Pyridine-catalyzed stereoselective addition of acyclic 1,2-diones to acetylenic esters: Synthetic and theoretical studies of an unprecedented rearrangement

Article abstract of DOI:10.1002/chem.200800250

A systematic study of the addition of various 1,2-acyclic diones to activated acetylenic esters catalyzed by pyridine under mild conditions is described. This reaction provides a new protocol for the stereoselective synthesis of 1,2-diaroyl maleates. The exclusive formation of the cis isomer is especially noteworthy. This reaction occurs through the initial generation of a pyridine-dimethyl acetylene dicarboxylate zwitterion and its addition to the dione followed by an unprecedented benzoyl migration. Pyridine and substituted pyridines, such as 4-dimethylaminopyridine (DMAP) and 3-methoxypyridine, are the best catalysts and anhydrous 1,2-dimethoxyethane is found to be the solvent of choice. Structural, electronic, energetic and mechanistic details of the reaction are also revealed by density functional theory calculations, which strongly support the exclusive formation of the cis isomer of the 1,2-diaroyl maleates.

Full text of DOI:10.1002/chem.200800250

FULL PAPER
DOI: 10.1002/chem.200800250
Pyridine-Catalyzed Stereoselective Addition of Acyclic 1,2-Diones to
Acetylenic Esters: Synthetic and Theoretical Studies of an Unprecedented
Rearrangement
Abhilash N. Pillai,[a] Cherumuttathu H. Suresh,*[b] and Vijay Nair*[a]
Dedicated to Professor Gilbert Stork
Abstract: A systematic study of the ad-
dition of various 1,2-acyclic diones to
activated acetylenic esters catalyzed by
pyridine under mild conditions is de-
scribed. This reaction provides a new
protocol for the stereoselective synthe-
sis of 1,2-diaroyl maleates. The exclu-
sive formation of the cis isomer is espe-
cially noteworthy. This reaction occurs
through the initial generation of a pyri-
dine–dimethyl acetylene dicarboxylate
zwitterion and its addition to the dione
followed by an unprecedented benzoyl
migration. Pyridine and substituted
pyridines, such as 4-dimethylaminopyri-
dine (DMAP) and 3-methoxypyridine,
are the best catalysts and anhydrous
1,2-dimethoxyethane is found to be the
solvent of choice. Structural, electronic,
energetic and mechanistic details of the
reaction are also revealed by density
functional theory calculations, which
strongly support the exclusive forma-
tion of the cis isomer of the 1,2-diaroyl
maleates.
Keywords: diaroyl maleates · ben-
zoyl
migration
·
density
functional calculations
·
pyridine
catalysis · reaction mechanisms
Introduction
carbon–carbon bond formation. In the context of our recent
interest in multicomponent reactions involving zwitterions[2]
formed from isoquinoline and dimethyl acetylenedicarboxy-
late (DMAD), we began exploring the reactions of the zwit-
terion formed from pyridine and DMAD; this species has a
long history going back to the work of Diels and Alder[1a] in
1932, but its structure was established only three decades
later by Acheson,[3] Winterfeldt[4] and Huisgen[5] independ-
ently.
It was found that the zwitterion 1 reacts readily with aro-
matic aldehydes (Scheme 1) to afford aroyl fumarates by a
process tantamount to the trans addition of the aldehyde
across the p system of the acetylenic ester.[6] Similar results
were obtained in the reaction of the zwitterion 1, formed in
situ from pyridine and DMAD, with N-tosylimines[7] and
electron deficient styrenes[8] (Scheme 2).
Carbon–carbon bond-forming reactions play a crucial role in
organic synthesis. With the notable exception of pericyclic
reactions, most carbon–carbon bond-forming processes in-
volve the use of reactive intermediates, such as carbocations,
carbanions, radicals and carbenes. It was realized that a class
of potentially useful reactive intermediates, zwitterions,
which are readily formed by the addition of nucleophiles to
activated p-systems, although known for a long time,[1] have
not found much use in organic synthesis, especially in
[a] Dr. A. N. Pillai, Prof. Dr. V. Nair
Organic Chemistry Section
National Institute for Interdisciplinary Science
and Technology (CSIR), Trivandrum-695019 (India)
Fax : (+91)471-249-1712
E-mail: vijaynair_2001@yahoo.com
[b] Dr. C. H. Suresh
Computational Modelling and Simulation Section
National Institute for Interdisciplinary Science
and Technology (CSIR), Trivandrum-695019 (India)
Fax : (+91)471-249-1712
E-mail: sureshch@gmail.com
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Scheme 1. Pyridine–DMAD zwitterion.
Chem. Eur. J. 2008, 14, 5851 – 5860
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5851

Table 1. Effect of solvents.
Solvent
Conditions
Yield [%][a]
ꢀ108C!RT, 12 h
reflux, 12 h,
62 (92)
63 (88)
44 (66)
48 (69)
36 (61)
36 (61)
33 (53)
35 (54)
39 (61)
42 (65)
24 (43)
25 (42)
DME
ꢀ108C!RT, 12 h
THF
reflux, 12 h
Scheme 2. Reaction of the pyridine–DMAD zwitterion with aldehydes,
N-tosylimines and dicyanostyrenes.
ꢀ108C!RT, 12 h
reflux, 12 h
CH2Cl2
ꢀ108C!RT, 12 h
diethyl ether
toluene
reflux, 12 h
ꢀ108C!RT, 12 h
reflux, 12 h
This unprecedented reaction was impressive for the facili-
ty with which it occurred under very mild conditions
(Scheme 2). The realization that the process involved the
ꢀ108C!RT, 12 h
cyclohexane
reflux, 12 h
ꢀ
formal scission of the robust C H bond of aryl aldehydes
[a] Yields based on recovered starting material are given in parentheses.
(BE ꢁ97 kcalmolꢀ1; BE=Bond energy) led us to surmise
that a 1,2-dicarbonyl compound, such as benzil, with a rela-
tively weak 1,2-C C bond (BE ꢁ70 kcalmol ) was likely to
ꢀ1
ꢀ
DME as the standard (Scheme 3) and the results are pre-
sented in Table 2. Interestingly, pyridine, 4-dimethylamino-
pyridine (DMAP) and 3-methoxypyridine afforded the best
results.
engage in a reaction with the pyridine–DMAD zwitterion.
