DOI: 10.1002/chem.
201504016 Full Paper
& Density Functional Calculations
How Many Water Molecules Does it Take to Dissociate HCl?
Alba Vargas-Caamal,[a] Jose Luis Cabellos,[a] Filiberto Ortiz-Chi,[b] Henry S. Rzepa,[c]
Albeiro Restrepo,*[d] and Gabriel Merino*[a]
In memory of Jon Mikel Azpiroz—we dearly miss him
Abstract: The potential energy surfaces of the HCl(H2O)n (n
is the number of water molecules) clusters are systematically
explored using density functional theory and high-level ab
initio computations. On the basis of electronic energies, the
number of water molecules needed for HCl dissociation is
four as reported by some experimental groups. However,
this number is five owing to the inclusion of entropic fac-
tors. Wiberg bond indices are calculated and analyzed, and
the results provide a quadratic correlation and classification
Introduction
Water plays a central role in our understanding of chemical
and biological processes, and it has the special ability to disso-
ciate molecules. In particular, acid dissociation, leading to
proton transfer[1, 2] and solvation of the fragments,[3–13] is one of
the most important chemical processes. The dissociation of
HCl has been the focus of recent studies in ambient bulk
water,[1, 2, 14, 15] as well as at low temperatures in confined geo-
metries such as on amorphous ice surfaces, on ice nanocrys-
tals, and in microsolvation environments, which are relevant to
atmospheric or interstellar problems.[16–19] Very recently, after
measuring the dipole moments of DCl-doped water clusters by
[a] A. Vargas-Caamal, J. L. Cabellos, G. Merino
Departamento de F†sica Aplicada
Centro de Investigaciûn y de Estudios Avanzados, Unidad M¦rida
Km 6 Antigua Carretera a Progreso. Apdo. Postal 73
Cordemex, 97310, M¦rida, Yuc. (M¦xico)
E-mail: gmerino@mda.cinvestav.mx
[b] F. Ortiz-Chi
Instituto Tecnolûgico Superior de Calkin†
Av. Ah-Canul s/n, Carr. Fed. Calkin†-Campeche
CP 24900, Calkin†, Campeche (M¦xico)
[c] H. S. Rzepa
Department of Chemistry, Imperial College London
South Kensington campus, London, SW7 2AZ (UK)
[d] A. Restrepo
Instituto de Qu†mica, Universidad de Antioquia UdeA
Calle 70 No. 52–21, Medell†n (Colombia)
E-mail: albeiro.restrepo@udea.edu.co
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201504016. This includes the variation of
the dipole moments as a function of temperature for the HCl(H2O)n clusters,
and the Cartesian coordinates for all the local minimum structures found
by Bilatu.
Chem. Eur. J. 2016, 22, 2812 – 2818
of clusters according to the nondissociated, partially dissoci-
ated, and fully dissociated character of the H¢Cl bond. Our
computations show that if temperature is not controlled
during the experiment, the values obtained for the dipole
moment (or for any measurable property) are susceptible to
change, providing a different picture of the number of water
molecules needed for HCl dissociation in a nanoscopic drop-
let.
the beam deflection method, Guggemos et al.[20] observed that
“the addition of a DCl molecule to a water cluster results in
a strongly enhanced susceptibility. There is evidence for a noticea-
ble rise in the dipole moment occurring at n … 5–6. This size is
consistent with predictions for the onset of ionic dissociation”.
