kgrid

Official Resources

• Homepage: https://github.com/WMD-group/kgrid
• GitHub: https://github.com/WMD-group/kgrid
• PyPI: https://pypi.org/project/kgrid/
• License: MIT License

Overview

kgrid is a Python tool for calculating the required k-point density from input geometry for periodic quantum chemistry calculations. It uses a length cutoff approach to determine appropriate Monkhorst-Pack grids, ensuring consistent k-point density across different cell sizes and shapes.

Scientific domain: K-point sampling, DFT calculations, Brillouin zone integration Target user community: DFT practitioners needing consistent k-point grids for convergence studies

Theoretical Background

kgrid implements k-point selection based on:

• Real-space length cutoff: L_cutoff determines minimum sampling
• Monkhorst-Pack grid: N_i = ceil(L_cutoff / |a_i|)
• Reciprocal space density: Ensures consistent BZ sampling
• Accounts for cell shape and anisotropy

Capabilities (CRITICAL)

• K-point Generation: Calculate optimal Monkhorst-Pack grids
• Length Cutoff: Specify real-space cutoff for consistent density
• Multi-code Output: VASP KSPACING, CASTEP MP_SPACING formats
• Automatic Scaling: Adjusts for cell dimensions

Key Strengths

Consistent Sampling:

• Length-based approach ensures comparable accuracy
• Automatic adjustment for cell shape
• Reproducible k-point selection

Multi-Code Support:

• VASP KSPACING output
• CASTEP MP_SPACING format
• Easy integration into workflows

Simple Interface:

• Command-line tool
• Python API
• Minimal dependencies

Inputs & Outputs

• Input formats:

  • POSCAR (VASP structure)
  • Any ASE-readable structure
  • Length cutoff parameter

• Output data types:

  • KSPACING value (VASP)
  • MP_SPACING value (CASTEP)
  • k-point grid dimensions

Installation

pip install kgrid

Usage Examples

Command line:

# VASP format (default)
kgrid POSCAR 25  # 25 Å cutoff

# CASTEP format
kgrid --castep POSCAR 25

# Show grid dimensions
kgrid --verbose POSCAR 25

Python API:

from kgrid import calc_kpt_tuple
from ase.io import read

atoms = read("POSCAR")
kpts = calc_kpt_tuple(atoms, cutoff_length=25)
print(f"K-point grid: {kpts}")

Performance Characteristics

• Speed: Instant calculation
• Accuracy: Consistent density across systems
• Simplicity: Single parameter (cutoff length)

Limitations & Known Constraints

• Monkhorst-Pack only: No support for other schemes
• Gamma-centered: Default behavior
• Odd/even grids: May need manual adjustment

Comparison with Other Tools

• vs SeeK-path: kgrid for grids, SeeK-path for paths
• vs manual selection: kgrid ensures consistency
• Unique strength: Length-based consistent sampling

Application Areas

• Convergence testing
• High-throughput calculations
• Consistent k-point selection
• Workflow automation

Best Practices

• Use same cutoff for related calculations
• Verify convergence with increasing cutoff
• Consider symmetry for efficiency
• Document cutoff used for reproducibility

Verification & Sources

Primary sources:

1. GitHub: https://github.com/WMD-group/kgrid

Confidence: VERIFIED

Verification status: ✅ VERIFIED

• Source code: OPEN (GitHub, MIT)
• Developer: WMD Group (Walsh Materials Design)