ABINIT

Official Resources

• Homepage: https://www.abinit.org/
• Documentation: https://docs.abinit.org/
• Source Repository: https://github.com/abinit/abinit
• License: GNU General Public License v3.0

Overview

ABINIT is a comprehensive open-source package for electronic structure calculations based on density-functional theory, using pseudopotentials and a plane-wave or wavelet basis set. It is particularly strong in linear-response calculations, many-body perturbation theory (GW), and excited-state methods.

Scientific domain: Condensed matter physics, materials science, electronic structure
Target user community: Academic researchers, materials scientists requiring advanced response properties

Theoretical Methods

• Density Functional Theory (DFT)
• Plane-wave pseudopotentials and PAW
• Wavelet basis sets (BigDFT integration)
• Density Functional Perturbation Theory (DFPT)
• Many-Body Perturbation Theory (GW approximation)
• Bethe-Salpeter Equation (BSE)
• Time-Dependent DFT (TDDFT)
• Dynamical Mean-Field Theory (DMFT)
• Constrained DFT
• Hybrid functionals
• DFT+U for correlated systems

Capabilities (CRITICAL)

• Ground-state electronic structure calculations
• Geometry optimization and cell relaxation
• Molecular dynamics (Born-Oppenheimer, Langevin, Nosé-Hoover)
• Phonon calculations via DFPT (full q-point grid)
• Dielectric response and Born effective charges
• Elastic constants via response functions
• Electron-phonon coupling via DFPT
• GW quasiparticle energies (G₀W₀, self-consistent GW)
• BSE for optical absorption including excitonic effects
• TDDFT for excited states
• Non-linear optical properties
• Temperature-dependent electronic structure
• Spin-orbit coupling and non-collinear magnetism
• Electric field responses (Berry phase)
• Wannier function generation (interface to Wannier90)
• DFT+DMFT for strongly correlated systems
• PAW datasets generation

Sources: Official ABINIT documentation, tutorials, cited in 7/7 source lists

Inputs & Outputs

• Input formats:

  • Main input file (.in or .abi) with structured keywords
  • Pseudopotential files (.psp8, .xml for PAW)
  • Density/wavefunction files for restarts
  • Structure files (various formats via ASE/pymatgen conversion)

• Output data types:

  • Main output (.out or .abo) with energies, forces, stresses
  • NetCDF files for density, wavefunctions, response functions
  • Phonon data (dynamical matrices, IFCs)
  • GW and BSE outputs
  • Formatted text files for band structures, DOS

Interfaces & Ecosystem

• Framework integrations:

  • ASE - calculator interface
  • AiiDA - aiida-abinit plugin for workflows
  • pymatgen - structure I/O and analysis
  • Phonopy - phonon post-processing from force constants
  • Wannier90 - tight-binding Hamiltonian generation

• Post-processing tools:

  • abipy - Python package for ABINIT workflows and analysis
  • AbiPy post-processing utilities
  • cut3d - for visualizing densities and potentials
  • Built-in analysis tools (abicheck, abiview, etc.)

• Related codes:

  • BigDFT - wavelet basis integration
  • Yambo - alternative GW/BSE post-processor
  • BerkeleyGW - can read ABINIT wavefunctions

Limitations & Known Constraints

• Learning curve: Input file syntax complex with many keywords; steep learning curve
• Documentation: Extensive but can be difficult to navigate; variable quality across topics
• Performance: Parallelization efficiency depends on calculation type; k-point parallelization most efficient
• Memory: GW and BSE calculations memory-intensive for large systems
• Pseudopotentials: Quality depends on pseudopotential table; requires validation
• Convergence: Response function calculations require careful convergence testing
• Hybrid functionals: Expensive; limited to smaller systems
• Installation: Build process can be complex with many optional dependencies
• File I/O: NetCDF files can become very large for big systems

Computational Cost

• Ground State (PW): $O(N^3)$.
• GW/BSE: Highly expensive ($O(N^4)$); requires massive memory.
• Wavelets (BigDFT): $O(N)$ linear scaling mode available via BigDFT integration.

Comparison with Other Codes

• vs Quantum ESPRESSO: ABINIT has a longer history with GW/response functions; QE is often faster for standard MD.
• vs Yambo: Yambo acts as a post-processor for QE; ABINIT has GW built-in, offering a more unified but sometimes monolithic experience.
• vs VASP: ABINIT is open-source and has more diverse basis set options (wavelets); VASP is faster for simple relaxation.

Best Practices

• Parallelization: Study the autoparal feature (automatic parallelization tuning).
• Pseudopotentials: Use PseudoDojo tables (standard for ABINIT).
• Convergence: GW calculations require convergence of empty states (nband), which is much harder than ground state.

Community and Support

• Forum: Official ABINIT Forum (forum.abinit.org).
• Events: ABINIT schools held annually (often in Europe/Louvain).
• Development: Hosted on GitHub (switched from diverse repos).

Verification & Sources

Primary sources:

1. Official website: https://www.abinit.org/
2. Documentation: https://docs.abinit.org/
3. X. Gonze et al., Comput. Phys. Commun. 180, 2582 (2009) - ABINIT overview
4. X. Gonze et al., Comput. Mater. Sci. 25, 478 (2002) - ABINIT first paper
5. F. Bottin et al., Comput. Mater. Sci. 42, 329 (2008) - PAW implementation

Secondary sources:

1. ABINIT tutorials: https://docs.abinit.org/tutorial/
2. abipy documentation: http://abinit.github.io/abipy/
3. ASE calculator: https://wiki.fysik.dtu.dk/ase/ase/calculators/abinit.html
4. Confirmed in 7/7 source lists (claude, g, gr, k, m, q, z)

Confidence: CONFIRMED - Appears in all 7 independent source lists

Verification status: ✅ VERIFIED

• Official homepage: ACCESSIBLE
• Documentation: COMPREHENSIVE and ACCESSIBLE
• Source code: OPEN (GitHub)
• Community support: Active (forum, mailing list)
• Academic citations: >3,000 (main papers)
• Active development: Regular releases