Kurate.org
Rankings
Back to Rankings

The concentration conjecture for Dirac--Coulomb operators

Habib Ammari, Jiayu Qiu

Oct 20, 2026arXiv:2610.28193v1
math-phmath.SP
Share
Scorecard· 16/16
8.0/10 impact

Resolves a 40-year-old named open problem in relativistic mathematical physics with a new, reusable two-center resolvent estimate, tempered by proof length, reliance on computer-checkable polynomial verifications and AI-assisted derivations.

Abstract

We prove the concentration conjecture for the three-dimensional Dirac operator with mass m>0m>0 and an attractive Coulomb potential generated by finitely many point charges. For a fixed total charge G∈(0,1]G\in(0,1], the lowest eigenvalue in the spectral gap is at least m1−G2m\sqrt{1-G^2}, with equality precisely when all the charge is concentrated at one point. The critical one-center operator is understood as its distinguished extension. The result also extends to arbitrary nonnegative finite charge measures. The proof uses a strict two-center estimate for the weighted free Dirac resolvent and a Birman-Schwinger comparison. We establish the two-center estimate through an auxiliary axial operator, combining a zero-energy bound and a scalar resolvent estimate with a radial decomposition, an angular Schur complement, and exact polynomial inequalities.

AI Impact Assessments

(1 model)

Core Contribution. The paper claims to prove the concentration conjecture for 3D Dirac–Coulomb operators. This is an open problem posed by Klaus (1980) and formulated in general form by Esteban–Lewin–Séré, who discussed it in a survey of open problems. For fixed total charge G ≤ 1, the lowest in-gap eigenvalue of Dm − V is at least m√(1−G²), with equality only for a single point charge. The result extends to arbitrary finite nonnegative measures. The strategy has three parts. First, a Birman–Schwinger reduction at the atomic energy E_G, using monotonicity of the free resolvent. Second, a block-norm matrix argument that reduces the problem to a strict two-center estimate ‖w_B(D_m−E)⁻¹w_A‖ < m/κ for all separations. Third, a proof of that estimate by mapping singular values of the two-center weighted resolvent to eigenvalues of an auxiliary axial operator in prolate spheroidal coordinates. That operator's spectrum must avoid [−1,1]. This last step is the technical core.

Methodological Rigor. The architecture is logical and the reduction steps (Birman–Schwinger formula, Perron–Frobenius-type matrix argument, measure extension via Cauchy–Schwarz on cₘ(A,B)) look sound. The two-center estimate is proved in three regimes:

  • Large separation (k ≥ 2/5): a zero-energy gap via exact radial/angular separation of H², a scalar Coulomb-weight resolvent estimate using the explicit hydrogen spectrum, and concavity in q.
  • Small separation (k ≤ 3/16): a Schur complement on the radial kernel, with an explicit angular spectrum from Jacobi polynomials.
  • The intermediate rectangle, handled by explicit polynomial inequalities (Bernstein expansions with tabulated rational coefficients).

    • The mixture of elegant structure (exact anticommutation, ground-state factorizations) and brute-force polynomial verification is typical of a hard computational proof. Concerns:

  • The coefficient tables and the Bernstein verification are not human-checkable at a glance. They are rational-arithmetic checks that could be verified by computer algebra, but no code is supplied.
  • The paper is long and technically dense, with many delicate domain and self-adjointness arguments.
  • The disclosure that an AI system (the "ChatGPT 6 Astra" named in the paper) helped work out details of Sections 5–7 raises a verification burden. Community scrutiny will be needed before acceptance.
  • Some steps are asserted tersely ("direct calculation", "one can check"). A few are potentially delicate, such as the reduction to j = 1/2 via (3.38) and the use of norm-continuity.

    • The overall logic appears coherent and checkable in principle, but full correctness depends on the heavy computational pieces.

    Potential Impact. Resolving a named open problem in relativistic quantum mechanics is significant for mathematical physics. The physical statement is that the "united atom" binds a relativistic electron most strongly. Technically, the paper provides:

  • a new two-center weighted Dirac resolvent bound, which is a reusable tool;
  • the axial-operator reduction via prolate coordinates;
  • a Birman–Schwinger approach that sidesteps the lack of a variational principle for Dirac operators.

    • The approach could potentially extend to related problems, such as the stronger monotonicity under nuclear contraction (explicitly left open), other dimensions, magnetic fields, or the many-body/Dirac–Fock setting. Direct real-world application is minimal. It is a pure mathematical-physics result, with no computational chemistry consequences beyond rigorous confirmation of numerics already supporting the conjecture.

    Timeliness & Relevance. The conjecture has been open for over 40 years and was highlighted in recent surveys. It is a well-recognized open question in the Esteban–Lewin–Séré program. Interest in rigorous relativistic quantum chemistry is steady rather than surging. If the proof is verified, it is a notable milestone for the community.

    Strengths.

  • It resolves a precisely stated, long-standing conjecture, including the critical case G = 1 and general measures, with a strict equality characterization.
  • The Birman–Schwinger plus two-center-estimate framework is conceptually clean.
  • The exact separation of variables in prolate coordinates, with sharp constants such as the µ* gap and the explicit ρ bounds, is clever.
  • The result is complete: existence of the lowest eigenvalue and an equality case.

    • Limitations.

  • The proof is long and relies partly on verification-by-expansion that cannot be quickly audited.
  • The AI-assistance disclosure raises extra concerns about verification.
  • Theorem 1.1 is not the stronger monotonicity-under-contraction statement.
  • The approach is specific to Coulomb 1/|x| potentials in 3D, so the tools may not generalize easily.
  • No numerical sanity checks of the key inequalities (for example, plots of the margin in the intermediate regime) are presented. The margins are small in places, for example 31439/2250000 in Corollary 4.3.
  • Presentation: well organized with a clear roadmap, but notation-heavy, and some steps are left to the reader.

    • Overall. If correct, this is a high-significance result within mathematical physics and spectral theory of Dirac operators. It solves a classical conjecture with a novel technical machinery. Its audience is narrow (math-phys, spectral theory, PDE), and the foundationality is moderate, driven by the two-center resolvent estimate.

    Rating:8/ 10
    Significance 7.5Rigor 7Novelty 8Clarity 6.5

    Generated Oct 21, 2026

    Comparison History (0)

    No comparisons yet.