Chemical Bonding in Large Systems Using Projected Population Analysis from Real-Space Density Functional Theory...

Area of Science:

• Computational materials science
• Quantum chemistry
• Solid-state physics

Background:

• Extracting chemical bonding information from large-scale density functional theory (DFT) calculations is computationally challenging.
• Existing methods may struggle with systems containing thousands of atoms or diverse boundary conditions (periodic, semiperiodic, nonperiodic).

Purpose of the Study:

• To present an efficient and scalable computational approach for projected population analysis within real-space finite-element (FE)-based Kohn-Sham DFT (DFT-FE).
• To enable the extraction of detailed chemical bonding information from large-scale materials simulations.

Main Methods:

• Derivation of mathematical expressions for projected overlap and Hamilton populations.
• Development of scalable numerical implementation procedures for multinode CPU architectures.
• Projection of FE-discretized Kohn-Sham orbitals or Hamiltonian onto localized atom-centered basis sets within the DFT-FE code.

Main Results:

• Implementation of a unified framework for ground-state DFT calculations and population analysis on the same FE grid.
• Benchmarking against the LOBSTER code demonstrates accuracy and performance for periodic and nonperiodic systems.
• Successful application to a case study of hydrogen chemisorption in silicon-carbon nanoparticles.

Conclusions:

• The developed DFT-FE approach offers a scalable and efficient method for quantitative chemical bonding analysis in large material systems.
• This facilitates the investigation of complex materials, such as those relevant for hydrogen storage.
• The unified framework streamlines the process from DFT calculation to bonding analysis.