Nature of transition states
Nitrone 1 preferably exists in Z configuration [36]. It is also
well known that anisalacetophenone (5) exists exclusively
in the trans form and hence was used as such in the
computations. We attempted to locate the transition states
(ts) for endo and exo approaches of 5 over (Z)-1 in the
present study along the regioisomeric reaction channels.
tnm and txm indicate the endo and exo transition states for
the uncatalyzed reaction along the meta pathway leading to
products 8 (pn) and 9 (px), respectively (Fig. 3).
Analysis based on the interaction energies
between the reacting species
In the present study, we calculated the interaction energies
between reactants to finally predict the regioselectivities by
invoking a perturbative orbital independent theoretical
123

DFT analysis augmented by X-ray and NMR studies
1693
Table 5 Interaction energy values/kJ mol-1 for global–global interactions (DEint), local–global interactions between dipoles and dipolarophiles
k
l
(ðDEint
DE
Þ
dipole), and between dipolarophiles and dipoles (ðDEintÞdipolarophile
)
Global–global Local–global interactions between dipole
interactions
Local–global interactions between dipolarophile 2
and dipole 1
1 and dipolarophile 2
Nucleophilic attack
at atom
Electrophilic attack
at atom
Nucleophilic attack
at atom
Electrophilic attack
at atom
O1
C3
O1
C3
Ca
Cb
Ca
Cb
Reaction 1 ? 5
NPA
DEt
DEl0
-0.001
-0.0002
-0.0003
-0.0005
-0.0003
-0.0001
-0.0003
-0.0003
-0.0000
-343.113
-633.261
k
-619.514
-570.265
-618.347
-656.703
l
ðDEintÞdipolarophile
-601.351
-601.351
-669.429
Interaction DEint
energy
ðDEintÞdipole
-0.0002 - -0.0003 - -0.0005 - -0.0003 - -0.0001 - -0.0003 - -0.0003 - -0.0000 -
-0.001 -
343.113k
633.261k
619.514k
570.265k
618.347k
656.703k
601.351k
601.351k
669.429k
MK
DEt
DEl0
-0.001
-343.113
-0.0003
-0.0004
-575.515
-0.0004
-585.004
-0.0003
-622.156
-0.0000
-0.0004
-581.893
-0.0003
-586.311
-0.0000
-671.827
-622.156
k
-678.424
l
ðDEintÞdipolarophile
Interaction DEint
energy
ðDEintÞdipole
-0.0003 - -0.0004 - -0.0004 - -0.0003 - -0.0000 - -0.0004 - -0.0003 - -0.0000 -
-0.001 -
343.113k
622.156k
575.515k
585.004k
622.156k
678.424k
581.893k
586.311k
671.827k
Reaction 1 ? 5-II
NPA
DEt
DEl0
-1.098
-0.1943
-0.2365
-577.795
-0.3861
-534.725
-0.2399
-576.779
-0.0389
-0.2809
-586.311
-0.1087
-640.366
-0.0459
-689.108
-329.920
-589.735
k
-686.944
l
Interaction DEint
energy
ðDEintÞdipole
ðDEintÞdipolarophile
-0.1943 - -0.2365 - -0.3861 - -0.2399 - -0.0389 - -0.2809 - -0.1087 - -0.0459 -
-1.098 -
329.920k
589.735k
577.795k
534.725k
576.779k
686.944k
586.311k
640.366k
689.108k
MK
DEt
DEl0
-1.098
-329.920
-0.0586
-0.2416
-576.273
-0.2416
-576.273
-0.0171
-640.678
-0.0353
-0.3207
-573.754
-0.0841
-648.272
-0.0375
-662.670
-629.013
k
-663.340
l
Interaction DEint
energy
ðDEintÞdipole
ðDEintÞdipolarophile
-0.0586 - -0.2416 - -0.2416 - -0.0171 - -0.0353 - -0.3207 - -0.0841 - -0.0375 -
-1.098 -
329.920k
629.013k
576.273k
576.273k
640.678k
663.340k
573.754k
648.272k
662.670k
Reaction 1-II ? 5
NPA
DEt
DEl0
-1.0678
-0.1898
-0.2879
-601.351
-0.1201
-657.690
-0.0540
-680.181
-0.0570
-0.2336
-588.677
-0.2336
-588.677
-0.0165
-654.086
-339.037
-634.485
k
-641.932
l
ðDEintÞdipolarophile
Interaction DEint
energy
ðDEintÞdipole
-0.1898 - -0.2879 - -0.1201 - -0.0540 - -0.0570 - -0.2336 - -0.1087 - -0.0165 -
-1.0678 -
339.037k
634.485k
601.351k
657.690k
680.181k
641.932k
588.677k
640.366k
654.086k
MK
DEt
DEl0
-1.0678
-339.037
-0.3245
-0.4904
-533.205
-0.0598
-678.073
-0.0722
-673.896
0.0125
-0.2963
-570.265
-0.2823
-574.508
-0.0083
-656.375
-588.677
k
-662.670
l
ðDEintÞdipolarophile
Interaction DEint
energy
ðDEintÞdipole
-0.3245 - -0.4904 - -0.0598 - -0.0722 - 0.0125 -
-1.0678 -
339.037k
-0.2963 - -0.2823 - -0.0083 -
588.677k
533.205k
678.073k
673.896k
662.670k 570.265k 574.508k 656.375k
tno and txo indicate the corresponding transition states
(Fig. 4). The Wiberg and atom–atom overlap weighted
natural atomic orbitals (NAO) bond order values indicate
asynchronicity for the concerted cycloaddition pro-
along the ortho pathway. For tnm and txm, the lengths of
˚
C–O forming bonds (2.132 and 2.118 A) are shorter than
˚
those of C–C forming bonds (2.137 and 2.153 A) (Table 7)
´
cesses. Jasinski et al. [37] also reported that B3LYP/6-
123

