A High-Performance Mesoscale Molecular Dynamics Framework for Polar Ordering in Dipolar Liquid Crystals

Abstract Polar ordering in liquid crystals arises from the interplay between molecular anisotropy and long-range dipole–dipole interactions, making molecular dynamics simulations of formation kinetics computationally demanding due to the cost of evaluating these interactions. Here, we develop a high-performance mesoscale molecular dynamics framework for efficient large-scale simulations of dipolar liquid-crystal systems. The framework incorporates GPU- and DCU-accelerated algorithms for efficiently computing anisotropic Gay-Berne forces and long-range dipole–dipole interactions, and supports generalized Gay-Berne particles carrying an arbitrary number of embedded dipoles with freely assignable positions and orientations. The algorithm employs a multithreaded parallelization strategy with adaptive optimization, yielding significant speedups on heterogeneous computing platforms. We validate the framework against representative ferroelectric and antiferroelectric systems, faithfully reproducing known phase behavior while maintaining computational efficiency. This work provides a versatile molecular dynamics framework for investigating polar ordering and its formation kinetics across a wide range of liquid-crystal systems and dipolar assemblies.