These results exceed by two the number of water molecules
required to dissociate HCl reported in earlier experimental and
theoretical works.[21, 22] For instance, Gutberlet et al. reported
the experimental observation of a nanoscopic aqueous droplet
of acid formed within a superfluid helium cluster at 0.37 K. By
using high-resolution mass-selective infrared laser spectrosco-
py, these authors found that successive aggregation of HCl
with water molecules, HCl(H2O)n, readily results in the forma-
tion of hydronium at n = 4.[23]
The literature related to the dissociation of acids in aqueous
environments and the microsolvation of the resulting anions
and protons is extensive.[15, 20, 24–44] In 2011, Leopold published
a comprehensive review of this topic,[45] in which Table 1 col-
lects experimental and theoretical evidence about the number
of water molecules needed to dissociate some simple acids,
suggesting four water molecules for the case of HCl. A note ac-
companying Table 1 explicitly calls for the reader to “note excel-
lent agreement with the experimental work”, making reference
to papers by the groups of Maillard[24] and Suhm,[26, 46] who
used several IR spectroscopy techniques to draw their conclu-
sions. As a generalization, Leopold summarized the results
found in the literature in the following way: “the overall picture
is that for acids that ultimately ionize in aqueous solution, three
to five water molecules typically are needed to induce ionization
in the lowest-energy isomer of a given cluster. This range is en-
tirely consistent with available experimental evidence”.[45]
High-level ab initio and DFT computations on molecular clus-
ters face a tremendous challenge in producing accurate poten-
2812 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

tial energy surfaces (PESs) for intermolecular interactions
owing to the exponential dependency of the number of local
minima on the number of molecules.[47–49] In most cases, chem-
ical intuition is the weapon of choice to guess the best candi-
dates. However, this biased approach quickly becomes imprac-
tical because of the overwhelming number of possibilities,
which, even for clusters of relatively small sizes, is bound to
miss relevant regions of the corresponding PESs. This is a very
sensitive issue, especially in the cases where several local
minima are very close in energy to the global minimum with-
out being structurally related. Water clusters are the archetypal
example of complicated energy landscapes populated by mul-
titudes of structurally unrelated local minima,[47, 50–57] which in
several cases are not discernible from the corresponding
global minimum using either experimental or theoretical meth-
ods. The situation is further complicated by low interconver-
sion barriers and by consideration of temperature, entropy,
and internal degrees of freedom,[58–62] which usually shuffle sta-
bility orders in favor of more open structures as the tempera-
ture increases, eventually leading to ideal gas behavior at suffi-
ciently high temperatures.[47, 52] Computations on interactions
of water with HCl, dissociated or not, have been performed
through intuitive construction of initial guesses for cluster
structures[27, 28, 30, 50] (see Leopold’s review[45] for a more compre-
hensive list of theoretical calculations on the subject), with the
aim of identifying the structural preferences for all possible
combinations of interacting species that could be found in
HCl/water mixtures.
In view of the preceding discussion, in this article we have
attempted to understand the microsolvation process of HCl,
and specifically, to clarify why the number of water molecules
needed to achieve dissociation may vary depending on the ex-
perimental technique employed. To this end, we adopt a sto-
chastic, structurally unbiased approach to sample the potential
energy surfaces exhaustively for interactions between hydro-
chloric acid and one to six water molecules. Our results, based
on the electronic energies, indicate that the number of water
molecules needed to dissociate HCl is four. However, this con-
clusion changes upon consideration of entropic factors. At
room temperature, the number of water molecules needed to
dissociate HCl is five. Therefore, the effects of temperature on
the stability and dipole moments are discussed in detail.
Computational Details
With the steady growth in computational power at affordable
cost, stochastic explorations of moderately sized atomic and
molecular clusters using quantum Hamiltonians have become
a viable choice for obtaining detailed information on the
nature of intermolecular interactions and to overcome the
problems associated with biased initial guesses. Our computa-
tional procedure employs a modified kick algorithm called
Bilatu[48, 63–65] to generate starting structures, establishing a hier-
archical screening at the PBE0[66]/D95 V level, and the
B2PLYPD3 dispersion-corrected double-hybrid functional[67, 68] in
conjunction with the def2-TZVP[69] basis set. Therefore, the
geometrical and energetic discussion is based on the
Chem. Eur. J. 2016, 22, 2812 – 2818 www.chemeurj.org 2813
Full Paper
B2PLYPD3/def2-TZVP results. All computations are performed
using the very-tight integral accuracy and ultrafine grid op-
tions available in the Gaussian 09 (revision D.01) package.[70]
Each structure is characterized as a true minimum by harmonic
vibrational frequency analysis.
To validate our methodology, we optimized and character-
ized the structures of the HCl–H2O and HCl(H2O)2 clusters (one
and seven local minima, respectively) at the CCSD(T)/def2-
TZVP level of theory, obtaining very similar structures and rela-
tive energies to those found at the B2PLYPD3 level, supporting
the use of this approach.