1694
N. Acharjee et al.
Table 6 Interaction energy values/kJ mol-1 for global–global interactions (DEint; Eq. (14)), local–global interactions between dipoles and dipol-
k
l
arophiles (ðDEint
DE
Þ
dipole; Eq. (15)), and between dipolarophiles and dipoles (ðDEint
Þ
dipolarophile; Eq. (15)) (k calculated from Eq. (16) in each case)
Global–global Local–global interactions between dipole 1 and Local–global interactions between dipolarophile 2
interactions
dipolarophile 2
and dipole 1
Nucleophilic attack at Electrophilic attack at Nucleophilic attack at Electrophilic attack at
atom
O1
atom
O1
atom
Ca
atom
Ca
C3
C3
Cb
Cb
Reaction 1 ? 5
NPA
DEt
DEl
-0.001
-3.603
-0.0002
-6.649
-0.0003
-6.505
-0.0005
-5.988
-0.0003
-6.493
-0.0001
-0.0003
-6.314
-0.0003
-6.314
-0.0000
-7.029
-6.895
l
k
Interaction energy DEint
-3.604
ðDEintÞdipole
-6.649
ðDEintÞdipolarophile
-6.505
-5.988
-6.493
-6.895
-6.314
-6.314
-7.029
MK
DEt
DEl
-0.001
-3.603
-0.0003
-6.533
-0.0004
-6.043
-0.0004
-6.143
-0.0003
-6.533
-0.0000
-0.0004
-6.110
-0.0003
-6.157
-0.0000
-7.054
-7.123
l
k
Interaction energy DEint
-3.603
ðDEintÞdipole
-6.533
ðDEintÞdipolarophile
-6.043
-6.143
-6.533
-7.123
-6.110
-6.157
-7.054
Reaction 1 ? 5-II
NPA
DEt
DEl
-1.098
-2.474
-0.1943
-4.423
-0.2365
-4.333
-0.3861
-4.010
-0.2399
-4.326
-0.0389
-0.2809
-4.397
-0.1087
-4.803
-0.0459
-5.168
-5.152
l
k
Interaction energy DEint
-3.572
ðDEintÞdipole
-4.617
ðDEintÞdipolarophile
-4.570
-4.397
-4.566
-5.191
-4.678
-4.911
-5.214
MK
DEt
DEl
-1.098
-2.474
-0.0586
-4.718
-0.2416
-4.322
-0.2416
-4.322
-0.0171
-4.805
-0.0353
-0.3207
-4.303
-0.0841
-4.862
-0.0375
-4.970
-4.975
l
k
Interaction energy DEint
-3.572
ðDEintÞdipole
-4.776
ðDEintÞdipolarophile
-4.564
-4.564
-4.822
-5.010
-4.624
-4.946
-5.010
Reaction 1-II ? 5
NPA
DEt
DEl
-1.0678
-1.356
-0.1898
-2.538
-0.2879
-2.405
-0.1201
-2.630
-0.0540
-2.721
-0.0570
-0.2336
-2.355
-0.2336
-2.355
-0.0165
-2.616
-2.568
l
k
Interaction energy DEint
-2.424
ðDEintÞdipole
-2.728
ðDEintÞdipolarophile
-2.693
-2.751
-2.774
-2.624
-2.588
-2.588
-2.633
MK
DEt
DEl
-1.0678
-1.356
-0.3245
-2.355
-0.4904
-2.133
-0.0598
-2.712
-0.0722
-2.695
0.0125
-0.2963
-2.281
-0.2823
-2.298
-0.0083
-2.626
-2.651
l
k
Interaction energy DEint
-2.424
ðDEintÞdipole
-2.679
ðDEintÞdipolarophile
-2.623
-2.772
-2.768
-2.638
-2.577
-2.580
-2.626
to explain the bond formation in cycloaddition reactions of
2-nitro-1-propene to (Z)-C,N-diarylnitrones. They further
defined asymmetry indices Da as the absolute values of the
differences between IC3–C4 and IC5–O1. In the present study,
the calculated IC3–C4 and IC5–O1 (Table 7) indicate that for
tnm, txm, tno, and txo, the C3–C4 bond forms more rapidly
than the C5–O1 bond. The asymmetry indices Da follow
the order tno [ tnm [ txm [ txo for the cycloaddition
reaction. The index t was estimated by the relation given by
31G(d) calculations perfectly illustrate the structure of
transition states in cycloadditions between diarylnitrones
and 1-EWG-2-arylethenes.
´
Jasinski et al. [38] reported the application of bond
development indices IC3–C4 and IC5–O1 calculated by the
relations
IC3ꢁC4 ¼ 1 ꢁ ½frCTS3ꢁC4 ꢁ rCP3ꢁC4g=rCP3ꢁC4
IC5ꢁO1 ¼ 1 ꢁ ½frCTS5ꢁO1 ꢁ rCP5ꢁO1g=rCP5ꢁO1
ꢂ
ꢂ
123

DFT analysis augmented by X-ray and NMR studies
1695
Fig. 3 Endo/meta approach for
reaction systems 1/5, 1/5-II,
and 1-II/5
Cl
Cl
1
1
OMe
OMe
M
3
O
1
3
PhOC
H
N
2
O
H
PhOC
H
H
1
N
Ph
H
MeO
2
MeO
Ph
H
H
-endo
H
O
-endo
O
N
N
Ph
Ph
(β)
5
4
(α)
(β)
5
O
Cl
Cl
4
(α)
COPh
pn-II
pnm
C
M
Ph
5
5-II
Cl
1-II
OMe
3
O
M
1
PhOC
H
H
N
2
H
MeO
Ph
H
-endo
O
M
N
Ph
(β)
O
5
Cl
4
(α)
pn-III
C
Ph
5
˚
Table 7 Bond lengths (r/A), Wiberg bond indices (a), atom–atom overlap weighted NAO bond orders (b), charge transfer (CT), bond
development indices (I), and asymmetry indices (Da) of the located transition states
ts
O1–N1
N2–C3
C3–C4
C4–C5
C5–O1
aC3–C4
aC5–O1
bC3–C4
bC5–O1
IC3–C4
IC5–O1
Da
CT
tnm
txm
tno
1.311
1.313
1.301
1.306
1.351
1.325
1.363
1.364
1.372
1.367
1.345
1.371
2.137
2.153
2.093
2.197
2.384
2.034
1.415
1.417
1.413
1.413
1.451
1.417
2.132
2.118
2.126
2.070
1.717
2.252
0.406
0.409
0.446
0.396
0.278
0.484
0.333
0.342
0.327
0.362
0.597
0.258
0.332
0.328
0.366
0.314
0.211
0.403
0.202
0.214
0.201
0.228
0.440
0.150
0.640
0.635
0.666
0.597
0.494
0.708
0.535
0.546
0.522
0.553
0.818
0.473
0.105
0.089
0.144
0.044
0.324
0.235
0.029
0.022
0.024
txo
0.001
tnII
tnIII
0.148
-0.153
Leroy et al. [39]; t = -RqA where qA is the net charge and
the sum is taken over all the atoms of the dipolarophile.
The charge transferred t for both endo (0.029 e, 0.024 e)
and exo (0.022 e, 0.001 e) reaction channels reveal an
almost neutral reaction and an electron flow from the
nitrone to the dipolarophile. This is in agreement with the
electronic chemical potential values (Table 2). We then
focused our attention on the lithium catalyzed process. The
endo/meta pathway (leading to the major adduct) has been
studied with dipolarophile–Lewis acid (P1) and dipole–
Lewis acid (P2) coordination possibilities. tn-II and tn-III
represent transition states for P1 and P2 reaction channels.
The electron flow indicated by t values for tn-II (0.148 e)
and tn-III (-0.153 e) agree well with the electronic
chemical potentials and electrophilicities (Table 2). The
transition state tn-II showed considerable asynchronicity
compared to tn as is evident from the bond orders of the
forming C–O and C–C bonds. The forming C–O bond in
˚
tn-II has a shorter bond length (1.717 A) value than that of
˚
the forming C–C bond (2.384 A). However, the forming
C3–C4 bond is shorter than C5–O1 for the tn-III transition
state. Bond development indices indicate more rapid
C5–O1 bond formation than C3–C4 for tn-II. On the
other hand, IC3–C4 is calculated to be greater than IC5–O1 for
tn-III. It is interesting to note here that asynchronicity of
the forming C–O and C–C bonds experiences a crossover for
the uncoordinated to dipolarophile-coordinated transition
state but not for the dipole-coordinated channel. The
catalyzed transition states tn-II and tn-III have greater
asymmetry indices Da compared to tnm.
