On the asymptotics of ground states for a boundary value problem for the equation $-\varepsilon \Delta_p u = a|u|^{q-2}u - b|u|^{\gamma-2}u$

This paper investigates the singularly perturbed Dirichlet problem for the $p$-Laplacian with competing superlinear terms, establishing the existence of critical parameters that determine solution nonexistence or multiplicity, and proving that positive ground states converge strongly to an explicit profile as the perturbation parameter vanishes.

The Gist

Imagine you are trying to find the perfect "balance point" for a system that is being pulled in two opposite directions by invisible forces. This is the core story of the paper by Il'yasov and Turianova. They are studying a complex mathematical puzzle involving a specific type of equation (the $p$-Laplacian) that describes how things spread or settle in a space, like heat in a metal plate or a population in a territory.

Here is the breakdown of their discovery using simple analogies:

1. The Setup: A Tug-of-War with a "Friction" Knob

Imagine a rubber sheet (the domain $\Omega$) stretched over a frame. The edges of the sheet are pinned down to zero (the boundary condition).

On this sheet, two invisible giants are pulling:

• Giants of Growth (Term $a|u|^{q-2}u$): They want to push the sheet up.
• Giants of Damping (Term $b|u|^{\gamma-2}u$): They want to pull the sheet down.

The paper looks at a special situation where the "Growth" giant is weaker than the "Damping" giant in terms of how fast they grow as the sheet gets higher, but they are both pulling harder than the sheet's natural tension (which is the $p$-Laplacian part).

There is also a small knob labeled $\epsilon$ (epsilon).

• When the knob is turned up (large $\epsilon$), the sheet has a lot of "stiffness" or "friction." It resists moving easily.
• When the knob is turned down (small $\epsilon$), the sheet becomes very "slippery" and sensitive. The stiffness almost disappears.

2. The Critical Thresholds: The "Tipping Points"

The authors discovered that there are two specific "tipping points" for the knob $\epsilon$ that determine what happens to the sheet:

• The "No-Go" Zone ($\epsilon > \epsilon^*$): If the knob is set too high (too much stiffness), the two giants cancel each other out perfectly, and the sheet just stays flat. There is no solution where the sheet moves up or down; the only answer is "nothing happens."
• The "Sweet Spot" ($\epsilon < \epsilon^*_e$): If you turn the knob down low enough, the system wakes up. Suddenly, the sheet can settle into two different stable shapes:
  1. The Ground State (The Deep Valley): This is the most stable, lowest-energy shape. It's like the sheet settling into the deepest possible dip.
  2. The Second State (The High Hill): A second, less stable shape where the sheet is pushed higher up.

The paper proves that if you are in the "Sweet Spot," you will definitely find these two shapes. If you are in the "No-Go" zone, you find nothing.

3. The Big Discovery: What Happens When the Knob is Almost Zero?

The most exciting part of the paper is what happens when you turn the knob $\epsilon$ almost all the way down to zero.

Usually, in physics and math, when you remove the "stiffness" (the derivative term) from an equation, things get messy. You might expect the sheet to form sharp spikes, bubbles, or chaotic patterns near the edges.

But this paper says: No.

Instead of forming spikes or chaotic bubbles, the sheet settles into a smooth, predictable pattern that looks exactly like a recipe written on the sheet itself.

As the knob $\epsilon$ approaches zero, the shape of the sheet ($u$) converges to a specific formula:
$$\bar{u}_0(x) = \left( \frac{a(x)}{b(x)} \right)^{\text{power}}$$

The Analogy:
Imagine the sheet is a map. The "Growth" giant ($a$) and the "Damping" giant ($b$) have different strengths at different locations on the map.

• Where the Growth giant is strong and the Damping giant is weak, the sheet rises high.
• Where the Damping giant is strong, the sheet stays low.

The paper proves that as the "stiffness" vanishes, the sheet doesn't wiggle or spike. It simply becomes a perfect map of the ratio between these two giants. The sheet stops being a "physics problem" about movement and becomes a simple "algebra problem" about balancing two numbers at every single point.

4. Why This Matters (According to the Paper)

The authors emphasize that this is a rare case where the "limit" (what happens when the knob is zero) is not a chaotic mess or a single point, but a distributed equilibrium.

• The "Measure" Convergence: They prove that the sheet gets closer and closer to this perfect recipe shape everywhere, except perhaps for a few tiny, insignificant spots.
• The "Strong" Convergence: For most practical measurements (like the average height of the sheet), the sheet matches the recipe perfectly.

Summary

In short, the paper solves a puzzle about a rubber sheet pulled by two competing forces.

