Theoretical characterization of NV-like defects in 4H-SiC using ADAQ with the SCAN and r2SCAN meta-GGA functionals

TL;DR

This paper evaluates the SCAN and r2SCAN meta-GGA functionals within DFT for accurately characterizing NV-like defects in 4H-SiC, offering a computationally efficient alternative to more expensive methods like HSE.

Contribution

It demonstrates that SCAN and r2SCAN functionals provide improved accuracy over PBE in defect characterization, approaching HSE precision at lower computational costs.

Findings

SCAN and r2SCAN outperform PBE in defect energy predictions.

They approach HSE accuracy while maintaining PBE-like computational efficiency.

The results support using SCAN family functionals for large-scale defect screening.

Abstract

Kohn-Sham density functional theory (DFT) is widely used for screening color centers in semiconductors. While the Perdew-Burke-Ernzerhof (PBE) functional is efficient, it often lacks precision in describing defects. The Heyd-Scuseria-Ernzerhof (HSE) functional is more accurate but computationally expensive, making it impractical for large-scale screening. However, third-rung functionals of "Jacob's ladder" remain largely under explored in this context. This study evaluates the Strongly Constrained and Appropriately Normed (SCAN) family of meta-GGA functionals as potential alternatives to PBE for characterizing NV-like color centers in 4H-SiC using the Automatic Defect Analysis and Qualification (ADAQ) framework. We examine nitrogen, oxygen, fluorine, sulfur, and chlorine vacancies in 4H-SiC, focusing on applications in quantum technology. Our results show that SCAN and r2SCAN achieve greater accuracy than PBE, approaching HSE's precision at a lower computational cost. This suggests that the SCAN family offers a practical improvement for screening new color centers, with computational demands similar to PBE.