123

1696
N. Acharjee et al.
Fig. 4 DFT/B3LYP/6-
31G(d) calculated transition
states for the cycloaddition
reactions
Analysis based on activation parameters and rate
constants
activation are a necessary condition for multicentered
concerted cycloaddition reactions [40]. This is evident
for the calculated entropies of activations (-195 to
-202 J mol-1 K-1) for all four reaction channels. Analy-
sis of the relative energies presented in Table 8 indicates
that B3LYP calculations do not clearly predict the regi-
oselectivity of this 1,3DC. This inability to predict the
regioselectivity of 1,3DCs reliably has also come up in the
study of other related reactions [41]. Domingo et al. [42]
reported similar results for nitrone cycloaddition reactions
to acrylonitrile. Domingo et al. [43] further concluded that
regioselectivity predictions are highly dependent on the
computational levels but all computational levels predict
correct stereoselectivities for the preferred channel.
The activation energy of transition state tnm is
6.6 kJ mol-1 lower than that of txm (Table 8). The free
energy of activation DG# for the predicted endo/meta
channel along the uncatalyzed process is lower than the
corresponding exo attack by 8.3 kJ mol-1 (Table 9). These
calculations agree well with the experimentally determined
84 % de of the endo adduct 8 (pnm) over the exo isomer 9
(pxm). Figure 5a presents DE and DG reaction profiles for
these two reaction channels.
The endo product pnm is thermodynamically more sta-
ble than pxm by 13.7 kJ mol-1. Large entropies of
123

DFT analysis augmented by X-ray and NMR studies
1697
Table 8 B3LYP/6-31G(d) calculated energies for the reaction channels
Configuration of the
approaching reactants
Optimized product
energies/au
Transition states
ts
Optimized energies/au
Activation energy/kJ mol-1
Gas phase
Toluene
1 ? 5 (endo/meta)
1 ? 5 (exo/meta)
-1,860.084916
-1,860.079711
-1,860.086701
-1,860.088029
-2,327.924386
-2,327.923059
tnm
txm
tno
-1,860.036118
-1,860.033620
-1,860.039146
-1,860.029347
-2,327.893474
-2,327.884103
90.887
97.445
82.937
108.664
75.200
89.074
101.629
106.334
93.788
1 ? 5 (endo/ortho)
1 ? 5 (exo/ortho)
1 ? 5-II (endo/meta)
1-II ? 5 (endo/meta)
txo
118.247
96.101
tn-II
tn-III
121.823
Table 9 DFT/B3LYP/6-31G(d) calculated free energy and enthalpy
changes along the different paths
other. These values suggest faster reactions for dipolaro-
phile coordination and retardation in the case of dipole
coordination for the favored endo/meta channel.
Reacting channel DG#/
kJ mol-1
DH#/
DS#/J mol-1 K-1
kJ mol-1
Domingo et al. [42, 46–48] reported solvent effect
considerations by B3LYP/6-31G(d) single point calcula-
tions on the gas-phase structures using a self-consistent
reaction field (SCRF) [49] based on the polarizable con-
tinuum model (PCM) of Tomasi’s group [50–52] to
calculate DE values for cycloaddition reactions of diphe-
nylnitrone to acrolein [46] and maleimide [46], carbonyl
ylides to 1,2-benzoquinone [48], cyclic nitrone to methyl
propiolate and acrylonitrile [42], etc. We have also repor-
ted similar calculations for the reactions of nitrone 1 to
benzylideneacetophenone [21]. We similarly calculated DE
for the present reaction channels in toluene with the
dielectric constant 2.379 at 298.15 K. These are collected
in Table 8. tnm has a lower activation energy than txm by
4.7 kJ mol-1 in toluene. The dipole coordination increases
the activation barrier of the endo/meta channel by
20.2 kJ mol-1 in toluene. On the contrary, the energy of
activation is lowered by 5.5 kJ mol-1 by dipolarophile
coordination. It is also evident that B3LYP calculations do
not clearly predict regioselectivities for this 1,3DC even
after inclusion of solvent effects. However, regiochemist-
ries predicted by local electrophilicities, softness matching
indices, and interaction energy calculations are found to be
in conformity with the experimental findings.
Endo/meta [5/1]
Exo/meta [5/1]
151.657
159.903
93.119
99.690
84.704
110.232
76.982
91.081
-196.432
-202.058
-198.123
-195.461
-217.622
-208.781
Endo/ortho [5/1] 143.746
Exo/ortho [5/1] 168.559
Endo/meta [5-II/1] 139.199
Endo/meta [1-II/5] 155.934
The observed endo selectivity is probably the outcome
of the steric interactions between the aryl group in the C3
position and the electron-withdrawing benzoyl group in the
C4 position and the stabilizing secondary orbital interac-
tions between pz orbitals on the aryl rings in the 3 and 5
positions. This has been pointed out for the reactions of
(Z)-C,N-diphenylnitrone and (E)-2-arylnitroethenes [43].
Gandolfi et al. [44, 45] critically analyzed the importance
of secondary orbital interactions in favoring the endo mode
for dipolar cycloadditions.
We also noted the generation of pre-reaction complexes
at the initial stage on all of the investigated pathways. The
relative energies of these complexes are shown in Fig. 5.
The presence of coordinated dipolarophile 5 decreases
DG# by 12.5 kJ mol-1 along the endo/meta channel
(Table 9). However, the dipole coordination increases the
free energy of activation by 4.3 kJ mol-1. The small dif-
ferences between DE# and DH# of tnm and tn-III do not
allow the quantitative evaluation of the results. However,
DE# and DH# of tn-II are lower those that of tnm by 15.7
and 16.1 kJ mol-1, respectively. This is clearly evident
from the reaction profile Fig. 5b. In the present study, the
rate constant ratios k1/k2, k3/k1, k4/k1 (Table 10) were cal-
culated from both partition functions (Eq. 18) and free
energy of activations (Eq. 19) and are listed in Table 10.
The results were found to be in good agreement with each
Although from the molecular orbital point of view the
cycloaddition reaction appears to be Sustmann type II and
hence should be accelerated by both dipole and dipolaro-
phile coordinations, it is only the latter coordination which
accelerates the reaction by lowering the activation param-
eters and dipole coordination retards the cycloaddition
process. Probably, the limited rate increase at high tem-
perature (110 °C) is an outcome of the competitive
coordination of the dipole and dipolarophile with the
employed Lewis acid catalyst and hence does not seem to
be compatible with the typical catalytic effects.