Energetics and structures
We uncovered minima on the PESs for the interactions of HCl
with up to six water molecules of previously unnoticed com-
plexity and structural diversity. A total of 431 different local
minima were located, distributed as 1, 7, 22, 82, 148, and 171
isomers for n = 1, 2, 3, 4, 5, and 6, respectively (n is the
number of water molecules). Despite the staggering number
of structures found here, the PESs in question cannot be con-
sidered fully characterized; this is not an attainable goal with
currently available experimental or computational methods be-
cause the number of possible local minima increases exponen-
tially with the size of the system. Nonetheless, we are confi-
dent that our searches capture the most important structural
features of the HCl(H2O)n clusters.
We found structures containing nondissociated, partially dis-
sociated, and fully dissociated HCl. A wide variety of interac-
tions stabilizing the clusters are at play: water–water and HCl–
water hydrogen bonds (HBs) with the possibility of HCl acting
as donor and as acceptor of one or two HBs, long-range Cl···H
interactions, and microsolvation of the H3O + cation in the
form of Eigen cations (each proton interacting with a lone pair
from a vicinal water molecule), and what we refer to as quasi-
Eigen cations, in which Cl¢ takes over the stabilizing role of
one of the water molecules (in well-defined Eigen cations,[27, 45]
the H3O + moiety is not in direct contact with the Cl¢ anion).
No Zundel (H2O···H···OH2) + cations were located. However, as
in the case of the Eigen cations, a number of partially dissociat-
ed structures, in which the stabilizing role of the second water
molecule is assumed by Cl¢ , were identified (Cl¢···H + ···OH2).
Figure 1 A shows the global minimum structures of the
HCl(H2O)n (n = 1–6) clusters optimized using the B2PLYPD3 ap-
proach considering the zero-point energy (ZPE) corrections. In
our nomenclature, HCl-nx-y, x indicates the number of water
molecules, and y denotes the energetic relative ordering posi-
tion with respect to the lowest free energy structures. The
most noteworthy structural variation up to n = 3 involves the
key O···H interaction contracting rapidly (from 1.863 to 1.569 æ)
until a hydrogen atom is transferred completely to form a hy-
dronium ion for n = 4 (structure HCl-n4-36). So, the overall pic-
ture provided by these computations is that for HCl, four
water molecules are needed to induce full dissociation.
Interestingly, our computations show that acid dissociation
depends on the inclusion of the entropic factors. Figure 1 also
collects the lowest free energy forms at 298.15 K and 1 atm. A
Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 Full Paper

Figure 1. A) Lowest energy structures (DEZPE) and B) lowest free energy structures at 298.15 K and 1 atm (DG298.15) for HCl(H2O)n, n = 1–6. Wiberg Bond Indices
for the H···Cl interactions (WBIHCl) are also shown. Isomer populations (%xi) were estimated from standard Boltzmann distributions using the Gibbs free ener-
gies (298 K, 1 atm). ClH···O distances in æ. Data taken from the B2PLYPD3/def2-TVZP optimized geometries.

comparison between the lowest energy and the lowest free Guggemos et al.[20] reported the addition of a DCl (not HCl)
energy structures reveals small structural differences for n = 2 molecule to a water cluster. In principle, this substitution could
and n = 3, in that the hydrogen atoms of the water molecules affect the energetics. Table 1 summarizes the free energy dif-
point in different directions. Strictly speaking, the structures ferences obtained for the n = 2 case upon substitution of the
are different, but the motif is the same. The first notable HCl unit by DCl. It is apparent that the substitution of HCl by
change is perceived for n = 4. At room temperature, the most DCl does not alter the relative energies.
stable structure is an associated pentameric form with an O···H
interaction distance of 1.516 æ (structure HCl-n4-1). The dissoci-
Bonding analysis
ation is complete for the n = 5 case (structure HCl-n5-1), which
is structurally closely related to the global minimum in the A convenient way to analyze proton transfer structurally is to
water pentamer.[54] It is clear that entropy factors are important correlate the distances between all atoms involved in the
for explaining the size evolution of these clusters. Cl···H···O hydrogen bonds (see Figure 2).[71] Cases of the lowest
Thus, temperature is a key variable in determining the equi- free energy form for n = 6 are not included in this correlation.
librium structure of a hydrated molecule. For n = 4, isomer
populations estimated through Boltzmann distributions sug-
gest that at 298.15 K and 1 atm, the lowest energy form con- Table 1. Comparison of free relative energies obtained with HCl and DCl
tributes only approximately 0.14 % (DG = 2.6 kcal mol¢1), where- for n = 2.
