Coupling Between Moiré Patterns in Twisted Bilayer Graphene and Nanofluidics: A Deep Neural Network Approach

Twisted bilayer graphene at the magic angle has garnered attention due to its intriguing superconductivity and correlated insulator behavior resulting from strong electronic correlations. However, the impact of the electronic properties due to Moiré Patterns in twist bilayer graphene on the structural and dynamic properties of water remains largely unexplored. This knowledge gap stems from computational challenges associated with simulating large unit cells using density functional theory. In this study, we present an approach utilizing a deep neural network potential (DP) model. The DP model is trained using a dataset obtained from ab initio molecular dynamics simulations of water on various large twist angle bilayer graphene. Our DP model accurately characterizes key water properties, such as OH bond length, HOH bond angle, and power spectra, on top of magic angle twisted bilayer graphene. Leveraging this model, we investigate the structural and dynamical properties of water on bilayer graphene with various twist angles, ranging from 1.08° to 9.43°. By analyzing the effects of these twist angles and their corresponding electronic properties, we gain insights into the nanofluidic behavior of water. This exploration opens avenues for future research, focusing on harnessing the unique properties of twisted bilayer graphene to control and optimize nanofluidic behavior.