Official Resources
- Homepage: https://github.com/lechifflier/PyGW
- Documentation: https://github.com/lechifflier/PyGW#readme
- Source Repository: https://github.com/lechifflier/PyGW
- License: Open Source
Overview
PyGW is an electronic structure code for performing G0W0 and GW0 quasiparticle calculations on realistic materials. Implemented as a hybrid Fortran/Python code, it bridges computational efficiency with modern scripting capabilities for GW calculations in condensed matter physics.
Scientific domain: Quasiparticle band structures, band gap calculations, electronic excitations
Target user community: Condensed matter physicists studying electronic properties of materials
Theoretical Methods
- G0W0 approximation
- GW0 (eigenvalue self-consistent)
- Many-body perturbation theory
- Screened Coulomb interaction
- Quasiparticle corrections
- Plane-wave / pseudopotential framework
Capabilities (CRITICAL)
- G0W0 quasiparticle energies
- GW0 self-consistent eigenvalues
- Band structure calculations
- Band gap predictions
- Quasiparticle corrections to DFT
- Bulk materials
- Semiconductor and insulator systems
Sources: Official GitHub repository
Key Strengths
Fortran/Python Hybrid:
- Computational efficiency from Fortran
- User-friendly Python interface
- Modern workflow integration
- Scriptable calculations
GW Implementations:
- Standard G0W0 calculations
- GW0 self-consistency
- Proven methodology
- Materials science focus
Realistic Materials:
- Production-quality calculations
- Plane-wave accuracy
- Pseudopotential efficiency
- Solid-state applications
Inputs & Outputs
-
Input formats:
- DFT wavefunctions and eigenvalues
- Pseudopotential files
- Python configuration
-
Output data types:
- Quasiparticle energies
- Band structures
- Band gaps
- Self-energy data
Interfaces & Ecosystem
-
DFT Integration:
- Requires DFT input (wavefunctions, eigenvalues)
- Interfaces with plane-wave DFT codes
-
Python scripting:
- Python driver scripts
- Workflow automation
- Post-processing capabilities
Performance Characteristics
- Speed: Fortran computational core
- Accuracy: Standard GW precision
- System size: Typical GW scaling
- Memory: Plane-wave requirements
Computational Cost
- G0W0: Single-shot calculation
- GW0: Multiple iterations for self-consistency
- Scaling: O(N^4) typical GW scaling
Limitations & Known Constraints
- Documentation: Limited compared to major codes
- DFT interface: Requires specific input format
- Development: Smaller community
Comparison with Other Codes
- vs BerkeleyGW: PyGW smaller, BerkeleyGW more features
- vs Yambo: Different interface and workflow
- Unique strength: Fortran/Python hybrid design
Application Areas
Semiconductors:
- Band gap calculations
- Quasiparticle corrections
- Electronic structure
Insulators:
- Accurate band gaps
- Beyond-DFT corrections
- Materials screening
Community and Support
- Open-source on GitHub
- Academic development
- Limited but growing documentation
Verification & Sources
Primary sources:
- Official GitHub: https://github.com/lechifflier/PyGW
- Active development (2024 commits)
Confidence: VERIFIED
- GitHub repository: ACCESSIBLE
- Active development: Yes
- Working implementation: Confirmed