Transiting the molecular potential energy surface along low energy pathways: the TRREAT algorithm

Area of Science:

• Computational chemistry
• Molecular modeling
• Chemical kinetics

Background:

• Characterizing molecular pathways is crucial for understanding chemical reactions and conformational changes.
• Existing methods may require predefined collective variables or struggle with high-dimensional spaces.
• Accurate identification of transition states is essential for building reliable kinetic models.

Purpose of the Study:

• Introduce the Transition Rapidly exploring Random Eigenvector Assisted Tree (TRREAT) algorithm.
• Apply TRREAT to identify conformational changes in molecular systems without predefined collective variables.
• Demonstrate TRREAT's utility in improving pathway identification for transition state calculations and reaction mechanism generation.

Main Methods:

• TRREAT combines local potential energy surface (PES) curvature with an iterative Rapidly exploring Random Tree (RRT) algorithm.
• The method utilizes Cartesian coordinates for pathway searches, avoiding a priori definition of collective variables.
• Pathway identification was analyzed for alanine dipeptide, cyclohexane, and glycine using various force fields.

Main Results:

• TRREAT successfully identified low-curvature pathways on the PES for molecular systems.
• The identified pathways serve as effective input for double-ended methods like the Nudged Elastic Band (NEB) to determine transition state energies.
• The algorithm demonstrated its capability in handling both nonreactive and reactive force fields.

Conclusions:

• TRREAT offers an efficient approach for exploring pathways on PES, particularly for conformational changes.
• The method enhances the accuracy and scope of transition state energy calculations.
• TRREAT can significantly improve reaction databases used in automated chemical reaction mechanism generators for kinetic modeling.