| name | magnetic-properties |
| description | Magnetic Properties (3 sub-skills: magnetic-anisotropy, magnetic-ordering, spin-polarized) |
Magnetic Properties
Overview
This skill group covers first-principles calculations of magnetic properties using
Quantum ESPRESSO (QE) with Python-based pre/post-processing (pymatgen, ASE, matplotlib).
All workflows are designed for a Docker container environment with QE 7.5 installed.
Sub-Skills
| Sub-Skill | Directory | Description |
|---|
| Spin-Polarized DFT | spin-polarized/ | Collinear spin-polarized SCF calculations (nspin=2), magnetic moments, spin-resolved DOS, and spin density visualization via pp.x. |
| Magnetic Ordering | magnetic-ordering/ | Comparison of FM, AFM, and non-magnetic configurations to determine ground-state magnetic order. Estimation of exchange coupling constants. |
| Magnetic Anisotropy | magnetic-anisotropy/ | Magnetic anisotropy energy (MAE) via noncollinear DFT with spin-orbit coupling (noncolin=.true., lspinorb=.true.). Requires fully relativistic pseudopotentials. (VASPKIT 621) |
| Magnetic Moments | magnetic-moments/ | Extract and visualize local magnetic moments. QE: Lowdin/Mulliken from projwfc.x. VASP: OUTCAR parsing. Generate MAGNETIC_MOMENTS.cif. Spin density visualization. (VASPKIT 629) |
When to Use
- spin-polarized/ -- You need to check whether a material is magnetic, compute atomic
magnetic moments, or obtain spin-resolved electronic structure (DOS, charge density).
- magnetic-ordering/ -- You need to determine whether a material prefers FM, AFM, or
non-magnetic order, or estimate exchange coupling constants from total-energy differences.
- magnetic-anisotropy/ -- You need the magnetic anisotropy energy (easy/hard axis),
which requires spin-orbit coupling and noncollinear magnetism.
- magnetic-moments/ -- You need per-atom magnetic moments, spin density maps, or a
CIF file annotated with magnetic moment vectors for visualization.
General Prerequisites
- Quantum ESPRESSO 7.5 (
pw.x, pp.x, projwfc.x, bands.x, dos.x)
- Python packages:
pymatgen, ase, numpy, scipy, matplotlib
- Appropriate pseudopotentials (SSSP Efficiency or PseudoDojo; fully relativistic for SOC)
- Sufficient computational resources (magnetic calculations are more expensive than non-magnetic)
