Correcting Delocalization Error in Materials with Localized Orbitals and Linear-Response Screening

Area of Science:

• Computational chemistry
• Materials science
• Quantum mechanics

Background:

• Density functional theory (DFT) is a powerful tool for materials science.
• Delocalization error in DFT leads to inaccurate predictions, such as underestimated band gaps and misaligned interface energy levels.
• Accurate modeling of materials and molecules is crucial for scientific advancement.

Purpose of the Study:

• To introduce a new method, lrLOSC, to correct delocalization error in DFT.
• To improve the accuracy of predicting fundamental band gaps and energy levels in materials.
• To enable unified DFT modeling of molecules, materials, and their interfaces.

Main Methods:

• Development and implementation of the lrLOSC method.
• Application of lrLOSC to predict fundamental gaps for eleven materials.
• Parallel implementation of lrLOSC for molecular property calculations.

Main Results:

• lrLOSC corrects delocalization error across a wide range of material band gaps.
• Predicted fundamental gaps for eleven materials are accurate to within 0.22 eV.
• Nonzero total energy corrections are provided, and molecular properties show improvement.

Conclusions:

• lrLOSC effectively addresses delocalization error in DFT calculations.
• The method enhances the predictive accuracy for material properties and molecular behavior.
• lrLOSC represents a significant step towards unified DFT modeling across different systems.