as the lowest DG structure (see Figure 1 B) contributes around Isomer DGHCl DGDCl
11 %. There is no ambiguity about the fully dissociated charac- [kcal mol¢1] [kcal mol¢1]
ter of the global minima and for n = 5, 6: the lowest DG struc- 1 0.00 0.00
ture for n = 5 is a quasi-Eigen cation (structure HCl-n5-1), and 2 0.00 0.00
the structure for n = 6 contains an Eigen cation (structure HCl- 3 0.16 0.14
n6-1). The presence of Eigen and quasi-Eigen cations and the 4 0.16 0.14
5 4.35 4.10
absence of Zundel cations are remarkable because it is known 6 5.07 5.04
that Eigen cations are the dominant forms present in bulk 7 5.07 5.04
samples of aqueous solutions of dissociated acids. In summary,
dissociation of HCl occurs with four water molecules without
correction for entropic effects and ZPE, as reported by Mail- For linear hydrogen bonds, q1, defined as (r1¢r2)/2, is the dis-
lard[24] and Suhm,[26, 46] but occurs with five water molecules if tance from the hydrogen atom to the center of the hydrogen
such effects are included. bond, and q2 is the distance between the Cl and O atoms (r1

Chem. Eur. J. 2016, 22, 2812 – 2818 www.chemeurj.org 2814 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Figure 2. Correlation between q2 and q1. The squares indicate the positions
of the lowest free energy structures for each molecularity. q2 and q1 are in æ.
Data taken from the B2PLYPD3/def2-TVZP geometries.
and r2 are the Cl¢H and O¢H distances, respectively). High
negative (positive) values of q1 correspond to hydrogen atoms
that are clearly attached to the chlorine (oxygen) atom, and q1
… 0 indicates partially transferred protons, as the hydrogen
atoms are equidistant from Cl and O. As in other proton trans-
fer cases,[72] a beautiful quadratic correlation is obtained re-
gardless of the number of water molecules present. Figure 2
provides solid evidence for proton transfer beyond subjective
visual inspection of the structures. Clearly, the lowest free
energy structures for n ‹ 4 are in the attached region, and
complete transfer is evident for n = 5.
Further evidence about the state of HCl dissociation in gas-
phase water clusters is provided in Figure 3; here, we analyze
the Wiberg bond indices for the H¢Cl interactions (as a refer-
ence, the computed bond index for isolated H¢Cl is 0.94) for
the same clusters as selected for Figure 2. Three distinct areas
can be discerned. The undissociated region covers the [0.8–
0.9] interval. There are two intervals of partial proton transfer:
[0.65–0.75], corresponding to early to intermediate stages of
the transfer process, and [0.25–0.45], corresponding to inter-
mediate to late stages of proton transfer. Finally, the [0.1–0.2]
interval encloses structures that all exhibit fully transferred pro-
tons. Figure 3 provides clear-cut evidence for proton transfer,
which is consistent with the results discussed previously in
Figure 2.
Dipole moment distribution
Guggemos and co-workers[20] wrote “we point out that the dy-
namically averaged dipole moment within a finite highly fluxio-
nal system can be qualitatively different from that computed for
a static minimum-energy framework”. We complement this
statement by adding that the differences may also be quantita-
tive. This issue is further complicated, because chemical and
physical processes are not exclusive domains of the lowest
energy structures in a given PES, and as mentioned above,
many more structures may still be missing because no explora-
Chem. Eur. J. 2016, 22, 2812 – 2818 www.chemeurj.org
Full Paper
Figure 3. Wiberg bond indices (WBI) for the H¢Cl interaction in clusters re-
ported in this work. Partially transferred protons appear in two categories:
the right peak shows those in the early stages of H¢Cl dissociation, and the
left peak comprises structures for advanced HCl dissociation. Data taken
from the B2PLYPD3/def2-TVZP optimized geometries.
tion of complex energy landscapes should be considered com-
plete. This is particularly applicable to our case because of the
large number of structural possibilities arising from the relative
positions of hydrogen atoms not involved in the stabilizing hy-
drogen-bonding networks, which change the individual dipole
moments significantly. Additionally, it is clear that the entropic
effects play a key role in determining the equilibrium structure
of the title clusters.
