Pseudo-fragment approach for extended systems derived from linear-scaling DFT

Area of Science:

• Computational physics and chemistry
• Materials science
• Condensed matter theory

Background:

• Density functional theory (DFT) is crucial for materials simulation.
• Simulating large, extended systems with DFT is computationally intensive.
• Accurate description of localized electronic states is challenging in extended systems.

Purpose of the Study:

• To develop a computational approach for reducing the complexity of extended systems in DFT.
• To create a method for generating optimized localized basis functions.
• To enable accurate and efficient nanoscale simulations.

Main Methods:

• A recipe for generating localized basis functions optimized for bulk-like or defective regions.
• Identification of regions requiring specific optimization of Kohn-Sham degrees of freedom.
• Wavelet-based method for calculations across different dimensionalities (nanotubes, slabs, periodic systems).

Main Results:

• The method effectively reduces computational degrees of freedom for nanoscale simulations.
• Provides reliable error estimation for localized approaches.
• Allows straightforward relation to effective models like tight-binding Hamiltonians.
• Successfully tested on SiC nanotube-like cages.

Conclusions:

• The developed computational approach significantly simplifies DFT calculations for extended systems.
• Offers a balance between computational efficiency and accuracy for nanoscale materials.
• Applicable to various system dimensionalities, enhancing its versatility.