On the W 2,p solvability for mixed boundary value problems

arXiv:2609.30090 · math.AP · Submitted 2026-09-24 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.

Ada: We're looking at the paper: "On the W 2,p solvability for mixed boundary value problems".

Emmy: Here's the gist of the paper. Global W2,p solvability for mixed boundary value problems is established for the Poisson equation with mixed Dirichlet–Neumann boundary conditions in two classes of domains in all dimensions n ≥ 2 The authors obtain optimal range 1 1 − 1/p For Lipschitz polyhedra with facewise boundary decompositions, solvability is obtained for p close to 1 In both settings, they establish endpoint W 2,1 solvability for data in an adapted atomic Hardy space.

Ada: Solvability Results on Regular Domains Theorem 1.1 provides the optimal W2,p solvability on regular domains in every dimension For C1,α domains with a Reifenberg-flat interface, there exists θ0 > 0 such that for p = 1 and f ∈ H1 D(omega), there exists a unique solution u ∈ W2,1(omega) ∩ W1,1 D (omega) of (P), and the estimate is given by∥D2u∥L1(omega) + ∣δ−1Γ∇u∥L1(omega) ≤ C∥f∥H1 D(omega) For 1 < p < 4/3 and f ∈ Lp(omega), there exists a unique solution u ∈ W2,p(omega) ∩ W1,p D (omega) of (P), and the estimate is given by∥D2u∥Lp(omega) + ∣δ−1Γ∇u∥Lp(omega) ≤ C∥f∥Lp(omega) Remark 1.2 states that the endpoint 4/3 is sharp, as the harmonic function u(r, ϕ) = r(1/2) sin(ϕ/2), 0 < ϕ < π, has D2u ≃ r(-3/2).

Emmy: Solvability on Polyhedral Domains Theorem 1.3 establishes W2,p solvability on Lipschitz polyhedra with a facewise mixed decomposition (D, N) For p close to 1, there exists p0 = p0(omega) > 1 such that for f ∈ Lp(omega), there exists a unique solution u ∈ W2,p(omega) ∩ W1,p D (omega) of (P), and the estimate is given by∥D2u∥Lp(omega) + ∣δ−1Σ∇u∥Lp(omega) ≤ C∥f∥Lp(omega) Geometric Reduction Strategy The proofs of both theorems follow a common strategy involving a Whitney-type decomposition that reduces second-order estimates to weighted gradient estimates This reduction leads to the endpoint H1 D → W2,1 estimate using Hölder estimates and duality The W2,p estimates are then obtained by a real-variable interpolation argument.

Ada: Proposition 3.1 reduces the local W2,p estimate to a weighted estimate for the gradient Q∩omegaD2up ≤ C κQ∩omegafp + δ−pΣ + l(Q)−p∇up.

Paper summary: Ada: So looking at the paper "On the W 2,p solvability for mixed boundary value problems", what's the big picture here <ref:2609.30090#pg1,p solvability for mixed boundary value problems>? They've essentially provided a comprehensive framework for when you can guarantee a solution with this specific second-order regularity under these mixed boundary conditions across different geometric settings.

Emmy: I think it really highlights how important the structure of the domain matters, whether it's smooth or piecewise flat, and how that structure dictates what kind of input data you need to get a good solution. The paper shows that even when things get complicated near the interface between Dirichlet and Neumann parts, there's a predictable way to handle it if you stick to those right geometric assumptions.

Kolya: From my side, I’m looking at how they define those optimal ranges for p, like one less than four thirds on smooth domains and close to one on polyhedra. It suggests that the regularity of the solution is intrinsically tied to the geometry of the boundary interface itself.

Ada: It means that for someone working on these types of problems, knowing which geometric class you're in tells you immediately what kind of input data—what kind of function f—you can expect to get a solution with that W 2,p regularity <ref:2609.30090#pg1>.

Emmy: The implication is that we have established a robust way to find solutions for mixed boundary value problems in all dimensions, even when the domain isn't perfectly smooth or curved. It gives us tools to move from simple existence proofs toward actual estimates for the second derivatives of our solutions.

Conclusion: Ada: So, we've seen how this paper tackles mixed boundary value problems for the Poisson equation, looking at those W two p solvability conditions across different shapes of domains and boundaries.

Emmy: Exactly. The authors are establishing a framework for when you can guarantee a solution with that specific regularity under these mixed Dirichlet and Neumann conditions in all dimensions two or higher.

Ada: It sounds like they're tying the existence of a solution directly to how regular the boundary interface is, whether it's smooth or piecewise flat.

Kolya: What I want to know is what those optimal ranges for p actually mean for someone trying to solve a problem in real life. Is it just theoretical?

