REVIEW 2 major objections 5 minor
A maximal Hohenberg-Kohn theorem for non-interacting systems via potential theory
T0 review · 2 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash
Pith's one-line read This paper proves the Hohenberg–Kohn theorem for non-interacting systems in the largest admissible class of singular potentials, with quasi-everywhere positivity of the density as the exact criterion for uniqueness.
desk verdict A serious, novel proof that the non-interacting HK theorem holds in the maximal Laplace form-bounded class; the main chain is sound, with one terse step in Lemma 5.1 that a referee should ask to be completed. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the Sobolev capacity of Ω: it assigns a size to sets via the minimal H¹ energy of a test function that exceeds 1 on the set, and a condition holds 'quasi-everywhere' when it fails only on sets of zero capacity. The proof's engine is the characterization of H¹-regular states: when the density is q.e. positive, any form-bounded potential with zero expectation must send each natural orbital to a linear combination of the finitely many orbitals; a quasi-continuous representative of the potential then takes only finitely many values, and the local connectedness of the quasi-topology forces it to be constant; a dense-range lemma for multiplication operators identifies the con
What would settle it
Find a distributional potential v that is Laplace form-bounded on some connected open Ω, such that the Rayleigh quotient (5.2) has a minimizer whose precise representative vanishes on a compact set of positive capacity (or two linearly independent minimizers). Such an example would refute Lemma 5.1 and, with it, Theorems 2.5 and Corollary 2.6; conversely, a Slater determinant with q.e.-positive density and a nonzero form-bounded v with ⟨v, ρ⟩ = 0 would refute Theorem 2.9.
Extended reading notes
Core claim
The paper's central claim is the characterization of H¹-regular states: a finite-rank state is regular—no nonzero Laplace form-bounded distributional potential has zero expectation value in it—if and only if its density is strictly positive quasi-everywhere. From this, the Hohenberg–Kohn theorem follows by the variational argument for any finite-rank ground state whose density is q.e.-positive, and fails when the density vanishes on a set of positive capacity. For non-interacting systems, the paper shows the lowest single-particle eigenfunction of −Δ+v is unique and q.e.-positive for form-bounded distributional v, so the many-body density inherits q.e.-positivity; consequently any two non-in
Load-bearing premise
The load-bearing premise is Lemma 5.1's strong maximum principle: a single-particle Schrödinger operator with a merely form-bounded distributional potential has a unique ground state that is strictly positive quasi-everywhere; if this fails for any admissible potential, the non-interacting density need not be q.e.-positive and the Hohenberg–Kohn corollary collapses.
Editorial extensions
If this is right
- For weakly correlated states with finite occupation numbers, the density-to-potential map is injective exactly when the density is positive quasi-everywhere; densities vanishing on a positive-capacity set admit distinct potentials with the same ground state.
- Any non-interacting system with discrete ground-state energy satisfies that condition, so uniqueness of the Kohn–Sham potential holds within the full Laplace form-bounded class.
- The universal density functional cannot have a unique subgradient at any density that vanishes on a set of positive capacity, so q.e.-strict positivity is necessary for its differentiability.
- The classical approach through unique continuation of many-body wavefunctions is neither necessary nor sufficient here; the paper's mechanism is (quasi-)unique continuation of the density.
Reading between the lines
- If the finite-rank assumption is removed, the criterion would extend to interacting systems with infinitely many occupied orbitals; the necessary direction already holds without it, so the gap is confined to the sufficiency step.
- The proof constructs a quasi-continuous representative of the potential from ratios of natural orbitals, which suggests a practical reconstruction algorithm for Kohn–Sham potentials from densities—something the paper does not address.
