Band Advanced

Method Decision Guide

Need advanced band structure analysis?
  |
  +--> Full BZ dispersion for 2D material?
  |      YES --> 3d-band-structure/
  |
  +--> Accurate band gap (hybrid functional)?
  |      YES --> hybrid-dft-bands/
  |
  +--> Supercell with defect/alloy/interface, want primitive-cell-like bands?
         YES --> band-unfolding/

Common Prerequisites

• Converged structure: Always start from a well-relaxed structure (use scf-relax skill).
• Pseudopotentials: QE calculations require pseudopotential files. Use SSSP or PseudoDojo libraries.
• Python environment: pymatgen, ase, numpy, scipy, matplotlib are pre-installed. Install extras with pip install seekpath as needed.
• QE executables: pw.x, bands.x, pp.x, projwfc.x (Quantum ESPRESSO 7.5).
• MACE limitation: MACE is a machine-learning force field and cannot compute electronic band structures. Use it only for structure pre-relaxation before DFT.

Important Notes

• Standard PBE band gaps are underestimated by 30--50%. Use hybrid-DFT bands for quantitative gap values.
• 3D band structure calculations require dense k-meshes and are significantly more expensive than standard k-path calculations.
• Band unfolding is essential when studying defects, alloys, or interfaces in supercells -- without it, the zone-folded bands are uninterpretable.
• All three sub-skills share the same SCF foundation. A well-converged SCF charge density is the starting point for each method.