Insight into the solvation and isomerization of 3-halo-1-azaallylic anions from ab initio metadynamics calculations and NMR experiments

Article abstract of DOI:10.1002/chem.200800948

A study performed ab initio metadynamics calculations and nuclear magnetic resonance investigations, to demonstrate the solvation and isomerization of 3-halo-1-azaallylic anions. The 3-halo-1-azaallylic anions were solvated and isomerized from the N-isopr

Full text of DOI:10.1002/chem.200800948

DOI: 10.1002/chem.200800948
Insight into the Solvation and Isomerization of 3-Halo-1-azaallylic Anions
from Ab Initio Metadynamics Calculations and NMR Experiments
[
a]
[a]
[a]
[b]
Reinout Declerck, Bart De Sterck, Toon Verstraelen, Guido Verniest,
[a]
[
b]
[b]
[b]
Sven Mangelinckx, Jan Jacobs, Norbert De Kimpe,* Michel Waroquier, and
[
a]
Veronique Van Speybroeck*
In organic synthesis, it is observed experimentally that the
nature of the solvent can influence tremendously the reac-
tivity and overall product selectivity. In this communication,
we report on the E/Z isomerization of a typical solvated
species, that is, the stable lithiated 3-chloro-3-methyl-1-
azaallylic anion readily accessible from the N-isopropyl-
chemistry of 1-azaallylic anions leads to basic heterocyclic
systems such as aziridines, azetidines, pyrrolidines, pyrroles,
piperidines, oxiranes, oxolanes, and higher functionalized
ring systems, currently of interest for pharmaceutical
chemistry and agrochemistry. The application of certain
halogenated counterparts, that is, the 3-chloro-3-methyl-1-
azaallylic anions, in particular by the group of De Kimpe
and more general by the group of Florio, which incorporat-
ed the 3-chloro-3-methyl-1-azaallylic moiety into heterocy-
clic structures, has led to the synthesis of various important
ACHTUNGTRENNUNGi mine of a-chloropropiophenone, from a theoretical and a
subsequent NMR study. To date, the configurational proper-
ties of these 3-chloro-3-methyl-1-azaallylic anions are poorly
understood. Our main emphasis is devoted to the monomer-
ic species as these are believed to be the most important for
further reactivity studies (see below). It will be shown that
the investigated species is a particular example in which the
inclusion of the solvent in the modeling study is of the
utmost importance to determine the proper chemical
[5]
classes of compounds such as cyclopropanes, tetrahydro-
[
6]
[6c]
[7]
[7b,d,8]
furans,
tetrahydropyrans,
oxiranes,
aziridines,
[9]
[10]
[11]
chloroimines, pyrroles and pyridines, steroids, alkenyl-
[12]
[13]
heterocycles, and oxazetidines. As mentioned, 3-chloro-
3-methyl-1-azaallylic anions can be used for the synthesis of
functionalized oxiranes and aziridines since the former
anions behave as nucleophiles in Darzens- and aza-Darzens-
ACHTUNGTRENNUNGb ehavior.
The 3-chloro-3-methyl-1-azaallylic anion was chosen as a
model compound to obtain a deeper insight into the struc-
tural features of 3-halo-1-azaallylic anions and to get a
better understanding, and eventually a better control, of the
stereochemical outcome of the reactions in which these
anions are involved. Since their first use in the early
[7,8]
type reactions with carbonyl compounds and imines.
One
of the determining factors in the stereochemical outcome of
these Darzens-type reactions is the E/Z stereochemistry of
[14]
the starting 1-azaallylic anion. Therefore, it is important
to know and understand the configurational properties of 3-
chloro-3-methyl-1-azaallylic anions in order to perform
aldol- and Mannich-type reactions with these intermediates
in a stereocontrolled manner.