Ada: They give us specific numbers, like the optimal range one minus one over p, and they get really close to one when the boundary has a nice facewise decomposition.

Emmy: It suggests that you can get good second-order estimates if you have a certain structure on the boundary interface itself. The geometry dictates what kind of input data you need for a good solution.

Kolya: If we're talking about practical applications, are we looking at problems where this specific regularity helps us predict something tangible? Or is this purely about the mathematical machinery?

Ada: Well, it gives us tools to move from just knowing a solution exists to actually getting estimates on how big the second derivatives of that solution can get. It’s about controlling the behavior near those tricky boundary spots.

Emmy: It opens up a path for researchers working on physical systems where boundaries aren't perfectly smooth but still have some predictable structure. We need to see how this translates into actual modeling.

Kolya: So, the paper sets the rules for when we can expect controlled error terms in our calculations, based on what's happening at the boundary interface. That leads us next to how they actually prove these things, which involves a specific reduction strategy.

ROHIT KHANDELWAL, ZONGYUAN LI

South Asian University · City University of Hong Kong

math.AP

Submitted: 2026-09-24

Updated: 2026-11-08

Comments: 20 pages, 1 figure

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 81/100

The gist: Global W2,p solvability for mixed boundary value problems is established for the Poisson equation with mixed Dirichlet–Neumann boundary conditions in two classes of domains in all dimensions n ≥…

Terminology

Summary

Global W2,p solvability for mixed boundary value problems is established for the Poisson equation with mixed Dirichlet–Neumann boundary conditions in two classes of domains in all dimensions n ≥ 2 The authors obtain optimal range 1

1 − 1/p For Lipschitz polyhedra with facewise boundary decompositions, solvability is obtained for p close to 1 In both settings, they establish endpoint W 2,1 solvability for data in an adapted atomic Hardy space.

Solvability Results on Regular Domains

Theorem 1.1 provides the optimal W2,p solvability on regular domains in every dimension For C1,α domains with a Reifenberg-flat interface, there exists θ0 > 0 such that for p = 1 and f ∈ H1 D(omega), there exists a unique solution u ∈ W2,1(omega) ∩ W1,1 D (omega) of (P), and the estimate is given by∥D2u∥L1(omega) + ∣δ−1Γ∇u∥L1(omega) ≤ C∥f∥H1 D(omega) For 1 < p < 4/3 and f ∈ Lp(omega), there exists a unique solution u ∈ W2,p(omega) ∩ W1,p D (omega) of (P), and the estimate is given by∥D2u∥Lp(omega) + ∣δ−1Γ∇u∥Lp(omega) ≤ C∥f∥Lp(omega) Remark 1. p. 1

Where the paper says it

There exists θ0 > 0 such that, if the interface Γ is θ0-Reifenberg flat, the following assertions hold. (a) If p = 1 and f ∈ H1 D(Ω), there exists a unique solution u ∈ W2,1 (Ω) ∩ W 1,1 D (Ω) of (P), and (1.

2 states that the endpoint 4/3 is sharp, as the harmonic function u(r, ϕ) = r(1/2) sin(ϕ/2), 0 < ϕ < π, has D2u ≃ r(-3/2). p. 1
Where the paper says it

The endpoint 4/3 is sharp. Indeed, in the upper half-plane the harmonic function u(r, ϕ) = r 1/2 sin(ϕ/2), 0 < ϕ < π, has zero Dirichlet data on one half of the boundary and zero Neumann data on the other, while |D2u| ≃ r −3/2 .

Solvability on Polyhedral Domains

Theorem 1.3 establishes W2,p solvability on Lipschitz polyhedra with a facewise mixed decomposition (D, N) For p close to 1, there exists p0 = p0(omega) > 1 such that for f ∈ Lp(omega), there exists a unique solution u ∈ W2,p(omega) ∩ W1,p D (omega) of (P), and the estimate is given by∥D2u∥Lp(omega) + ∣δ−1Σ∇u∥Lp(omega) ≤ C∥f∥Lp(omega)

Geometric Reduction Strategy

The proofs of both theorems follow a common strategy involving a Whitney-type decomposition that reduces second-order estimates to weighted gradient estimates This reduction leads to the endpoint H1 D → W2,1 estimate using Hölder estimates and duality The W2,p estimates are then obtained by a real-variable interpolation argument. Proposition 3. p. 6

Where the paper says it

MIXED PROBLEMS 7 3. A Whitney reduction The following proposition reduces the local W2,p estimate to a weighted estimate for the gradient.

