Energy Conservation in Adaptive Hybrid Atomistic/Coarse-Grain Molecular Dynamics

Area of Science:

• Computational chemistry
• Molecular dynamics simulations
• Multiscale modeling

Background:

• Complex molecular systems require multiscale simulations spanning diverse time and length scales.
• Coupling different descriptive levels in spatial regions presents a significant challenge (the 'hand-shaking' problem).
• Dynamically changing material representation across boundaries in adaptive multiscale simulations is technically difficult but highly desirable.

Purpose of the Study:

• To present a novel molecular dynamics simulation algorithm that is multiscale in both time and space.
• To address the challenge of energy conservation in simulations where the number of degrees of freedom changes dynamically.
• To enable the tuning of adaptive hybrid algorithms for optimized performance.

Main Methods:

• Developed a multiscale molecular dynamics simulation algorithm.
• Supplemented potential and kinetic energy expressions with auxiliary terms.
• Ensured total energy conservation despite changes in the total number of degrees of freedom.

Main Results:

• The algorithm successfully recovers total energy as a conserved quantity.
• Demonstrated the ability to handle dynamic changes in the number of degrees of freedom.
• Provided a method for assessing and optimizing adaptive hybrid algorithms.

Conclusions:

• The presented algorithm offers a robust solution for multiscale molecular dynamics simulations.
• It effectively addresses the critical issue of energy conservation in adaptive hybrid methods.
• Facilitates the optimization of simulation parameters for improved accuracy and efficiency.