Trio of improvements allows for fast, realistic molecular simulations of electrochemical systems

Researchers use molecular simulations to evaluate processes occurring at solution-electrode interfaces in electrochemical devices. Classical Density Functional Theory (DFT) is an efficient way to simulate electrochemical systems, but its approximations are not realistic descriptions of electrochemical systems. Molecular Dynamics (MD) simulations are more realistic, but this method is slower because it has a higher computational cost.

Jeanmairet et al. proposed a new way to do calculations for molecular simulations of graphene electrodes separated by liquid water that offers accuracy comparable to MD and computational cost on par with DFT. They improved classical DFT by removing three approximations: First, the authors described electrodes with realistic atomic resolution. They also described water with a realistic molecular model. Finally, instead of having a uniform charge density on each electrode, they fixed the potential difference between electrodes, which is how actual experiments are conducted.

When the authors applied these three changes to a water-graphene capacitor system, they found their calculations remained within the efficient framework of DFT. Their results provided insight into how the water behaves at the graphene interface, and how applied voltage affects water. Their results also agree with a MD simulation of the same system.

This study is one of the first steps toward simulating realistic electrochemical systems with classical DFT. Next, the authors want to simulate electrochemical cells containing ions. Author Guillaume Jeanmairet said the numerical efficiency of their new method would allow for systematic studies of energy storage devices or energy production devices, which is out of reach for the relatively expensive MD approach.

Source: “Study of a water-graphene capacitor with molecular density functional theory,” by Guillaume Jeanmairet, Benjamin Rotenberg, Daniel Borgis, and Mathieu Salanne, Journal of Chemical Physics (2019). The article can be accessed at https://doi.org/10.1063/1.5118301 .