Variational Solutions of the Dirichlet Problem

• Variational Solutions of the Dirichlet Problem are functions that minimize energy functionals subject to boundary conditions, unifying classical and modern analytic methods.
• They extend traditional Laplace formulations to incorporate nonlinear, degenerate, and nonconvex models by leveraging relaxation, duality, and finite element discretization techniques.
• Applications span metric measure spaces and fractional as well as vectorial problems, offering robust frameworks for analyzing existence, uniqueness, and regularity.

A variational solution of the Dirichlet problem is a function that minimizes an appropriately chosen energy functional among all admissible functions agreeing with given boundary data. This formulation serves as a unifying approach across classical linear elliptic equations, nonlinear and degenerate operators, nonconvex problems, and geometric minimization in diverse analytic settings. The concept draws on minimization principles, critical point theory, relaxation in nonsmooth frameworks, and recent extensions to problems involving degenerate growth, nonlocal and fractional operators, and even metric measure spaces.

1. Classical Formulation of Variational Dirichlet Problems

The prototype case is the Laplace (Poisson) equation on a bounded domain Ω ⊂ ℝⁿ, with prescribed Dirichlet boundary data g on ∂Ω. The solution is characterized via the Dirichlet energy functional:

$E(u) = \int_\Omega |\nabla u|^2\,dx$

minimized over functions u in

$$H^1_g(\Omega) = \{\, u\in H^1(\Omega):\; \gamma u = g \text{ on }\partial\Omega\,\}$$

where γ denotes the trace operator. The unique minimizer u∈H¹(Ω) satisfies the weak form:

$$\int_\Omega \nabla u \cdot \nabla v\,dx = 0,\quad \forall v\in H^1_0(\Omega)$$

which corresponds to the standard weak solution of the Poisson equation when f=0. This variational approach underpins the classical well-posedness theory, as well as its discretization via finite elements (Goga et al., 2021).

Extensions accommodate general linear elliptic equations (with or without nonhomogeneous terms f), more general Sobolev or Orlicz–Sobolev spaces, and nonlinear operators. For example, the nonlinear Dirichlet problem for −div (a(x,∇u))=f is handled by variational minimization of

$$J(u) = \int_\Omega A(x,\nabla u)\,dx - \int_\Omega F(u)\,dx$$

over an appropriate Sobolev space with prescribed boundary trace (Bisci et al., 2016).

Recent work addresses the definition of variational solution for general continuous boundary data (not necessarily Sobolev-trace regular) (Arendt et al., 17 Dec 2025). If g ∈ C(∂Ω), there always exists an extension Φ ∈ C(Ω̄) with ΔΦ ∈ H^{-1}(Ω); the variational solution is constructed as u = Φ − v, with v ∈ H¹₀(Ω), and Δv = ΔΦ in H^{-1}(Ω). This solution agrees with the Perron solution and is unique within the class of continuous harmonic functions attaining the boundary datum in the trace sense (Arendt et al., 17 Dec 2025).

2. Variational Principles in Nonlinear and Degenerate Models

For nonlinear elliptic equations, variational methods generalize to include energy functionals of the form:

$I[u] = \int_\Omega F(|\nabla u|)\,dx,$

where F has non-standard growth, e.g., p-growth, (1 < p < ∞), or even linear growth (e.g., minimal surfaces). Globally Lipschitz minimizers exist, under sharp necessary and sufficient conditions on the integrand, such as the Serrin-type barrier criterion:

$$\int_1^\infty tF''(t)\,dt = \infty \implies \text{solvability for all regular data and domains}$$

(Beck et al., 2016).

Degenerate and double-phase energies, such as those involving the 1-Laplacian or double-phase functionals (mixing linear and superlinear growth), are handled in the space of functions of bounded variation (BV). For instance, the double-phase Dirichlet problem

$I(u) = \int_\Omega |Du| + a(x)|Du|^q$

posed in BV ∩ W^{1,q}(Ω), admits unique weak solutions for appropriate weights a(x) and Dirichlet data g in W^{1-1/q,q}(∂Ω); the solution minimizes I in the set {u ∈ BV(Ω) ∩ W^{1,q}(Ω): Tr u = g}, and satisfies an Euler–Lagrange system involving Anzellotti's pairings and measure-divergence (Matsoukas et al., 13 May 2025).

