Adaptively Accelerating Reactive Molecular Dynamics Using Boxed Molecular Dynamics in Energy Space.

Area of Science:

• Computational chemistry
• Molecular dynamics
• Chemical kinetics

Background:

• Observing rare events in molecular dynamics is challenging due to long timescales.
• Current rare event acceleration methods often require pre-defined event specifications.
• Previous work introduced boxed molecular dynamics in energy space for chemical master equations.

Purpose of the Study:

• To adapt and apply the boxed molecular dynamics algorithm to the energy space of classical trajectory simulations.
• To accelerate the observation of generic rare events, such as chemical reactions.
• To validate the accuracy of the accelerated simulations by comparing product ratios.

Main Methods:

• Application of the boxed molecular dynamics in energy (BXDE) algorithm to classical trajectory simulations.
• Simulating rare events, specifically chemical reactions, using the BXDE approach.
• Comparison of simulation results with unbiased simulations at identical temperatures.

Main Results:

• The BXDE approach accelerated the observation of rare events by multiple orders of magnitude compared to unbiased simulations.
• Product ratios obtained from BXDE simulations were found to be consistent with those from unbiased simulations.
• Demonstrated the effectiveness of BXDE for accelerating generic rare events in molecular dynamics.

Conclusions:

• The boxed molecular dynamics in energy (BXDE) method is effective for accelerating rare events in molecular dynamics simulations.
• BXDE provides a significant speedup for simulating chemical reactions without compromising the accuracy of product distributions.
• This approach offers a powerful tool for studying complex molecular processes that occur over long timescales.