123

1698
N. Acharjee et al.
Fig. 5 Reaction profiles for the investigated reaction paths: a DG and DE for the uncatalyzed reaction channels; b DG and DE for the
uncatalyzed (I), dipole-coordinated (II), and dipolarophile-coordinated (III) reaction channels along the endo/meta pathway
spectra were recorded on a Bruker DRX500 NMR spec-
trometer. Mass spectra are recorded with a JEOL JMS600
mass spectrometer.
Table 10 Rate constant ratios for the reaction paths
Ratios
Calculated Calculated
from Eq. (18) from Eq. (19)
Endo/meta [5/1] (k1):exo/meta
[5/1] (k2)
2.7874 9 101 2.6992 9 101
Procedure for the cycloaddition reactions
Endo/meta [1-II/5] (k3):endo/meta 1.7459 9 10-1 1.7782 9 10-1
[5/1] (k1)
Endo/meta [5-II/1] (k4):endo/meta 1.5200 9 102 1.5707 9 102
[5/1] (k1)
Each of the dipoles (0.0044 mol; 1.02 g 1, 1.06 g 2, 1.06 g
3, and 0.75 g 4 [53]) was allowed to react with 0.0066 mol
of the dipolarophiles (1.57 g 5 [54, 55], 1.37 g 6, and
1.38 g 7) in 5 cm3 refluxing dry thiophene-free toluene
under a nitrogen atmosphere. Dipolarophiles 5 and 6 were
obtained from mixed aldol condensation of benzaldehyde/
anisaldehyde and acetophenone [54, 55]. Compound 7 was
prepared from aldol condensation of benzaldehyde and
2-acetylpyridine at 5 °C. Each cycloaddition reaction used
0.00088 mol of the catalysts (0.14 g lithium triflate, 0.16 g
magnesium bromide, 0.55 g ytterbium triflate, and 0.32 g
copper triflate). Magnesium bromide was dissolved in a
minimum quantity of diethyl ether and then added to the
solution. Each reaction was monitored at regular time
Experimental
UV spectra were recorded in dry acetonitrile using a Shi-
madzu UV-3101 PC spectrophotometer. IR spectra were
recorded in KBr pellets using a Perkin Elmer RX-9 FT-IR
spectrophotometer. Melting points were recorded on an
1
electrically heated Kofler Block apparatus. 300 MHz H
NMR and 75.5 MHz 13C NMR spectra were recorded on a
1
Bruker AV-300 NMR spectrometer. 500 MHz H NMR,
1
intervals by analyzing aliquots by TLC and 300 MHz H
125.5 MHz 13C NMR, DQF-COSY, HMQC, and HMBC
NMR spectroscopy. After the run time was over, toluene
123

DFT analysis augmented by X-ray and NMR studies
1699
(C2,6(B), C2,6(C), C3,5(C), C2,6(D)), 195.80 (C=O) ppm;
MS: m/z = 480 [C29H24N2O5, M?], 239 [C16H14O2H?], 131
[C9H7O?], 105 [C7H5O?], 91 [C7H7?].
was removed under reduced pressure and the crude reac-
tion mixtures were subjected to column chromatography
over neutral alumina to separate the cycloadducts. 1H NMR
signals for compounds 12, 14, 15, and 17 were detected in
1
the 300 MHz H NMR spectra of the crude reaction mix-
(3a,4b,5a)-4-Benzoyl-3-(4-nitrophenyl)-2,5-
diphenylisoxazolidine (13, C28H22N2O4)
tures prior to column chromatography. The column was
initially eluted in each case with petroleum ether (b.p. 60–
80 °C) and then with petroleum ether/benzene mixtures
progressively increasing the proportion of benzene in the
eluent. We recently reported [21] the isolation, properties,
and spectroscopic data for cycloadducts 8, 9, 21, 22, and 23
and hence have not repeated them here.
Pale yellow crystals; yield 1.13 g (57 %); isolated from
20 % benzene in petroleum ether/benzene = 4:1 eluates;
m.p.: 132 °C; UV–Vis (acetonitrile, c = 5 9 10-5 mol
dm-3): kmax (log e) = 249 (4.02) nm; IR (KBr): m = 3,057
ꢀ
(w), 1,679 (s), 1,519 (s), 1,343 (s), 827 (m), 755 (s), 694
(s) cm-1
;
1H NMR (300 MHz, CDCl3): d = 5.61
(d, J = 6.8 Hz, H3), 4.58 (dd, J = 6.8, 9.0 Hz, H4), 5.38
(d, J = 9.0 Hz, H5), 7.07 (d, J = 7.9 Hz, H2,6(A)),
7.21–7.31 (dist. t, m, H3,5(A)), 7.05 (dist. d, H4(A)),
7.73 (d, J = 8.6 Hz, H2,6(B)), 8.25 (d, J = 8.6 Hz, H3,
5(B)), 7.46 (dist. d, H2,6(C)), 7.29–7.40 (m, H3,4,5(C),
H2,3,4,5,6(D)) ppm; 13C NMR (75.5 MHz, CDCl3):
d = 73.87 (C3), 69.39 (C4), 85.09 (C5), 150.92 (C1(A)),
114.18 (C2,6(A)), 127.32 (C3,5(A)), 122.31 (C4(A)),
149.07 (C1(B)), 124.32 (C3,5(B)), 147.54 (C4(B)),
135.76, 135.98 (C1(C), C1(D)), 133.96 (C4(C)), 126.81
(C2,6(D)), 129.13 (C4(D)), 128.56, 128.67, 128.87, 129.20
(C2,6(B), C2,6(C), C3,5(C), C3,5(D)), 195.80 (C=O) ppm;
MS: m/z = 450 [C28H22N2O4, M?], 242 [C13H10N2O3],
209 [C15H12OH?], 131 [C9H7O?], 105 [C7H5O?], 93
[C6H5O?], 77 [C6H5?], 51 [C4H3?].
(3a,4b,5a)-4-Benzoyl-5-(4-methoxyphenyl)-3-
(4-nitrophenyl)-2-phenylisoxazolidine (10, C29H24N2O5)
Bright yellow crystals; yield 1.22 g (58 %); isolated from
30 % benzene in petroleum ether/benzene = 4:1 eluates;
ꢀ
m.p.: 146 °C; IR (KBr): m = 3,077 (w), 1,670 (s), 1,520
(s), 1,347 (s), 1,251 (s), 827 (m), 752 (s), 692 (s) cm-1; 1H
NMR (300 MHz, CDCl3): d = 3.77 (s, OCH3), 5.51 (d,
J = 6.6 Hz, H3), 4.44 (dd, J = 6.6, 9.0 Hz, H4), 5.21 (d,
J = 9.0 Hz, H5), 6.97 (d, J = 7.8 Hz, H2,6(A)), 6.93
(t, J = 7.4 Hz, H4(A)), 7.62 (d, J = 8.8 Hz, H2,6(B)),
8.18 (d, J = 8.8 Hz, H3,5(B)), 7.36 (d, J = 7.0 Hz,
H2,6(C)), 7.11–7.27 (m, H3,5(A), H3,4,5(C), H2,6(D)),
6.76 (d, J = 7.8 Hz, H3,5(D)) ppm; 13C NMR (75.5 MHz,
CDCl3): d = 55.30 (OCH3), 73.83 (C3), 69.27 (C4), 85.06
(C5), 151.01 (C1(A)), 114.18, 114.07 (C2,6(A), C3,5(D)),
127.26 (C3,5(A)), 122.18 (C4(A)), 149.26 (C1(B)), 124.31
(C3,5(B)), 147.51 (C4(B)), 136.00 (C1(C)), 133.92
(C4(C)), 130.19 (C1(D)), 160.24 (C4(D)), 128.32, 128.56,
128.69, 129.19 (C2,6(B), C2,6(C), C3,5(C), C2,6(D)),
195.80 (C=O) ppm; MS: m/z = 480 [C29H24N2O5, M?],
239 [C16H14O2H?], 131 [C9H7O?], 105 [C7H5O?], 91
[C7H7?], 77 [C6H5?], 51 [C4H3?].