In the event, a facile reaction leading to the formation of
1,2-dibenzoyl maleate occurred and the preliminary results
have already been published.[9] Herein we report the details
of work that involved the screening of several nucleophiles
and solvents. Detailed computational studies aimed at gain-
ing insight into the mechanistic pathway for the reaction are
also included.
Table 2. Screening of catalysts.
Catalyst
pyridine
DMAP
Conditions
Yield [%][a]
ꢀ108C!rt, 12 h
reflux, 12 h
62 (92)
63 (88)
65 (87)
67 (92)
ꢀ108C!rt, 12 h
reflux, 12 h
ꢀ108C!rt, 12 h
reflux, 12 h
63 (90)
68 (92)
33 (51)
33 (51)
< 10
< 10
0
0
0
0
0
0
0
0
0
0
Results and Discussion
3-methoxypyridine
2-picoline
ꢀ108C!rt, 12 h
Synthetic studies: Our studies began with the addition of a
catalytic amount (20 mol%) of pyridine to a solution of
benzil 6 and DMAD (3) in anhydrous 1,2-dimethoxyethane
(DME) at ꢀ108C (Scheme 3). The reaction mixture was al-
reflux, 12 h
ꢀ108C!rt, 12 h
reflux, 12 h
2,6-lutidine
ꢀ108C!rt, 12 h
triphenylphosphine
triethyl amine
DABCO
reflux, 12 h
ꢀ108C!rt, 12 h
reflux, 12 h
ꢀ108C!rt, 12 h
reflux, 12 h
ꢀ108C!rt, 12 h
reflux, 12 h
DBU
ꢀ108C!rt, 12 h
potassium tert-butoxide
reflux, 12 h
Scheme 3. Pyridine-catalyzed addition of DMAD to benzil.
[a] Yields based on recovered starting materials are given in parentheses.
lowed to warm to room temperature (ꢁ308C) and was then
stirred for 8 h. The usual processing of the reaction mixture
followed by column chromatography afforded dibenzoyl
maleate 7 in 62% yield along with the unreacted benzil
(33%).
Taking the reaction shown in Scheme 3 as the standard,
several solvents were used instead of DME to examine
whether there is any change in the efficiency of the reaction
profile or the yield of product. The results of this study are
summarized in Table 1 and clearly indicate that the reaction
works well with polar solvents, which is presumably due to
the stabilization of the zwitterion in polar solvents relative
to non-polar solvents. Evidently, anhydrous DME is the sol-
vent of choice.
A dialkyl dione, 2,3-butanedione (8), also participated in
the reaction with the pyridine–DMAD zwitterion under the
usual conditions and product 9 was formed, albeit in low
yields (Scheme 4). This may be due to side reactions that
occur by deprotonation of the activated methyl groups of
dione 8.
A variety of nucleophiles were screened in place of pyri-
dine, with the reaction of benzil and DMAD in anhydrous
Scheme 4. Addition of DMAD to 2,3-butanedione catalyzed by pyridine.
5852
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5851 – 5860

Addition of Acyclic 1,2-Diones to Acetylenic Esters
FULL PAPER
To see if the size of the substituent in the ester group has
any impact on the reactivity pattern, a reaction with di-tert-
butyl acetylenedicarboxylate as the activated alkyne was
carried out. The reaction proceeded smoothly in the same
way as in the earlier cases to afford product 11 in 48% yield
(Scheme 5), thus suggesting that the bulkiness of the ester
group has no major effect on the yield of the product.
Figure 1. a) The active “cis” and b) the inactive “trans” form of 1 with
the lone pair at the carbanion-like carbon atom. c) Steric effects in the
2,6-lutidine-derived zwitterion and d) the expected steric effect in the
trans form of the zwitterion derived from di-tert-butyl acetylenedicarbox-
ylate as the activated alkyne.
Scheme 5. Addition of di-tert-butyl acetylenedicarboxylate to benzil cata-
lyzed by pyridine.
Electronic and steric effects: It should be noted that in
Scheme 1 we have adopted a cis structure[10] (with respect to
the ester moieties) for the zwitterionic intermediate 1. Con-
ceivably, the interaction of 1 with the electron-deficient
dione is enhanced when the lone-pair centre at the carban-
ion-like carbon atom of 1 (Figure 1a) is more electron rich.
The higher catalytic activity observed for DMAP and 3-me-
thoxypyridine (Table 2) can be used as evidence in support
of this electronic effect. Al-
the formation of the cis form of the zwitterion, the data in
Table 2 do not fully eliminate the possibility of the forma-
tion of the trans form.
Mechanism: Two probable mechanistic pathways for the
formal insertion of DMAD into benzil are depicted in
Scheme 6.[9] In path 1, the zwitterion adds to one of the car-
though the formation of a trans
structure is also possible for
this system (Figure 1b), on the
basis of the following evidence,
the existence of a cis structure
can be confirmed as the active
form. The first evidence came
from the data obtained during
the screening of catalysts
(Table 2). As we can see, the
catalytic efficiency of 2-picoline
and 2,6-lutidine is less pro-
nounced than that of pyridine,
DMAP and 3-methoxypyridine.