[1–3]
1960s,
non-halogenated 1-azaallylic anions have gained a
predominant role in organic synthesis due to their ability to
[4]
form new CÀC bonds with a lack of side products. The
NMR investigation and semiempirical calculations on the
stereochemistry of (2-(a-chloroethyl)benzothiazolyl)lithium
and (4,4-dimethyl-2-(a-chloroethyl)oxazolinyl)lithium have
demonstrated that internal coordination between lithium
and chlorine stabilizes the corresponding isomer with nitro-
gen and chlorine at the same side of the C=C double
[
a] Dr. R. Declerck, B. De Sterck, T. Verstraelen,
Prof. Dr. M. Waroquier, Prof. Dr. V. Van Speybroeck
Center for Molecular Modeling, Ghent University
Proeftuinstraat 86, 9000 Gent (Belgium)
Fax : (+32)9-264-6697
[15]
E-mail: Veronique.VanSpeybroeck@UGent.be
bond.
The lack of such structural investigations on 3-
[
b] Dr. G. Verniest, Dr. S. Mangelinckx, J. Jacobs, Prof. Dr. N. De Kimpe
Department of Organic Chemistry, Faculty of Bioscience Engineering
Ghent University, Coupure links 653, 9000 Gent (Belgium)
Fax : (+32)9-264-6243
chloro-3-methyl-1-azaallylic anions in sensu stricto urged us
to study their stereochemical properties.
The E/Z isomerism for non-halogenated 1-azaallylic
anions has been observed and investigated quite frequently.
The facile carbonÀcarbon bond rotation in simple lithiated
E-mail: Norbert.DeKimpe@UGent.be
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200800948.
1
1-azaallylic anions was investigated using H NMR spectros-
580
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 580 – 584

COMMUNICATION
[16]
copy. The rotational activation free energy was found to
À1
be (74.1Æ1.3) kJmol at 313 K. E/Z isomerization was also
observed upon deprotonation of ketimines of 2-butanone at
[17]
room temperature. For closely related non-chlorinated an-
alogues of the species 1–2 in Figure 1, that is, the lithiated
anion derived from the N-phenylimine of propiophenone,
isomerization from the kinetically favored E isomer with the
methyl group and the phenyl group at the same side of the
C2=C3 double bond (like in Z-isomer 1) to the thermody-
namically most stable Z isomer with the methyl group and
nitrogen at the same side of the C2=C3 double bond (like in
[18]
E-isomer 2), was observed. Therefore, it was assumed that
the lithiated 3-chloro-1-azaallylic anions of the present work
could also undergo a similar type of isomerization.
Figure 2. Details of the off-resonance ROESY spectrum of the Z-isomer
1
8
of the lithiated 3-chloro-1-azaallylic anion ([D ]THF). The resonance
signals at 0.96 ppm (doublet) and 2.84 ppm (septet) are from diisopropyl-
amine formed upon protonation of LDA.
lithium 1-azaallylic anions, higher aggregates are also ex-
[19]
pected here. It must, however, be stressed that the experi-
mental NMR spectroscopy data gives by no means any in-
formation on the solution aggregation number of the title
compounds. According to a series of NMR spectroscopic
studies and colligative measurements in THF, both mono-
meric and dimeric species will occur at low concentration of
Figure 1. Z-isomer 1 and E-isomer 2 of the lithiated (2-chloro-1-phenyl-
prop-1-en-1-yl)isopropylamide anion.
[20]
the lithium 1-azaallyl anions.
On the other hand, the
At first instance, the lithiated 3-chloro-3-methyl-1-azaal-
lylic anion was generated by deprotonation of N-(2-chloro-
monomeric lithium 1-azaallylic anions retain in most cases
their structural properties in the dimer form; the stereo-
ACHTUNGTRENNUNG
[21]
1
-phenylpropylidene)isopropylamine with lithium diisopro-
chemical preferences in reactions can readily be explained
1
[22]
pylamide (LDA) in [D ]THF at 273 K and analyzed by H
with the properties of the monomeric form.
