Are There Wavefunctions in Density Functional Theory?
At a glance
- 1. Identify the basic variable
In DFT, the electron density—not the many-electron wavefunction—is the central quantity.
- 2. Distinguish two approaches
Kohn–Sham DFT uses auxiliary orbitals; orbital-free DFT methods seek the density without them.
- 3. Interpret the orbitals carefully
Kohn–Sham orbitals help construct the density, but they are not generally the electrons’ physical orbitals.
- 4. Check what a result means
A density or total energy is a primary DFT result; orbital energies and shapes need more careful interpretation.
The phrase “there are no wavefunctions in DFT” captures an important distinction, but it is too absolute. Density functional theory (DFT) is formulated in terms of the electron density, not the full interacting wavefunction of all the electrons. Yet most calculations called DFT use Kohn–Sham orbitals, which are wavefunctions for an auxiliary, noninteracting system.
The useful distinction is between the fundamental variable of the theory and the computational tools used to calculate it. The density is central to DFT; orbitals appear in the widely used Kohn–Sham method to make the density and energy practical to calculate.
What is the basic variable in DFT?
For electrons, the interacting electronic wavefunction depends on the positions and spins of all electrons. In position-and-spin notation, it is . The electron density, by contrast, depends on position alone, with spin densities included when needed:
The Hohenberg–Kohn theorems establish, for a system’s ground state under their assumptions, that the density determines the external potential up to a constant and thus determines the ground-state properties. This makes the density a suitable foundation for a ground-state theory. It does not mean that an interacting wavefunction ceases to exist, or that every practical DFT method avoids using wavefunctions or orbitals.
Why do common DFT calculations have orbitals?
In Kohn–Sham DFT, the interacting system is represented through a constructed noninteracting system chosen to have the same ground-state density. Its one-electron orbitals are found from equations of the form
in atomic units. The effective potential includes the external potential, the classical electrostatic (Hartree) potential, and an exchange-correlation potential. The density is then assembled from the occupied orbitals:
where is the occupation of orbital , including spin according to the calculation’s convention. For a closed-shell calculation using spatial orbitals, each doubly occupied orbital has .
These are Kohn–Sham orbitals. They are solutions to an auxiliary problem, not generally the exact one-electron wavefunctions of the interacting molecule. Their purpose is to produce the correct density within the Kohn–Sham framework. The corresponding Slater determinant can be called a Kohn–Sham wavefunction, but it is likewise an auxiliary construction, not the exact interacting many-electron wavefunction.
A simple example: a doubly occupied orbital
Suppose a closed-shell calculation has just one occupied spatial Kohn–Sham orbital, with occupation two. Its density is
If the orbital is normalized, then , so integrating the density gives , the electron count. The orbital is a convenient way to represent the density in this example. Knowing its shape does not, by itself, make it the exact physical wavefunction for either electron in the interacting system.
In a real calculation, the orbitals and effective potential are determined self-consistently: an initial density gives a potential, the potential gives orbitals, and the occupied orbitals give an updated density. This process is repeated until the density and energy converge.
Are Kohn–Sham orbitals physically meaningful?
Their status depends on the quantity and the context. The ground-state density and total energy are central DFT outputs. Individual orbital shapes and energies can be useful for interpreting bonding, but they should not automatically be treated as direct observables or as exact molecular orbitals of the interacting system. In particular, approximate exchange-correlation functionals can make orbital energies quantitatively inaccurate.
There are important qualifications. For the exact Kohn–Sham system, the highest occupied orbital energy has a special relation to the first ionization energy, subject to the usual ground-state assumptions. Some orbital-based quantities can therefore have rigorous meaning in particular settings. But that does not make every Kohn–Sham orbital or eigenvalue a directly measurable property.
Also, not all DFT is Kohn–Sham DFT. Orbital-free DFT methods try to work directly with the density without solving for Kohn–Sham orbitals. They can be useful in some applications, though constructing accurate density functionals for the kinetic energy is challenging.
Common mistakes
- “DFT has no wavefunctions.” More accurately, the exact theory is based on density rather than the interacting many-electron wavefunction; standard Kohn–Sham implementations nevertheless use auxiliary orbitals.
- “Kohn–Sham orbitals are the actual electron wavefunctions.” They are generally auxiliary orbitals designed to reproduce the density, not the exact interacting orbitals.
- “DFT never uses a wavefunction.” Kohn–Sham calculations often construct a determinant from their orbitals. The issue is what that determinant represents, not whether it can be written down.
- “Orbital energies are always observable energies.” Their interpretation depends on the orbital, the functional, and the property being considered; they are not all direct experimental observables.
So, if “wavefunctions” means the interacting many-electron wavefunction as DFT’s basic variable, the phrase is a useful shorthand. If it means that DFT calculations contain no wavefunctions or orbitals at all, it is incorrect.
References
- Inhomogeneous Electron Gas — P. Hohenberg and W. Kohn, Physical Review (1964)
- Self-Consistent Equations Including Exchange and Correlation Effects — W. Kohn and L. J. Sham, Physical Review (1965)
- Density-Functional Theory of Atoms and Molecules — R. G. Parr and W. Yang
Spotted an error? Send a correction. Read how we write and check articles in our editorial policy.