(3a,4b,5a)-4-Benzoyl-3-(2-nitrophenyl)-2,5-
diphenylisoxazolidine (16, C28H22N2O4)
Pale yellow crystals; yield 1.11 g (56 %); isolated from
20 % benzene in petroleum ether/benzene = 4:1 eluates;
m.p.: 174 °C; UV–Vis (acetonitrile, c = 5 9 10-5 mol
dm-3): kmax (log e) = 243 (3.98) nm; IR (KBr): m = 3,062
ꢀ
(w), 1,677 (s), 1,523 (s), 1,343 (s), 859 (w), 749 (s), 691
(s) cm-1 1H NMR (500 MHz, CDCl3): d = 6.12 (d,
;
(3a,4b,5a)-4-Benzoyl-5-(4-methoxyphenyl)-3-
(4-nitrophenyl)-2-phenylisoxazolidine (11, C29H24N2O5)
Yellow amorphous powder; yield 0.13 g (6 %); isolated from
30 % benzene in petroleum ether/benzene = 4:1 eluates after
J = 4.3 Hz, H3), 4.56 (dd, J = 4.3, 8.5 Hz, H4), 5.27 (d,
J = 8.5 Hz, H5), 7.12 (dd, J = 7.7, 1.0 Hz, H2,6(A)), 7.34
(br. t, H3,5(A)), 7.01 (tt, J = 7.7, 1.0 Hz, H4(A)), 8.01 (dd,
J = 8.0, 1.3 Hz, H3 (B)), 7.51 (td, J = 8.0, 1.3 Hz, H4(B)),
7.79 (td, J = 8.0, 1.3 Hz, H5(B)), 8.41 (dd, J = 8.0, 1.3 Hz,
H6(B)), 7.44 (dist. d, H2,6(C)), 7.19 (t, J = 7.5 Hz,
H3,5(C)), 7.41 (dist. t, J = 7.5 Hz, H4(C)), 7.24–7.34 (m,
H2,3,4,5,6(D)) ppm; 13C NMR (125.5 MHz, CDCl3):
d = 71.71 (C3), 69.99 (C4), 86.44 (C5), 150.47 (C1(A)),
114.28 (C2,6(A)), 129.68 (C3,5(A)), 122.42 (C4(A)), 138.93
(C1(B)), 147.69 (C2(B)), 125.27 (C3(B)), 129.05 (C4(B)),
134.76 (C5(B)), 130.25 (C6(B)), 136.50, 136.48 (C1(C)),
C1(D)), 129.32 (C2,6(C)), 129.12 (C3,5(C)), 133.85
(C4(C)), 127.71 (C2,6(D)), 128.84 (C3,5(D)), 129.57
(C4(D)), 196.98 (C=O) ppm; MS: m/z = 450 [C28H22
N2O4, M?], 242 [C13H10N2O3], 209 [C15H12OH?], 131
ꢀ
removing the crystals of 10; IR (KBr): m = 3,078 (w), 1,672
(s), 1,525 (s), 1,349 (s), 1,254 (s), 828 (m), 748 (s), 690
(s) cm-1; 1H NMR (300 MHz, CDCl3): d = 3.80 (s, OCH3),
5.83 (d, J = 10.2 Hz, H3), 4.60 (dd, J = 10.2, 9.6 Hz, H4),
5.15 (d, J = 9.6 Hz, H5), 6.94 (d, J = 7.9 Hz, H2,6(A)), 6.96
(m, H4(A)), 7.60 (d, J = 8.9 Hz, H2,6(B)), 8.16 (d,
J = 8.9 Hz, H3,5(B)), 7.46 (d, J = 7.0 Hz, H2,6(C)),
7.09–7.24 (m, H3,5(A), H3,4,5(C), H2,6(D)), 6.72 (d,
J = 7.9 Hz, H3,5(D)) ppm; 13C NMR (75.5 MHz, CDCl3):
d = 54.99 (OCH3), 70.34 (C3), 65.85 (C4), 81.94 (C5),
148.73 (C1(A)), 115.82 (C2,6(A)), 127.51 (C3,5(A)), 122.85
(C4(A)), 146.82 (C1(B)), 123.47 (C3,5(B)), 147.50 (C4(B)),
135.85 (C1(C)), 134.14 (C4(C)), 129.87 (C1(D)), 114.07
(C3,5(D)), 160.45 (C4(D)), 128.69, 129.15, 129.38, 129.46
123

1700
N. Acharjee et al.
[C9H7O?], 105 [C7H5O?], 93 [C6H5O?], 77 [C6H5?], 51
[C4H3?].
X-ray structure determination
Compound 13 was isolated as bundles of imbricate crys-
tals. One of the small masses was dispersed in oil (Nujol)
and crunched. It appeared that small pieces of debris were
pure monocrystals under polarized light. One of these small
plates with approximate size 10 lm was captured with a
nylon loop whilst observing under a microscope. As it was
anticipated that a standard diffractometer would not be
sensitive enough, measurements were done at the ESRF
synchrotron source in Grenoble. The crystal was mounted
at the end of the BM 14 beam line operating a QUANTUM
(3a,4b,5a)-5-Benzoyl-3-(2-nitrophenyl)-2,4-
diphenylisoxazolidine (18, C28H22N2O4)
Yellow amorphous powder; yield 0.08 g (4 %); isolated
from 30 % benzene in petroleum ether/benzene = 4:1
ꢀ
eluates; IR (KBr): m = 3,060 (w), 1,676 (s), 1,522 (s),
1,340 (s), 860 (w), 751 (s), 690 (s) cm-1 1H NMR
;
(500 MHz, CDCl3): d = 5.70 (d, J = 5.9 Hz, H3), 4.33
(t, J = 5.9 Hz, H4), 5.45 (d, J = 5.9 Hz, H5), 7.06 (d,
J = 7.7 Hz, H2,6(A)), 7.23 (br. t, H3,5(A)), 6.98 (t,
J = 7.7 Hz, H4(A)), 7.99 (dd, J = 8.0, 1.2 Hz, H3(B)),
7.43 (td, J = 8.0, 1.2 Hz, H4(B)), 7.72 (td, J = 8.0,
1.2 Hz, H5(B)), 8.34 (dd, J = 8.0, 1.2 Hz, H6(B)), 7.39
(dist. d, H2,6(C)), 7.11 (t, J = 7.5 Hz, H3,5(C)), 7.35 (dist.
