ShengBTE

Official Resources

• Homepage: http://www.shengbte.org/
• Repository: https://github.com/ShengBTE/ShengBTE
• License: GNU General Public License v3.0

Overview

ShengBTE is a widely used software package for solving the Phonon Boltzmann Transport Equation (BTE) to calculate the lattice thermal conductivity of crystalline materials. It operates on a fully ab initio basis, taking second-order (harmonic) and third-order (anharmonic) interatomic force constants (IFCs) from density functional theory (DFT) calculations as input. By solving the BTE iteratively, it accurately captures phonon-phonon scattering processes beyond the relaxation time approximation (RTA), making it a standard tool for investigating heat transport in bulk materials and nanowires.

Scientific domain: Thermal Transport, Phononics, Materials Science Target user community: Researchers in thermoelectrics, thermal management, and condensed matter physics

Theoretical Methods

• Iterative BTE Solver: Solves the linearized BTE self-consistently to include Normal (N) scattering processes which conserve crystal momentum.
• Scattering Mechanisms:
  • Three-phonon scattering (absorption and emission).
  • Isotopic scattering (mass variance).
  • Boundary scattering (for nanowires/domains).
• Relaxation Time Approximation (RTA): Also provides the RTA solution for comparison.

Capabilities

• Thermal Properties:
  • Lattice thermal conductivity tensor ($\kappa_{\alpha\beta}$).
  • Temperature dependence of $\kappa$.
  • Cumulative thermal conductivity with respect to phonon mean free path (MFP).
• Microscopic Analysis:
  • Mode-resolved scattering rates and lifetimes.
  • Gruneisen parameters.
  • Phase space available for scattering.
• Dimensionality:
  • Bulk 3D crystals.
  • Nanowires (via diffusive boundary terms).
  • 2D materials (with appropriate thickness normalization).

Key Strengths

• Accuracy: The iterative solution is essential for high-thermal-conductivity materials (like Diamond, Graphene) where N-processes play a major role.
• Efficiency: Highly optimized for symmetry reduction, allowing calculations on complex unit cells.
• Ecosystem: Works seamlessly with thirdorder.py for generating anharmonic IFCs.

Inputs & Outputs

• Inputs:
  • CONTROL: Main input file.
  • FORCE_CONSTANTS_2ND: Harmonic force constants.
  • FORCE_CONSTANTS_3RD: Anharmonic force constants.
• Outputs:
  • BTE.kappa: Final thermal conductivity.
  • T_P_lifetimes.dat: Phonon lifetimes.
  • cumulative_kappa.dat: MFP analysis.

Interfaces & Ecosystem

• Upstream:
  • VASP / QE: Generate forces.
  • Phonopy: Often used to prepare supercells and 2nd order IFCs.
  • thirdorder.py: Standard script to generate 3rd order IFCs.
• Visualization: Output data is simple text, easily plotted with Python/Gnuplot.

Workflow and Usage

Typical ShengBTE Workflow:

# 1. Generate 2nd order force constants (phonopy)
phonopy -d --dim="2 2 2" -c POSCAR
# Run DFT on displaced structures
phonopy --fc vasprun.xml

# 2. Generate 3rd order force constants (thirdorder.py)
thirdorder.py sow POSCAR
# Run DFT on displaced structures
thirdorder.py reap POSCAR

# 3. Run ShengBTE
ShengBTE

Advanced Features

• Iterative BTE: Full solution beyond RTA for accurate transport
• Spectral analysis: Frequency-resolved thermal conductivity
• Cumulative functions: Mean free path accumulation
• Size effects: Grain boundary and nanostructure scattering
• Isotope scattering: Natural isotope disorder effects
• Nanowire support: Diffusive boundary scattering for 1D systems

Computational Cost

• Force constant calculations (DFT): Dominant cost (hundreds of calculations for 3rd order)
• ShengBTE BTE solution: Minutes to hours
• Iterative BTE more expensive than RTA
• Dense q-grids increase cost significantly

Best Practices

• Converge supercell size for force constants
• Systematic q-point grid convergence
• Test RTA vs iterative BTE
• Validate against experimental data
• Appropriate cutoff distances for 3rd order IFCs
• Use symmetry to reduce DFT calculations

Performance Characteristics

• Computational Cost: The BTE solution is fast (seconds to minutes on a single core). The bottleneck is generating the 3rd-order IFCs (hundreds of DFT runs).
• Parallelism: MPI parallelization over the q-point grid.

Limitations & Known Constraints

• Higher-Order Scattering: Only considers 3-phonon processes; 4-phonon scattering (important at high T) is not included in the standard version (extensions exist).
• Q-grid Convergence: Requires careful convergence of the q-point mesh for accurate results.

Comparison with Other Codes

• vs. Phono3py: Similar capabilities; ShengBTE's iterative solver was historically faster/more robust for N-processes, though Phono3py has caught up. ShengBTE is Fortran-based, Phono3py is Python/C.
• vs. almaBTE: almaBTE extends the BTE approach to space-dependent problems (devices), whereas ShengBTE is primarily for bulk/homogeneous systems.

Application Areas

• Thermoelectrics: Screening for low-$\kappa$ materials ($PbTe$, $SnSe$).
• Heat Management: High-$\kappa$ materials ($BAs$, Diamond) for electronics cooling.
• Isotope Engineering: Tailoring thermal properties via isotope enrichment.

Community and Support

• Development: Developed by Wu Li (CEA/CAS) and collaborators.
• Source: GitHub / Bitbucket.

Verification & Sources

• Repository: https://github.com/ShengBTE/ShengBTE
• Primary Publication: W. Li et al., Comp. Phys. Comm. 185, 1747 (2014).
• Verification status: ✅ VERIFIED
  • Gold standard code in the field.