Classical reactive molecular dynamics implementations: state of the art.

Area of Science:

• Computational Materials Science
• Chemical Physics
• Molecular Modeling

Background:

• Reactive Molecular Dynamics (RMD) is crucial for simulating material behavior, including chemical reactions and time evolution.
• Existing reviews by Srivastava and Garrison (1995) and Brenner (2000) provide foundational context for RMD methods.
• The development of force field approaches has enabled RMD to model complex chemical transformations in materials.

Purpose of the Study:

• To review and categorize reactive molecular dynamics (RMD) implementations developed over the last decade and earlier.
• To classify RMD methods into two distinct categories based on their approach to simulating chemical reactions.
• To highlight the applications and capabilities of different RMD classes in materials science and chemistry.

Main Methods:

• Classification of RMD methods into two classes: those using a reaction cutoff distance with sudden transitions, and those employing empirical reactive force fields with smooth transitions.
• Focus on RMD implementations that simulate time evolution and chemical reactions, excluding methods without time evolution (e.g., ab initio, DFT, MC) and certain hybrid QM/MM methods.
• Discussion of force field types used, including many-body, bond-order, and modified standard force fields (e.g., CHARMM, AMBER, MM3).

Main Results:

• The first RMD class, using cutoff distances, is suitable for generating equilibrated polymer structures and provides qualitative insights into polymerization.
• The second RMD class, based on empirical reactive force fields, allows for smooth transitions and is aimed at investigating reaction kinetics, mechanisms, and transition states.
• RMD methods reviewed are specifically those that simulate the time evolution of chemical systems, distinguishing them from static computational chemistry approaches.

Conclusions:

• RMD methods offer powerful tools for understanding material reactions at the atomic and molecular levels.
• The choice of RMD approach depends on the specific research question, ranging from structural generation to detailed mechanistic studies.
• Continued development in reactive force fields enhances the accuracy and scope of RMD simulations for chemical reactions in materials.