This suggests that steric interac-
Scheme 6. Proposed mechanistic rationalization for the reaction. a) Path 1, b) path 2; Py=pyridine
tions as presented in Figure 1c
may be operative in ortho-sub-
stituted pyridines, which results
in the shielding of the lone pair at the carbanion-like carbon
atom, thus reducing its interaction with the dione. The cis
form of the zwitterion is also supportive of the smooth reac-
tion of di-tert-butyl acetylenedicarboxylate given in
Scheme 5 because in such a configuration, the tBu group is
unable to exert a steric shielding effect on the lone-pair
region of the active carbon atom. On the other hand, attri-
bution of a trans structure to the zwitterion would have pre-
dicated it to lower reactivity towards di-tert-butyl acetylene-
dicarboxylate due to the presence of a steric shielding effect
from the tBu group in the lone-pair region of the active
carbon (Figure 1d). Although the steric effect is in favor of
bonyls of benzil to form an alkoxide intermediate A. The
latter can transform into an epoxy derivative B, which then
unravels to the pyridinium ylide C. It is conceivable that in-
tramolecular benzoyl transfer followed by elimination of
pyridine can afford the final product. In path 2, a 1,2-migra-
tion of the benzoyl anion in A to form a pyridinium ylide E
and subsequent generation of a cyclopropane intermediate
F is conceived. Elimination of pyridine from F furnishes the
product.
To obtain support for the postulated reaction mechanism,
we resorted to theoretical calculations by using DFT. DFT
methods have been accepted as efficient and reasonably ac-
Chem. Eur. J. 2008, 14, 5851 – 5860
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5853

V. Nair, C. H. Suresh, and A. N. Pillai
ꢀ
curate methods and they have become a powerful tool in
transition-state modelling, particularly for ground-state reac-
tions.[11,12] We have selected the reaction of 2,3-butanedione
with DMAD and pyridine (Scheme 4) for the theoretical
study. Details of these calculations are discussed in the fol-
lowing section.
trans arrangement. The Csp2 Csp2 length of 1.550 Š in the
dione is unusually long for a bond between sp2-hybridized
carbon atoms, which suggests its electron deficiency due to
the localization of carbonyl groups. In pyridine, the nitrogen
lone pair is the HOMO, which indicates its nucleophilic
character. Attractive electrostatic interactions arise when
the pyridine nitrogen atom approaches the electron deficient
ꢀ
Theoretical calculations: All the molecular geometries were
optimized at the DFT level by using Beckeꢁs three-parame-
ter exchange functional in conjunction with the Lee–Yang–
Parr correlation functional (B3LYP)[13] as implemented in
the Gaussian 03 suite of programs.[14] For H, C, N and O
atoms, the 6-31G(d) basis functions were selected for solving
the Schrçdinger equation.[15] Normal coordinate analysis has
been performed to characterise all the stationary points. The
energy-minimum structures showed positive eigenvalues of
the Hessian matrix, whereas transition states (TSs) showed
one negative eigen value. All the transition states were com-
puted by using the Transit-Guided Quasi-Newton
(STQN;QST3) method.[16,17] For most TSs, the analysis in-
cluding the visualization of the negative frequency was suffi-
cient to specify the corresponding reaction path. In the case
of TS2, intrinsic reaction coordinate (IRC) calculations near
the TS region followed by geometry optimisation of both re-
actants and products were performed to confirm its connec-
tivity to reactant and product.[18] Zero-point energy (ZPE),
entropy corrections, thermal correction to enthalpy, and
thermal correction to Gibbs free energy were derived by
using unscaled frequencies and the usual approximations of
statistical thermodynamics (ideal gas, harmonic oscillator
and rigid rotor) at a temperature of 298.15 K and a pressure
of 1.00 atm. Furthermore, the solvation free-energy values
(Esolvation) along the reaction coordinate were evaluated from
the self-consistent reaction field (SCRF) procedure.[19] All of
the solvation calculations were performed at the B3LYP/6-
Csp2 Csp2 bond of the dione, which in turn leads to the for-
mation of a weak pyridine–dione complex (12). The enthal-
py change DH (enthalpy of productꢀenthalpy of reactant)
accompanying this associative reaction is found to be
ꢀ3.2 kcalmolꢀ1, whereas the free-energy change DG at the
B3LYP/6-31G(d) level has a positive value of 3.2 kcalmolꢀ1
due to the negative entropy factor. Furthermore, activation
ꢀ
of the Csp2 Csp2 bond of the dione in 12 is expected to lead
to the formation of the pyridine–dione covalent adduct con-
taining a N C bond. However, all attempts to optimise the
adduct structure resulted in the formation of the weak com-
ꢀ
plex 12 (Figure 2). If such an adduct is formed, the negative
Figure 2. Pyridine–dione and pyridine–diester complexes. Bond lengths
are given in angstroms [Š].
311++G(2d,2p) level by using the geometries optimised at
A
the B3LYP/6-31G(d) level in the gas phase. The selected
SCRF procedure is the polarisable continuum model (PCM)
by using Klamtꢁs form of the conductor reaction field
(COSMO).[19–22] The standard state of a solution is taken as
1 molLꢀ1 at 298.15 K, and an additional correction to the
free-energy terms is needed to convert the standard state of
an ideal gas (P0=1 atm) to the standard state of the solu-
tion (c0=1 molLꢀ1) and this was obtained by adding the
term RTln(RT) to each free energy term (the value of this
correction is 1.8943 kcalmolꢀ1 at 298.15 K). The relative
Gibbs free-energy values in the standard state of 1 molLꢀ1
suitable for solution reactions have been used throughout
the text to describe the energetics. For zwitterionic systems,
the electron-localization features were explored by analyz-
ing their molecular electrostatic potential (MESP) at the
B3LYP/6-31G(d) level.[23–25]
charge will be localized on the dione oxygen due to its con-
nectivity to the newly forming sp3 carbon atom, a situation
quite unfavourable for thermodynamic stability. Therefore,
the pyridine–dione interaction is irrelevant from a mechanis-
tic point of view.
Similar to the pyridine–dione complex, pyridine–diester
complex 13 is formed when the pyridine nitrogen atom is at-
tracted towards the ester group of the diester (Figure 2).
This complex formation is exothermic by 1.6 kcalmolꢀ1,
whereas the negative entropy factor gives an endergonic
value of 4.4 kcalmolꢀ1 at the B3LYP/6-31G(d) level.