t, J = 7.5 Hz, H4(C)), 7.14–7.32 (m, H2,3,4,5,6(D)) ppm;
13C NMR (125.5 MHz, CDCl3): d = 74.11 (C3), 68.05
(C4), 85.42(C5), 151.21 (C1(A)), 114.74 (C2,6(A)), 121.78
(C4(A)), 137.89 (C1(B)), 148.11 (C2(B)), 125.42 (C3(B)),
129.11 (C4(B)), 134.81 (C5(B)), 130.76 (C6(B)), 135.95,
135.98 (C1(C)), C1(D)), 128.61, 129.86, 129.34, 128.79
(C3,5(A), C2,6(C), C3,5(C), (C3,5(D)), 133.91 (C4(C)),
127.84 (C2,6(D)), 129.32 (C4(D)), 197.00 (C=O) ppm;
MS: m/z = 450 [C28H22N2O4, M?], 209 [C15H12OH?],
131 [C9H7O?], 105 [C7H5O?], 93 [C6H5O?], 77 [C6H5?].
˚
ADSC detector, tuned to a wavelength of 0.9615 A. Dif-
fraction data were processed using the HKL suite of
programs. The structure was refined with the SHELXL [56]
program, using isotropic then anisotropic thermal param-
eters. The hydrogen atoms were localised on Fourier
difference maps and introduced with an isotropic thermal
factor equal to that of the bonded atom. Compounds 16 and
19 were dispersed in paraffin oil and selected crystals were
mounted in cryo-loops and stored at 100 K in liquid
nitrogen. Recording parameters on beam line ID 29 were
adjusted to adapt with the atomic resolution diffracting
crystals. The structures were refined using the same pro-
tocol: isotropic atoms first, then location of hydrogens on
difference-Fourier maps, then refinements with anisotropic
thermal factors for non-hydrogen atoms.
(3a,4b,5a)-3-(4-Chlorophenyl)-4-(2-pyridinylcarbonyl)-
2,5-diphenylisoxazolidine (19, C27H21ClN2O2)
13: System: triclinic (space group P1, Z = 2); cell
˚
parameters: a = 9.390(1) A, b = 11.540(1) A, c = 11.701
˚
˚
White crystals; yield 1.20 g (62 %); isolated from 10 %
benzene in petroleum ether eluates; m.p.: 66 °C; UV–Vis
(acetonitrile, c = 5 9 10-5 mol dm-3): kmax (log e) =
(1) A, a = 100.74(4)°, b = 100.66(4)°, c = 108.53(4)°,
3
V = 1139.7(2) A .
˚
16: System: monoclinic (space group P21/n, Z = 4);
˚
cell parameters: a = 10.304(1) A, b = 17.286(2) A, c =
˚
ꢀ
245 (4.00) nm; IR (KBr): m = 3,051 (w), 1,684 (s), 1,569
(s), 1,484 (s), 1,348 (s), 1,245 (s), 834 (m), 760 (s), 690
˚
3
˚
13.027(1) A, a = 90°, b = 97.74(5)°; c = 90°, V =
2299.2(3) A .
(s) cm-1
;
1H NMR (300 MHz, CDCl3): d = 5.33 (d,
J = 5.9 Hz, H3), 5.27 (dd, J = 5.9, 7.6 Hz, H4), 5.56 (d,
J = 7.6 Hz, H5), 7.05 (dt, J = 7.8, 1.0 Hz, H2,6(A)),
7.24–7.28 (m, H3,5(A)), 6.97 (tt, J = 7.3, 1.0 Hz, H4(A)),
7.27–7.32 (m, H3,5(B)), 7.50 (d, J = 8.4 Hz, H2,6(B)),
7.97 (dt, J = 7.7, 1.0 Hz, H3(C)), 7.76 (td, J = 7.7,
1.7 Hz, H4(C)), 7.37 (ddd, J = 7.7, 4.7, 1.2 Hz, H5(C)),
8.36 (ddd, J = 4.7, 1.7, 0.9 Hz, H6(C)), 7.43 (dist. d,
J = 7.7 Hz, H2,6(D)), 7.24–7.31 (m, H3,4,5(D)) ppm; 13C
NMR (125.5 MHz, CDCl3): d = 73.47 (C3), 67.04 (C4),
83.94 (C5), 151.48 (C1(A)), 114.88 (C2,6(A)), 122.36
(C4(A)), 140.73 (C1(B)), 133.62 (C4(B)), 152.66 (C2(C)),
123.16 (C3(C)), 137.33 (C4(C)), 127.86 (C5(C)), 149.14
(C6(C)), 137.33 (C1(D)), 127.51 (C2,6(D)), 128.80 (C4
(D)), 128.79, 128.83, 129.23, 129.41 (C3,5(A), C2,6(B),
C3,5(B), C3,5(D)), 198.22 (C=O) ppm; MS: m/z = 440
[C27H21ClN2O2, M?], 231 [C13H10ClNO], 210 [C14H11
NOH?], 139, 131 [C9H7O?], 111 [C6H4Cl?], 105
[C7H5O?], 91 [C7H7?], 77 [C6H5?], 51 [C4H3?].
19 System: monoclinic (space group P21/c, Z = 4); cell
˚
parameters: a = 5.980(1) A, b = 11.750(1) A, c = 30.301
˚
3
˚
˚
(2) A, a = 95.14(4)°, V = 2120.5(4) A .
X-ray data of 13, 16, and 19 have been deposited at the
Cambridge Structural Data Centre under CCDC numbers
765672, 787238, and 787239, respectively. The data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/
data_request/cif or by e-mailing data_request@ccdc.
cam.ac.uk, or by contacting The Cambridge Crystallo-
graphic Data Centre, 12, Union Road, Cambridge CB2
1EZ, UK; fax: ?44-1223-336033.
Computational methods
The electronic chemical potential l [7] and chemical
hardness g [8] can be evaluated from the ionization
potential (I) [57] and electron affinity (A) [57] as
123

DFT analysis augmented by X-ray and NMR studies
1701
k
The interaction energy ðDEintÞA can be rewritten in
l ꢃ ꢁðI þ AÞ=2
g ꢃ I ꢁ A
The global electrophilicity index x [9–12] can be simply
expressed by
ð1Þ
ð2Þ
shorthand notation as
k
ðDEintÞA ꢃ DEt þ k DEl0 with DEl0 ¼ DElðk ¼ 1Þ:
ð13Þ
x ¼ l2=2g
ð3Þ
ð4Þ
The parameter k has been related to the deviation of
total softness of the interacting system AB from the sum of
the softnesses of individual systems A and B. It has been
defined somewhat arbitrarily in the literature [31–33]. In
the present study, following the work of Mendez et al. [18]
we initially assumed here that k is close to 1 and DEl is
much more important than DEt.