1 reduces the local W2,p estimate to a weighted estimate for the gradient Q∩omegaD2up ≤ C κQ∩omegafp + δ−pΣ + l(Q)−p∇up. p. 6
Where the paper says it

MIXED PROBLEMS 7 3. A Whitney reduction The following proposition reduces the local W2,p estimate to a weighted estimate for the gradient.

Estimate in Endpoint Hardy Spaces

Lemma 4.1 proves a W1,2+ε estimate for solutions u ∈ W1,2 D (omega) when f ∈ Lp(omega) with p > 2n/(n + 2) This follows from duality and the Sobolev-Poincaré inequality. p. 8

Where the paper says it

Here, p∗ = np/(n + p) and p ∗ = np/(n − p). This follows from a standard W1,2+ϵ estimate for equation with divergence-form right-hand side, duality, and the Sobolev-Poincar´e inequality.

Proposition 4. p. 6
Where the paper says it

MIXED PROBLEMS 7 3. A Whitney reduction The following proposition reduces the local W2,p estimate to a weighted estimate for the gradient.

2 shows that for f = a, an (1, ∞)D-atom, if u ∈ W1,2 D (omega) solves (P), then∥D2u∥L1(omega) + ∣δ−1∇u∥L1(omega) ≤ C. p. 9
Where the paper says it

Let Ω ⊂ R n be a bounded domain satisfying one of the following two conditions: (a) Ω ∈ C 1,α for some α > 0 with the Dirichlet–Neumann interface Γ is θ0-flat for some small θ0; (b) Ω is a Lipschitz polyhedron with a facewise Dirichlet–Neumann decomposition and Σ being the full polyhedron skeleton. Let u ∈ W 1,2 D (Ω) solves (P) with f = a, an adapted (1, ∞)D-atom.

Solvability with Hardy Data

Proposition 4.3 shows that solvability when f ∈ H1 D easily follows from Proposition 4. p. 6

Where the paper says it

In the proposition, only local regularity away from Σ is required. The estimate follows by summing over Whitney cubes, and, in particular, ∞ ≤ ∞ is allowed.

2 Let f = Pjλjaj be an atom decomposition with Pjλj ≤ 2∥f∥H1 D, then fn ∈ L∞(omega), and the limit u ∈ W2,1 is a weak solution satisfying the desired estimate. p. 9
Where the paper says it

Let Ω ⊂ R n be a bounded domain satisfying one of the following two conditions: (a) Ω ∈ C 1,α for some α > 0 with the Dirichlet–Neumann interface Γ is θ0-flat for some small θ0; (b) Ω is a Lipschitz polyhedron with a facewise Dirichlet–Neumann decomposition and Σ being the full polyhedron skeleton. Let u ∈ W 1,2 D (Ω) solves (P) with f = a, an adapted (1, ∞)D-atom.

Uniqueness in Endpoint Spaces

Proposition 5.1 establishes uniqueness in the endpoint space W2,1(omega) that works on C1,α domains with mild assumptions on Γ and Lipschitz polyhedrons with facewise decomposition It relies on Lemma 5. p. 3

Where the paper says it

Section 3 develops a Whitney-type reduction that is central to our approach, while Section 4 establishes the required estimates in endpoint Hardy spaces. In Section 5, we prove uniqueness, and in Section 6, we establish an interpolation lemma and use it to obtain solvability and global W2,p estimates for the Poisson equation with L p data.

2, which shows that if h ∈ W1,1(omega) is harmonic in omega, then∥N(h)∥L1(∂omega) ≤ C∥h∥W1,1(omega). p. 1
Where the paper says it

The endpoint 4/3 is sharp. Indeed, in the upper half-plane the harmonic function u(r, ϕ) = r 1/2 sin(ϕ/2), 0 < ϕ < π, has zero Dirichlet data on one half of the boundary and zero Neumann data on the other, while |D2u| ≃ r −3/2 .

Interpolation for Lp Data

Section 6 presents an interpolation lemma that upgrades the endpoint estimate with H1 D(omega) data to optimal Lp estimates This is achieved by a real-valuable interpolation argument. Lemma 6.2 proves an optimal reverse Hölder estimate for homogeneous solutions v ∈ W1,2 D (2Q∩omega) when dist(2Q, Γ) ≤ Cl(Q)

Lame Systems Extension

Theorem 7.1 extends these methods to the Lam´e system in two dimensions for bounded Lipschitz polygons with a facewise decomposition (D, N) where D ≠ ∅ The estimate is given by∥D2u∥L1(omega) + ∣δ−1Σ∇u∥L1(omega) ≤ C∥f∥H1 D(omega;R2)

Discussion for Higher Dimensions

The discussion for higher dimensions notes that the main missing ingredient is a local boundary Hölder estimate for homogeneous solutions In dimension three, such estimates are available for the pure Dirichlet and pure traction problems on general Lipschitz domains For n ≥ 4, even the weak maximum principle for the Dirichlet Lam´e problem on an arbitrary Lipschitz domain remains open.

