Skip to content
LESSON NOTES · 01

1. Density, constrained search, and the variational principle

Position in the course: Lesson 1 of 10. Complete the preceding derivation and use the explained exercises to check understanding.

1. Purpose and assumptions

A density records the expected number of electrons per volume, rather than a list of electron positions. Integrating over all space gives the electron count. The many-electron wavefunction has spatial and spin coordinates for every particle; taking its squared modulus and integrating out all but one coordinate produces the density. The prefactor N follows from indistinguishability. This reduction discards explicit pair information, but the ground-state variational problem can nevertheless be expressed using a density functional. A functional accepts a whole function, not a single number.

Work at fixed nuclei, fixed electron number, zero temperature, and a scalar external potential in atomic units. The external potential is normally the electron–nuclear attraction; the electron–electron interaction is kept in W. In the original nondegenerate Hohenberg–Kohn argument, assume two different potentials produce the same density. Applying the strict variational inequality in both directions and adding them gives an impossibility. Potentials differing only by a constant are excluded because they shift energy without changing the state.

The constrained-search construction makes the logic particularly transparent. First choose a candidate density with integral N; among antisymmetric normalized wavefunctions yielding that density, minimize the internal energy. Then minimize over the densities. The nested minimizations cover the same physical states as a direct wavefunction minimization. This establishes an exact formal framework but supplies neither an inexpensive algorithm nor the exact functional. Nonnegative normalization alone is not a complete representability criterion.

To obtain stationarity, vary E minus a multiplier times the particle-number constraint. The Euler equation is δF/δn plus v equals μ wherever the variation is allowed. The multiplier is a chemical potential, not a freely chosen extra force. Degenerate states, fractional populations, spin dependence, and nondifferentiability require refinements of this elementary picture. The value of this lesson is knowing exactly what DFT claims before introducing its practical approximations.

2. Derivation step by step

Read each equality with its assumptions. Atomic units are used for DFT equations unless another unit is stated; TB parameters retain explicit energy and length units. The conjugate transpose is denoted by a dagger, and a prime on a coordinate denotes a separate integration variable.

\[ \begin{aligned} n(\mathbf r)&=N\sum_{\sigma_1}\int|\Psi(\mathbf r\sigma_1,x_2,\ldots,x_N)|^2dx_2\cdots dx_N,\\ F[n]&=\min_{\Psi\to n}\langle\Psi|\hat T+\hat W|\Psi\rangle,\\ E_v[n]&=F[n]+\int v(\mathbf r)n(\mathbf r)d\mathbf r,\\ \min_n E_v[n]&=\min_\Psi\langle\Psi|\hat H_v|\Psi\rangle=E_0. \end{aligned} \]

Here each integration over \(x_j\) includes its spatial integral and spin sum; the remaining explicit sum is only over \(\sigma_1\).

2.1. Why the order of minimization matters

The inner search must keep the entire candidate density fixed, not only its integral. If only N were fixed, every inner search would return the same global ground state and F would cease to distinguish densities. The universal functional is universal with respect to the scalar external potential while the kinetic operator and electron interaction remain fixed. Switching to a different interaction or relativistic Hamiltonian changes that functional. Nuclear repulsion is outside the electronic search because it is independent of the electronic state at fixed geometry. These distinctions make the constrained search a definition with precise inputs rather than a slogan about replacing a wavefunction by three coordinates.

3. Worked example

For a normalized orbital φ with two opposite-spin electrons, n=2|φ|² and ∫n=2. Replacing n by |φ|² accidentally describes a one-electron system. If v is shifted by C, E shifts by CN: the internal constrained search is unchanged, while ∫Cn=CN. This simple identity explains why an absolute scalar potential is not observable.

4. Exercises with explained solutions

Exercise. Derive the particle-number Euler equation and explain why μ is constant in space.

Explained solution. Write L=F[n]+∫vn−μ(∫n−N). Its variation is ∫[δF/δn+v−μ]δn. A single global constraint introduces one scalar μ. Setting the integrand to zero on allowed directions gives the equation; inequality boundaries or nondifferentiability need additional care.

Further check. State the units and the allowed regime for every parameter in the worked example. Change one assumption and identify which derivation step must be revisited. A correct explanation names the affected constraint, operator, or boundary condition rather than merely saying that the answer changes.

5. Misconceptions and limitations

The theorem is about ground states under specified conditions, not a proof that every measured spectrum equals a Kohn–Sham spectrum. It also does not make approximate energies rigorous upper bounds to the exact energy.

Self-consistent density cycle with residual check

The illustration is an original teaching schematic. It is not output from a numerical materials simulation.

6. Connections and sources

Related: localized-basis theory · Molecular electronic structure

ABACUS · VASP · Quantum ESPRESSO · CP2K

The explanations, algebra, and invented worked examples are original teaching synthesis. The cited papers establish the underlying theories, not the numerical toy values.

Quantum mechanics · Molecular methods · Tight binding · Molecular dynamics · VASP · Quantum-Espresso · CP2K · ABACUS


Course overview · Next lesson