Zwitterionic complexes: From 13, a zwitterionic complex 14
(Figure 3a) is formed when the pyridine nitrogen interacts
covalently with the acetylenic triple bond. This can even be
considered as the Michael-type addition of the nitrogen
atom of pyridine to one of the two conjugated positions of
DMAD. Complex 14 is a Huisgen zwitterion-type species
and similar species have been proposed as intermediates in
polar cycloaddition reactions.[12a] In complex 14, the ester
groups are in a cis arrangement with respect to the C=C
Electronic features of pyridine, diester, and dione and the
formation of pyridine–dione and pyridine–diester adducts:
In the diester, the ester groups are in an orthogonal arrange-
ment whereas in the dione, the carbonyl groups prefer a
5854
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5851 – 5860

Addition of Acyclic 1,2-Diones to Acetylenic Esters
FULL PAPER
Figure 3. a) Pyridine–diester zwitterionic adduct 14. Bond lengths are
given in angstroms [Š]. b) MESP pasted on the van der Waals surface of
14. The color coding from blue to red is ꢀ10.0 to +60.0 kcalmolꢀ1
.
double bond (a trans arrangement of the ester groups is less
stable than the cis arrangement by 2.0 kcalmolꢀ1). The
MESP picture of 14 (Figure 3b) gives a clear idea of its zwit-
terionic character, wherein the pyridine ring is positively
charged (red region) and the negatively charged carbon
atom (blue region) is evident on the DMAD moiety. A
MESP minimum value of ꢀ61.9 kcalmolꢀ1 is observed for
this carbon atom, which is nearly equal to the MESP mini-
mum value calculated for pyridine (ꢀ62.4 kcalmolꢀ1); this
clearly suggests that the DMAD moiety possesses the char-
acter of a stabilised carbanion. Therefore, a high nucleophil-
ic reactivity in the s-plane is expected for the negatively
charged carbon atom (blue region in Figure 3b), whereas
the p-delocalization of the positive charge through the ring
atoms and the connected ester moiety (red region mainly
centred on the pyridine ring and ester moiety) contributes
towards its stability.
It can be noted that the formation of 14 from free pyri-
dine and diester is endothermic by 10.4 kcalmolꢀ1. Because
it is an association reaction, the entropy factor is negative
and, therefore, a high value for the free-energy change
(22.3 kcalmolꢀ1) is observed, which suggests that the ther-
modynamic factors are not very favourable for the reaction
in the gas phase and the observed formation of complex 14
in the experiment can be explained by invoking solvation ef-
fects and kinetic factors. Therefore, 14 is expected to be a
highly reactive short-lived zwitterionic intermediate. The ex-
periments conducted in the present work and the previous
reports on similar reactions are in favour of the above-men-
tioned observation.
The nucleophilic attack of Huisgen zwitterion 14 by its
carbanion on the electron deficient carbonyl carbon atoms
of the dione is considered as the next step of the reaction
and this would lead to the formation of zwitterionic complex
15 (Figure 4). Energetically, complex 15 is quite stable as its
formation from 14 and dione is exothermic by 7.9 kcalmolꢀ1,
whereas the free energy of the reaction shows a decrease by
1.5 kcalmolꢀ1. It can be noted that with respect to the elec-
tronic energy, enthalpy and free energy of (pyridine+
dione+diester), the corresponding relative values of 15 are
ꢀ0.5, 2.5 and 23.8 kcalmolꢀ1, respectively. The free-energy
change observed for 15 in the gas phase is quite large as it is
built from three small molecules. By starting from complex
15, two mechanistic possibilities will be discussed.
Figure 4. 1,4-dipolar cycloaddition. Bond lengths are given in angstroms
[Š].
Mechanism 1: In the reactant complex 15, a further ap-
proach of the dione to the negatively charged carbon atom
ꢀ
in the zwitterion is expected to produce a C C bond be-
tween the two moieties. However, a transition state (TS) lo-
cated for this process produced the structure TS1 (Figure 4),
ꢀ
ꢀ
which showed simultaneous formation of a C C and C O
bond. The structure of TS1 is quite similar to a TS typically
observed in a cycloaddition reaction and the present case
may be better described as resulting from a 1,4-dipolar cy-
cloaddition. The free energy of activation (DG#) for this
step is only 8.3 kcalmolꢀ1 and the free energy of product 16
was found to be 18.2 kcalmolꢀ1 lower than 15. It can be
noted that in 15, the pyridine moiety carries a positive
ꢀ
charge and when the C C bond starts to develop in TS1 at
2.138 Š (see the structure of TS1), the negative charge will
be transferred to the carbonyl group of the dione moiety
ꢀ
and, correspondingly, the activation of this C O bond
(1.260 Š in TS1) will occur. Hence, in TS1 there is a favour-
able electrostatic interaction between the negatively charged
dione oxygen atom and the positively charged pyridine
moiety. This may be the reason for the low activation barrier
for this reaction. Similar electrostatic stabilisation was re-
ported recently for a zwitterionic transition state corre-
sponding to the formation of a 1,3-dipolar cycloadduct of bis-
(phenylazo)stilbene.[24a] Compared to TS1, the charge is well
balanced in 16 and it is a normal neutral molecule, in which
the nitrogen shows a slight amount of pyramidalization. It
can be noted that the loss of aromaticity in the pyridine ring
of 16 is well compensated by the formation of two new
bonds, which accounts for the high exothermicity of this
step of the reaction. We have also determined the structure
of transition-state TS1 in the solvent phase (solvent=THF)
by using the PCM method and it is found that gas phase
structure is nearly preserved in solution (see the Supporting
Chem. Eur. J. 2008, 14, 5851 – 5860
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5855

V. Nair, C. H. Suresh, and A. N. Pillai
Information). In other words, the formation of a zwitterionic
complex that would be more stable than the neutral bicyclic
compound 16 is not observed.[12e]
A structure similar to intermediate 16 was isolated in an
earlier work, in which the pyridine–DMAD zwitterion was
intercepted with N-substituted isatins,[7] which supports the
mechanism discussed so far. Further rearrangement of the
intermediate 16 is anticipated and the possibility of the
atom is exposed to more electron density because the three
bonds on this carbon atom are more like single bonds.