Acknowledgement

Generative AI tools were used to assist with language editing and improving the presentation of the manuscript The authors take full responsibility for the content of the manuscript.

References

The references listed are [1] S. Agmon, A. Douglis, and L. Nirenberg [2] Constantin B˘acut¸˘a, Anna L. p. 2

Where the paper says it

2] established weighted L p solvability on polyhedral cones and a Fredholm theory on bounded three-dimensional polyhedral domains under spectral conditions. In arbitrary dimension, B˘acut¸˘a, Mazzucato, Nistor, and Zikatanov [2] developed regularity

Mazzucato, Victor Nistor, and Ludmil Zikatanov [3] Hugo Beir˜ao da Veiga [4] Susanne C. p. 2
Where the paper says it

2] established weighted L p solvability on polyhedral cones and a Fredholm theory on bounded three-dimensional polyhedral domains under spectral conditions. In arbitrary dimension, B˘acut¸˘a, Mazzucato, Nistor, and Zikatanov [2] developed regularity

Brenner and L. Ridgway Scott [5] Luis A. p. 3
Where the paper says it

For instance, let Vh,D be the space of continuous piecewise affine functions vanishing on D, on a shape-regular simplicial mesh that respects the boundary decomposition [8], and let h denotes the maximum element diameter. The Scott–Zhang approximation theorem [32, Theorem 4.

Caffarelli and Ireneo Peral [6] Jongkeun Choi, Hongjie Dong, and Zongyuan Li [7] Philippe G. p. 0
Where the paper says it

ON THE W2,p SOLVABILITY FOR MIXED BOUNDARY VALUE PROBLEMS ROHIT KHANDELWAL AND ZONGYUAN LI Abstract. We establish global W2,p solvability for the Poisson equation with mixed Dirichlet– Neumann boundary conditions in two classes of domains in all dimensions n ≥ 2.

Ciarlet [8] Philippe G. Ciarlet [9] Ronald R. Coifman and Guido Weiss [10] Björn E. p. 5
Where the paper says it

Here C depends only on n, s, and the Lipschitz character of Ω. The proof is a simple adaption of the standard Coifman–Weiss decomposition of mean-zero atoms.

J Dahlberg [11] Björn E. p. 2
Where the paper says it

First, additional geometric structure is needed for global W2,p estimates: such estimates cannot be expected on arbitrary Lipschitz domains, even under pure Dirichlet boundary conditions. See Dahlberg’s classical counterexample [10].

J Dahlberg [12] Björn E. p. 2
Where the paper says it

First, additional geometric structure is needed for global W2,p estimates: such estimates cannot be expected on arbitrary Lipschitz domains, even under pure Dirichlet boundary conditions. See Dahlberg’s classical counterexample [10].

J Dahlberg and Carlos E Kenig [13] Björn E. p. 2
Where the paper says it

First, additional geometric structure is needed for global W2,p estimates: such estimates cannot be expected on arbitrary Lipschitz domains, even under pure Dirichlet boundary conditions. See Dahlberg’s classical counterexample [10].

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

1

Improve solver for Poisson equation with mixed boundary conditions by establishing global W2p solvability for arbitrary right-hand sides f in Lp(Omega) in all dimensions n >= 2. This allows solving problems without requiring additional integrability assumptions on f.

2

Enable numerical analysis of diffusion problems on Lipschitz polyhedra by providing optimal convergence rates for finite element approximations, specifically showing that inf vh∈Vh,D u − vhLp(Omega) + hnabla(u − vh)Lp(Omega) <= Ch2fLp(Omega).

3

Develop a framework for approximating solutions on rough geometries by utilizing W2p estimates with p < 2, which is essential for approximation in Lp and W1,p without requiring H2 regularity.

Abstract

We establish global W 2,p solvability for the Poisson equation with mixed Dirichlet--Neumann boundary conditions in two classes of domains in all dimensions n at least 2. For C 1,α domains with a Reifenbeg flat interface, we obtain the optimal range 1<p<4/3, provided that α>1-1/p. For Lipschitz polyhedra with facewise boundary decompositions, we obtain solvability for p close to 1. In both settings, we establish endpoint W 2,1 solvability for data in an adapted atomic Hardy space. Applications of the methods to Lamé systems are also discussed.

Related papers