Official Resources
- Source Repository: https://github.com/yufengliang/mbxaspy
- Documentation: Included in repository
- License: Open source
Overview
mbxaspy is a Python software package for predicting X-ray spectra using the determinant formalism based on the independent-electron approximation as used in DFT. It interfaces with the ShirleyXAS Fortran package for DFT and XAS calculations at the one-body level, and can also work with tight-binding models for many-body XAS calculations.
Scientific domain: X-ray absorption spectroscopy, many-body spectroscopy
Target user community: Researchers computing XAS spectra from DFT and tight-binding models, including many-body extensions
Theoretical Methods
- Determinant formalism for XAS
- Independent-electron approximation (DFT)
- Many-body extension beyond DFT
- Tight-binding models
- Core-hole treatment
- Bethe-Salpeter-like corrections
- ShirleyXAS integration
Capabilities (CRITICAL)
- X-ray absorption spectroscopy (XAS)
- X-ray emission spectroscopy (XES)
- Many-body XAS calculations
- Tight-binding XAS
- DFT-level XAS (via ShirleyXAS)
- Core-hole effect simulation
- Polarization-dependent spectra
- K-edge and L-edge calculations
Sources: GitHub repository
Key Strengths
Determinant Formalism:
- Systematic improvement beyond DFT
- Can include many-body effects
- Flexible Hamiltonian choices
- From DFT to tight-binding models
ShirleyXAS Integration:
- Well-established XAS code
- Full DFT-level calculations
- Core-hole pseudopotentials
- Production-quality results
Python Interface:
- Scriptable workflow
- Jupyter notebook compatible
- Easy post-processing
- Visualization tools
Inputs & Outputs
-
Input formats:
- ShirleyXAS DFT outputs
- Tight-binding model parameters
- Hamiltonian specifications
-
Output data types:
- XAS spectra
- XES spectra
- Many-body corrected spectra
- Polarization-dependent spectra
Interfaces & Ecosystem
- ShirleyXAS: DFT/XAS Fortran backend
- Python: Scripting and analysis
- DFT codes: Via ShirleyXAS interface
Performance Characteristics
- Speed: Depends on backend (ShirleyXAS or TB)
- Accuracy: DFT-level or better with many-body
- System size: Limited by backend
- Memory: Moderate
Computational Cost
- DFT XAS: Hours (ShirleyXAS)
- TB XAS: Minutes
- Many-body: Hours to days
- Typical: Moderate
Limitations & Known Constraints
- ShirleyXAS dependency: Requires external Fortran code
- Documentation: Limited
- Community: Small
- Installation: Can be complex (Fortran + Python)
Comparison with Other Codes
- vs FEFF: mbxaspy can do many-body, FEFF is single-particle
- vs OCEAN: mbxaspy is more flexible, OCEAN is BSE-based
- vs xspectra: mbxaspy has many-body extension, xspectra is DFT-only
- Unique strength: Determinant formalism for XAS with both DFT and many-body capabilities
Application Areas
Transition Metal XAS:
- K-edge and L-edge spectra
- Many-body effects in XAS
- Core-hole screening
- Charge transfer satellites
Battery Materials:
- Transition metal redox
- Oxygen K-edge XAS
- Charge state analysis
- Cycling effects
Correlated Oxides:
- Multiplet-like features
- Hubbard model XAS
- Charge transfer insulators
- Metal-insulator transitions
Best Practices
DFT Parameters:
- Use well-converged ShirleyXAS calculations
- Appropriate core-hole treatment
- Test k-point convergence
- Validate against experiment
Many-Body Extensions:
- Start with DFT-level results
- Add many-body corrections systematically
- Validate against BSE if possible
- Compare with experiment
Community and Support
- Open source on GitHub
- Developed at LBNL Molecular Foundry
- Limited documentation
- Research code with active development
Verification & Sources
Primary sources:
- GitHub repository: https://github.com/yufengliang/mbxaspy
- Y. Liang et al., related publications from LBNL
Confidence: VERIFIED
Verification status: ✅ VERIFIED
- Source code: ACCESSIBLE (GitHub)
- Documentation: Limited
- Active development: Research code
- Specialized strength: Determinant formalism XAS with DFT and many-body capabilities, ShirleyXAS integration