In metric measure spaces with a doubling measure and a (1,1)-Poincaré inequality, the Dirichlet problem for BV-energy minimizers splits into two variational notions: (A) boundary-extension minimization, and (B) trace-penalty minimization incorporating the inner perimeter measure and the trace operator. The latter is obtained as the limit of p-harmonic minimizers as p→1, reflecting the lack of lower-semicontinuity for the boundary term (Korte et al., 2016).

3. Variational Solutions in Nonconvex, Non-Smooth, and Vector Valued Settings

Variational Dirichlet problems for nonconvex functionals, such as those with double-well or multiwell structure, admit multiple non-smooth (possibly non-minimizing) solutions. The canonical duality–triality framework analytically continues the variational PDE into the dual space by Legendre (Fenchel) dualization, transforming the Euler–Lagrange equation into a system of algebraic equations, whose roots correspond to different critical points (global minimizers, local minimizers, and maximizers). Analytical classification is realized via the second-variation (Hessian) in the primal and dual variables (Lu et al., 2016).

For vectorial problems or L^∞ variational problems, the classical Euler–Lagrange framework is insufficient. The notion of $\mathcal{D}$-solution leverages Young measures to extend differentiability and captures generalized solutions to fully nonlinear PDE systems. For e.g. supremal functionals

$$E_{\infty}(u;Ω) = \operatorname{ess\,sup}_{x∈Ω} H(Du(x)),$$

the $\mathcal{D}$-solutions are constructed via singular value constraints on Du and are recognized as critical points in an appropriate relaxed sense, without the necessity of convexity or even quasiconvexity (Croce et al., 2016).

In selection among multiple nonunique formal solutions (e.g. for the eikonal system |∂u/∂x_i|=1 with Dirichlet data), a secondary variational criterion is introduced to regularize the minimization: the total cost of gradient discontinuities is penalized by a weighted Hausdorff measure, which selects "regular" solutions with minimal jump sets and specified structural properties (Croce et al., 2011).

4. Variational Frameworks for Nonlocal and Fractional Operators

Nonlocal Dirichlet problems involving fractional powers of the Laplacian or anisotropic integral operators admit variational characterizations in fractional Sobolev or Orlicz–Sobolev spaces. For the Cauchy-Dirichlet problem for the nonlinear fractional equation

$$∂_t(|u|^{q-2}u) + (−Δ)^θ u = 0, \quad u|_{\mathbb{R}^d\setminus Ω} = 0,$$

the energy space is defined using Gagliardo seminorms, and solutions are constructed via a time-discretization minimizing movement scheme over the energy functional (gradient flow formulation). Uniqueness and quantitative decay and convergence to asymptotic self-similar profiles are derived variationally in this setting (Akagi et al., 29 Mar 2024).

For Dirichlet problems with fractional Orlicz–g-Laplacians,

$(−Δ_g)^s u = f(x,u),$

posed in W^{s,G}_0(Ω), existence and multiplicity (positive, negative, and sign-changing) results are obtained via direct variational minimization, Palais-Smale compactness arguments, and sub-supersolution methods, under suitable Orlicz-geometric and growth conditions (Ochoa et al., 2021, Frassu, 2018). Regularity, boundary behavior (e.g., Hopf-type lemmas), and equivalence of minimization in natural and boundary-weighted topologies are established.

5. Key Analytical Tools and Existence Methodologies

Direct Method: Existence of variational solutions is typically obtained via the direct method of the calculus of variations: coercivity of the energy functional, weak lower semicontinuity, closure of the admissible set under weak convergence, and compactness or tightness (often via Sobolev/BV/embedding theorems and maximal principles) guarantee the existence of minimizers (Galewski et al., 2015, Beck et al., 2016, Korte et al., 2016, Matsoukas et al., 13 May 2025).