Pal and Chandrakumar [34, 35] related the parameter k
to the change in the electron densities at the interacting site
before and after the interaction process and calculated the
global and local interaction energies as given by Eqs. (16)
and (17).
The global softness S is given by [11]
S ¼ 1=2g
Fukui functions [13, 14] (for nucleophilic [f?k ] and
electrophilic [f-k ] attack) of an atom in a molecule can be
written in terms of the respective electron populations of
the cationic, neutral, and anionic systems.
For nucleophilic attack
fkþ ¼ qkðN þ 1Þ ꢁ qkðNÞ
For electrophilic attack
ð5Þ
ð6Þ
ꢀ
ꢁ
ꢀ
ꢁ
2
ꢁðlA ꢁ lBÞ
SASB
1
4
k
fkꢁ ¼ qkðNÞ ꢁ qkðN ꢁ 1Þ
The local electrophilicity index xk [12–15] is given by
xk = xf?k . The local softnesses [s?, s-] are calculated as
the product of global softness S and the respective Fukui
functions.
DEint
ꢃ
ꢁ
ð14Þ
2
SA þ SB
SA þ SB
m
l
ꢀ
ꢁ
2
ꢁðlA ꢁ lBÞ
SAfAxSB
ðDEintÞAx
ꢃ
2
SAfAx þ SB
m
ꢀ
ꢁ
ð15Þ
1
4
k
ꢁ
SAfAx þ SB
sþk ¼ Sfkþ; sꢁk ¼ Sfk
ð7Þ
l
kl
ij
This change will give the effective number of valence
electrons that have participated in the interaction process.
Gazquez also pointed out that the factor k is proportional to
an effective number of valence electrons [31–33] taking
part in the reaction. An expression for the term k can be
written as the difference of electron densities of the system
A before and after the interaction [34, 35]:
The softness matching index D is given by
2
2
Dikjl ¼ ðsiꢁ ꢁ sþk Þ þ ðsjꢁ ꢁ slþÞ
ð8Þ
where atoms i and j of the nucleophile interact with atoms k
and l of the electrophile, respectively, to give rise to the
preferred regioisomer. The reaction pathway involving
kl
ij
lower value of D will be the favored one.
p
p
X
X
When a stable molecule A (formed by bonding of K
atoms, NA being the total number of electrons in A) and a
stable molecule B (formed by bonding of L atoms, NB
being the total number of electrons in B) interact with each
other, the interaction energy [18] from DFT is given by
eq
Ai
kA ¼
q
ꢁ
q0Ai
ð16Þ
i¼1
i¼1
Alternately, the term k can be defined as the difference
of electron densities for system B
q
q
X
X
DEint ¼ DEt þ DEl
where DEt and DEl are the energy changes at constant
external potential and constant chemical potential,
respectively, and are written as
eq
Bj
kB ¼
q
ꢁ
q0Bj
ð17Þ
j¼1
j¼1
where the first terms of the right-hand side of the Eqs. (16)
and (17) refer to the sum of the electron densities of each
atom in A and B in the molecule AB at equilibrium,
respectively, and the second terms in Eqs. (16) and (17)
refer to electron densities of each atom in the isolated
systems A and B, respectively. The indices p and q are the
number of atoms of the systems A and B, respectively.
With this approach, k appears to be much less than 1 and
DEl and DEt may be quite comparable.
2
DEt ꢃ ꢁ1=2[f SASB½lA ꢁ lBꢂ g/f SA þ SBg]
DEl ꢃ ꢁ1=2½k=fSA þ SBgꢂ
ð9Þ
ð10Þ
where l, g, and S are derived from Eqs. (1), (2), and (6),
respectively. When we invoke the local viewpoint of one
reactant, DEt, and DEl in terms of condensed Fukui
function fk become
2
The geometries were optimized by DFT with Becke’s
[58] three-parameter hybrid exchange functional in com-
bination with the gradient-corrected correlation functional
DEt ꢃ ꢁ1=2 [f SASBfk½lA ꢁ lBꢂ g=fSAfk þ SBgꢂ
DEl ꢃ ꢁ1=2½k=ffkSA þ SBgꢂ
ð11Þ
ð12Þ
123

1702
N. Acharjee et al.
2. Huang ZZ, Kang YB, Zhou J, Ye MC, Tang T (2004) Org Lett
6:1677
3. Suga H, Nakajima T, Itoh K, Kakehi A (2005) Org Lett 7:1431
4. Chow SS, Nevalainen M, Evans CA, Johannes CW (2007) Tet-
rahedron Lett 48:277
5. Padwa A, Pearson WH (2002) Synthetic applications of 1,3-
dipolar cycloaddition chemistry towards heterocycles and natural
products. Wiley, New York
6. Gothelf KV (2002) In: Kobayashi S, Jorgensen KA (eds) Cyclo-
addition reactions in organic chemistry. Wiley, Weinheim, p 211
7. Parr RG, Yang W (1989) In: Density functional theory of atoms
and molecules. Oxford University Press, New York
8. Parr RG, Pearson RG (1983) J Am Chem Soc 105:7512
9. Parr RG, Szentpaly LV, Liu S (1999) J Am Chem Soc 121:1922
of Lee et al. [59] (B3LYP) using the 6-31G(d) basis set.
The transition states were localized at DFT/B3LYP/6-
31G(d) level of theory. The stationary points were char-
acterized through vibrational frequency analysis done at
298.15 K at the DFT/B3LYP/6-31G(d) level. All the sta-
tionary points are definitely identified for minima (number
of imaginary frequencies = 0) or transition states (number
of imaginary frequencies = 1). In accordance with our
previous studies [60], intrinsic reaction coordinate (IRC)
[61, 62] calculations were performed to verify that the
energy curve connecting the optimized reactants and the
products passes through the correct and the lowest TS
which must be a first-order saddle point. The electron
affinity and ionization potential were obtained at similar
level using UB3LYP theory for the anion and cation. The
geometries of the neutral species are used to calculate the
electronic structure of the charged species in order to fulfill
the demand for constant external potential. The electronic
populations were computed from NPA [29] and also by the
charges derived from the electrostatic potential according
to MK algorithm [28].
