Learn more about FHI-aims

 

FHI-aims is an all-electron electronic structure code based on numeric atom-centered orbitals. It enables first-principles simulations with very high numerical accuracy for production calculations, with excellent scalability up to very large system sizes (thousands of atoms) and up to very large, massively parallel supercomputers (ten thousand CPU cores).

 

Broad accuracy on par with the best available benchmarks e.g demonstrated in:

Science 351.6280 (2016)

J. Phys. Chem. 8, 1449 (2016)

Phys. Rev. Materials 1, 033803 (2017)

 

Scalability to very large systems and CPU numbers is demonstrated, for example, in:

J. Chem. Phys. 161, 024112 (2024)

JCOMEL 26, 213201 (2014)

Phys. Rev. Lett. 111, 065502 (2013)

Comp. Phys. Commun. 192, 60 (2015)

Versatile

Versatile

  • Non-periodic and periodic systems for LDA, GGA, and hybrid functionals, including:
    • Dipole corrections for surface slab calculations
    • Large vacuum regions for surface slabs with practically no overhead
  • Preconstructed, hierarchical numeric atom-centered basis sets for elements 1-102, for systematic convergence from fast qualitative to meV-level total energy convergence
  • Numeric atom-centered valence correlation consistent basis sets for systematic convergence of many-body perturbation methods (currently, for H-Ar)
  • Support for Gaussian basis sets of quantum chemistry
Precise & Accurate

Precise & Accurate

  • Preconstructed default settings for the most important numerical choices (grids, Hartree potential, basis cutoff, etc.): "light", "intermediate", "tight", and “really_tight".
  • Density-functional approximations for molecules and solids, including gradients (forces):
    • LDA: Perdew/Wang (1992), Perdew/Zunger (1981), Vosko/Wilk/Nusair (1980) 
    • GGA: AM05, BLYP, PBE, PBEsol, RPBE, revPBE, PBEint 
    • meta-GGA: M06-L, TPSS, revTPSS, TPSSloc, SCAN
    • Many more functionals available through the integration of LibXC 
  • Hartree-Fock and hybrid functionals
    • Non-periodic and periodic, including analytical gradients and stress): Hartree-Fock, PBEh (incl. PBE0), HSE03, HSE06, B3LYP, M06 and M11
  • Tkatchenko-Scheffler interatomic dispersion (van der Waals) correction
  • Many-body dispersion correction (integration of LibMBD)
  • "XYG3" doubly-hybrid functional (currently, post-SCF only) 
  • A self-consistent version of Langreth-Lundvist non-local vdW-DF (Ville Havu et al., Helsinki) 
  • Many-body perturbation theory (post-SCF correction for single-point non-periodic geometries):
    • Second-order Moller-Plesset theory (MP2), Random Phase Approximation (RPA), and renormalized second-order perturbation theory 
    • G0W0 approximation for single-carrier like excited energy levels 
    • Self-consistent GW (only for molecules) 
Scalable

Scalable

  • DFT-LDA/GGA: All-electron accuracy at a computational cost comparable to plane-wave/pseudopotential implementations
  • System size range up to thousand(s) of atoms, with O(N) like scaling for the most expensive operations [limiting factor: Conventional O(N3) eigensolver beyond this range]
  • Seamlessly parallel (time and memory) from desktop up to currently (ten)thousands of CPUs
  • Specifically optimized, massively parallel conventional eigensolver ELPA to minimize scalability barriers independent of the type of system treated
  • Unified access to eigensolvers and density matrix solvers through ELSI 
  • Partial support for GPU acceleration (CUDA)
  • Memory-efficient Hartree-Fock/exact exchange implementation due to use of MPI-3 intra-node shared memory arrays
  • Accurate, fast, and more memory-efficient "localized" resolution of identity method for exchange-correlation beyond LDA/GGA (currently functions as a base method for Hartree-Fock and hybrid functionals, for more than 1,000 heavy atoms on large supercomputers)