Critical Point Theory: For functionals not strictly convex or with multiple wells, one employs mountain-pass, Morse, or three-critical-point theorems to produce several distinct solutions (Bisci et al., 2016, Frassu, 2018, Ochoa et al., 2021, Lu et al., 2016). Palais–Smale or Cerami compactness conditions are pivotal in validating these methods.

Relaxation and Trace Methodologies: In low-regularity domains or metric spaces, suitable notions of boundary traces, perimeter, and inner perimeter measures are developed to give meaning to the minimization principle and Dirichlet data (Korte et al., 2016, Arendt et al., 17 Dec 2025).

Discretization and Numerical Schemes: Variational finite element discretizations provide computational solvability, with error control and convergence to weak solutions established for smooth and rough data alike (Goga et al., 2021, Galewski et al., 2015).

6. Regularity, Uniqueness, and Non-Local Properties

Regularity of minimizers is closely tied to the structure of the functional and the domain. For classical elliptic problems, minimizers inherit interior and boundary regularity from elliptic theory. In degenerate, BV, or nonlocal contexts, fine properties of the jump set, boundary traces, and structure of singularities (e.g., Lebesgue’s cusp) lead to a rich variety of phenomena: lack of continuity at isolated singular boundary points is generically present, and the solution's regularity depends in a nonlocal fashion on data prescribed far from a singularity (Arendt et al., 17 Dec 2025).

Uniqueness is guaranteed under strict convexity and coercivity conditions; otherwise, multiple variational solutions (local minimizers, global minimizers, saddle points, maximizers) can exist. In particular, for the total variation or minimal surface problem, uniqueness can fail without additional constraints (Korte et al., 2016, Matsoukas et al., 13 May 2025).

7. Open Problems and Research Directions

Despite broad success, open issues remain. These include uniqueness and finer regularity (e.g., character of minimizing jump sets in eikonal systems (Croce et al., 2011)), extension to higher dimensions, development of precise Euler–Lagrange characterizations for discontinuity- or BV-penalized functionals, and analysis of continuity and stability for solutions under low regularity of domains or data (Arendt et al., 17 Dec 2025, Korte et al., 2016). Nonlocal sensitivity and generic discontinuity near singular points remain active and challenging areas, linking the analytic and geometric aspects of variational Dirichlet problems.

---

References

• "Variational solutions of the Dirichlet problem, Lebesgue's cusp and non-local properties" (Arendt et al., 17 Dec 2025)
• "On the selection of a particular class of solutions to a system of eikonal equations" (Croce et al., 2011)
• "$\mathcal{D}$-solutions to the system of vectorial Calculus of Variations in $L^\infty$ via the singular value problem" (Croce et al., 2016)
• "On the solutions of a double-phase Dirichlet problem involving the 1-Laplacian" (Matsoukas et al., 13 May 2025)
• "Notions of Dirichlet problem for functions of least gradient in metric measure spaces" (Korte et al., 2016)
• "Globally Lipschitz minimizers for variational problems with linear growth" (Beck et al., 2016)
• "Multiple nonsmooth solutions for nonconvex variational boundary value problems in $\mathbb{R}^n$" (Lu et al., 2016)
• "Multiple solutions for elliptic equations involving a general operator in divergence form" (Bisci et al., 2016)
• "The variational approach of an elliptic problem and its solution by finite elements" (Goga et al., 2021)
• "Non-spurious solutions to discrete boundary value problems through variational methods" (Galewski et al., 2015)
• "Energy solutions of the Cauchy-Dirichlet problem for fractional nonlinear diffusion equations" (Akagi et al., 29 Mar 2024)
• "Existence and multiplicity of solutions for a Dirichlet problem in Fractional Orlicz- Sobolev spaces" (Ochoa et al., 2021)
• "Nonlinear Dirichlet problem for the nonlocal anisotropic operator $L_K$" (Frassu, 2018)
• "A bipolynomial fractional Dirichlet-Laplace problem" (Idczak, 2018)