Structure and thermodynamics of nondipolar molecular liquids and solutions from integral equation theory

ABSTRACT

Solvent-induced solute polarisation of nondipolar solvents originates mainly from specific directional interactions and higher electrostatic multipole moments. Popular continuum solvation models such as the polarisable continuum models ignore such interactions and, therefore, cannot adequately model solvation effects on electronic structure in these environments. Important examples of nondipolar solvents that are indistinguishable by continuum methods are benzene and hexafluorobenzene. Both substances have very similar macroscopic properties, while solutes dissolved in either benzene or hexafluorobenzene behave differently due to their inverted electrostatic quadrupole moments and slightly different size. As a first step towards a proper and computationally feasible description of nondipolar molecular solvents, we present here integral equation theory results based on various forms of the reference interaction site model coupled to quantum-chemical calculations for benzene and hexafluorobenzene solutions of small molecules. We analyse solvation structures, also in comparison with molecular dynamics simulations, and show that predictions of transfer Gibbs energies, which define partition constants, benefit substantially from considering the exact, wave function-derived electrostatic field distribution beyond a simple point charge solute model in comparison with experimental data. Moreover, by constructing artificial uncharged and charge-inverted toy models of the solvents, it is possible to dissect the relative importance of dispersion and quadrupolar electrostatic effects on the partitioning equilibria. Such insight can help to design specifically optimised solvents to control solubility and selectivity for a wide range of applications.

KEYWORDS:

• RISM
• integral equations
• benzene
• hexafluorobenzene
• dispersion
• quadrupolar solvation

Acknowledgments

Figures 1, 5, 6 and 10 have been created with the Molcad software (http://www.molcad.de). We thank the Stiftung Mercator for financial and the IT and Media Center (ITMC) of the TU Dortmund for computational support. The authors declare no competing financial interests.

Supplemental data for this article can be accessed at http://dx.doi.org/10.1080/00268976.2016.1167266.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Stiftung Mercator [grant number Pr-2010-0002].