OctoInformBeta

OpenAI preprint, September 2026 · §1. Introduction · §1

§1 (setting): quadratic forms, negative eigenvalue moments, and optimal constants: Variational negative moments and optimal constants with the extremizer potential for γ∈(1/2,3/2)\gamma \in (1/2, 3/2)

Schrödinger quadratic forms and the sharp Lieb-Thirring assertion.

Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its declaration both in Lean and in words.

The Setting

Consider a one-dimensional Schrödinger operator −d2/dx2−W-d^2/dx^2 - W on ℝ\mathbb{R} with an almost everywhere nonnegative potential WW. For a spectral exponent γ∈(1/2,3/2)\gamma \in (1/2, 3/2), the problem is to find the sharp constant bounding the sum of the γ\gamma-th powers of the magnitudes of its negative eigenvalues by the integral of Wγ+1/2W^{\gamma+1/2}. One also asks whether maximizing the ratio using only a single bound state gives the same optimal constant, and which potential attains equality.

Quadratic Forms and Negative Moments

The Sobolev space H1(ℝ;ℂ)H^1(\mathbb{R}; \mathbb{C}) consists of pairs (u,u′)(u, u') of complex-valued functions u,u′∈L2(ℝ;ℂ)u, u' \in L^2(\mathbb{R}; \mathbb{C}) with respect to Lebesgue measure whose weak derivative u′u' satisfies (1) for every smooth compactly supported test function φ∈Cc∞(ℝ;ℂ)\varphi \in C_c^\infty(\mathbb{R}; \mathbb{C}): ∫ℝu(x)φ′(x)dx=−∫ℝu′(x)φ(x)dx.\int_{\mathbb{R}} u(x)\varphi'(x)\,dx = -\int_{\mathbb{R}} u'(x)\varphi(x)\,dx.(1)✦

The L2L^2 inner product ⟨u,v⟩L2∈ℂ\langle u, v \rangle_{L^2} \in \mathbb{C} of u,v∈H1(ℝ;ℂ)u, v \in H^1(\mathbb{R}; \mathbb{C}) is the integral (2): ⟨u,v⟩L2=∫ℝu(x)¯v(x)dx.\langle u, v \rangle_{L^2} = \int_{\mathbb{R}} \overline{u(x)}\,v(x)\,dx.(2)✦

The Schrödinger quadratic form h−W(u,v)∈ℂh_{-W}(u, v) \in \mathbb{C} of a real-valued potential W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} for u,v∈H1(ℝ;ℂ)u, v \in H^1(\mathbb{R}; \mathbb{C}) is the complex number (3): h−W(u,v)=∫ℝu′(x)¯v′(x)dx−∫ℝW(x)u(x)¯v(x)dx.h_{-W}(u, v) = \int_{\mathbb{R}} \overline{u'(x)}\,v'(x)\,dx - \int_{\mathbb{R}} W(x)\,\overline{u(x)}\,v(x)\,dx.(3)✦

A weak eigenfunction u∈H1(ℝ;ℂ)u \in H^1(\mathbb{R}; \mathbb{C}) of h−Wh_{-W} at negative energy −k2-k^2, for a potential W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} and k∈ℝk \in \mathbb{R}, is an element of H1(ℝ;ℂ)H^1(\mathbb{R}; \mathbb{C}) such that k>0k > 0 and (4) holds for every v∈H1(ℝ;ℂ)v \in H^1(\mathbb{R}; \mathbb{C}): h−W(v,u)=−k2⟨v,u⟩L2.h_{-W}(v, u) = -k^2 \langle v, u \rangle_{L^2}.(4)✦

An orthonormal family (ui)1≤i≤N(u_i)_{1 \le i \le N} in L2L^2 of elements in H1(ℝ;ℂ)H^1(\mathbb{R}; \mathbb{C}), for an integer N≥0N \ge 0, is a family in H1(ℝ;ℂ)H^1(\mathbb{R}; \mathbb{C}) satisfying (5) for all 1≤i,j≤N1 \le i, j \le N: ⟨ui,uj⟩L2={1,i=j,0,i≠j.\langle u_i, u_j \rangle_{L^2} = \begin{cases} 1, & i = j, \\ 0, & i \ne j. \end{cases}(5)✦

