A growing string method for determining transition states: comparison to the nudged elastic band and string methods

Area of Science:

• Computational Chemistry
• Chemical Physics

Background:

• Calculating minimum energy pathways and transition states is crucial for understanding chemical reactions.
• Existing methods like nudged elastic band (NEB) and string methods require an initial pathway guess, which can lead to convergence issues.

Purpose of the Study:

• To introduce a novel growing string method for determining minimum energy pathways and transition states.
• To eliminate the need for an initial pathway guess in reaction pathway calculations.

Main Methods:

• The growing string method starts with two separate string fragments for reactants and products.
• These fragments grow independently until they converge and join.
• The combined string then refines the pathway using the string method algorithm.

Main Results:

• The growing string method successfully identified the saddle point for the alanine dipeptide rearrangement.
• It required significantly fewer electronic structure force calculations compared to the standard string method and NEB.
• This demonstrates the method's efficiency, especially when initial guesses are suboptimal.

Conclusions:

• The growing string method offers a robust and efficient alternative for calculating reaction pathways.
• It overcomes the limitations of initial guess dependency in traditional methods.
• This advancement can accelerate computational studies of chemical reactions.