Therefore, the C···C distance of 3.152 Š noted in the struc-
ture of 17 is expected to be the next hot region for a chemi-
cal change. This is indeed true because at the C···C distance
of 2.138 Š, TS3 is located, which yields complex 18
(Figure 6). Compared to 17, complex 18 is further stabilized
ꢀ
breaking of the newly formed C O bond is assessed by lo-
cating TS2 (Figure 5). It can be seen that in TS2, the oxygen
ꢀ
ꢀ
Figure 6. C O bond breaking leading to COCH3 migration. Bond
lengths are given in angstroms [Š].
by 3.0 kcalmolꢀ1 in free energy. The 17 to 18 conversion is
ꢀ
Figure 5. C O bond breaking leading to a zwitterionic intermediate.
Bond lengths are given in angstroms [Š].
ꢀ
the result of a C O bond breaking that leads to the migra-
tion of the CH3C=O group. The DG# of this step is only
4.4 kcalmolꢀ1. The dissociation of pyridine from 18 is ex-
pected to be spontaneous as it releases 16.4 kcalmolꢀ1 of
energy, which leads to the formation of the desired cis-1,2-
dialkoyl enoate 19.
atom disconnected from the pyridine ring interacts with the
C C bond of the dione moiety at a distance of 2.253 Š. In
this TS, the original dione C C bond is further activated to
a distance of 1.568 Š. The product from TS2 is expected to
be an epoxide. However, the geometry optimisation pro-
duced the structure 17, in which the original dione C C
bond is found to be completely cleaved. The 16 to 17 con-
version required
ꢀ
ꢀ
The reaction profile, which shows the changes of enthalpy,
free energy and Esolvation is depicted in Figure 7.[26] The over-
all reaction is highly exothermic and exergonic and the rate-
determining step corresponds to the conversion of the bicy-
ꢀ
a
DG# of
26.9 kcalmolꢀ1 and this process
is exergonic by 9.6 kcalmolꢀ1.
It can be noted that in 17, the
pyridine ring carries a positive
charge, which means that it is
zwitterionic in nature. Further,
17 is 9.6 kcalmolꢀ1 more stable
than 16. The energetic stabilisa-
tion is interesting because nor-
mally
a
charge separated
system is expected to be less
stable than its neutral isomer.
In 17, we can see two ester
ꢀ
groups and one
OCOCH3
group, and each of these groups
is connected to a different sp2
carbon atom. This means that a
major share of the negative
charge of the zwitterion can de-
localise over all of these groups
via the sp2 carbon atoms and
the connected oxygen atoms.
From the bond-length features
of 17, it is felt that the carbon
&
*
Figure 7. Relative Gibbs free-energy values in the gas phase ( ) and PCM level relative Esolvation values ( ) ob-
~
atom connected to the nitrogen tained for mechanism 1. The relative enthalpy profile in the gas phase is also given ( ).
5856
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5851 – 5860

Addition of Acyclic 1,2-Diones to Acetylenic Esters
FULL PAPER
clic intermediate 16 to the open system 17. Solvation plays
an important role as it gives more stability to transition
states relative to other minimum-energy structures. The sol-
vation effect is the largest for the transition-state TS2 locat-
ed for the rate-determining step and this in turn reduces the
activation barrier to a feasible value of 16.1 kcalmolꢀ1.
Mechanism 2: We have also considered the possibility of the
migration of the COCH3 group in 16 to the nearby sp2
ꢀ
carbon atom. TS4 was located for this process, which yields
the 1,2-migration product 20 (Figure 8). The DG# of this
step is 27.2 kcalmolꢀ1, which is only 0.3 kcalmolꢀ1 higher in
energy than the 16 to 17 conversion discussed in mecha-
nism 1. Furthermore, the 16 to 20 conversion is endergonic
by 3.0 kcalmolꢀ1, whereas the 16 to 17 conversion is exer-
ꢀ1
ꢀ
gonic by 9.6 kcalmol . From 20, migration of the COCH3
ꢀ
Figure 8. Mechanism 2 showing the 1,2-migrations of COCH3. Bond
lengths are given in angstroms [Š].
group to the sp2 carbon connected to the nitrogen atom is
also feasible as a TS located for such a reaction (TS5) re-
quired only a DG# value of 16.8 kcalmolꢀ1 (Figure 8). The
product obtained from TS5 is the same compound 18 al-
ready located in mechanism 1. The reaction profile depicted
in Figure 9 also suggests that even in the presence of sol-
vent, this reaction required the highest activation energy of
24.3 kcalmolꢀ1 (conversion of 16 to 20 through TS4).
It can be noted that TS2 of mechanism 1 yields a more
stable intermediate (17) than that which results from TS4 of
mechanism 2 (20). The conversion of 17 to the final product
19 is more facile than the conversion of 20 to 19 because the
former required an activation energy of only 2.4 kcalmolꢀ1
compared with 11.3 kcalmolꢀ1 for the latter (solvent phase
values). The solvation effect is clearly in favour of mecha-
nism 1 as the rate determining
existence of complex 14 in the system. Although the gas-
phase relative free energy of 22.1 kcalmolꢀ1 observed for
this system is rather high, the relative Esolvation value of
6.5 kcalmolꢀ1 supports the formation of this species in the
reaction mixture. The high solvation effect is observed be-
cause of the zwitterionic character of the molecule. To gain
further support for this argument, we have analyzed the ki-
netic stability of 14 by simulated annealing techniques that
involve semiempirical Born–Oppenheimer molecular dy-
namics (BOMD) calculations at the PM3 level.[27] The simu-
lations were done for 1155 fs at a simulated temperature of
273 K. The default criteria used in the Gaussian 03 pro-
step, which shows an activation
barrier of 16.1 kcalmolꢀ1 that is
8.2 kcalmolꢀ1 smaller than the
corresponding value obtained
for
mechanism 2
(solvent
phase). Furthermore, if we
follow mechanism 2, the disso-
ciation of 16 to the starting ma-
terials is found to be easier
than the formation of inter-
mediate 20. On the basis of the
above results, it can be conclud-
ed that mechanism 1 is pre-
ferred over the competitive
mechanism 2 in both the gas
and solvent phases.