Using our calculated structures, we attempted to reproduce
the dipole moment distributions reported by Guggemos
et al.[20] Following the postulates of quantum mechanics, we
computed the expected dipole, < mn > , as the average over
dipole moments of all individual structures in the given PES,
min, weighted by xin, the estimated isomer populations on that
surface obtained from standard Boltzmann distributions of the
Gibbs free energies. The results at 298.15 K are plotted in
Figure 4. Experimental points were taken directly from the
work by Guggemos et al.[20]
There is a perfect match between the expected dipole
moment and the experimental value (Figure 4 A) only for n = 3.
For larger n, there is no clear correlation between the two
quantities. However, if < mn > are computed including only the
undissociated (or dissociated) structures, the picture changes
drastically. The expected dipole moments for the undissociated
forms increase considerably from n = 4 to n = 6, whereas
< mn > for the dissociated arrangements decrease in the same
range (see Figure 4 B). In the particular case of n = 4, the
weight of the undissociated structures is very low. Interestingly,
the expected dipole moments for the associated clusters are
very close to those obtained experimentally for n = 4 and 5.
On the contrary, for n = 6, the computed value for the dissoci-
ated forms is very close to that reported by Guggemos et al.
The n = 5 case is interesting. Besides our stochastically locat-
ed structures, we took the lowest DG isomer shown in
Figure 1 and optimized the molecular geometries for all possi-
ble combinations of the “above” and “under” relative positions
2815 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Figure 4. Expected dipole moments for the HCl(H2O)n clusters. A) Expected
dipole moments for all structures in a given PES. B) Expected dipole mo-
ments for undissociated and dissociated structures in a given PES. In both
cases the expected dipoles moments are estimated from standard Boltz-
mann distributions of the Gibbs free energies at 298.15 K, 1 atm.
of the four peripheral protons (16 combinations in total). The
resulting structures were energetically very close to the lowest
DG isomer, with DG values no higher than 0.43 kcal mol¢1 rela-
tive to the reference structure. Without these additional struc-
tures, isomers had populations with higher values, with other
geometries having smaller significance. Inclusion of the new
structures complicates the population picture with 11 isomers
now having contributions in the 3–6 % range. Individual dipole
moments are also more susceptible to changes in the orienta-
tion of peripheral protons: amongst the new structures, ex-
pected dipole moments cover the 2.46–5.09 D range, (com-
pared with 3.39 D for the reference structure). However, the ex-
pected averaged dipole moment only changed from 3.77 to
3.80 D. These results illustrate clearly the preceding discussion
regarding contributions from individual structures to the prop-
erties of complex energy landscapes, and at the same time
provide solid support for our approach to calculating expected
properties as statistical averages weighted by isomer popula-
tions derived from Boltzmann distributions.
What about the temperature effect on the dipole moment
distribution? Clearly, the isomer populations will change as
Chem. Eur. J. 2016, 22, 2812 – 2818 www.chemeurj.org
Full Paper
a result of a temperature variation, and this could affect the ex-
pected dipole moment. In this sense, it has been shown that
the number of water molecules needed to dissociate NaCl, for
example, is sensitive to temperature.[73] It is apparent from
Figure 5 that changes in temperature induce very strong
Figure 5. Variation in the expected dipole moments as a function of temper-
ature for the HCl(H2O)n clusters. The expected dipoles moments are estimat-
ed from standard Boltzmann distributions of the Gibbs free energies at
1 atm.
changes in the dipole moments. At very low temperatures, the
variation is almost negligible (except for n = 6). However, it is
not possible to predict a lowering or increasing of the dipole
moment as a consequence of the temperature even when the
population is separated into dissociated and associated forms
(see Supporting Information). Therefore, if the temperature is
not controlled during the experiment, the values obtained for
the dipole moments are susceptible to change, providing a dif-
ferent picture of the number of water molecules needed for
HCl dissociation in water. The main reason is that at high tem-
peratures, the number of hydrogen bonds decreases, and less
compact clusters are obtained. In other words, we recover the
gas behavior (no intermolecular interactions, and thus, no clus-
ter formation) at high temperatures.
