| Density functional theory | |
|---|---|
| Name | Density functional theory |
| Field | Quantum mechanics |
| Introduced | 1960s |
| Known for | Electronic structure calculations |
Density functional theory
Density functional theory (DFT) is a computational quantum mechanical modelling method used to investigate the electronic structure of many-body systems, especially atoms, molecules, and solids. It reduces the many-electron problem to functionals of the electron density and is central to modern studies in condensed matter physics, theoretical chemistry, and materials science. DFT matters because it provides a practical balance between accuracy and computational cost for predicting ground-state properties and guiding experimental and industrial research.
Density functional theory originated in the work of Hohenberg and Kohn with the Hohenberg–Kohn theorems (1964) and was made practically usable by the Kohn–Sham equations (1965). Early theoretical roots also trace to the Thomas–Fermi model developed by Llewellyn Thomas and Enrico Fermi in the 1920s. The development of DFT accelerated with improved exchange–correlation approximations and the rise of high-performance computing in the late 20th century. Influential figures include John Pople, Veltman is less directly related but important in quantum theory discourse, and later contributors such as Jean-Pierre Perdew, Kieron Burke, Axel Becke, and Sham who developed key functional forms and pedagogical expositions. National laboratories such as Argonne National Laboratory, Lawrence Berkeley National Laboratory, and university groups at Massachusetts Institute of Technology, University of California, Berkeley, and Cambridge University contributed major code development and methodological advances.
DFT is founded on the principle that the ground-state energy is a unique functional of the electron density. The Hohenberg–Kohn theorems provide existence and variational principles, while the Kohn–Sham formalism maps the interacting many-electron problem onto a system of noninteracting particles moving in an effective potential, recovering the kinetic energy via single-particle orbitals. Key quantum mechanical concepts invoked include the Schrödinger equation, electronic exchange interaction, and electron correlation. The theory is connected to many-body techniques such as Green's functions, coupled cluster methods, and quantum Monte Carlo through attempts to benchmark and improve exchange–correlation descriptions. Formal extensions include time-dependent DFT (TDDFT) for excited states and current-density functional theory for magnetic and transport phenomena.
The exchange–correlation functional encapsulates all many-body effects beyond classical electrostatics and single-particle kinetic energy. Common approximations are the local density approximation (LDA), popularized using homogeneous electron gas results, and the generalized gradient approximation (GGA) with prominent examples like the Perdew–Burke–Ernzerhof (PBE) functional and Becke exchange combined with Lee–Yang–Parr (BLYP). Hybrid functionals incorporate portions of Hartree–Fock exchange; notable hybrids include B3LYP and PBE0. Meta-GGA and range-separated hybrids (e.g., HSE06) improve specific properties. Empirical and nonempirical strategies compete: functionals from groups led by John Perdew, Axel Becke, and Kristian Burke are widely used. Benchmarks rely on reference data from high-level quantum chemistry such as CCSD(T) and experimental databases like those produced by the NIST and collaborative initiatives (e.g., the Materials Project).
Practical DFT calculations use basis sets and numerical schemes implemented in software packages. Plane-wave pseudopotential approaches are common for periodic solids and are implemented in codes like VASP, Quantum ESPRESSO, and ABINIT. Localized basis and all-electron methods appear in packages such as Gaussian, NWChem, CRYSTAL, and WIEN2k. Algorithms exploit pseudopotential theory, projector augmented-wave (PAW) methods, and iterative diagonalization or density-matrix purification for large systems. High-performance computing centers and initiatives—Oak Ridge National Laboratory, NERSC, and the European Centre for Medium-Range Weather Forecasts (as an HPC example)—support large-scale DFT projects. Integration with workflows uses materials databases including the Materials Project, AFLOW, and Open Quantum Materials Database (OQMD).
DFT underpins predictions of structural, electronic, magnetic, vibrational, and thermodynamic properties. In chemistry, DFT guides reaction mechanisms, catalysis design, and spectroscopy simulation; widely cited applications include studies of heterogeneous catalysis at surfaces such as Pt (platinum) and Au (gold). In materials science, DFT aids discovery of semiconductors, battery electrode materials, and topological insulators. In condensed matter physics, DFT calculations elucidate band structures, Fermi surfaces, and magnetic order in systems like graphene, transition metal oxides, and perovskite materials. Industrial applications span pharmaceuticals, agrochemicals, and energy technologies where codes from companies like Schrödinger (company) and Materials Design support commercial workflows.
Despite successes, DFT faces limitations: approximate exchange–correlation functionals can yield self-interaction errors, poor treatment of van der Waals dispersion (partly mitigated by DFT-D corrections), and inaccurate band gaps in semiconductors and insulators. Strongly correlated systems such as Mott insulators often require beyond-DFT methods like DFT+U, dynamical mean field theory (DMFT), or embedding approaches. Current research directions emphasize nonlocal correlation functionals (e.g., vdW-DF), machine-learning assisted functionals and potentials, quantum embedding, and rigorous benchmarking against quantum Monte Carlo and experimental spectroscopies. Community efforts through organizations like the American Physical Society and conferences such as the APS March Meeting foster standards and collaboration to preserve scientific rigor and practical utility for technology and national innovation.
Category:Computational chemistry Category:Condensed matter physics Category:Quantum mechanics