OctoInformBeta

OpenAI preprint, September 2026 · Spectral and planar preliminaries

Lemma 2.1: For a planar domain Ω\Omega with Dirichlet Sobolev space H01(Ω)H_0^1(\Omega), attainment of the Dirichlet energy minimum and subcritical coercivity

A Dirichlet minimum above the Neumann value.

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

Let Ω⊂ℝ2\Omega\subset\mathbb R^2 be a nonempty bounded open planar domain. The Dirichlet energy is considered on functions obtained by closing smooth compactly supported functions in the Sobolev norm. The spectral quantities are defined using real functions and Lebesgue measure on Ω\Omega.

The test space 𝒟(Ω)\mathscr D(\Omega) consists of the smooth functions on the Euclidean plane whose support is compact and contained in Ω\Omega: 𝒟(Ω)={ψ∈C∞(ℝ2;ℝ):supp⁡ψ is compact,supp⁡ψ⊂Ω}.\mathscr D(\Omega) =\{\psi\in C^\infty(\mathbb R^2;\mathbb R): \operatorname{supp}\psi\text{ is compact},\quad \operatorname{supp}\psi\subset\Omega\}.(1)✦ For a test function ψ∈𝒟(Ω)\psi\in\mathscr D(\Omega) and a direction e∈ℝ2e\in\mathbb R^2, the test derivative ∂eψ∈L2(Ω;ℝ)\partial_e\psi\in L^2(\Omega;\mathbb R) and test vector field ψe∈L2(Ω;ℝ2)\psi e\in L^2(\Omega;\mathbb R^2) are the functions in (2). ∂eψ(x)=Dψ(x)[e],(ψe)(x)=ψ(x)e.\partial_e\psi(x)=D\psi(x)[e],\qquad (\psi e)(x)=\psi(x)e.(2)

The ambient Hilbert space ℋ(Ω)\mathcal H(\Omega) for a scalar function and its gradient is the product of the scalar and vector L2L^2 spaces on Ω\Omega, with the norm in (3). ℋ(Ω)=L2(Ω;ℝ)×L2(Ω;ℝ2),∥(u,g)∥ℋ2=∥u∥22+∥g∥22.\mathcal H(\Omega)=L^2(\Omega;\mathbb R)\times L^2(\Omega;\mathbb R^2), \qquad \|(u,g)\|_{\mathcal H}^2=\|u\|_2^2+\|g\|_2^2.(3) A weak gradient of a function u:ℝ2→ℝu:\mathbb R^2\to\mathbb R is a vector field g:ℝ2→ℝ2g:\mathbb R^2\to\mathbb R^2, with both functions square-integrable on Ω\Omega, satisfying the identity in (4) for every test function and every direction.✦ The Sobolev space H1(Ω)H^1(\Omega) is the weak gradient graph in (3): H1(Ω)={(u,g)∈ℋ(Ω):∫Ωu∂eψ=−∫Ω(g⋅e)ψfor all ψ∈𝒟(Ω),e∈ℝ2}.H^1(\Omega)= \left\{(u,g)\in\mathcal H(\Omega): \int_\Omega u\,\partial_e\psi =-\int_\Omega (g\cdot e)\psi \quad\text{for all }\psi\in\mathscr D(\Omega),\ e\in\mathbb R^2 \right\}.(4) A function together with a weak gradient determines its Sobolev class by taking their L2L^2 classes.✦

