EPWpy

Official Resources

• Homepage: http://epwpy.org/
• Repository: https://github.com/Sponce24/EPWpy (or related grouping)
• License: Open Source (GPL compatible).
• Developers: H. Lee, S. Poncé, et al. (EPW Collaboration).

Overview

EPWpy is the official high-level Python interface for EPW (Electron-Phonon Wannier), the leading code for calculating electron-phonon coupling properties using Maximally Localized Wannier Functions. EPWpy is designed to democratize access to complex electron-phonon calculations by automating the often-intricate workflows involving Quantum ESPRESSO, Wannier90, and EPW, and providing modern data structures for analysis.

Scientific domain: Electron-phonon physics, superconductivity, transport, optics. Target user community: Users of EPW requiring automated workflows and Python-based post-processing.

Theoretical Methods

• Electron-Phonon Coupling:
  • Computation of vertex $g_{mn\nu}(k,q)$ in Wannier basis.
  • Interpolation to fine k/q-grids.
• Superconductivity:
  • Anisotropic Migdal-Eliashberg theory.
  • Calculation of $T_c$ and superconducting gap functions $\Delta(k)$.
• Transport:
  • Carrier mobilities via Boltzmann Transport Equation (BTE).
  • Self-energy relaxation times.

Capabilities

• Workflow Automation:
  • End-to-end management of scf $\to$ phonons $\to$ wannier $\to$ epw pipeline.
  • provenance tracking of calculation steps.
• Post-Processing:
  • Calculation of spectral functions and linewidths.
  • Visualization of electron-phonon matrix elements.
  • Fermi surface and phonon dispersion plotting.
• Data Management:
  • Parsing complex EPW output files into accessible Python objects / NetCDF.

Key Strengths

• Usability: drastically reduces the barrier to entry for running EPW, which historically required complex manual file management.
• Reproducibility: Encourages scripted, reproducible science compared to ad-hoc shell scripts.
• Community: Supported by the core developers of the EPW code itself.
• Visualization: Leverages Python's rich plotting ecosystem (Matplotlib, etc.) for high-quality figures.

Inputs & Outputs

• Inputs:
  • QE input files (pw.in, ph.in).
  • EPW input parameters (epw.in).
  • Python script configuration.
• Outputs:
  • Structured data files (NetCDF/HDF5).
  • Visualization plots.

Interfaces & Ecosystem

• EPW: The core engine.
• Quantum ESPRESSO: Full integration with QE suite.
• Wannier90: Manages the Wannierization step.
• AiiDA: Similar goals, but EPWpy is a lighter, code-specific solution compared to the full AiiDA database approach.

Performance Characteristics

• Overhead: Minimal; the heavy computation is done by the Fortran EPW binaries.
• Efficiency: Accelerates the "time-to-result" for researchers by eliminating manual file handling errors.

Comparison with Other Codes

• vs [AiiDA-EPW](file:///home/niel/git/Indranil2020.github.io/scientific_tools_consolidated/Workflow_Managers/AiiDA.md): AiiDA provides a full database-driven provenance framework (heavier setup); EPWpy is a lightweight script-based interface easier for individual projects.
• vs Manual EPW: EPWpy prevents common errors in input consistency between PHonon and EPW steps.

Application Areas

• Superconductors: Determining critical temperatures of new materials.
• Photovoltaics: Analyzing phonon-assisted optical absorption.
• Thermoelectrics: Calculating carrier lifetimes and conductivities.

Verification & Sources

• Primary Source: EPWpy Website
• Citation: Lee, H. et al., "Electron–phonon physics from first principles using the EPW code", npj Comput. Mater. (2023) [Referencing EPWpy].
• Verification Status: ✅ VERIFIED.