There are
8
13
and C NMR spectroscopy. The a-chloropropiophenone
imine was deprotonated under these conditions to a single
stereoisomer as demonstrated by the presence of a single set
of characteristic H NMR chemical shifts of the methyl
group on the double bond (s, d=1.77 ppm), methine func-
also several indications that the monomeric forms are the
most likely reactive forms of these molecules. Studies by
Streitwieser and co-workers have shown that the reactive
form of lithiated compounds, present in solution for the
larger extent as higher aggregates, can be the monomeric
form due to a kinetic advantage because of lower-energy
transition states as compared to the less reactive higher ag-
1
tion (septet, 2.96 ppm) and isopropyl methyl groups (d,
1
3
0.83 ppm). Also a single set of characteristic C NMR
[23]
chemical shifts of the lithiated 3-chloro-1-azaallylic anion
were observed (see Supporting Information). The stereo-
chemistry of the Z anion 1 was determined by off-resonance
ROESY spectroscopy showing ROE effects between the
methyl group and the ortho-protons of the phenyl ring posi-
tioned at the same side of the carbonÀcarbon double bond
gregates.
Based on these arguments and the expensive
computational resources needed for treating higher argu-
ments, the theoretical part of this communication will focus
only on the configuration of the monomeric species, being
aware that these might be part of higher aggregates under
experimental conditions.
(
Figure 2). Furthermore, the observed ROE effects between
Despite the huge amount of theoretical studies that ap-
peared the last years, modeling of complex phenomena such
as chemistry in liquids remains a challenge as standard opti-
mization techniques and ab initio molecular dynamics meth-
the N-isopropyl substituent and the phenyl group support
the anti stereochemistry of the 3-chloro-1-azaallylic anion,
that is, the N-isopropyl group is oriented anti with respect to
the C2=C3 double bond. In contrast to the non-chlorinated
[24,25]
ods are often not suitable.
The first set of methods is
[19]
species,
the aforementioned NMR experiments indicate
routinely performed nowadays, but for our systems in which
the solvent participates actively, a single optimized structure
does not resemble the configurational distribution at finite
temperature. First-principle molecular dynamics simulations
are often restricted by short simulation times. As such, inter-
esting regions of phase space are often so high in free
energy that their sampling during a standard MD simulation
is a rare event. Enhanced sampling techniques have become
that in THF only the Z/anti isomer 1 occurs and that both
amide and C=C double bond rotations are inhibited. This
particular behavior of chlorinated 1-azaallylic anions de-
manded a theoretical interpretation. Before elaborating on
the theoretical results, it is important to focus on the tenden-
cy of 1-azaallylic anions to form higher aggregates in solu-
tion. From the extensive work that has been performed on
Chem. Eur. J. 2009, 15, 580 – 584
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
581

V. Van Speybroeck, N. De Kimpe et al.
[25]
an active research domain.
The relatively new metady-
dral angles is displayed in
namics method has particularly attracted our attention. It
was first proposed by Laio and Parrinello and enables an
enhanced sampling of separated regions in phase space, si-
multaneously mapping the underlying free-energy landscape
Figure 4. The Gibbs free energy
barriers for E-to-Z and Z-to-E
isomerization
amount
to
and
À1
(107.1Æ12.1) kJmol
[26]
À1
as a function of a limited number of collective variables.
The particular implementation is based on the work of Ian-
(128.6Æ12.1) kJmol , respec-
[32]
tively. These barriers are
[27]
nuzi et al.
high, preventing isomerization
at the experimental tempera-
ture. The Z-isomer 1 is more
stable than the E-isomer 2 by
Prior to the modeling of the 1-azaallylic anions, we mod-
eled the liquid structure of pure THF by using first-principle
molecular dynamics calculations. The liquid structure of
THF was recently assessed via hydrogen/deuterium isotopic
substitution neutron-diffraction techniques by Bowron,
À1
DGZ–E =(21.5Æ12.1) kJmol ,
which indicates that the experi-
mentally observed Z-isomer 1
is thermodynamically favored.
