An ab initio approach to free-energy reconstruction using logarithmic mean force dynamics

Area of Science:

• Computational Chemistry
• Physical Chemistry
• Molecular Dynamics

Background:

• Accurate free energy profiles are crucial for understanding chemical reactions and molecular behavior.
• Traditional methods often rely on empirical force fields, which can limit accuracy and transferability.
• First-principles calculations offer higher accuracy but can be computationally expensive for dynamics simulations.

Purpose of the Study:

• To introduce and validate a novel ab initio approach for free energy profile calculations.
• To combine the strengths of first-principles molecular dynamics and logarithmic mean force dynamics (LogMFD).
• To assess the accuracy and applicability of the new method compared to existing techniques.

Main Methods:

• Developed a first-principles LogMFD (FP-LogMFD) method.
• Utilized density functional theory (DFT) to estimate mean forces.
• Applied FP-LogMFD to a glycine dipeptide system for free energy profile reconstruction.

Main Results:

• Successfully reconstructed one- and two-dimensional free energy profiles for glycine dipeptide using DFT.
• Demonstrated that FP-LogMFD provides accurate free energy profiles.
• Validated the method by comparing results with thermodynamic integration and empirical force-field based LogMFD.

Conclusions:

• FP-LogMFD is a promising ab initio method for calculating free energy profiles.
• The approach eliminates the need for empirical force fields, enhancing accuracy.
• This method offers a robust tool for computational studies in chemistry and related fields.