Conclusion
On the basis of electronic energies, the number of water mole-
cules needed for HCl dissociation is four as reported by the
groups of Maillard,[24] Suhm,[26, 46] and others. However, on the
basis of Gibbs free energies, this number is five. For n = 4, the
lowest DG structure is also partially dissociated. The nondisso-
ciated, partially dissociated, and fully dissociated character of
the H¢Cl bond can be characterized unambiguously by Wiberg
bond indices, and supported by quadratic correlations be-
tween the distance between Cl and O atoms and the distance
from the proton to the center of the corresponding Cl···H···O
interaction. Very narrow energy windows for the partially and
fully dissociated structures in the corresponding PES preclude
the assignment of macroscopic experimental observables to in-
2816 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

dividual cluster structures. Although our search is comprehen-
sive and exhaustive, it is by no means complete because of
the large number of structural possibilities arising from the rel-
ative positions of peripheral hydrogen atoms in the clusters.
Paraphrasing Guggemos et al., we point out that the dynamical-
ly averaged dipole moment within a finite highly fluxional
system can be qualitatively and quantitatively different from that
computed for a static minimum-energy framework. It is manda-
tory to separate the undissociated from the dissociated struc-
tures in order to reproduce the overall experimental trend. Our
computations show that if temperature is not controlled
during the experiment, the values obtained for the dipole
moment (or for any measurable property) are susceptible to
change, providing a different picture of the number of water
molecules needed for HCl dissociation in water.
Acknowledgements
The Moshinsky Foundation, Tecnolûgico Nacional de M¦xico
(069-14-PD) and PRODEP ( ITESCAM-PTC-026) supported the
work in M¦xico. A. R. thanks Universidad de Antioquia for pro-
viding partial funding for this work via “Estrategia de sostenibi-
lidad”. AVC and JLC thank Conacyt for the PhD and postdoctor-
al fellowship, respectively.
Keywords: ab initio calculations · dissociation · global
optimization · hydrogen bonds · microsolvation
[1] D. Marx, ChemPhysChem 2006, 7, 1848 – 1870.
[2] D. Marx, ChemPhysChem 2007, 8, 209 – 210.
[3] W. Domcke, A. L. Sobolewski, Science 2003, 302, 1693 – 1694.
[4] S. M. Hurley, T. E. Dermota, D. P. Hydutsky, A. W. Castleman, Science
2002, 298, 202 – 204.
[5] M. Miyazaki, A. Fujii, T. Ebata, N. Mikami, Science 2004, 304, 1134 – 1137.
[6] W. H. Robertson, E. G. Diken, E. A. Price, J. W. Shin, M. A. Johnson, Sci-
ence 2003, 299, 1367 – 1372.
[7] W. H. Robertson, M. A. Johnson, Science 2002, 298, 69.
[8] J. W. Shin, N. I. Hammer, E. G. Diken, M. A. Johnson, R. S. Walters, T. D.
Jaeger, M. A. Duncan, R. A. Christie, K. D. Jordan, Science 2004, 304,
1137 – 1140.
[9] A. Stace, Science 2001, 294, 1292 – 1293.
[10] A. F. Voegele, K. R. Liedl, Angew. Chem. Int. Ed. 2003, 42, 2114; Angew.
Chem. 2003, 115, 2162 – 2164.
[11] J. M. Headrick, Science 2005, 309, 1326.
[12] J. M. Headrick, E. G. Diken, R. S. Walters, N. I. Hammer, R. A. Christie, J.
Cui, E. M. Myshakin, M. A. Duncan, M. A. Johnson, K. D. Jordan, Science
2005, 308, 1765 – 1769.
[13] T. S. Zwier, Science 2004, 304, 1119 – 1120.
[14] K. Ando, J. T. Hynes, Adv. Chem. Phys. 1999, 110, 381 – 430.
[15] K. Laasonen, M. L. Klein, J. Am. Chem. Soc. 1994, 116, 11620 – 11621.
[16] R. Bianco, J. T. Hynes, Acc. Chem. Res. 2006, 39, 159 – 165.
[17] J. P. Devlin, N. Uras, J. Sadlej, V. Buch, Nature 2002, 417, 269 – 271.
[18] T. Huthwelker, M. Ammann, T. Peter, Chem. Rev. 2006, 106, 1375 – 1444.
[19] V. F. McNeill, T. Loerting, F. M. Geiger, B. L. Trout, M. J. Molina, Proc. Natl.
Acad. Sci. USA 2006, 103, 9422 – 9427.
[20] N. Guggemos, P. Slavicek, V. V. Kresin, Phys. Rev. Lett. 2015, 114, 043401.
