Box length search algorithm for molecular simulation of systems containing periodic structures

Area of Science:

• Computational Chemistry
• Materials Science
• Statistical Mechanics

Background:

• Accurate molecular simulations require appropriate box dimensions, especially for periodic structures.
• Constant-volume simulations are crucial for quantitative comparisons with experimental and theoretical data for incompressible systems.
• Traditional methods for determining box dimensions can be computationally intensive.

Purpose of the Study:

• To develop an efficient algorithm for determining optimal box dimensions in constant-volume molecular simulations.
• To ensure quantitative comparability of simulation results by maintaining a fixed total volume.
• To address systems with preferred dimensions, such as phase-separated block copolymers.

Main Methods:

• Developed a box length search algorithm that equalizes pressure in each direction while preserving constant total volume.
• Implemented and tested a Monte Carlo algorithm capable of adjusting box lengths under constant volume conditions.
• Validated the algorithm's performance on a phase-separated block copolymer system.

Main Results:

• The developed box length search algorithm converges significantly faster (over two orders of magnitude) than the variable box length Monte Carlo method.
• The algorithm successfully identifies the box length that minimizes the system's free energy.
• Despite not being ergodic, the algorithm provides accurate results for the target systems.

Conclusions:

• The novel box length search algorithm offers a highly efficient approach for constant-volume molecular simulations of periodic structures.
• This method facilitates reliable quantitative comparisons by maintaining constant volume.
• The algorithm is effective in finding optimal box dimensions that minimize free energy, even for complex systems.