The variational negative moment ℳγ(W)∈[0,∞]\mathcal{M}_\gamma(W) \in [0, \infty] of a potential W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} at exponent γ∈ℝ\gamma \in \mathbb{R} is the supremum (6) over all integers N≥0N \ge 0, all L2L^2-orthonormal families u=(ui)1≤i≤Nu = (u_i)_{1 \le i \le N} in H1(ℝ;ℂ)H^1(\mathbb{R}; \mathbb{C}), and all k=(ki)1≤i≤N∈ℝNk = (k_i)_{1 \le i \le N} \in \mathbb{R}^N such that each uiu_i is a weak eigenfunction of h−Wh_{-W} at negative energy −ki2-k_i^2: ℳγ(W)=supN,u,k∑i=1Nki2γ.\mathcal{M}_\gamma(W) = \sup_{N,\, u,\, k} \sum_{i=1}^N k_i^{2\gamma}.(6)✦

The variational one-state moment ℳγ(1)(W)∈[0,∞]\mathcal{M}_\gamma^{(1)}(W) \in [0, \infty] of a potential W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} at exponent γ∈ℝ\gamma \in \mathbb{R} is the supremum (7) over all normalized functions u∈H1(ℝ;ℂ)u \in H^1(\mathbb{R}; \mathbb{C}) with ⟨u,u⟩L2=1\langle u, u \rangle_{L^2} = 1 and all k∈ℝk \in \mathbb{R} such that uu is a weak eigenfunction of h−Wh_{-W} at negative energy −k2-k^2: ℳγ(1)(W)=supu,kk2γ.\mathcal{M}_\gamma^{(1)}(W) = \sup_{u,\, k} k^{2\gamma}.(7)✦

Admissible Potentials and Optimal Constants

With a+=max⁡(a,0)a_+ = \max(a, 0) denoting the positive part of a∈ℝa \in \mathbb{R}, an admissible potential W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} at exponent γ∈ℝ\gamma \in \mathbb{R} is a potential satisfying (8): W(x)≥0 for almost every x∈ℝ,W∈L(γ+1/2)+(ℝ).W(x) \ge 0 \;\text{ for almost every } x \in \mathbb{R}, \qquad W \in L^{(\gamma + 1/2)_+}(\mathbb{R}).(8)✦

The potential mass mγ(W)∈[0,∞]m_\gamma(W) \in [0, \infty] of a potential W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} at exponent γ∈ℝ\gamma \in \mathbb{R} is the Lebesgue integral (9): mγ(W)=∫ℝW+(x)γ+1/2dx.m_\gamma(W) = \int_{\mathbb{R}} W_+(x)^{\gamma + 1/2}\,dx.(9)✦

The semiclassical constant Lγcl∈ℝL_\gamma^{\mathrm{cl}} \in \mathbb{R} at exponent γ∈ℝ\gamma \in \mathbb{R} is the real number defined in terms of Euler’s gamma function Γ\Gamma by (10): Lγcl=Γ(γ+1)2πΓ(γ+3/2).L_\gamma^{\mathrm{cl}} = \frac{\Gamma(\gamma + 1)}{2\sqrt{\pi}\,\Gamma(\gamma + 3/2)}.(10)✦

The sharp constant Lγ∈ℝL_\gamma \in \mathbb{R} at exponent γ∈ℝ\gamma \in \mathbb{R} is the real number given in terms of LγclL_\gamma^{\mathrm{cl}} by (11): Lγ=2(γ−1/2γ+1/2)γ−1/2Lγcl.L_\gamma = 2\left(\frac{\gamma - 1/2}{\gamma + 1/2}\right)^{\gamma - 1/2} L_\gamma^{\mathrm{cl}}.(11)✦

The variational optimal constant Λγ∈[0,∞]\Lambda_\gamma \in [0, \infty] at exponent γ∈ℝ\gamma \in \mathbb{R} is the supremum (12) over all potentials W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} admissible at exponent γ\gamma with 0<mγ(W)0 < m_\gamma(W): Λγ=supWℳγ(W)mγ(W).\Lambda_\gamma = \sup_{W} \frac{\mathcal{M}_\gamma(W)}{m_\gamma(W)}.(12)✦

The variational one-state constant Λγ(1)∈[0,∞]\Lambda_\gamma^{(1)} \in [0, \infty] at exponent γ∈ℝ\gamma \in \mathbb{R} is the supremum (13) over all potentials W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} admissible at exponent γ\gamma with 0<mγ(W)0 < m_\gamma(W): Λγ(1)=supWℳγ(1)(W)mγ(W).\Lambda_\gamma^{(1)} = \sup_{W} \frac{\mathcal{M}_\gamma^{(1)}(W)}{m_\gamma(W)}.(13)✦