The Sobolev value map val⁡:H1(Ω)→L2(Ω;ℝ)\operatorname{val}:H^1(\Omega)\to L^2(\Omega;\mathbb R) and gradient map ∇:H1(Ω)→L2(Ω;ℝ2)\nabla:H^1(\Omega)\to L^2(\Omega;\mathbb R^2) are the two projections in (5). val⁡(u,g)=u,∇(u,g)=g.\operatorname{val}(u,g)=u,\qquad \nabla(u,g)=g.(5) In formulas involving integrals, a Sobolev element is written as its value under (5). The smooth test inclusion j:𝒟(Ω)→H1(Ω)j:\mathscr D(\Omega)\to H^1(\Omega) is the map sending a test function to its value and classical gradient.✦ The Dirichlet Sobolev space H01(Ω)H_0^1(\Omega) is the closed subspace of H1(Ω)H^1(\Omega) obtained as the closure of j𝒟(Ω)j\mathscr D(\Omega) in (6). j:𝒟(Ω)→H1(Ω),jψ=(ψ,∇ψ),H01(Ω)=j𝒟(Ω)¯H1(Ω).j:\mathscr D(\Omega)\longrightarrow H^1(\Omega),\qquad j\psi=(\psi,\nabla\psi),\qquad H_0^1(\Omega)=\overline{j\mathscr D(\Omega)}^{\,H^1(\Omega)}.(6) The Dirichlet value map val⁡:H01(Ω)→L2(Ω;ℝ)\operatorname{val}:H_0^1(\Omega)\to L^2(\Omega;\mathbb R) and gradient map ∇:H01(Ω)→L2(Ω;ℝ2)\nabla:H_0^1(\Omega)\to L^2(\Omega;\mathbb R^2) are the restrictions of the Sobolev projections in (5) to H01(Ω)H_0^1(\Omega).

The set ℛ(Ω)\mathcal R(\Omega) of mean-zero Rayleigh values consists of the quotients in (7), formed from functions in H1(Ω)H^1(\Omega) with zero mean and nonzero L2L^2 value.✦ The first positive Neumann value μ\mu of Ω\Omega is the infimum of the mean-zero Rayleigh values in (7): ℛ(Ω)={∫Ω|∇u|2∫Ωu2:u∈H1(Ω),∫Ωu=0,∫Ωu2>0},μ=inf⁡ℛ(Ω).\begin{split} \mathcal R(\Omega) &=\left\{ \frac{\int_\Omega |\nabla u|^2}{\int_\Omega u^2}: u\in H^1(\Omega),\quad \int_\Omega u=0,\quad \int_\Omega u^2>0 \right\},\\ \mu&=\inf\mathcal R(\Omega). \end{split}(7)

The Claim

Theorem 1. Let Ω⊂ℝ2\Omega\subset\mathbb R^2 be nonempty, bounded and open. There are a number λD>0\lambda_D>0 and an element ϕ∈H01(Ω)\phi\in H_0^1(\Omega) with nonzero L2L^2 value such that λD∥v∥22≤∥∇v∥22for every v∈H01(Ω),⟨∇ϕ,∇v⟩L2=λD⟨ϕ,v⟩L2for every v∈H01(Ω).\begin{aligned} \lambda_D\|v\|_2^2&\leq\|\nabla v\|_2^2 &&\text{for every }v\in H_0^1(\Omega),\\ \langle\nabla\phi,\nabla v\rangle_{L^2} &=\lambda_D\langle\phi,v\rangle_{L^2} &&\text{for every }v\in H_0^1(\Omega). \end{aligned}(8)✦ For any λ>0\lambda>0 and any ϕ∈H01(Ω)\phi\in H_0^1(\Omega) with nonzero L2L^2 value, suppose that ⟨∇ϕ,∇v⟩L2=λ⟨ϕ,v⟩L2\langle\nabla\phi,\nabla v\rangle_{L^2} =\lambda\langle\phi,v\rangle_{L^2} for every v∈H01(Ω)v\in H_0^1(\Omega). Then μ<λ.\mu<\lambda.(9)✦ There is also a number λ>0\lambda>0 satisfying μ<λ,λ∥v∥22≤∥∇v∥22for every v∈H01(Ω).\mu<\lambda,\qquad \lambda\|v\|_2^2\leq\|\nabla v\|_2^2 \quad\text{for every }v\in H_0^1(\Omega).(10)✦

In the paper, the result reads: for a nonempty bounded simply connected planar domain Ω\Omega as above, the first positive Neumann eigenvalue μ\mu is strictly below the first Dirichlet eigenvalue λD\lambda_D.

What Is Checked

The assertions (8), (9) and (10) are stated over nonempty bounded open subsets of ℝ2\mathbb R^2, with the Dirichlet space defined by (6) and the Neumann value defined by (7).

The value and gradient maps on H1(Ω)H^1(\Omega) and H01(Ω)H_0^1(\Omega) are continuous and linear, and the smooth test inclusion jj is linear.