10
$\begingroup$

I have asked a similar question but after thinking about it I have a more specific question.

According to Ullrich, Carsten A.. Time-Dependent Density-Functional Theory : Concepts and Applications, the Hohenberg–Kohn theorem states

In a finite, interacting N-electron system with a given particle–particle interaction there exists a one-to-one correspondence between the external potential $V(r)$ and the ground-state density $n_0(r)$. In other words, the external potential is a unique functional of the ground-state density, $V[n_0](r)$, up to an arbitrary additive constant.

The way I understand it, assuming V differs by more than a constant and psi differs by more than a phase, the logic is: one potential (V) yields one hamiltonian (H) which yields a wave function (Ψ) which yields a density (n). V -> Ψ -> n.

V -> Ψ (ignoring constant) This is proven in HK theorem via proof by contradiction

Ψ -> n (ignoring phase factor) This is proven in HK theorem via proof by contradiction.

Then they conclude that: We have thus shown that $Ψ_0$ and $Ψ′_0$ give different densities $n_0$ and $n′_0$; but in the first step we showed that $Ψ_0$ and $Ψ′_0$ also come from different potentials $V$ and $V′$. Therefore, a unique one-to-one correspondence exists between potentials and ground-state densities, which can be formally expressed by writing $V[n_0](r)$, and thus $V[n_0]$.

This confuses me because they have only proven "one direction." They have proven that two V's cannot give the same Ψ but they haven't proven that one V cannot yield more than one Ψ. Likewise they have proven that two Ψ's cannot give the same n but haven't proven that one Ψ cannot yield more than one n. Perhaps I'm missing something obvious but any insight would be appreciated.

$\endgroup$
  • 3
    $\begingroup$ Isn't one $\Psi$ yielding multiple $n$ trivially dealt with by the fact that $ n\left( \textbf{r} \right) = {q\left|\psi\left( \textbf{r} \right)\right|^{2}} $? The other part ($V$ yielding multiple $\Psi$) seems to go back to your other question about uniqueness of solutions to the Schrodinger equation. $\endgroup$ – Kevin J. M. Oct 21 at 1:20
8
$\begingroup$

I'm not familiar with the Carsten Ullrich text you mention. However, one possibility is that he followed Hohenberg and Kohn's example of assuming a non-degenerate ground state. If the ground state $\Psi$ is non-degenerate, $V$ can by definition only produce one $\Psi$.

You're correct that the proofs have to be modified when there is a degenerate ground state manifold. There's a nice discussion in K. Capelle, C. A. Ullrich, G. Vignale, Degenerate ground states and nonunique potentials: breakdown and restoration of density functionals, Physical Review A 76, p. 012508 (2007) (arXiv link) that may answer some of your questions.

I agree with Kevin's comment that the density $n$ is fixed by the inner product $|\psi|^2$.

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.