Force-Free Identification of Minimum-Energy Pathways and Transition States for Stochastic Electronic Structure...

Area of Science:

• * Computational Chemistry
• * Quantum Mechanics
• * Physical Chemistry

Background:

• * Accurate potential energy surface (PES) mapping is vital for understanding chemical reactions and conformational changes.
• * Stochastic electronic structure theories like Quantum Monte Carlo (QMC) offer high accuracy but struggle with force and Hessian calculations needed for pathway identification.
• * Traditional methods for minimum-energy pathway (MEP) and transition state (TS) identification often rely on computationally expensive force calculations.

Purpose of the Study:

• * To develop a force-free QMC approach for efficient identification of MEPs and TSs.
• * To enable accurate PES mapping without direct computation of forces at the QMC level.
• * To introduce a hybrid DFT-QMC method for improved thermodynamic and kinetic calculations.

Main Methods:

• * Utilization of a surrogate Hessian line-search method adapted for QMC structural optimization.
• * Modification of the surrogate Hessian algorithm to operate in path-orthogonal subspaces and at saddle points.
• * Development of a hybrid DFT-QMC scheme for calculating thermodynamic and kinetic properties.

Main Results:

• * Successful identification of MEPs and TSs for ammonia inversion and S_N2 reaction using a force-free QMC approach.
• * Validation of QMC results against established Density Functional Theory (DFT) and Coupled Cluster (CCSD, CCSD(T)) methods.
• * Demonstrated improvement in accuracy for thermodynamic and kinetic calculations using the hybrid DFT-QMC scheme.

Conclusions:

• * The developed force-free QMC strategy enables efficient and accurate PES mapping and TS determination.
• * This methodology significantly reduces computational cost for high-accuracy electronic structure calculations.
• * The approach is generalizable to other systems and high-accuracy theories facing gradient computation challenges.