| Perdew–Burke–Ernzerhof | |
|---|---|
| Name | Perdew–Burke–Ernzerhof |
| Field | Density functional theory |
| Developers | John P. Perdew; Kieron Burke; Matthias Ernzerhof |
| Introduced | 1996 |
| Related | Local density approximation, Generalized gradient approximation, Hybrid functional |
Perdew–Burke–Ernzerhof
Perdew–Burke–Ernzerhof (commonly abbreviated PBE) is a widely used non-empirical exchange–correlation functional within density functional theory (DFT) for electronic-structure calculations. Designed to satisfy known exact constraints on the exchange energy and correlation energy, PBE provides a balance between computational efficiency and predictive accuracy for atoms, molecules, and solids, making it central to modern computational quantum physics and materials modeling.
PBE sits within the family of generalized gradient approximations (GGA) that improve upon the local density approximation (LDA) by incorporating local gradients of the electron density. In quantum physics, DFT with functionals like PBE enables tractable solutions to the many-electron Schrödinger equation for systems with dozens to thousands of electrons, underpinning research in condensed matter physics, physical chemistry, and materials science. The non-empirical construction by John P. Perdew, Kieron Burke, and Matthias Ernzerhof emphasizes satisfaction of uniform scaling, exchange–correlation hole, and other constraints derived from quantum many-body theory and the Hartree–Fock method.
PBE decomposes the exchange–correlation energy into an exchange part and a correlation part that depend on the density n(r) and its gradient ∇n. The exchange enhancement factor was derived to respect exact properties known from the homogeneous electron gas studied in quantum Monte Carlo simulations by researchers such as David M. Ceperley and Bernard J. Alder. The correlation component builds on the Perdew–Wang parameterizations of the correlation energy for the uniform electron gas. The functional is formulated without empirical fitting to molecular data, relying instead on constraints from many-body theory, sum rules, and the Lieb–Oxford bound, which grounds its transferability across chemical environments.
Several variants address specific shortcomings of PBE. PBEsol (PBE for solids) modifies gradient coefficients to improve equilibrium properties of densely packed solids and surfaces; it was developed with contributions from condensed-matter theorists seeking improved lattice constants and bulk moduli. The revised PBE (RPBE) adjusts the exchange enhancement to better describe adsorption energies in surface science, used by groups at institutions like the Max Planck Society and Lawrence Berkeley National Laboratory. Hybrid functionals such as PBE0 mix a fraction of exact exchange from the Hartree–Fock method with PBE correlation to reduce self-interaction errors; PBE-based hybrids are implemented in packages developed at IBM Research and university codes. Meta-GGA and range-separated hybrids further extend the approach, linking PBE ideas with functionals like SCAN and HSE.
PBE is implemented in virtually all major electronic-structure software: VASP (Vienna Ab initio Simulation Package), Quantum ESPRESSO, ABINIT, Gaussian, WIEN2k, GPAW, and CASTEP. These codes provide plane-wave, augmented-plane-wave, localized basis, and real-space representations, enabling calculations for molecules, surfaces, and bulk crystals. Pseudopotential and projector-augmented wave (PAW) datasets compatible with PBE are distributed by projects such as the PSLibrary and PSPGEN communities, facilitating reproducible simulations across high-performance computing centers and national labs like Argonne National Laboratory.
PBE generally yields improved atomization energies, geometries, and elastic properties versus LDA, but it tends to overestimate bond lengths and underestimate band gaps in semiconductors and insulators due to self-interaction and missing derivative discontinuity. Benchmarks against higher-level methods—coupled-cluster (e.g., CCSD(T)), quantum Monte Carlo, and many-body perturbation theory methods like GW—quantify PBE errors. For adsorption and dispersion-dominated interactions, PBE often requires corrections such as empirical dispersion (DFT-D) or nonlocal van der Waals functionals (e.g., vdW-DF). The balance between accuracy and computational cost makes PBE the default choice in high-throughput materials discovery projects like the Materials Project and AFLOW.
PBE underlies predictive studies of crystal structures, phase stability, defect energetics, surface chemistry, and catalytic mechanisms. It has been instrumental in discovering novel two-dimensional materials, modeling battery electrode materials investigated at institutions like the Argonne Battery Hub, and simulating electronic structure in complex oxides studied by the National Institute of Standards and Technology. Coupled with workflows for high-throughput screening, PBE enables explorations of equitable technology transitions—screening cheaper, less resource-intensive materials for renewable-energy technologies and informing policy-relevant assessments of material criticality.
Widespread use of PBE in computational materials design raises ethical questions about equity in access to computational resources, open data, and reproducibility. Projects employing PBE influence industrial decisions on battery materials, catalysts, and semiconductors, which have socioeconomic and environmental impacts. Equity-focused scholars and computational scientists advocate for open-source implementations (e.g., Quantum ESPRESSO), FAIR data practices, and inclusive collaborations with communities affected by extraction and supply chains. Responsible deployment of PBE-based predictions demands transparency about uncertainty, limitations relative to higher-level quantum methods, and attention to social justice in technology adoption.
Category:Density functional theory Category:Computational chemistry Category:Materials science