Single-ended transition state finding with the growing string method.

Area of Science:

• Computational Chemistry
• Chemical Dynamics

Background:

• Reaction path finding and transition state (TS) searching are crucial in computational chemistry.
• Existing double-ended string methods require known reactant and product structures, limiting systematic exploration.

Purpose of the Study:

• Introduce a novel single-ended growing string method (GSM) for reaction path finding.
• Enable reaction path searches starting from a single structure.
• Enhance computational efficiency and automate reactivity exploration.

Main Methods:

• The single-ended GSM sequentially adds nodes along internal coordinates (bonds, angles, torsions) to map reaction paths.
• String optimization refines the path and locates the exact transition state.
• Utilizes internal coordinates, eigenvector optimization, and Hessian estimates for fast convergence.

Main Results:

• The single-ended GSM demonstrates computational efficiency comparable to or exceeding double-ended methods.
• Successfully applied to transition metal reactivity and automated searches for unknown reactions.
• An automated search identified 165 reaction paths for NH3 BH3 and (LiH)4 without user input.

Conclusions:

• The single-ended GSM offers a more flexible and efficient approach to reaction path finding and TS searching.
• This method significantly advances automated exploration of chemical reactivity.
• The approach is robust and applicable to complex chemical systems.