Non-self-consistent Density-Functional Theory Exchange-Correlation Forces for GGA Functionals.

Area of Science:

• Computational Chemistry
• Biomolecular Modeling
• Materials Science

Background:

• Density Functional Theory (DFT) is vital for accurate biomolecular property prediction.
• Generalized Gradient Approximation (GGA) functionals are often required for high accuracy.
• Non-self-consistent (NSC) methods offer potential computational savings.

Purpose of the Study:

• To derive and implement a GGA-compatible NSC force expression for DFT.
• To investigate the utility of NSC structural relaxations for improving convergence.
• To assess the computational efficiency of NSC pre-relaxations for biomolecules.

Main Methods:

• Derivation of GGA NSC force expression for local orbital methods.
• Implementation within the Conquest linear scaling DFT code.
• Application of NSC structural relaxations prior to full self-consistent relaxations.

Main Results:

• A novel GGA NSC force expression was derived and implemented.
• NSC pre-relaxations were found to improve convergence of structural relaxations.
• Calculations on glycine and alanine peptides demonstrated reduced computational effort.

Conclusions:

• GGA NSC force calculations are feasible and beneficial for DFT.
• NSC pre-relaxations offer a practical strategy to accelerate convergence.
• This approach can significantly reduce computational cost in biomolecular simulations.