- A direct test of sharpness is to take densities with explicit polar zero sets (for instance |x|^α in dimension d ≥ 3) and check whether any form-bounded distributional potential annihilates them; the necessary direction predicts such densities are never regular.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proves a Hohenberg–Kohn theorem for non-interacting Schrödinger systems with external potentials in the maximal class of Laplace form-bounded distributions. The main results are: (i) Theorem 2.1, an if-and-only-if criterion for the HK theorem under a finite-rank assumption on the single-particle density matrix, phrased in terms of quasi-everywhere positivity of the density; (ii) Theorem 2.9, a characterization of H^1-regular states via q.e. strict positivity of the density; and (iii) Theorem 2.5, which shows that for non-interacting systems with a discrete ground state energy the ground-state density is q.e. positive. The proof of Theorem 2.5 relies on Lemma 5.1, a strong maximum principle for a single-particle Schrödinger operator with form-bounded distributional potential. Corollary 2.6 then asserts uniqueness of the Kohn–Sham potential within the class of Laplace form-bounded potentials. The paper also develops potential-theoretic tools (capacitary measures, Maz'ya–Verbitsky criteria, quasi-topology) and argues that the fundamental mechanism is unique continuation of the density rather than of the wavefunction.
Significance. If the results hold, they constitute a substantial advance: the HK theorem is extended from the locally-L^p class of Garrigue to the natural maximal class of potentials preserving the H^1 form domain, and the proof introduces a new potential-theoretic mechanism (q.e. positivity of the density) that is both necessary and sufficient under the finite-rank assumption. The paper is conceptually novel in replacing unique continuation of the many-body wavefunction with quasi-strict positivity of the density, and the necessary part of Theorem 2.9 is shown without the finite-rank restriction. The proofs are detailed and self-contained, with explicit uses of classical potential theory; there are no fitted constants or circular dependencies. The central claims are therefore significant for the mathematical foundations of DFT, provided the identified gaps in the proof of Lemma 5.1 and the final step of Theorem 2.9 are repaired.
major comments (2)
- [§5, Lemma 5.1, after Eq. (5.11)] The final step of the proof is not justified as written. After assuming that the zero set Z_u has positive capacity and choosing a compact K⊂Z_u of positive capacity, the argument shows that u=0 a.e. in every bounded connected Lipschitz domain ω with K⊂⊂ω⊂Ω. The text then says 'As ω is arbitrary, we must have u=0'. This is not immediate, because ω is constrained to contain K; the conclusion only covers subdomains meeting K. A rigorous proof requires an exhaustion argument: for every x∈Ω, choose a connected compact set containing K and x (using connectedness of Ω), then a bounded connected Lipschitz ω with that set compactly inside Ω, and apply the previous argument to conclude u=0 in a neighbourhood of x. This yields u=0 a.e. in Ω, contradicting the non-triviality of u. Since Lemma 5.1 is the load-bearing step for Theorem 2.5 and Corollary 2.6, this gap must be completed explicitly.
- [§4.2, Step 4 of the proof of Theorem 2.9] The final step concludes that 'v−λ=0 in the distributional sense, which completes the proof'. However, the desired conclusion of H^1-regularity is v=0, not v=λ. If v is the constant distribution λ, then v̂Γ=λ n Γ, which is non-zero for λ≠0, contradicting the standing assumption v̂Γ=0. The proof should explicitly use (4.19) (or its trace) to infer λ=0. This is a short argument, but as written the proof is incomplete at a point necessary for the statement of Theorem 2.9 and hence Theorem 2.1.
minor comments (5)
- [§4.1, Eq. (4.9)] There is a typo in the integration variable: the final expression should be ∫_{F∩L}(G_2*μ|_F)(x)dμ(x) = ||G_1*(μ|_F)||^2_{L^2}. Also, the sentence 'since the support of μ is contained in Ω, it suffices to verify (4.5) for compact subsets of Ω' deserves a one-line justification: for any compact F⊂R^d, μ(F)=μ(F∩K) and Cap(F)≥Cap(F∩K).
- [§5, Lemma 5.3] The proof is only a sketch. In particular, the 'only if' direction via Weyl's criterion is asserted without detail. Since Lemma 5.3 is used to guarantee finite rank and the fact that the lowest single-particle eigenfunction is a natural orbital, it would be helpful to expand the min-max argument, especially the claim that less than n discrete one-particle eigenvalues forces the n-particle ground-state energy to be non-discrete.
- [§5, after Eq. (5.9)] In the display following (5.9), '∥φ^2∥_{L^2}' should be '∥φ∥^2_{L^2}'. Also, in the proof of Lemma 5.1 the statement says (5.1) is extended to H^1_0 'by Lemma 4.2'; Lemma 4.2 is stated for measures, though the same polarization argument applies to general form-bounded distributions. Please clarify the reference.