[28]
Finney, and Soper. A periodic cubic simulation cell was
filled with 64 THF molecules. This choice represents an op-
timal compromise between computational cost and a proper
embedding of the solute in the solvent. The simulation cell
size was chosen to correspond with the experimental density
Within a static cluster approach
using a combined explicit/im- Figure 3. Characteristic snap-
plicit solvent model we were
unable to determine the transi-
shot of the MD simulation of
the Z-isomer 1 (A) and the E-
isomer 2 (B) solvated in THF.
À3 [29]
of 0.88 kgdm . The performance of the THF model was
validated by calculating the radial distribution function
tion state for E/Z isomerization
(
RDF) of the molecular centers, which was found to be in
excellent agreement with the benchmark RDF reported in
reference [28] (see Supporting Information). Moreover, the
MD simulations yielded a conformational distribution of
5
9% twisted and 41% oxygen envelope, indicating a thor-
[28]
ough sampling of the system.
After the THF model had been successfully assessed, it
was applied to study the degree of coordination of the 3-
chloro-1-azaallylic anions in solution. The coordination
number for lithium enolates in ethereal solvents is rather
difficult to establish but four-coordinate lithium cations
have been clearly recognized in NMR studies of solvent sep-
[30]
arated ion pairs. For contact ion pairs, coordination is ex-
pected less important because of the electrostatic effect of
the counter ion. Theoretically the structures of a variety of
organic lithium compounds were determined in the gas
phase and in solvation using microsolvation with explicit
[31]
ethereal ligands and/or continuum models.
For the 1-
azaallylic anions as encountered here which are subject to
large steric crowding, the degree of coordination is not a
priori clear and can not be deduced straightforwardly from
the experimental data. Isothermal–isobaric (NPT) molecular
dynamics simulations during a period of 2.5 ps show that the
Z-isomer 1 is monocoordinated whereas the E-isomer 2 fea-
tures a two-fold coordination with THF (illustrated in
Figure 3). In the E-isomer 2 the halogen–lithium coordina-
tion is not present which allows a second THF molecule to
coordinate with the counter ion.
À1
Figure 4. Gibbs free energy profile (in kJmol ) governing the E–Z iso-
merization of the lithiated 3-chloro-1-azaallylic anion in THF. The posi-
tions of both stable isomers E (2) and Z (1) and the saddle point (E–Z)
are added. Note that the two collective variables feature a 2p periodicity.
°
as the coordination number varies during the chemical
transformation. Moreover, the stability of the Z-isomer 1
with respect to the E-isomer 2 was 20 kJmol too high
À1
In order to obtain insight into the occurrence of only one
stereoisomer in case of 3-chloro-3-methyl-1-azaallylic anions
compared to the metadynamics calculations. By capturing
the movement of both dihedral angles, we were able to ob-
2
3
1
and 2, we decided to construct the free-energy landscape
serve the sp to sp hybridization transition of the C3 carbon
atom upon rotation, a well-known feature of rotations about
allylic bonds. This is reflected in the fact that the saddle
connecting the basins of the two isomers. To this end we ap-
plied the metadynamics method in which the dihedral
angles Cl-C3-C2-N and C4-C3-C2-N were chosen as collec-
tive variables. This choice guarantees the independent
movement of the methyl and chlorine substituents. The re-
sulting free-energy landscape as a function of the two dihe-
°
point, denoted as (E–Z) , does not lie on the linear pathway
connecting both isomers, which confirms a posteriori the im-
portance of capturing the movement of both dihedral
angles.
582
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 580 – 584

3
-Halo-1-azaallylic Anions
COMMUNICATION
Finally, we infer from both first principle metadynamics
Acknowledgements
and NMR experiments that the Z isomer is the only config-
uration formed upon the deprotonation of the starting imine
to the lithium 3-chloro-1-azaallylic anion. The interaction
between the counter ion and the halogen, an effect that is
not present in the non-halogenated 1-azaenolates, stabilizes
The authors thank Professor J. Martins (Department of Organic Chemis-
try, NMR and Structure Analysis, Ghent University) for recording the
ROESY spectra, and Dr. D. T. Bowron (STFC Rutherford Appleton
Laboratory) for providing the EPSR-derived RDF data. This work was
supported by the Fund for Scientific Research-Flanders and the Research
Board of Ghent University.
