Using molecular simulations to study anode surfaces

Using molecular simulations to study anode surfaces

Experiments have so many factors affecting their outcomes. Even with no contamination, we typically don’t fully understand the reactions occurring and the transfer and movement of ions and molecules in the battery. Molecular simulations can help simplify our system and allow us to study individual aspects of the battery.

I use molecular dynamics (MD) to study surface mobility in a Na-K anode at the anode-electrolyte interface. Sodium metal anodes often suffer from needle- or mossy-like structures called dendrites that form from irregular deposition of sodium ions. Dendrites reduce the battery’s capacity by using up active material and can penetrate through the separator, causing a short circuit. If surface mobility is increased, ions can migrate away from these dendrite structures more easily, helping alleviate the issue. When potassium is added to sodium, the melting temperature is significantly reduced, increasing the mobility of atoms. We want to study diffusion and component and phase separation in this Na-K system in detail using MD.

The main user input in MD is an interatomic potential and initial atom positions and velocities. From these inputs, the force acting on each atom can be computed. Then, using Newton’s laws of motions over a small time, the new atom positions and velocities can be calculated. Interatomic potentials follow a specific function or equation form and are typically fit to experimental data or quantum calculations. For metals, potentials of the embedded atom model family are often used. This model uses the idea that an atom is embedded or placed in an existing electron cloud, formed by the electrons belonging to the other atoms in the system. You might have learned in chemistry class that metal bonding is like having metal atoms in a sea of electrons – this is that same idea! I have investigated pure Na and K surface melting and mobility using embedded atom model potentials published by other researchers and am working on developing a Na-K potential to be able to accurately study the Na-K anode surface.

atoms in a box sea of electrons