´
10. Perez P, Domingo LR, Aurell MJ, Contreras R (2003) Tetrahe-
dron 59:3117
11. Benchouk W, Mekelleche SM (2008) J Mol Struct (Theochem)
852:46
12. Chattaraj PK, Sarkar U, Roy DR (2006) Chem Rev 106:2065
13. Yang W, Mortier WJ (1986) J Am Chem Soc 108:5708
14. Chandra AK, Nguyen MT (1997) J Chem Soc, Perkin Trans
2:1415
15. Zhang YL, Yang ZZ (2000) J Mol Struct (Theochem) 496:139
´
16. Aurell MJ, Domingo LR, Perez P, Contreras R (2004) Tetrahe-
dron 60:11503
17. Merino P, Revuelta J, Tejero T, Chiacchio U, Rescifina U,
Romeo G (2003) Tetrahedron 60:3581
Apart from activation barrier calculations, it would be
very desirable to theoretically evaluate the rate constant
ratios for elementary bimolecular reactions in the gas
phase. According to the theory of reaction rates [20], the
rate constant for reaction ½A þ B ꢀ AB# ! productsꢂ
between A and B is given by
´
18. Mendez F, Tamariz J, Geerlings P (1998) J Phys Chem A
102:6292
19. McQuarrie DA, Simon JD (1998) Physical chemistry, 1st South
Asian edn. Viva Books, New Delhi
20. Laidler KJ (2009) Chemical kinetics, 3rd edn. Pearson Education,
Singapore
´
21. Acharjee N, Banerji A, Prange T (2010) Monatsh Chem 141:1213
22. Banerji A, Maiti KK, Haldar S, Mukhopadhyay C, Banerji J,
kBT Q#
a
´
Prange T, Neumann A (2000) Monatsh Chem 131:901
k ¼
eꢁE =RT
ð18Þ
h QAQB
23. Sridharan V, Muthusubramanium S, Sivasubramanian S, Polborn
K (2004) Tetrahedron 60:8881
24. Kohn W, Sham L (1965) Phys Rev 140A:1133
where Q values represent the total partition constants and
Ea the activation energy. Rate constants can also be
calculated with the simple Eyring equation [18], derived
from transition state theory:
´
25. Domingo LR, Chamorro E, Perez P (2008) J Org Chem 73:4615
26. Domingo LR, Asensio A, Arroyo P (2002) J Phys Org Chem
15:660
27. Chandra AK, Nguyen MT (2002) Int J Mol Sci 3:310
28. Besler BH, Merz KM Jr, Kollman PA (1990) J Comput Chem
11:431
kBT
#=RT
k ¼
eꢁDG
ð19Þ
h
29. Reed AE, Weinhold F (1983) J Chem Phys 78:4066
30. Acharjee N, Banerji A (2011) Comput Theor Chem 967:50
where DG# represents the free energy of activation at
298 K.
´
31. Parr RG, Gazquez JL (1993) J Phys Chem 97:3939
32. Gazquez JL, Martinez A, Mendez F (1993) J Phys Chem 97:4059
33. Gazquez JL (2003) In: Sen KD (ed) Chemical hardness: structure
and bonding, vol 80. Springer, New York, p 27
34. Pal S, Chandrakumar KRS (2000) J Am Chem Soc 122:4145
35. Pal S, Chandrakumar KRS (2002) J Phys Chem A 106:5737
36. Koyano K, Suzuki H (1969) Bull Soc Chem Jpn 42:3306
All calculations were carried out using the Gaussian
2003 [63] set of programs along with the graphical inter-
face Gauss View 2003.
Acknowledgments Nivedita Acharjee is grateful to the University
Grants Commission, India (project no F.PSW-111/11-12 (ERO)) for
financial support to carry out the experimental and theoretical cata-
lytic investigations. The kind cooperation of Prof. Manas Banerjee of
Burdwan University, West Bengal, India is greatly acknowledged.
´
´
37. Baranski A, Lyubimtsev A, Jasinski R, Kwiatkowska M (2008)
Polish J Chem 82:1037
´
´
38. Jasinski R, Wa˛sika K, Mikulskaa M, Baranski A (2009) J Phys
Org Chem 22:717
39. Leroy G, Sana M, Burke LA, Nguyen MT (1980) Quantum
Theory Chem React 1:91
40. Husigen R (1976) J Org Chem 41:403
41. Cossio P, Morao I, Jiao H, Schleyer PVR (1999) J Am Chem Soc
121:6737
References
´
42. Carda M, Portoles R, Murga J, Uriel S, Marco JA, Domingo LR,
´
¨
1. Gothelf AS, Gothelf KV, Hazell RG, Jørgensen KA (2002)
Angew Chem Int Ed 41:4236
Zaragoza RJ, Roper H (2000) J Org Chem 65:7000
43. Baranski A, Jasinski R (2001) Chem Heterocycl Comp 37:1518
123

DFT analysis augmented by X-ray and NMR studies
1703
´
60. Acharjee N, Das TK, Banerji A, Banerjee M, Prange T (2010) J
Phys Org Chem 23:1187
44. Burdisso M, Gandolfi R, Gru¨nanger P (1990) J Org Chem
55:3429
45. Baldwin JE, Reddy PV (1989) J Org Chem 54:5264
46. Domingo LR, Benchouk W, Mekelleche SM (2007) Tetrahedron
63:4464
61. Gonzalez C, Schlegel HB (1989) J Phys Chem 90:2154
62. Gonzalez C, Schlegel HB (1990) J Phys Chem 94:5523
63. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Montgomery JA Jr, Vreven T, Kudin KN, Burant
JC, Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B,
Cossi M, Scalmani G, Rega N, Petersson GA, Nakatsuji H, Hada
M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nak-
ajima T, Honda Y, Kitao O, Nakai H, Klene M, Li X, Knox JE,
Hratchian HP, Cross JB, Adamo C, Jaramillo J, Gomperts R,
Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C,
Ochterski JW, Ayala PY, Morokuma K, Voth GA, Salvador P,
Dannenberg JJ, Zakrzewski VG, Dapprich S, Daniels AD, Strain
MC, Farkas O, Malick DK, Rabuck AD, Raghavachari K,
Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clifford S, Cio-
slowski J, Stefanov BB, Liu G, LiashenkoA, Piskorz P,
Komaromi I, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng
CY, Nanayakkara A, Challacombe M, Gill PMW, Johnson B,
Chen W, Wong MW, Gonzalez C, Pople JA (2003) Gaussian 03,
revision B.03. Gaussian, Pittsburgh, PA
´ ´
47. Domingo LR, Aurell MJ, Arno M, Saez JA (2007) J Mol Struct
(Theochem) 811:125
48. Domingo LR, Aurell MJ, Perez P, Saez JA (2012) RSC Adv
2:1334
´
´
49. Tomasi J, Persico M (1994) Chem Rev 94:2027
50. Cances E, Mennucci B, Tomasi J (1997) J Chem Phys 107:3032
51. Cossi M, Barone V, Cammi R, Tomasi J (1996) Chem Phys Lett
255:327
52. Barone V, Cossi M, Tomasi J (1998) Comput Chem 19:404
53. Banerji A, Biswas PK, Sengupta P, Dasgupta S, Gupta M (2004)
Indian J Chem 43B:880
54. Kohler EP, Chadwell HM (1922) Org Synth 2:1
55. Kohler EP, Chadwell HM (1941) Org Synth Coll 1:78
56. Sheldrick GM, Schneider TR (1997) Methods Enzymol 277:319
´
57. Domingo LR, Saez JA, Perez P (2007) Chem Phys Lett 438:341
58. Becke AD (1993) J Chem Phys 98:5648
´
59. Lee C, Yang W, Parr RG (1988) Phys Rev B 37:785
123