Starts 11 Mar 2026 11:00
Ends 11 Mar 2026 12:00
Central European Time
CMSP Seminar (Atomistic Simulation Seminar Series): DFT+α: An alternative to computationally heavy hybrid functionals
Euler Lecture Hall (Leonardo Building) and via Zoom
Abdulgaffar Abdurrazaq
(SISSA)
Abstract:
Accurately predicting the electronic band gap of semiconductors remains a well known limitation of standard Density Functional Theory (DFT). Conventional exchange-correlation approximations, such as the Local Density Approximation (LDA) and the various forms of the Generalized Gradient Approximation (GGA), systematically underestimate semiconductor band gaps and can even yield spurious metallic ground states. This issue is particularly pronounced in materials like germanium and α-Sn, for which many ab-initio studies incorrectly predict metallic behavior. In this work, we introduce DFT+α, a computational approach tailored for materials exhibiting significant s-p orbital mixing. The method provides a computationally efficient alternative to hybrid functionals while maintaining strong accuracy. We demonstrate that DFT+α reliably captures electronic, structural, and dielectric properties across group IV semiconductors, as well as in III-V and II-VI dipolar compounds. In particular, DFT+α markedly improves band gap predictions, offering a practical solution to the long standing gap underestimation problem in standard DFT. The method also achieves near chemical precision for lattice constants. Improvements in dielectric constant predictions are substantial for most systems, although enhancements in Born effective charges remain modest. The consistent performance across a broad materials set highlights both the robustness and the transferability of the DFT+α framework, especially for systems dominated by valence s-p mixing. However, the method does not fully resolve the band gap problem for materials in which both the conduction band minimum and valence band maximum occur at the same point in the Brillouin zone, such as silicon and carbon. For these materials, DFT+α still yields accurate lattice parameters and ground state properties, but fails to reproduce the correct electronic gaps. This limitation suggests that additional projection patterns or methodological extensions are required to properly describe such systems.