The extremizer potential W*:ℝ→ℝW_* \colon \mathbb{R} \to \mathbb{R} at spectral exponent γ∈ℝ\gamma \in \mathbb{R}, with r=(γ−1/2)−1r = (\gamma - 1/2)^{-1}, is the potential defined for x∈ℝx \in \mathbb{R} by (14): W*(x)=r+1cosh⁡2(rx).W_*(x) = \frac{r + 1}{\cosh^2(rx)}.(14)✦

The Definition

Definition 1. For a spectral exponent γ∈(1/2,3/2)\gamma \in (1/2, 3/2) and a potential W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} admissible at exponent γ\gamma, the sharp Lieb-Thirring assertion states that the extremizer potential W*W_* at exponent γ\gamma is admissible and (15) holds in [0,∞][0, \infty]: ℳγ(W)≤Lγmγ(W),Λγ=Lγ,Λγ(1)=Lγ,0<mγ(W*),ℳγ(W*)=Lγmγ(W*).\begin{aligned} \mathcal{M}_\gamma(W) &\le L_\gamma\, m_\gamma(W), \\ \Lambda_\gamma &= L_\gamma, \\ \Lambda_\gamma^{(1)} &= L_\gamma, \\ 0 < m_\gamma(W_*), &\qquad \mathcal{M}_\gamma(W_*) = L_\gamma\, m_\gamma(W_*). \end{aligned}(15)✦

In the paper, the result reads: on the line, for γ>1/2\gamma > 1/2 and 0≤W∈Lγ+1/2(ℝ)0 \le W \in L^{\gamma+1/2}(\mathbb{R}), h−W[v]=∫ℝ|v′(x)|2dx−∫ℝW(x)|v(x)|2dxh_{-W}[v] = \int_{\mathbb{R}} |v'(x)|^2\,dx - \int_{\mathbb{R}} W(x)|v(x)|^2\,dx on H1(ℝ;ℂ)H^1(\mathbb{R}; \mathbb{C}) and Tr⁡(H−W)−γ=∑λj(H−W)<0|λj(H−W)|γ\operatorname{Tr}(H_{-W})_-^\gamma = \sum_{\lambda_j(H_{-W}) < 0} |\lambda_j(H_{-W})|^\gamma for the negative eigenvalues λj(H−W)\lambda_j(H_{-W}) of H−W=−d2/dx2−WH_{-W} = -d^2/dx^2 - W, where Lγ,1L_{\gamma,1} (here Λγ\Lambda_\gamma) is the optimal constant in Tr⁡(H−W)−γ≤Lγ,1∫ℝW(x)γ+1/2dx\operatorname{Tr}(H_{-W})_-^\gamma \le L_{\gamma,1} \int_{\mathbb{R}} W(x)^{\gamma+1/2}\,dx, Lγ,1cl=Γ(γ+1)/(2πΓ(γ+3/2))L_{\gamma,1}^{\mathrm{cl}} = \Gamma(\gamma+1)/(2\sqrt{\pi}\,\Gamma(\gamma+3/2)) (here LγclL_\gamma^{\mathrm{cl}}), and Lγ,1(1)L_{\gamma,1}^{(1)} (here Λγ(1)\Lambda_\gamma^{(1)}) is the optimal constant when retaining only the lowest eigenvalue.

What Is Checked

The sharp Lieb-Thirring assertion (15) is stated over spectral exponents γ∈(1/2,3/2)\gamma \in (1/2, 3/2) and admissible real-valued potentials W:ℝ→ℝW \colon \mathbb{R} \to \mathbb{R} with potential mass mγ(W)m_\gamma(W), using the variational negative moment ℳγ(W)\mathcal{M}_\gamma(W) and variational one-state moment ℳγ(1)(W)\mathcal{M}_\gamma^{(1)}(W) defined over L2L^2-orthonormal families and normalized weak eigenfunctions of the Schrödinger quadratic form h−Wh_{-W} on the Sobolev space H1(ℝ;ℂ)H^1(\mathbb{R}; \mathbb{C}) equipped with the L2L^2 inner product, together with the semiclassical constant LγclL_\gamma^{\mathrm{cl}}, the sharp constant LγL_\gamma, the variational optimal constant Λγ\Lambda_\gamma, the variational one-state constant Λγ(1)\Lambda_\gamma^{(1)}, and the extremizer potential W*W_*.