Merging metadynamics into hyperdynamics: accelerated molecular simulations reaching time scales from microseconds to...

Area of Science:

• Computational Chemistry
• Materials Science
• Chemical Physics

Background:

• Simulating slow atomic-level processes is computationally challenging.
• Existing hyperdynamics methods require complex potential construction.
• Accelerated molecular dynamics is crucial for studying complex chemical and physical phenomena.

Purpose of the Study:

• To develop a generally applicable and versatile hyperdynamics algorithm.
• To enhance the efficiency and simplicity of simulating slow atomic processes.
• To create a self-learning accelerated molecular dynamics method.

Main Methods:

• Implementation of a collective variable-driven hyperdynamics (CVHD) algorithm.
• Integration of metadynamics principles for on-the-fly bias potential construction.
• Modular design allowing customization of local system properties and biasing methods.

Main Results:

• Demonstrated applicability on diverse model systems: surface diffusion, catalytic decomposition, and polymer folding.
• Achieved significant acceleration (boost factors up to 10^9), enabling simulations of seconds-long dynamics.
• Accurate reproduction of system dynamics was confirmed.

Conclusions:

• The CVHD method offers a versatile, transparent, and self-learning approach to accelerated molecular dynamics.
• It simplifies the simulation of slow processes by abstracting system-specific details from the biasing algorithm.
• CVHD significantly advances the capability to study complex atomic-level phenomena.