Molecular dynamics calcula-
tions: Although our experi-
ments and related studies in the
literature[10] strongly support
the formation of the zwitterion
14 as a transient species, the
credibility of the reported theo-
&
*
Figure 9. Relative Gibbs free-energy values in the gas phase ( ) and PCM level relative Esolvation values ( ) ob-
~
retical results depends on the tained for mechanism 2. The relative enthalpy profile in the gas phase is also given ( ).
Chem. Eur. J. 2008, 14, 5851 – 5860
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5857

V. Nair, C. H. Suresh, and A. N. Pillai
gram[14] were used for setting up the initial conditions. At
about 1100 fs, the complex dissociated to pyridine and die-
ster units (Figure 10). Interestingly, the simulation also
between the zwitterion, generated from pyridine and
DMAD, and the dione is located by calculations (Figure 4).
The bicyclic adduct 16 is a locked system and the breaking
ꢀ
of the new C O bond can only
happen in such a way that it
will lead to the cis product.
This is validated as the product
formed is exclusively the cis
isomer. Similarly, in place of
epoxy derivative B, a transition-
state TS2 is found in which the
ꢀ
ꢀ
C C and C O bonds are parti-
ally broken; this TS then pro-
ceeds to 17 (C in Scheme 6) in
ꢀ
which the original C C dione
bond is completely cleaved.
Then, the acyl-group migration
takes place by the breaking of
ꢀ
the C O bond (18 or D) fol-
lowed by elimination of pyri-
dine to give the products, which
are similar in both the suggest-
ed postulates and calculations.
In the second postulated
mechanism, the key intermedi-
ate in the calculation is also the
adduct 16. From 16, a 1,2-acyl
migration occurs which gives
the intermediate 20 (E). Instead
of the cyclopropane system F in
ꢀ
Figure 10. BOMD simulation showing the histogram of the C N bond lengths. A proton is seen to migrate
ꢀ
from the pyridine ring to the negatively charged carbon atom after 70 fs. C N bond lengths indicated are
a) 3.370, b) 1.455 and c) 1.479 Š.
showed the migration of a proton from the pyridine ring to
the negatively charged carbon atom (see the model shown
in the middle of Figure 10). The product from the proton
migration retained its structure for around 70 fs. The proton
migration was also studied at the B3LYP/6-31G(d) level,
which confirmed that the migration can take place with an
activation energy of 14.5 kcalmolꢀ1 and the product of the
reaction was 4.8 kcalmolꢀ1 less stable than 14.
The BOMD calculations are rather computationally ex-
pensive and more sophisticated ab initio or DFT simulations
were unaffordable. The results of the PM3 level BOMD
simulation support at least a 1 ps lifetime for the zwitterion
system 14. The proton migration in 14 indicates the high re-
activity of its negatively charged carbon atom. The proton
migration may be advantageous for the reaction because it
would increase the concentration of the active species in the
reaction mixture by preventing the quick dissociation of the
the proposed mechanism (Scheme 6), the transition state
TS5 is located by calculations in which simultaneous break-
ꢀ
ꢀ
ing of one C C bond and formation of another C C bond is
observed. The rest are the same in both the postulate and
the calculation. But the theoretical results predict that
mechanism 1 is preferred.
ꢀ
It can be noted that with respect to the central C C bond
of 18, the ester groups have a cis arrangement and, there-
fore, the dissociation of pyridine from 18 would give the cis
product 19. However, considering the single-bond character
ꢀ
of the central C C bond, there exists the possibility of the
rotation of this bond leading to the formation of a trans-
structure. The B3LYP/6-31G(d) level calculations showed a
value of 10.0 kcalmolꢀ1 for the rotation of the C C bond
ꢀ
for the formation of the trans isomer of 18, which was
0.7 kcalmolꢀ1 less stable than the cis form (see the Support-
ing Information). Thus the thermodynamics are in favour of
the formation of cis product 19.
ꢀ
C N bond.
On comparing the mechanistic postulates proposed with
those from theoretical calculations, the following conclu-
sions can be made. The experimental results and theoretical
calculations support the proposed formation of the zwitter-
ion 14 between pyridine and DMAD as the initial step. The
BOMD simulation strongly suggests that 14 is a highly reac-
tive species with a short lifetime in the order of pico sec-
onds. Instead of the zwitterion A as proposed in the mecha-
nistic postulate (Scheme 6), the formation of an adduct 16
Conclusion
We have uncovered a novel reaction, in which acyclic 1,2-
diones add to an activated acetylene providing a unique syn-
thetic protocol for the stereoselective synthesis of 1,2-diaro-
yl maleates. The screening of various catalysts showed that
ꢀ
pyridine and substituted pyridines act as mediators for C C
5858
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5851 – 5860

Addition of Acyclic 1,2-Diones to Acetylenic Esters
FULL PAPER
DMAD (3) (198 mg, 1.3953 mmol) at ꢀ108C. The reaction mixture was
allowed to attain room temperature and was then stirred for 12 h. Re-
moval of the solvent followed by column chromatography as described in
the general procedure afforded product 9 as a viscous yellow liquid
(43 mg, 16%). 1H NMR: d=3.86 (s, 6H), 2.24 ppm (s, 6H); 13C NMR:
d=189.5, 165.1, 134.6, 44.0, 28.5 ppm; IR (KBr): n˜ =2967, 1743, 1682,
bond formation in these reactions. It was also shown that
the reaction works well with pyridine, 4-dimethylaminopyri-
dine (DMAP) and 3-methoxypyridine. A detailed study
with various solvents has shown that the reaction works well
in anhydrous 1,2-dimethoxyethane. It is noteworthy that the
reaction occurs by means of an unprecedented benzoyl/acyl
migration. The structural, electronic, energetic and mecha-
nistic details of the reaction are also unravelled by theoreti-
cal calculations, which strongly supported the exclusive for-
mation of the cis isomer, relative to 1,2-diaroyl maleates.
