🔢 mathematics

Sharp Neumann eigenvalue estimates and $C^2$ elliptic regularity in non-obtuse polyhedral domains

This paper establishes the optimal Neumann eigenvalue lower bound and $C^2$ elliptic regularity for non-obtuse Riemannian polyhedral domains in all dimensions by proving a mutual implication between a spectral characterization on spherical domains and a regularity result for conical polyhedral domains.

Original authors: Nick Edelen, Chao Li

Published 2026-08-19

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Nick Edelen, Chao Li

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a world built not of smooth curves, but of sharp corners and flat faces, like a crystal or a stack of blocks. In mathematics, these shapes are called polyhedral domains. They appear everywhere in geometry and physics, from the structure of molecules to the way heat spreads through a building. For decades, mathematicians have studied how things behave on these shapes, particularly how waves or fields settle down when they hit the edges. A central question has been: if you know the rules for how a field behaves inside such a shape, and you know how it interacts with the walls, can you predict exactly how smooth and well-behaved that field will be? The answer depends heavily on the angles where the walls meet. If the corners are too sharp, the field can become jagged and unpredictable. But if the corners are gentle enough, the field remains smooth.

This paper tackles a precise version of that problem. The researchers, Nick Edelen and Chao Li, focused on a specific type of shape where every corner is "non-obtuse," meaning the angle between any two meeting walls is no wider than a right angle. They wanted to prove that in these specific shapes, the mathematical fields describing physical phenomena are not just smooth, but possess a very high degree of smoothness, known as $C^2$ regularity. This level of smoothness is crucial because it allows scientists to calculate second derivatives, which represent how a field curves or accelerates. Without this smoothness, many advanced calculations in geometry and physics would break down. The authors did not just guess that this smoothness exists; they proved it by connecting two seemingly different mathematical worlds: the study of vibrating membranes on spherical shapes and the study of smooth fields in sharp-cornered cones.

The journey to this proof began with a deep dive into the behavior of vibrations on spherical shapes that are cut by flat planes. Imagine a sphere, like a globe, but instead of being a perfect ball, it is sliced into a shape by flat planes that meet at angles no wider than ninety degrees. The researchers asked a fundamental question about the vibrations, or eigenvalues, that can exist on such a shape. They discovered that the possible frequencies of these vibrations are extremely restricted. In a range of values that one might expect to be filled with many different possibilities, the vibrations can only take on three specific values: zero, a specific number related to the dimension of the space, and a larger number related to the square of that dimension. Furthermore, they proved that if a vibration corresponds to the middle value, it must be a simple, straight-line pattern. If it corresponds to the larger value, it must be a pattern shaped like a quadratic curve. This finding was a sharp, precise limit on what is possible in these geometric settings.

The brilliance of the paper lies in how the authors used this finding about vibrations to solve the problem of smoothness in the sharp-cornered cones. They established a two-way bridge between these two ideas. First, they showed that if you can prove the vibrations on the spherical shapes are so restricted, then you can guarantee that the fields in the sharp-cornered cones are perfectly smooth. Second, they showed the reverse: if you assume the fields in the cones are smooth, you can deduce that the vibrations on the spheres must be restricted in exactly the way they found. This circular logic allowed them to prove both statements simultaneously. By starting with the known smoothness of fields in simpler cases and climbing up through dimensions, they demonstrated that the high degree of smoothness holds true for all dimensions, provided the corners are non-obtuse.

The result is a definitive confirmation of a long-standing intuition in the field. For years, mathematicians suspected that the condition of having non-obtuse angles was the exact threshold needed to ensure that solutions to these equations remain smooth and well-behaved. If the angles were any wider, the smoothness would break down. Edelen and Li proved that this suspicion was correct. They showed that in any dimension, as long as the dihedral angles—the angles between the faces of the shape—are no wider than a right angle, the solutions to these elliptic equations are not just continuous, but possess the specific second-order smoothness required for complex geometric analysis. This work provides a solid foundation for future research in curvature conditions and rigidity theorems, ensuring that the mathematical tools used to describe the shape of the universe are reliable even in the presence of sharp, angular boundaries.

Technical Summary: Sharp Neumann Eigenvalue Estimates and $C^2$ Elliptic Regularity in Non-Obtuse Polyhedral Domains

Problem Statement
The paper addresses two interconnected problems in the analysis of second-order elliptic equations on Riemannian polyhedral domains with Neumann boundary conditions:

Spectral Gap: Determining the sharp lower bounds for the Neumann eigenvalues of the Laplacian on spherical polyhedral domains with non-obtuse dihedral angles.
Regularity: Establishing optimal $C^{2,\alpha}$ regularity for weak solutions to elliptic equations in Euclidean polyhedral cones with non-obtuse dihedral angles.

