2024 Winter Molecular Simulation Seminar
Room 302, the 2nd experimental building at POSTECH
Presenter: Sangmin Lee
Understanding the behavior of excess protons in aqueous solution (H3O+) is deeply involved with the integral part of energy conversion in biology and technology. Especially, the transport properties of H3O+ turns out to be much more interesting with a diffusion coefficient that is roughly 1 order of magnitude larger than that of other small cations [1]. While the latter undergo vehicular diffusion through Brownian motion, the driving force of fast proton transport in aqueous solution is originated from both Brownian motion and Grotthuss diffusion. The Grotthuss mechanism allows excess protons in system to rapidly migrate through water by swapping covalent bonds for hydrogen bonds. Although computational modeling of these Grotthuss events is expected to play an important role in uncovering many atomistic details, classical MD simulation based on conventional force field has its own inability to describe the bond breaking/formation process. As such, a handful of MD studies only treat the H3O+ ions as Brownian particles without Grotthuss mechanism and without allowing for protonation/deprotonation events. It should be emphasized that a number of different computational approaches were previously used to dynamically simulate proton transport based on first principle calculations. However, the computational overhead of such approaches severly limited the spatiotemporal scale which can be accessible through simulation, covering time scales of 10-20 ps simulation time. This warrants the accurate and efficient simulation techniques that enable the description of Grotthuss mechanism in diverse range of chemical system with extended spatiotemporal scale.
In this talk, I'll discuss the reactive MD simulation techniques used for describing proton transport. The most commonly used technique is to incorporate the MC steps in MD simulation that attempts to break/form the bond on the fly [2,3]. The othe approaches will be discussed as well such as fluctuating charge RMD (FCRMD), lambda dynamics [4, 5], and coarse-grained RMD [6]. Lastly, I'll briefly discuss the issues regarding hybrid MCMD such as low acceptance ratio, breaking of the detailed balance condition, and kinetic control of the reaction rate.