- [§5, final paragraph] Typo: 'Saard's theorem' should be 'Sard's theorem'.
- [§2, Remark 2.2] The phrase 'maximal class of potentials preserving the form domain' is slightly over-stated: the infinitesimal form bound in V(Ω) is sufficient but not necessary for the form domain to be H^1_0. The later Remark 2.7(1) correctly refers to 'Laplace form-bounded potentials' with a ground state, but the abstract and Corollary 2.6 could be more precise about the exact class used.
Circularity Check
No significant circularity: the central characterization is proved by independent potential-theoretic constructions; self-citations are motivational only.
full rationale
The derivation chain is not circular. Theorem 2.9 characterizes H1-regular states (injectivity of v ↦ v̂Γ) by q.e. strict positivity of the density through genuinely independent constructions: necessity builds infinitesimally form-bounded measures on positive-capacity zero sets via Maz'ya–Verbitsky and Bessel-capacity duality (Lemma 4.1), and sufficiency uses a quasi-continuous representative of v, Fuglede's quasi-connectedness, and the dense-range Lemma 4.7. Theorem 2.1 then follows by the variational principle plus Theorem 2.9, and Theorem 2.5/Corollary 2.6 rest on Lemma 5.1, whose maximum-principle proof is adapted from Orsina–Ponce / Brezis–Ponce rather than from the target HK statement. There are no fitted constants renamed as predictions, and no load-bearing uniqueness theorem is imported from the author's own prior work; the self-citations occur only as motivation (Section 1.1) and in an illustrative footnote. The main caveats are non-circular: Remark 2.4 explicitly flags the finite-rank/weak-correlation limitation, and the final step of Lemma 5.1 ('As ω is arbitrary, we must have u=0') is terse and would require a careful exhaustion argument to be fully rigorous, but this is a completeness/correctness gap, not a reduction of the conclusion to the hypotheses.
Assumptions & free parameters
assumptions (5)
- standard math KLMN theorem permits definition of H_n(v,w) via its quadratic form for distributions satisfying the infinitesimal form bound (2.5).
- standard math Classical potential theory facts: dual capacity formula (Lemma 4.3), Maz'ya–Verbitsky infinitesimal form-bound criterion (Lemma 4.4), precise representatives/quasi-continuity (Thm 3.4, Props 3.5–3.12), Fuglede quasi-connectedness (Lemma 4.9).
- standard math Fermionic second quantization and CAR: identity (4.17) for commutators with single-particle excitations.
- ad hoc to paper Finite-rank (weak-correlation) assumption on γ_Γ.
- domain assumption For v∈V(Ω), the ground state of H_n(v,0) exists only if h(v) has a minimizer; Lemmas 5.1 and 5.3 assume the relevant minimizers/eigenvalues exist.
Cite this review
Pith. "Pith review of A maximal Hohenberg-Kohn theorem for non-interacting systems via potential theory." pith.science (2026). https://pith.science/paper/EEEPVHQK
@misc{pith2026260712852,
author = {Pith},
title = {Pith review of: A maximal Hohenberg-Kohn theorem for non-interacting systems via potential theory},
year = {2026},
howpublished = {\url{https://pith.science/paper/EEEPVHQK}},
note = {Machine review of arXiv:2607.12852}
}
read the original abstract
We show that for Schr\"odinger operators in a connected domain, the Hohenberg-Kohn theorem holds within the class of Laplace form-bounded external potentials if and only if the single-particle density is strictly positive quasi-everywhere. Furthermore, we show that this condition is satisfied for non-interacting Schr\"odinger operators whenever a ground-state exists. Consequently, we establish the Hohenberg-Kohn theorem for non-interacting systems, and thereby the uniqueness of the Kohn-Sham potential, within the maximal class of Laplace form-bounded distributions. The main ingredient to establish these results is a characterization of regular states, whose proof relies on tools from classical potential theory. Moreover, this characterization reveals that, in the continuum setting, the fundamental mechanism underlying the Hohenberg-Kohn theorem is the (quasi)-unique continuation of the density rather than of the many-body wavefunction.
Reviewed August 2, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.