The negatively charged carbon atom of the transient species,
cis-zwitterion 14, was identified as the reactive centre that
upon coordination to the dione gives the fairly stable inter-
mediate product 16. This means that the transformation of
the transient species 14 to the stable intermediate 16 drives
the reaction in the forward direction. As we can see from
mechanism 1 (Figure 7), once 16 is formed, its decomposi-
tion to 14 and dione is less-favoured relative to the forma-
tion of 17 as the former process is endergonic, whereas the
later one is exergonic. Furthermore, the forward reaction
from 16 is more facile than the backward reaction. The ste-
reochemistry of the product 19 is thus mainly decided at the
beginning, when 16 is formed from the cis zwitterion. Al-
1564, 1441, 1351, 1132 cmꢀ1
228.0634 [M+]; found: 228.0663.
; HRMS (EI): m/z: calcd for C10H12O6:
Di-tert-butyl (2Z)-2,3-dibenzoylbut-2-enedioate (11): Pyridine (20 mol%)
was added to a solution of benzil 6 (100 mg, 0.4762 mmol) and di-tert-
butyl acetylenedicarboxylate 10 (129 mg, 0.5714 mmol) at ꢀ108C. The re-
action mixture was allowed to attain room temperature and was then
stirred for 12 h. Removal of the solvent followed by column chromatog-
raphy afforded product 11 as a crystalline white solid (110 mg, 48%).
M.p.: 145–1478C (recrystallized from
a CH2Cl2/hexane mixture);
1H NMR: d=8.01 (d, J=7.32 Hz, 4H), 7.58–7.51 (m, 6H), 1.14 ppm (s,
18H); 13C NMR: d=191.2, 161.8, 142.2, 136.0, 133.7, 128.9, 128.7,
27.3 ppm; IR (KBr): n˜ =2943, 1751, 1672, 1579, 1443, 1406, 1126 cmꢀ1
HRMS (EI): m/z: calcd for C26H28O6: 436.1886 [M+]; found: 436.1904.
;
Acknowledgements
We thank the Council of Scientific and Industrial Research (CSIR) and
Department of Science and Technology (DST), New Delhi for financial
assistance. We also thank S. Mathew for the spectral data and S. Viji for
all the mass and elemental analysis.
ꢀ
though the rotation of the central C C bond in 18 can give
rise to a trans isomer of 19, the rotation itself requires an ac-
tivation energy of 10.0 kcalmolꢀ1 and, therefore, the pre-
dominant formation of the cis isomer 19 is expected. The
solvent effect plays a major role in the facile formation of
the products as there is a substantial reduction in the activa-
tion barrier in all steps of the reaction, which is mainly at-
tributed to the increased stability of the highly charge-sepa-
rated transition states in the solvent phase.
[1] a) O. Diels, K. Alder, Liebigs Ann. Chem. 1932, 498, 16; b) R. M.
Acheson, Adv. Heterocycl. Chem. 1963, 1, 125; c) E. Winterfeidt,
Angew. Chem. 1967, 79, 389; Angew. Chem. Int. Ed. Engl. 1967, 6,
423.
[2] V. Nair, C. Rajesh, A. U. Vinod, S. Bindu, A. R. Sreekanth, J. S.
Mathen, L. Balagopal, Acc. Chem. Res. 2003, 36, 899.
[3] R. M. Acheson, A. O. Plunkett, J. Chem. Soc. 1964, 2676.
[4] E. Winterfeldt, Chem. Ber. 1965, 98, 3537.
[5] R. Huisgen, M. Morikawa, K. Herbig, E. Brunn, Chem. Ber. 1967,
100, 1094.
[6] a) V. Nair, A. R. Sreekanth, A. U. Vinod, Org. Lett. 2001, 3, 3495;
b) V. Nair, A. R. Sreekanth, A. U. Vinod, Org. Lett. 2002, 4, 2807.
[7] V. Nair, A. R. Sreekanth, N. Abhilash, A. T. Biju, B. Remadevi,
R. S. Menon, N. P. Rath, R. Srinivas, Synthesis 2003, 1895.
[8] V. Nair, B. Remadevi, N. Vidya, R. S. Menon, N. Abhilash, N. P.
Rath, Tetrahedron Lett. 2004, 45, 3203.
Experimental Section
General information: Melting points were recorded on a Buchi melting
point apparatus and are uncorrected. NMR spectra were recorded at 300
(1H) and 75 MHz (13C), respectively, on a Bruker Avance DPX-300 MHz
NMR spectrometer. Chemical shifts are reported (d) relative to TMS
(1H) and CDCl3 (13C) as the internal standards. Coupling constants (J)
are reported in hertz (Hz). Mass spectra were recorded under EI/HRMS
(at 5000) resolution by using an Auto Spec. mass spectrometer. IR spec-
tra were recorded on Bomem MB Series FTIR spectrophotometer. Ele-
mental analyses were performed on Perkin–Elmer 2400 Elemental Ana-
lyzer.
[9] V. Nair, A. N. Pillai, R. S. Menon, E. Suresh, Org. Lett. 2005, 7,
1189.
[10] Indirect evidence for the cis configuration of the zwitterion was ac-
crued from literature reports,[1,3] which show that the zwitterion 1
adds to another molecule of acetylenic ester to form a second zwit-
terion, which then cyclizes to form 4H-quinolizine. Evidently the ad-
dition product from a transoid zwitterion would not undergo such a
cyclization.
[11] a) J. M. Seminario, P. Politzer, Modern Density Functional Theory. A
Tool for Chemistry, Elsevier, Amsterdam, 1995; b) K. N. Houk in
1,3-Dipolar Cycloaddition Chemistry (Ed:. A. Padwa), Vol. 2, Wiley,
New York, 1984; c) C. F. Christian, T. Takeya, M. J. Szymanski,
D. A. Singleton, J. Org. Chem. 2007, 72, 6183; d) D. A. Singleton,
S. R. Merrigan, J. Am. Chem. Soc. 2000, 122, 11035; e) M. H. El-
Badri, D. Willenbring, Tantillo, D. J. Gervay-Hague, J. Org. Chem.
2007, 72, 4663.
[12] a) L. R. Domingo, M. Arnꢂ, R. Contreras, P. Pꢃrez, J. Phys. Chem.
