Intermolecular bonding in O2-O2: evolution from the gas to high

CODECS 2013 Workshop. San Lorenzo de El Escorial, Madrid, 18th –22nd April, 2013
Intermolecular bonding in O2-O2: evolution from the gas to high
pressure solid phases
M. I. Hernándeza, E. Carmona-Novilloa, M. Bartolomeia, J. Campos-Martíneza and
R. Hernández-Lamonedab
a
Instituto de Física Fundamental (IFF-CSIC), C/ Serrano 123, Madrid 28006, Spain
b
Centro de Investigaciones Químicas (UAEM), 62210, Cuernavaca, Mor. Mexico
[email protected]
The nature of the bonding between two oxygen molecules has been a subject of debate since
the conjecture by Lewis of the formation of a singlet O4 especies to explain the magnetic
properties of liquid oxygen.1 Indeed, a O2-O2 dimer more strongly bonded than a typical van der
Waals (vdW) molecule could be expected, given the open-shell character of the ground state of
the monomers (3g-). Experiments in the gas phase have shown that the dimer behaves as a vdWlike complex with a rectangular geometry. This bonding in the gas phase must also play a role in
the condensed phases, since it is known that the closest neighbors in the lower pressure phases of
molecular oxygen are also arranged forming a rectangular shape.2
In this communication, we report recent accurate calculations of the rovibrational bound
states of O2-O2 in its singlet electronic state,3 based in high level multiconfigurational ab initio
calculations of the potential energy surface.4 The vibrational ground state has a rectangular
geometry, in agreement with previous studies, but it is found that it behaves as a semi-rigid
molecule with a hindered internal rotation (Fig.1). This feature is due to exchange interactions
that become rapidly smaller as the configuration of the dimer departs from the rectangular
geometry. In spite of this, such interactions are not strong enough so as to lead to chemical
bonding. Comparison with high resolution measurements5 is puzzling, indicating the need of
further studies: there is an excellent agreement regarding rotational constants but the calculated
frequency of the torsional mode is larger than the experimental determinations.
We have also investigated6 the modification of the O2-O2 intermolecular bonding when the
molecular oxygen is a solid under a very high pressure (tens of Gpa), as in the phase. It has
been discovered that this phase consists of layers of well defined clusters composed of four
oxygen molecules.7,8 Based on multiconfigurational ab initio calculations of the (O2)4 complex in
its singlet state, we have considered a simple model of the cluster surrounded by oxygen
molecules (mimicking a layer of -O2) and, with this, we have been able to reproduce the
observed7 dependence with pressure of the intra- and inter-cluster distances. Several analyses6,9
indicate that, for geometrical sizes of the (O2)4 complex close to those encountered in the solid
phase, an incipient intermolecular chemical bond is formed. This is a key point for the
understanding of several of the characteristic features of the  phase.
CODECS 2013 Workshop. San Lorenzo de El Escorial, Madrid, 18th –22nd April, 2013
Fig. 1. Variation of the interaction energy between two O2 molecules near the equilibrium geometry, with
respect to the torsional angle. The two lowest vibrational bound energy levels are shown. Internal
(torsional) rotation is hindered for the ground state.
Acknowledgments
The work has been funded by the Spanish Grant FIS2010-22064-C02-02. We also thank the
COST-CMTS Action CM1002 “CODECS” for support and CESGA for allocation of computing
time.
References
1. G. N. Lewis, J. Am. Chem. Soc. 46, 2027 (1924).
2. Y. A. Freiman and H. J. Jodl, Phys. Rep. 401, 1 (2004).
3. E. Carmona-Novillo, M. Bartolomei, M. I. Hernández, J. Campos-Martínez, and R. HernándezLamoneda, J. Chem. Phys. 137, 114304 (2012).
4. M. Bartolomei, E. Carmona-Novillo, M. I. Hernández, J. Campos-Martínez, and R. HernándezLamoneda, J. Chem. Phys. 133, 124311 (2010).
5. L. Biennier, D. Romanini, A. Kachanov, A. Campargue, B. Bussery-Honvault, and R. Bacis, J. Chem.
Phys. 112, 6309 (2000).
6. M. Bartolomei, E. Carmona-Novillo, M. I. Hernández, J. Pérez-Ríos, J. Campos-Martínez, and R.
Hernández-Lamoneda, Phys. Rev. B 84, 092105 (2011)
7. H. Fujihisa, Y. Akahama, J. Kawamura, Y. Ohishi, O. Shimomura, H. Yamawaki, M. Sakashita, Y.
Gotob, S. Takeya, and K. Honda, Phys. Rev. Lett. 97, 085503 (2006).
8. L. F. Lundegaard, G. Weck, M. I. McMahon, S. Desgreniers, and P. Loubeyre, Nature 443, 201 (2006).
9. M. García-Revilla, E. Francisco, A. Martín Pendás, J. M. Recio, M. Bartolomei, M. I. Hernández, J.
Campos-Martínez, E. Carmona-Novillo, and R. Hernández-Lamoneda, J. Chem. Theory Comput., in
press (2013).