VASP band structure calculation. Two-step workflow with SCF charge density followed by non-SCF band calculation along high-symmetry k-path.
Scanned 9/20/2026
Install to Claude Code
npx -y skills add Hello-QM/catgo-LRG --skill vasp-band --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Vasp Band?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/hello-qm-vasp-band)More formats (shields.io, HTML) on the badges page.
---
name: vasp-band
description: VASP band structure calculation. Two-step workflow with SCF charge density followed by non-SCF band calculation along high-symmetry k-path.
---
# VASP Band Structure Calculation
Compute electronic band structure along high-symmetry k-point paths. Requires a two-step process: self-consistent charge density, then non-SCF calculation along the k-path.
## Why Two Steps?
1. **Single point (SCF)** — compute self-consistent charge density with a uniform k-mesh
2. **Band calculation (non-SCF)** — read the converged CHGCAR and compute eigenvalues along the high-symmetry k-path without updating the charge density
This separation is necessary because the high-symmetry k-path does not provide uniform Brillouin zone sampling needed for SCF convergence.
## Full Band Structure Workflow
```python
from catgo.workflow import Workflow
from catgo.workflow.builtins import geo_opt, single_point
wf = Workflow("TiO2 band structure")
struct = wf.add_task("structure_input", structure=structure_json)
# Step 1: Optimize (skip if already relaxed)
opt = wf.add_task(geo_opt, structure=struct.output.structure,
ISIF=3, system_name="relax")
# Step 2: SCF single point to generate CHGCAR
scf = wf.add_task(single_point, structure=opt.output.structure,
LCHARG=True, # Write CHGCAR
EDIFF=1e-6, # Tight convergence
system_name="SCF")
# Step 3: Non-SCF band calculation
band = wf.add_task(single_point, structure=opt.output.structure,
ICHARG=11, # Read CHGCAR, do not update
LORBIT=11, # Projected band character
LCHARG=False,
LWAVE=False,
kpath_mode="auto", # Auto-detect high-symmetry path
kpath_density=40, # Points per segment
system_name="bands")
wf.submit()
```
## MCP Workflow
```
catgo_workflow_engine(action="create", params={"name": "Band structure"})
# Input structure
catgo_workflow_engine(action="add_task", params={
"workflow_id": "wf_xxx",
"task_type": "structure_input",
"structure": "<json>"
})
# SCF single point
catgo_workflow_engine(action="add_task", params={
"workflow_id": "wf_xxx",
"task_type": "single_point",
"software": "vasp",
"structure": "{{t_001.output.structure}}",
"LCHARG": true,
"EDIFF": 1e-6,
"system_name": "SCF"
})
# Non-SCF band calculation
catgo_workflow_engine(action="add_task", params={
"workflow_id": "wf_xxx",
"task_type": "single_point",
"software": "vasp",
"structure": "{{t_001.output.structure}}",
"ICHARG": 11,
"LORBIT": 11,
"kpath_mode": "auto",
"kpath_density": 40,
"system_name": "bands"
})
catgo_workflow_engine(action="submit", params={"workflow_id": "wf_xxx"})
```
## High-Symmetry K-Path
### Automatic Path Detection
Set `kpath_mode="auto"` to let the engine detect the Bravais lattice and generate the standard k-path. This works for most crystal systems.
### Manual K-Path
For custom paths, specify k-points explicitly:
```python
band = wf.add_task(single_point, structure=opt.output.structure,
ICHARG=11,
kpath_mode="manual",
kpath_points={
"G": [0.0, 0.0, 0.0],
"X": [0.5, 0.0, 0.0],
"M": [0.5, 0.5, 0.0],
"G2": [0.0, 0.0, 0.0],
"R": [0.5, 0.5, 0.5],
},
kpath_segments=["G-X", "X-M", "M-G2", "G2-R"],
kpath_density=40,
system_name="bands")
```
### Common K-Paths by Crystal System
| System | Path | Example |
|---|---|---|
| FCC | G-X-W-K-G-L-U-W-L-K | Cu, Al, Pt |
| BCC | G-H-N-G-P-H | Fe, W, Cr |
| HCP | G-M-K-G-A-L-H-A | Ti, Ru, Co |
| Tetragonal | G-X-M-G-Z-R-A-Z | TiO2 rutile |
| Simple cubic | G-X-M-G-R-X | SrTiO3 |
## Key Parameters
| Parameter | Value | Purpose |
|---|---|---|
| ICHARG | 11 | Read CHGCAR, non-self-consistent |
| LORBIT | 11 | Atom- and orbital-projected bands |
| NBANDS | auto | Number of bands (increase for unoccupied states) |
| LCHARG | False | Do not overwrite CHGCAR from SCF step |
| LWAVE | False | Do not write WAVECAR (saves disk) |
| kpath_density | 40 | K-points per segment (more = smoother bands) |
## Spin-Polarized Band Structure
For magnetic systems:
```python
scf = wf.add_task(single_point, structure=s,
LCHARG=True, ISPIN=2,
MAGMOM="2*5.0 4*0.6",
system_name="SCF_spin")
band = wf.add_task(single_point, structure=s,
ICHARG=11, ISPIN=2, LORBIT=11,
kpath_mode="auto", kpath_density=40,
system_name="bands_spin")
```
## Hybrid Functional Band Structure (HSE06)
HSE06 band structure is expensive but more accurate for band gaps:
```python
scf = wf.add_task(single_point, structure=s,
LCHARG=True, LHFCALC=True, HFSCREEN=0.2,
AEXX=0.25, ALGO="Damped", TIME=0.4,
system_name="SCF_HSE")
band = wf.add_task(single_point, structure=s,
ICHARG=11, LHFCALC=True, HFSCREEN=0.2,
AEXX=0.25, ALGO="Damped", TIME=0.4,
kpath_mode="auto", kpath_density=20,
system_name="bands_HSE")
```
**Note:** HSE band calculations are 10-100x more expensive than PBE. Use a lower kpath_density (20) and fewer NBANDS.
## Combined DOS + Band Structure
Run both from the same SCF calculation:
```python
scf = wf.add_task(single_point, structure=opt.output.structure,
LCHARG=True, EDIFF=1e-6, system_name="SCF")
# DOS branch
dos_sp = wf.add_task(single_point, structure=opt.output.structure,
ISMEAR=-5, NEDOS=3001, LORBIT=11,
system_name="DOS")
# Band branch
band = wf.add_task(single_point, structure=opt.output.structure,
ICHARG=11, LORBIT=11,
kpath_mode="auto", kpath_density=40,
system_name="bands")
```
## Troubleshooting
| Problem | Fix |
|---|---|
| Bands look wrong / discontinuous | CHGCAR from SCF may be on different k-mesh. Ensure SCF used uniform mesh |
| Band gap too small (PBE) | Expected — PBE underestimates gaps. Use HSE06 for accurate gaps |
| Missing unoccupied bands | Increase NBANDS (default may cut off conduction bands) |
| ICHARG=11 error | CHGCAR must exist from SCF step. Check SCF completed with LCHARG=True |
| Very slow HSE | Normal — reduce kpath_density, reduce NBANDS, use more nodes |
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!