QuantumATK Forum
QuantumATK => General Questions and Answers => Topic started by: Siang-Yu on May 15, 2025, 15:47
-
Dear Sir,
I am currently attempting to simulate the band structure and phonon dispersion of MoS₂ at high-symmetry points (Γ, M, K, Γ), following the approach described in the paper:
Gunst, Tue, et al. "First-principles method for electron-phonon coupling and electron mobility: Applications to two-dimensional materials." Physical Review B 93.3 (2016): 035414.
When I use the same exchange-correlation functional as the paper (LDA), my simulation results closely match those reported. However, when switching to GGA, the results deviate more noticeably from experimental values.
May I ask whether LDA is generally more suitable for simulating the band structure of transition metal dichalcogenides, especially in metallic or near-metallic conditions?
I would greatly appreciate your insights or any related recommendations.
Thank you very much for your time and assistance.
Best regards
-
I would say neither is suitable, at least for the semiconductor cases. If I recall correctly there might be some accidental cancellation of effects for MoS2 in LDA, which is not present in GGA. But neither is "supposed" to give the right result. For that you will need HSE or similar, although I am not sure which is the absolute best method; there is a vast literature on this to consult. For metallic cases GGA is probably ok, although I would look at PBESol rather than PBE, and possibly also r2SCAN (not so much for the band gap, but other properties like lattice constant).