Section: New Results

Surface chemistry and electrochemistry

Participant : Ismaila Dabo.

Rising energy imperatives have revived strong interest in electrochemical cells (e.g., fuel cells) and photoelectrochemical cells (e.g., dye-sensitized solar cells). Understanding and improving the electrical performance of such systems entails the accurate description of electrode-electrolyte interfaces at the microscopic level. Despite recent advances in the application of computational methods to study realistic electrode-electrolyte interfaces, capturing the effect of the applied electrical voltage and solvent electrical response remains a central challenge in computational electrochemistry. In order to address this difficulty, a comprehensive model for chemical systems embedded in ionic media has been developed. The model couples a quantum description of the electrode with a continuum representation of the electrolyte.

This year, in collaboration with O. Andreussi (MIT) and N. Marzari (University of Oxford), I. Dabo has worked on improving the predictive accuracy and numerical implementation of the continuum solvation model to describe molecular systems and metallic surfaces in interaction with a solvent [14] . The revised model overcomes some of the numerical limitations encountered in existing solvation models and extends their range of applicability. The approach proceeds by recasting the problem in terms of induced polarization charges that act as a direct mapping of the continuum dielectric solvent. The model is defined in a self-consistent manner in terms of the first-principles electronic density of the solute, thereby limiting the number of numerical parameters involved in existing solvation methods. The model accounts for additional pressure and cavitation contributions. The resulting self-consistent continuum solvation (SCCS) model provides an effective, compact fit of computational and experimental data, with solvation energies in error of 0.3-0.4 kcal/mol. The model is implemented in the widely used open-source program Quantum-Espresso , exploiting a numerical approach that is intrinsically parallel, robust, and straightforward to adapt to most electronic-structure codes.

In terms of practical applications in surface science and electrochemistry, the adsorption of arsenate (a severe contaminant in drinking water) at the surface of biogeochemical minerals has been studied in collaboration with M. Blanchard and G. Morin (Université Paris 6) [22] . The simulation of infrared spectroscopic experiments for carbon monoxide (a severe catalytic poison) at catalytic surfaces has also been addressed in [33] . Future work includes the generalization of the SCCS model to surfaces in contact with a reservoir of electrons, and molecular dynamics simulations of electrode surfaces.