À1
the Z isomer by 21 kJmol . Moreover, the transition from
the Z to the E isomer is very highly activated and features a
change in coordination for the lithium cation as the broken
lithium-chlorine interaction is replaced by a lithium–THF in-
teraction. These effects can only be seen because of the ex-
plicit inclusion of the large THF model in the QM simula-
tions. These results show that the stereochemistry of 3-
chloro-3-methyl-1-azaallylic anions is manifestly different
compared to their non-chlorinated counterparts and is the
result of their configurational stability which should be ben-
eficial during their synthetic use as functionalized intermedi-
ates in stereoselective reactions.
Keywords: azaallylic
metadynamics · NMR spectroscopy · solvent effects
anions
·
isomerization
·
[
[
1] G. Stork, S. R. Dowd, J. Am. Chem. Soc. 1963, 85, 2178–2180.
2] G. Wittig, H. D. Fommeld, P. Suchanek, Angew. Chem. 1963, 75,
78–979.
9
[3] G. Wittig, H. Reiff, Angew. Chem. 1968, 80, 8–15.
[4] S. Mangelinckx, N. Giubellina, N. De Kimpe, Chem. Rev. 2004, 104,
2
353–2400 and references therein.
[
5] a) N. De Kimpe, P. Brunet, R. Verhꢁ, N. Schamp, J. Chem. Soc.
Chem. Commun. 1988, 825–827; b) W. Aelterman, K. Abbaspour
Tehrani, W. Coppens, T. Huybrechts, N. De Kimpe, D. Tourwꢁ, J.-P.
Declercq, Eur. J. Org. Chem. 1999, 239–250; c) V. Capriati, S.
Florio, R. Luisi, M. T. Rocchetti, J. Org. Chem. 2002, 67, 759–763;
d) S. Florio, F. M. Perna, R. Luisi, J. Barluenga, F. Rodrꢂguez, F. J.
FaÇanꢃs, J. Org. Chem. 2004, 69, 5480–5482.
Experimental Section
All molecular dynamics calculations were performed within the cp2k/
[
33]
quickstep code, employing the Gaussian and plane-wave (GPW) densi-
[
34]
[6] a) N. De Kimpe, E. Stanoeva, N. Schamp, Tetrahedron Lett. 1988,
9, 589–592; b) N. De Kimpe, W. Aelterman, K. De Geyter, J.-P.
ty functional method and periodic boundary conditions. A BLYP gradi-
ent-corrected functional was used throughout, together with a TZVP-
2
[
35]
Declercq, J. Org. Chem. 1997, 62, 5138–5143; c) W. Aelterman, N.
Giubellina, E. Stanoeva, K. De Geyter, N. De Kimpe, Tetrahedron
Lett. 2004, 45, 441–444.
PSP basis set, a 400 Ry cutoff for the auxiliary plane wave grid, and
[
36]
pseudopotentials developed by Goedecker and co-workers. Isothermal-
isobaric (NPT) MD simulations of both isomers were conducted. The
species were properly embedded in the THF solvent model by determin-
ing, using atomic Pauling radii, the volume associated with their solvent
[
7] a) P. Sulmon, N. De Kimpe, N. Schamp, J.-P. Declercq, B. Tinant, J.
Org. Chem. 1988, 53, 4457–4462; b) V. Capriati, L. Degennaro, S.
Florio, R. Luisi, C. Tralli, L. Troisi, Synthesis 2001, 2299–2306;
c) F. M. Perna, V. Capriati, S. Florio, R. Luisi, J. Org. Chem. 2002,
[
37]
accessible surface. As the volume of THF is 2.96 times smaller com-
pared to the volume of the 3-chloro-1-azaallylic anion, three THF mole-
cules in the simulation cell were replaced by the 3-chloro-1-azaallylic spe-
cies. An equilibration time of 2.5 ps has been respected to allow the sol-
vent to accommodate to the presence of the solute and vice versa, fol-
lowed by a 40 ps metadynamics run. Accurate metadynamics parameter
values were determined from Gibbs free energy barrier predictions of
the lithiated 3-chloro-1-azaallylic anion in the gas phase (using a super-
cell approach), including only one THF molecule to impose the limited
6
7, 8351–8359; d) F. Bona, L. De Vitis, S. Florio, L. Ronzini, L.
Troisi, Tetrahedron 2003, 59, 1381–1387; e) L. Troisi, L. De Vitis, C.
Granito, P. Metrangolo, T. Pilati, L. Ronzini, ARKIVOC 2004, xiv,
6
1–73.
8] a) R. Luisi, V. Capriati, S. Florio, R. Ranaldo, Tetrahedron Lett.
003, 44, 2677–2681; b) L. De Vitis, S. Florio, C. Granito, L. Ronzi-
ni, L. Troisi, V. Capriati, R. Luisi, T. Pilati, Tetrahedron 2004, 60,
175–1182; c) R. Luisi, V. Capriati, S. Florio, P. Di Cunto, B. Musio,
[
2
1
freedom of the lithium cation. The set of parameter values w=
À1
Tetrahedron 2005, 61, 3251–3260; d) L. Troisi, C. Granito, C. Carluc-
ci, F. Bona, S. Florio, Eur. J. Org. Chem. 2006, 775–781.
2
.0 kJmol , s=0.33 rad and G=50 fs (notation: see ref. [26]) yielded an
À1
energy barrier within 1.0 kJmol of the convergence limit. According to
ref. [32] the estimated error using these parameters is e=6.1 kJmol
À1
[9] P. Sulmon, N. De Kimpe, N. Schamp, Synthesis 1989, 8–12.
10] W. Aelterman, N. De Kimpe, V. Tyvorskii, O. Kulinkovich, J. Org.
Chem. 2001, 66, 53–58.
.
[
Lithiated 3-chloro-1-azaallylic anion 1: To a stirred solution of diisopro-
pylamine (0.056 g, 0.55 mmol) in [D ]THF (1 mL), nBuLi (0.22 mL,
.55 mmol, 2.5 n in hexanes) was added slowly at 273 K. After 30 min of
stirring at 273 K, the solution was evaporated in vacuo to dryness, after
which, [D ]THF (0.5 mL) was added and a solution of N-(2-chloro-1-phe-
nylpropylidene)isopropylamine (0.11 g, 0.5 mmol) in [D ]THF (0.5 mL)
was dropped to the LDA solution at 273 K and stirring was continued for
h. Subsequently, the reaction mixture was allowed to reach room tem-
8
[
[
11] L. Troisi, S. Florio, C. Granito, Steroids 2002, 67, 687–693.
12] V. Capriati, L. Degennaro, S. Florio, R. Luisi, Eur. J. Org. Chem.
0
2
002, 2961–2969.
13] R. Luisi, V. Capriati, S. Florio, E. Piccolo, J. Org. Chem. 2003, 68,
0187–10190.
14] V. K. Aggarwal, D. M. Badine, V. A. Moorthie in Aziridines and Ep-
oxides in Organic Synthesis (Ed.: A. K. Yudin), Wiley-VCH, Wein-
heim, 2006, pp. 1–35.
8
[
[
8
1
1
1
13
perature during 15 min. H NMR (300 MHz) and C NMR (75 MHz)
spectra were taken from the prepared 3-chloro-1-azaallylic anion 1 at
[
[
[
15] A. Abbotto, S. Bradamante, S. Florio, V. Capriati, J. Org. Chem.
1
room temperature. H NMR (300 MHz, [D
8
]THF): d=0.83 (d, J=
), 2.96 (septet, J=6.1 Hz, 1H;
CH), 7.08–7.14 (m, 3H; o-CHar and p-CHar), 7.18–7.24 ppm (m,
1
997, 62, 8937–8940.
6
.33 Hz, 6H; (CH
3 2 3
) CH), 1.77 (s, 3H; CH
16] J. Y. Lee, T. J. Lynch, D. T. Mao, D. E. Bergbreiter, M. Newcomb, J.
Am. Chem. Soc. 1981, 103, 6215–6217.
17] J. K. Smith, D. E. Bergbreiter, M. Newcomb, J. Org. Chem. 1981, 46,
(
2
1
CH
3
)
2
1
3
8
H; m-CHar); C NMR (75 MHz, [D ]THF): d=22.3, 28.1, 48.6, 83.6,
25.9, 127.6, 130.0, 144.0, 155.0 ppm.
3
157–3158.
[
[
18] R. Knorr, P. Low, J. Am. Chem. Soc. 1980, 102, 3241–3242.
19] a) H. Ahlbrecht, E. O. Dꢄber, D. Enders, H. Eichenauer, P. Weuster,
Tetrahedron Lett. 1978, 19, 3691–3694; b) P. von Raguꢁ Schleyer, R.
Hacker, H. Dietrich, W. Mahdi, J. Chem. Soc. Chem. Commun.
1
985, 622–624; c) R. Glaser, C. M. Hadad, K. B. Wiberg, A. Streit-
Chem. Eur. J. 2009, 15, 580 – 584
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
583

V. Van Speybroeck, N. De Kimpe et al.
wieser, J. Org. Chem. 1991, 56, 6625–6637; d) R. Knorr, H. Dietrich,
W. Mahdi, Chem. Ber. 1991, 124, 2057–2063; e) P. B. Hitchcock,
M. F. Lappert, D.-S. Liu, J. Chem. Soc. Chem. Commun. 1994, 2637–
[26] A. Laio, M. Parrinello, Proc. Natl. Acad. Sci. USA 2002, 99, 12562–
12566.
[27] M. Iannuzzi, A. Laio, M. Parrinello, Phys. Rev. Lett. 2003, 90,
238302.
2
638; f) A. Abbotto, A. Streitwieser, P. von Raguꢁ Schleyer, J. Am.
Chem. Soc. 1997, 119, 11255–11268; Schleyer, J. Am. Chem. Soc.
[28] a) D. T. Bowron, J. L. Finney, A. K. Soper, J. Am. Chem. Soc. 2006,
ˇ
1
997, 119, 11255–11268; Schleyer, J. Am. Chem. Soc. 1997, 119,
128, 5119–5126; b) M. Strajbl, J. Floriꢃn, Theor. Chem. Acc. 1998,
9
9, 166–170; c) V. M. Rayꢅn, J. A. Sordo, J. Chem. Phys. 2005, 122,
1
1255–11268; g) L. M. Pratt, T. E. Hogen-Esch, I. M. Khan, Tetra-
2
04303.
hedron 1995, 51, 5955–5970; h) L. R. Liou, A. J. McNeil, A. Ramir-
ez, G. E. S. Toombes, J. M. Gruver, D. B. Collum, J. Am. Chem. Soc.
[
[
29] a) D. R. Lide in CRC Handbook of Chemistry and Physics, 84th ed.,
CRC Press, Wiley, New York, 2003; b) C. Carvajal, K. J. Tçlle, J.
Smid, M. Szwarc, J. Am. Chem. Soc. 1965, 87, 5548–5553.
2
008, 130, 4859–4868.
[
[
[
20] a) R. A. Wanat, D. B. Collum, G. Van Duyne, J. Clardy, R. T.
DePue, J. Am. Chem. Soc. 1986, 108, 3415–3422; b) N. Kallman,
D. B. Collum, J. Am. Chem. Soc. 1987, 109, 7466–7472; c) L. M.
Jackman, L. M. Scarmoutzos, B. D. Smith, P. G. Williard, J. Am.
Chem. Soc. 1988, 110, 6058–6063.
21] a) R. R. Fraser, N. Chuaqui-Offermanns, K. N. Houk, N. G. Rondan,
J. Organomet. Chem. 1981, 206, 131–138; b) R. Glaser, A. Streit-
wieser, J. Org. Chem. 1991, 56, 6612–6624; c) R. Glaser, C. M.
Hadad, K. B. Wiberg, A. Streitwieser, J. Org. Chem. 1991, 56, 6625–
30] a) L. M. Pratt, N. V aˇ n NguyÞn, B. Ramachandran, J. Org. Chem.
2
005, 70, 4279–4283; b) L. M. Pratt, B. Ramachandran, J. D. Xidos,
C. J. Cramer, D. G. Truhlar, J. Org. Chem. 2002, 67, 7607–7612;
c) P. L. Fast, M. L. Sanchez, D. G. Truhlar, Chem. Phys. Lett. 1999,
3
06, 407–410; d) J. A. Pople, M. Head-Gordon, K. Raghavachari, J.
Chem. Phys. 1987, 87, 5968–5975.
[
31] a) L. M. Pratt, A. Streitwieser, J. Org. Chem. 2003, 68, 2830–2838;
b) V. Van Speybroeck, K. Moonen, K. Hemelsoet, C. V. Stevens, M.
Waroquier, J. Am. Chem. Soc. 2006, 128, 8468–8478.
32] A. Laio, A. Rodriguez-Fortea, F. L. Gervasio, M. Ceccarelli, M. Par-
rinello, J. Phys. Chem. B 2005, 109, 6714–6721.
6
637.
[
22] a) R. Glaser, A. Streitwieser, J. Am. Chem. Soc. 1987, 109, 1258–
1260; b) R. Glaser, A. Streitwieser, Pure Appl. Chem. 1988, 60, 195–
204; c) R. Glaser, A. Streitwieser, J. Am. Chem. Soc. 1989, 111,
7340–7348; d) R. Glaser, A. Streitwieser, J. Am. Chem. Soc. 1989,
111, 8799–8809; e) R. Glaser, A. Streitwieser, J. Org. Chem. 1989,
54, 5491–5502; f) R. Glaser, A. Streitwieser, J. Mol. Struct. 1988,
163, 19–50.
[
[
33] http://cp2k.berlios.de.
34] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100; b) C. Lee, W.
Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[
[
35] G. Lippert, J. Hutter, P. Ballone, M. Parrinello, J. Phys. Chem. 1996,
1
00, 6231–6235.
36] a) S. Goedecker, M. Teter, J. Hutter, Phys. Rev. B 1996, 54, 1703–
[
[
23] A. Streitwieser, J. Mol. Model. 2006, 12, 673–680.
1
5
710; b) C. Hartwigsen, S. Goedecker, J. Hutter, Phys. Rev. B 1998,
8, 3641–3662.
24] a) A. V. Marenich, R. M. Olson, A. C. Chamberlin, C. J. Cramer,
D. G. Truhlar, J. Chem. Theory Comput. 2007, 3, 2055–2067; b) C. P.
Kelly, C. J. Cramer, D. G. Truhlar, J. Chem. Theory Comput. 2005, 1,
[37] a) J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999–
3093; b) Y. Takano, K. N. Houk, J. Chem. Theory Comput. 2005, 1,
70–77.
1
2
133–1152; c) J. Tomasi, M. Persico, Chem. Rev. 1994, 94, 2027–
094.
Received: May 18, 2008
[
25] B. Ensing, M. De Vivo, Z. Liu, P. Moore, M. L. Klein, Acc. Chem.
Res. 2006, 39, 73–81.
Revised: August 29, 2008
Published online: November 7, 2008
584
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 580 – 584