Multiscale reactive molecular dynamics.

Area of Science:

• Computational Chemistry
• Materials Science
• Biophysics

Background:

• Multiscale problems in chemistry, materials science, and biology require accurate methods for electronic and nuclear dynamics coupled to slower motions.
• Current methods struggle with computational efficiency and accuracy for these complex systems.

Purpose of the Study:

• To develop a computationally efficient and accurate method for multiscale modeling.
• To create reactive molecular dynamics models directly from ab initio simulations without empirical functions.

Main Methods:

• Utilized data from condensed phase ab initio simulations.
• Developed reactive models using linear combinations of interpolating functions optimized via linear least-squares.
• Minimized parameters requiring nonlinear optimization.

Main Results:

• Generated reactive models for hydrated excess proton and hydroxide ions.
• Faithfully reproduced water-ion structural properties and diffusion constants from ab initio simulations.
• Accurately reproduced free energy profiles for proton transfer.

Conclusions:

• The new method offers orders of magnitude greater computational efficiency than ab initio methods.
• Enables accurate modeling of general chemical reactions in condensed phase systems.
• Facilitates statistical sampling of couplings to large-scale system motions.