[21] Z. Kisiel, E. Bialkowska-Jaworska, L. Pszczûlkowski, A. Milet, C. Strunie-
wicz, R. Moszynski, J. Sadlej, J. Chem. Phys. 2000, 112, 5767 – 5776.
[22] D. Skvortsov, S. J. Lee, M. Y. Choi, A. F. Vilesov, J. Phys. Chem. A 2009,
113, 7360 – 7365.
[23] A. Gutberlet, G. Schwaab, ©. Birer, M. Masia, A. Kaczmarek, H. Forbert,
M. Havenith, D. Marx, Science 2009, 324, 1545 – 1548.
[24] C. Amirand, D. Maillard, J. Mol. Struct. 1988, 176, 181 – 201.
Chem. Eur. J. 2016, 22, 2812 – 2818 www.chemeurj.org
Full Paper
[25] V. E. Bondybey, M. Beyer, U. Achatz, S. Joos, G. Niedner-Schatteburg, Isr.
J. Chem. 1999, 39, 213 – 219.
[26] M. Farnik, M. Weimann, M. A. Suhm, J. Chem. Phys. 2003, 118, 10120 –
10136.
[27] A. A. Hassanali, J. Cuny, M. Ceriotti, C. J. Pickard, M. Parrinello, J. Am.
Chem. Soc. 2012, 134, 8557 – 8569.
[28] C. T. Lee, C. Sosa, M. Planas, J. J. Novoa, J. Chem. Phys. 1996, 104, 7081 –
7085.
[29] M. Letzner, S. Gruen, D. Habig, K. Hanke, T. Endres, P. Nieto, G. Schwaab,
L. Walewski, M. Wollenhaupt, H. Forbert, D. Marx, M. Havenith, J. Chem.
Phys. 2013, 139, 154304.
[30] S. Odde, B. J. Mhin, S. Lee, H. M. Lee, K. S. Kim, J. Chem. Phys. 2004, 120,
9524 – 9535.
[31] T. Schindler, C. Berg, G. Niedner-Schatteburg, V. E. Bondybey, Chem.
Phys. Lett. 1994, 229, 57 – 64.
[32] Y. K. Yoon, G. B. Carpenter, Acta Crystallogr. 1959, 12, 17 – 20.
[33] H. S. Rzepa, The Winnower 3:e142410.09115 (2015). DOI:10.15200/
winn.142410.09115.
[34] M. Masia, H. Forbert, D. Marx, J. Phys. Chem. A 2007, 111, 12181 – 12191.
[35] J. Zischang, D. Skvortsov, M. Yong Choi, R. A. Mata, M. A. Suhm, A. F. Vi-
lesov, J. Phys. Chem. A 2015, 119, 2636 – 2643.
[36] M. Onč‚k, P. Slav†ček, M. F‚rn†k, U. Buck, J. Phys. Chem. A 2011, 115,
6155 – 6168.
[37] A. M. Morrison, S. D. Flynn, T. Liang, G. E. Douberly, J. Phys. Chem. A
2010, 114, 8090 – 8098.
[38] V. Lesch, S. Jeremias, A. Moretti, S. Passerini, A. Heuer, O. Borodin, J.
Phys. Chem. B 2014, 118, 7367 – 7375.
[39] J. S. Mancini, A. K. Samanta, J. M. Bowman, H. Reisler, J. Phys. Chem. A
2014, 118, 8402 – 8410.
[40] Ł. Walewski, H. Forbert, D. Marx, ChemPhysChem. 2013, 14, 817 – 826.
[41] S. Re, Y. Osamura, Y. Suzuki, H. F. Schaefer, J. Chem. Phys. 1998, 109,
973 – 977.
[42] H. Forbert, M. Masia, A. Kaczmarek-Kedziera, N. N. Nair, D. Marx, J. Am.
Chem. Soc. 2011, 133, 4062 – 4072.
[43] E. M. Cabaleiro-Lago, J. M. Hermida-Ramûn, J. Rodr†guez-Otero, J. Chem.
Phys. 2002, 117, 3160 – 3168.
[44] J. S. Mancini, J. M. Bowman, Phys. Chem. Chem. Phys. 2015, 17, 6222 –
6226.
[45] K. R. Leopold, Annu. Rev. Phys. Chem. 2011, 62, 327 – 349.
[46] M. Weimann, M. Farnik, M. A. Suhm, Phys. Chem. Chem. Phys. 2002, 4,
3933 – 3937.