A 2002, 106, 952; b) L. R. Domingo, Eur. J. Org. Chem. 2004, 23,
4788; c) L. R. Domingo, M. T. Picher, Tetrahedron 2004, 60, 5053;
d) M. J. Aurell, L. R. Domingo, P. Pꢃrez, R. Contreras, Tetrahedron
2004, 60, 11503; e) L. R. Domingo, M. T. Picher, J. Andrꢃs, V. Mo-
liner, V. S. Safont, Tetrahedron 1996, 52, 10693.
Dimethyl (2Z)-2,3-dibenzoylbut-2-enedioate (7): Pyridine (20 mol%) was
added to a solution of benzil 6 (100 mg, 0.4762 mmol) and DMAD (3)
(82 mg, 0.5714 mmol) at ꢀ108C. The reaction mixture was allowed to
attain room temperature and was then stirred for 8 h. The solvent was
distilled off in vacuo by using a rotary evaporator, followed by column
chromatography as described in the general procedure to afford the
product 7 as a white crystalline solid (101 mg, 62%). M.p.: 138–1408C
1
(recrystallized from a CH2Cl2/hexane mixture); H NMR: d=7.80 (d, J=
7.34 Hz, 4H), 7.56 (t, J=7.46 Hz, 2H), 7.41 (d, J=7.78 Hz, 4H),
3.81 ppm (s, 6H); 13C NMR: d= 189.9, 163.7, 141.6, 135.3, 134.2, 129.5,
128.7, 53.3 ppm; IR (KBr): n˜ =2961, 1754, 1666, 1592, 1448, 1430,
1132 cmꢀ1; elemental analysis calcd (%) for C20H16O6: C 68.18, H 4.58;
found: C 68.32, H 4.52.
Dimethyl (2Z)-2,3-diacetylbut-2-enedioate (9): Pyridine (20 mol%) was
added to a solution of 2,3-butanedione (8) (100 mg, 1.1628 mmol) and
Chem. Eur. J. 2008, 14, 5851 – 5860
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5859

V. Nair, C. H. Suresh, and A. N. Pillai
[13] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) A. D. Becke, Phys.
Rev. A 1988, 38, 3098; c) C. T. Lee, W. T. Yang, R. G. Parr, Phys.
Rev. B 1988, 37, 785.
[22] F. Eckert, A. Klamt, AIChE J. 2002, 48, 369.
[23] a) P. Politzer, D. G. Truhlar, Chemical Applications of Atomic and
Molecular Electrostatic Potentials, Plenum, New York, 1981; b) S. R.
Gadre, R. N. Shirsat, Electrostatics of Atoms and Molecules, Univer-
sities Press, Hyderabad, 2000.
[14] Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc.: Wallingford
CT, 2004.
[24] a) C. H. Suresh, D. Ramaiah, M. V. George, J. Org. Chem. 2007, 72,
367; b) C. H. Suresh, S. R. Gadre, J. Phys. Chem. A 2007, 111, 710;
c) K. V. Radhakrishnan, S. Anas, K. K. Syam, V. S. Sajisha, K. S.
Anju, E. Suresh, C. H. Suresh, New J. Chem. 2007, 31, 237; d) C. H.
Suresh, Inorg. Chem. 2006, 45, 4982; e) M. C. Basheer, S. Alex, U.
Santhosh, V. Biju, C. H. Suresh, K. G. Thomas, S. Das, Tetrahedron,
2007, 63, 1617; f) C. H. Suresh, N. Koga, S. R. Gadre, J. Org. Chem.
2001, 66, 6883; g) C. H. Suresh, N. Koga, S. R. Gadre, Organometal-
lics 2000, 19, 3008; h) S. P. Gejji, C. H. Suresh, L. J. Bartalotti, S. R.
Gadre, J. Phys. Chem. A, 1997, 101, 5678; i) C. H. Suresh, S. R.
Gadre, S. P. Gejji, Theor. Chem. Acc. 1998, 99, 151.
[25] A. C. Limaye, S. R. Gadre, Curr. Sci. 2001, 80, 1296.
[26] We have also optimised the structures of transition states TS1 and
TS2 by using the basis set 6-31+G(d,p) consisting of the polarisa-
G
tion and diffused functions at the B3LYP level. These structures
were very similar to the B3LYP/6-31G(d) level structures and the
activation free energy values were nearly unchanged. Further, the
TS1 geometry showed only minor structural changes when opti-
mised with PCM at the B3LYP/6-301G(d) level. See the Supporting
Information for more details.
[15] a) W. R. Wadt, P. J. Hay, J. Chem. Phys. 1985, 82, 284; b) P. J. Hay,
W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
[16] C. Peng, P. Y. Ayala, H. B. Schlegel, M. J. Frisch, J. Comput. Chem.
1996, 17, 49.
[17] C. Peng, H. B. Schlegel, Isr. J. Chem. 1993, 33, 449.
[18] a) C. Gonzꢄlez, H. B. Schlegel, J. Chem. Phys. 1991, 95, 5853; b) K.
Fukui, Acc. Chem. Res. 1981, 14, 363.
[19] a) J. Tomasi, M. Persico, Chem. Rev. 1994, 94, 2027; b) C. J. Cramer,
D. G. Truhlar, Chem. Rev. 1999, 99, 2161.
[27] a) T. Helgaker, E. Uggerud, H. J. A. Jensen, Chem. Phys. Lett. 1990,
173, 145; b) E. Uggerud, T. Helgaker, J. Am. Chem. Soc. 1992, 114,
4265; c) K. Bolton, W. L. Hase, G. H. Peshlherbe, in Modern Meth-
ods for Multidimensional Dynamics Computation in Chemistry
(Ed.: D. L. Thompson), World Scientific, Singapore, 1998, 143.
[20] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003,
24, 669.
[21] a) V. Barone, M. Cossi, J. Tomasi, J. Chem. Phys. 1997, 107, 3210;
b) V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995.
Received: February 9, 2008
Published online: May 8, 2008
5860
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 5851 – 5860