Density Functional Theory (DFT)
Definition and meaning of Density Functional Theory (DFT) in chemistry.
DFT (density functional theory) is a quantum mechanical modeling method that calculates the electronic structure of atoms, molecules, and solids using the electron density rather than the many-electron wave function. Because the density depends on only three spatial coordinates instead of 3N coordinates for N electrons, DFT makes accurate calculations feasible for much larger systems than traditional wave-function methods.
In more detail
The method rests on the Hohenberg-Kohn theorems, which prove that the ground-state electron density uniquely determines all ground-state properties of a system, including its energy. In practice, chemists use the Kohn-Sham approach, which reintroduces a fictitious system of non-interacting electrons to make the density calculable, with an "exchange-correlation functional" approximating the tricky electron-electron interaction terms. Because the exact functional is unknown, results depend on the approximation chosen (e.g., B3LYP, PBE), so DFT is not fully ab initio but remains far cheaper than post-Hartree-Fock methods while often giving comparable accuracy.
Key facts
| Field | Physical Chemistry |
|---|---|
| Full name | Density Functional Theory |
| Key foundation | Hohenberg-Kohn theorems (1964) and Kohn-Sham equations (1965) |
| Common functionals | B3LYP, PBE, M06-2X |
A computational chemist uses DFT with the B3LYP functional and a 6-31G(d) basis set to optimize the geometry of a transition-metal catalyst and predict its reaction energy barrier, a calculation that would be far too costly with coupled-cluster methods for a molecule of that size.
Frequently asked questions
Is DFT an ab initio method?
Not strictly. DFT is exact in principle, but the exchange-correlation functional it relies on is approximated, so results depend on the functional chosen, unlike true ab initio wave-function methods.
Why is DFT so widely used in chemistry?
It offers a good balance of accuracy and computational cost, scaling much better with system size than methods like coupled-cluster theory, making it practical for molecules with dozens to hundreds of atoms.