[47] N. Acelas, G. Hincapie, D. Guerra, J. David, A. Restrepo, J. Chem. Phys.
2013, 139, 044310.
[48] M. Saunders, J. Comput. Chem. 2004, 25, 621 – 626.
[49] D. Guerra, J. David, A. Restrepo, J. Comput. Methods Sci. Eng. 2014, 14,
93 – 102.
[50] J. K. Gregory, D. C. Clary, J. Phys. Chem. A 1997, 101, 6813 – 6819.
[51] C. Z. Hadad, A. Restrepo, S. Jenkins, F. Ramirez, J. David, Theor. Chem.
Acc. 2013, 132, 1376 – 1387.
[52] G. HincapieÈ, N. Acelas, M. Castano, J. David, A. Restrepo, J. Phys. Chem.
A 2010, 114, 7809 – 7814.
[53] J. F. P¦rez, C. Z. Hadad, A. Restrepo, Int. J. Quantum Chem. 2008, 108,
1653 – 1659.
[54] F. Ram†rez, C. Z. Hadad, D. Guerra, J. David, A. Restrepo, Chem. Phys.
Lett. 2011, 507, 229 – 233.
[55] C. P¦rez, M. T. Muckle, D. P. Zaleski, N. A. Seifert, B. Temelso, G. C.
Shields, Z. Kisiel, B. H. Pate, Science 2012, 336, 897 – 901.
[56] Y. Wang, V. Babin, J. M. Bowman, F. Paesani, J. Am. Chem. Soc. 2012,
134, 11116 – 11119.
[57] V. Babin, F. Paesani, Chem. Phys. Lett. 2013, 580, 1 – 8.
[58] W. Lin, F. Paesani, J. Phys. Chem. A 2013, 117, 7131 – 7141.
[59] T. Takayanagi, K. Takahashi, A. Kakizaki, M. Shiga, M. Tachikawa, Chem.
Phys. 2009, 358, 196 – 202.
[60] L. Walewski, H. Forbert, D. Marx, J. Phys. Chem. Lett. 2011, 2, 3069 –
3074.
[61] S. Sugawara, T. Yoshikawa, T. Takayanagi, M. Tachikawa, Chem. Phys. Lett.
2011, 501, 238 – 244.
[62] A. Milet, C. Struniewicz, R. Moszynski, P. E. S. Wormer, J. Chem. Phys.
2001, 115, 349 – 356.
[63] J. L. Cabellos, F. Ortiz-Chi, A. Ram†rez, G. Merino, Bilatu 1.0. Cinvestav,
M¦rida, 2013.
2817 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

[64] R. Grande-Aztatzi, P. R. Martinez-Alanis, J. L. Cabellos, E. Osorio, A. Marti-
nez, G. Merino, J. Comput. Chem. 2014, 35, 2288 – 2296.
[65] A. Ramirez-Manzanares, J. Pena, J. M. Azpiroz, G. Merino, J. Comput.
Chem. 2015, 36, 1456 – 1466.
[66] C. Adamo, V. Barone, J. Chem. Phys. 1999, 110, 6158 – 6170.
[67] L. Goerigk, S. Grimme, J. Chem. Theory Comput. 2011, 7, 291 – 309.
[68] S. Grimme, J. Chem. Phys. 2006, 124, 034108.
[69] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297 – 3305.
[70] M. J. Frisch, G. W. T., 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. Moroku-
ma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dap-
Chem. Eur. J. 2016, 22, 2812 – 2818 www.chemeurj.org 2818
Full Paper
prich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, 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. Nanayak-
kara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C.
Gonzalez, J. A. Pople; Gaussian, Inc.: Wallingford CT, 2004.
[71] H. H. Limbach, P. M. Tolstoy, N. P¦rez-Hern‚ndez, J. Guo, I. G. Shendero-
vich, G. S. Denisov, Isr. J. Chem. 2009, 49, 199 – 216.
[72] J. D. Gonzalez, E. Florez, J. Romero, A. Reyes, A. Restrepo, J. Mol. Model.
2013, 19, 1763 – 1777.
[73] C.-K. Siu, B. S. Fox-Beyer, M. K. Beyer, V. E. Bondybey, Chem. Eur. J. 2006,
12, 6382 – 6392.
Received: October 7, 2015
Published online on January 15, 2016
Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim