Official Resources
- Homepage: https://mir-group.github.io/phoebe/
- Documentation: https://phoebe.readthedocs.io/
- Source Repository: https://github.com/mir-group/phoebe
- License: Apache License 2.0
Overview
Phoebe is a modern, high-performance code for calculating phonon and electron thermal transport properties from first principles. Developed at MIT, Phoebe solves the Boltzmann transport equation for both phonons and electrons with electron-phonon coupling, focusing on computational efficiency and advanced transport phenomena. The code features GPU acceleration, advanced algorithms, and handles coupled electron-phonon transport in a unified framework.
Scientific domain: Thermal transport, thermoelectrics, coupled electron-phonon dynamics
Target user community: Thermal transport researchers, thermoelectric materials, computational materials science
Theoretical Methods
- Phonon Boltzmann transport equation (BTE)
- Electron Boltzmann transport equation
- Coupled electron-phonon transport
- Iterative and variational BTE solutions
- Relaxation time approximation
- Phonon-phonon scattering (3-phonon processes)
- Electron-phonon scattering
- Phonon-boundary scattering
- Phonon-isotope scattering
- Electron-impurity scattering
- Wannier interpolation for electrons
Capabilities (CRITICAL)
- Lattice thermal conductivity from first principles
- Electronic thermal conductivity
- Coupled electron-phonon thermal transport
- Electrical conductivity
- Seebeck coefficient
- Phonon and electron lifetimes
- Spectral thermal conductivity
- Cumulative thermal conductivity
- Mode-resolved transport properties
- Temperature-dependent transport
- Nanostructure and boundary scattering
- GPU acceleration for large systems
- HPC parallelization (MPI + OpenMP)
- Wannier interpolation for efficient calculations
Sources: Official Phoebe documentation, Nature Communications 12, 2222 (2021)
Key Strengths
- GPU acceleration: Significant speedup for large-scale calculations
- Coupled transport: Unified electron-phonon treatment
- Modern architecture: C++ with Python interface, HPC-optimized
- Advanced algorithms: Variational and iterative BTE solvers
Inputs & Outputs
-
Input formats:
- Phonopy force constants (for phonons)
- Wannier90 data (for electrons)
- Electron-phonon matrix elements
- Crystal structure files
- Phoebe configuration files
-
Output data types:
- Thermal conductivity tensors
- Transport coefficients
- Scattering rates and lifetimes
- Spectral and cumulative properties
- Mode-resolved contributions
Interfaces & Ecosystem
- Phonopy: Import harmonic phonon properties
- Quantum ESPRESSO: Via phonopy interface for phonons
- Wannier90: For electronic structure and electron-phonon coupling
- Python: Python interface for workflow automation
- HDF5: Efficient data storage and exchange
Workflow and Usage
Phonon Transport Workflow:
# 1. Prepare phonon force constants (from phonopy/phono3py)
# 2. Create Phoebe input
# 3. Run Phoebe
mpirun -np 16 phoebe -in input.phoebe
# GPU acceleration
phoebe -in input.phoebe --useGPU
Coupled Electron-Phonon Transport:
# Requires both phonon and Wannier electron data
phoebe -in coupled_transport.phoebe
Advanced Features
- Variational BTE: Beyond relaxation time approximation
- GPU kernels: Optimized CUDA kernels for scattering calculations
- Adaptive grids: Smart q-point and k-point sampling
- Nanostructures: Boundary and grain scattering models
- Hydrodynamic phonons: Advanced transport regimes
Performance Characteristics
- GPU speedup: 10-100x faster than CPU-only for large systems
- Parallelization: Excellent scaling with MPI+OpenMP
- Memory: Optimized for large k/q grids
- Typical runtime: Hours with GPU; days CPU-only for production
Computational Cost
- Force constant calculations (DFT) most expensive
- Phoebe very efficient with GPU acceleration
- Iterative BTE more expensive than RTA
- Dense grids feasible with GPU
Limitations & Known Constraints
- GPU recommended: CPU-only slower for large systems
- Requires force constants: From external phonon codes
- Learning curve: Moderate; requires transport theory knowledge
- Documentation: Growing; some advanced features need expertise
- Platform: Linux; GPU support requires CUDA
Comparison with Other Codes
- vs ShengBTE/phono3py: Phoebe has GPU acceleration and coupled transport
- vs ALAMODE: Phoebe focuses on transport solver efficiency
- Unique strength: GPU-accelerated coupled electron-phonon transport
Application Areas
- Thermoelectrics: Figure of merit (ZT), optimizing transport properties
- Thermal management: Heat dissipation in electronics
- Nanostructures: Grain boundaries, interfaces, thin films
- Novel materials: Materials with complex transport physics
- High-throughput: Rapid screening with GPU acceleration
Best Practices
- Use GPU acceleration for production calculations
- Converge k/q-point grids systematically
- Test RTA vs iterative BTE convergence
- Validate against experimental data when available
- Appropriate boundary scattering parameters for nanostructures
Community and Support
- Open-source (Apache 2.0)
- GitHub repository
- Documentation website
- MIT development team
- Growing user community
- Research collaborations
Educational Resources
- Comprehensive documentation
- Tutorial examples
- Publication describing methodology
- Example input files
- Python API examples
Development
- MIT Materials Intelligence Research group
- Active development
- GPU optimization ongoing
- Feature additions for advanced transport
- Community contributions
Research Impact
Phoebe enables efficient first-principles thermal transport calculations with GPU acceleration, particularly valuable for coupled electron-phonon transport in thermoelectric materials and high-throughput materials screening.
Verification & Sources
Primary sources:
- Homepage: https://mir-group.github.io/phoebe/
- Documentation: https://phoebe.readthedocs.io/
- GitHub: https://github.com/mir-group/phoebe
- Publication: Nature Communications 12, 2222 (2021)
Confidence: VERIFIED
Verification status: ✅ VERIFIED
- Website: ACTIVE
- Documentation: COMPREHENSIVE
- Source: OPEN (GitHub, Apache 2.0)
- Development: ACTIVE (MIT)
- Applications: GPU-accelerated thermal transport, coupled electron-phonon BTE, variational transport solvers, thermoelectrics, high-performance computing, production quality