The authors define a Riemannian polyhedron as a space locally isometric to the intersection of smooth domains with boundary meeting transversely. The central geometric constraint is that all dihedral angles must be non-obtuse (at most $\pi/2$ ). The paper investigates whether this geometric condition is sufficient to guarantee that the first non-trivial Neumann eigenvalue is bounded below by $2(n+1)$ (in dimension $n$ ) and that weak solutions to $\Delta u = f$ are $C^{2,\alpha}$ .

Methodology
The core of the paper is a proof of equivalence between two dimension-dependent assertions, denoted $(P_n)$ and $(Q_n)$ :

Assertion $(P_n)$ : A sharp spectral estimate on spherical domains $\Omega \subset S^{n-1}$ . It posits that for non-obtuse domains, the Neumann spectrum in the interval $[0, 2(n+1)]$ contains only $\{0, n, 2(n+1)\}$ . Furthermore, it characterizes the eigenfunctions: eigenvalue $n$ corresponds to linear functions, and $2(n+2)$ corresponds to quadratic polynomials.
Assertion $(Q_n)$ : A regularity theorem stating that weak solutions to $\Delta u = f$ (with $f$ Hölder continuous) in a non-obtuse polyhedral cone in $\mathbb{R}^n$ are locally $C^{2,\beta}$ .

The authors prove the implications $(Q_n) \implies (P_n)$ and $(P_n) \implies (Q_{n+1})$ . By establishing this circular dependency, they inductively prove that both assertions hold for all dimensions $n \ge 1$ .

Key Technical Contributions

Implication $(Q_n) \implies (P_n)$ (Spectral Rigidity):
- Assuming $C^{2,\alpha}$ regularity (from $Q_n$ ), the authors analyze the traceless Hessian $T = \nabla^2 u - \frac{\Delta u}{n}g$ of an eigenfunction $u$ .
- They derive a pointwise tensor equation $\Delta T = (2n - \lambda)T$ and a divergence identity.
- Using a Reilly-type integration formula on the polyhedral domain, they integrate these identities. A critical step involves a sharp algebraic inequality (Lemma 2.6) for traceless tensors, which relates $|\nabla T|^2$ to $|\text{div } T|^2$ .
- This analysis yields the spectral gap $\lambda \ge 2(n+1)$ for non-constant eigenfunctions not corresponding to linear functions. The equality case is rigorously characterized, showing that equality implies the eigenfunction is a restriction of a quadratic polynomial.
Implication $(P_n) \implies (Q_{n+1})$ (Regularity via Spectral Gap):
- This direction is a local regularity theorem for polyhedral cones in $\mathbb{R}^{n+1}$ .
- The authors utilize the spectral information from $(P_n)$ regarding the spherical link $\Omega \cap S^{n-1}$ . Specifically, $(P_n)$ rules out the existence of homogeneous Neumann solutions with degrees strictly between 1 and 2.
- They employ an inductive argument on the symmetry dimension of the cone. If a cone is $l$ -symmetric (locally a product of a lower-dimensional cone and $\mathbb{R}^l$ ), the regularity of the solution is reduced to the regularity of the cross-section.
- By subtracting appropriate affine and quadratic approximations (justified by the spectral gap), they establish an "excess-decay" estimate. This decay, combined with the inductive hypothesis on lower-symmetry domains, yields the $C^{2,\beta}$ regularity.

Main Results

Theorem 1.2: The assertions $(P_n)$ and $(Q_n)$ are equivalent. Consequently, both hold for all integers $n \ge 1$ .
Corollary 1.3: For a non-obtuse spherical simplicial complex $\Omega \subset S^{n-1}$ , the first non-zero Neumann eigenvalue satisfies $\mu_1(\Omega) \ge 2(n+1)$ . Equality holds if and only if the eigenfunction is the restriction of a harmonic quadratic polynomial.
Optimal Regularity: Weak solutions to the Neumann problem in non-obtuse polyhedral domains are $C^{2,\alpha}$ , which is the optimal regularity expected for such singular domains.

Significance and Claims
The paper claims to provide the optimal Neumann eigenvalue lower bound and $C^2$ elliptic regularity theory for non-obtuse Riemannian polyhedral domains.

Geometric Connection: The work establishes an intimate connection between the spectral geometry of the spherical link and the analytic regularity of the elliptic operator in the cone.
Applications: The authors note that such $C^2$ estimates are indispensable for geometric comparison and rigidity arguments where curvature quantities involve second derivatives. They specifically mention relevance to polyhedral rigidity theorems proposed by Gromov.
Sharpness: The results are sharp; the non-obtuse condition is necessary. If a dihedral angle exceeds $\pi/2$ , the spectral gap closes (allowing eigenvalues between $n$ and $2(n+1)$ ), and $C^2$ regularity fails.

The authors emphasize that while the spectral gap argument in the smooth setting has precedents (Simon, Guan-Guo), their contribution lies in the sharp characterization of the equality case and the rigorous extension of these arguments to the singular setting of polyhedral domains via the interplay of spectral gaps and regularity.

More like this