1. If the sheet is too stiff, it stays flat.
2. If it's just right, it settles into two distinct shapes.
3. If you make it almost perfectly slippery (remove the stiffness), it doesn't go crazy. Instead, it instantly transforms into a smooth, predictable shape determined entirely by the local balance of the two pulling forces.

The authors used a clever mathematical tool called the "nonlinear Rayleigh quotient" (think of it as a specialized ruler that measures the balance of forces) to find these exact tipping points and prove this smooth behavior.

Technical Summary

Technical Summary: Asymptotics of Ground States for a Singularly Perturbed $p$-Laplacian with Competing Superlinear Terms

Problem Statement
The paper investigates the singularly perturbed Dirichlet problem for the $p$-Laplacian equation with competing superlinear terms:
$$\begin{cases} -\varepsilon \Delta_p u = a(x)|u|^{q-2}u - b(x)|u|^{\gamma-2}u, & x \in \Omega, \\ u = 0, & x \in \partial\Omega, \end{cases}$$
where $\Omega \subset \mathbb{R}^N$ is a bounded connected domain with $C^1$ boundary, $1 < p < q < \gamma < p^*$ (with $p^*$ being the critical Sobolev exponent), and $\varepsilon > 0$ is a small parameter. The coefficients satisfy $a, b \in L^\infty(\Omega)$, $a \geq 0$, $a \not\equiv 0$, and $b(x) \geq \sigma_b > 0$ almost everywhere.

The study focuses on the existence of weak solutions in the Sobolev space $W^{1,p}_0(\Omega)$ and, specifically, the asymptotic behavior of ground states (global minimizers of the associated energy functional) as $\varepsilon \to 0^+$. Unlike many singular perturbation problems that exhibit concentration phenomena (spikes or bubbles), this problem involves a competition between two superlinear terms where the limiting behavior is governed by a pointwise algebraic balance rather than an elliptic boundary value problem.

Methodology
The authors employ the nonlinear Rayleigh quotient method, a variational approach developed in prior works [18, 21]. The core of the methodology involves:

1. Energy Functional: The problem is framed as finding critical points of the functional:
   $$\Phi_\varepsilon(u) = \frac{\varepsilon}{p} \int_\Omega |\nabla u|^p dx - \frac{1}{q} \int_\Omega a|u|^q dx + \frac{1}{\gamma} \int_\Omega b|u|^\gamma dx.$$
2. Generalized Rayleigh Quotients: Two critical parameter values, $\varepsilon^*$ and $\varepsilon^*_e$, are introduced via nonlinear generalized Rayleigh quotients defined on the set $D = \{u \in W^{1,p}_0(\Omega) \setminus \{0\} : \int_\Omega a|u|^q > 0\}$.
   • $\varepsilon^*$ is associated with the Nehari manifold (where the derivative of the functional vanishes).
   • $\varepsilon^*_e$ is associated with the zero energy level (where the functional value vanishes).
     These values are derived as suprema of specific functionals involving the norms of the gradient and the $L^q$ and $L^\gamma$ weighted integrals.
3. Variational Analysis: The existence of solutions is established using the Mountain Pass Theorem and direct minimization arguments. The non-existence of solutions for large $\varepsilon$ is proven by contradiction using the properties of the Rayleigh quotient.
4. Asymptotic Analysis: To study the limit $\varepsilon \to 0^+$, the authors analyze the convergence of ground states $u_\varepsilon$ to the explicit profile $\bar{u}_0(x) = (a(x)/b(x))^{1/(\gamma-q)}$. This involves proving boundedness in $L^\gamma(\Omega)$, establishing convergence in measure via the strict convexity of the pointwise potential, and applying Vitali's convergence theorem for strong $L^r$ convergence.

Key Contributions and Results

• Existence and Non-existence Thresholds:
  The paper establishes sharp thresholds for the parameter $\varepsilon$:

  • Non-existence: If $\varepsilon > \varepsilon^*$, problem (1.1) admits no nontrivial weak solutions.
  • Existence of Ground State: For $0 < \varepsilon < \varepsilon^*_e$, there exists at least one positive ground state $u_\varepsilon$. This solution satisfies $\Phi_\varepsilon(u_\varepsilon) < 0$ and is a local minimizer of the functional restricted to the Nehari manifold.
  • Existence of Second Solution: For the same range $0 < \varepsilon < \varepsilon^*_e$, there exists a second positive weak solution $v_\varepsilon$ with positive energy ($\Phi_\varepsilon(v_\varepsilon) > 0$), obtained via the Mountain Pass Theorem.
  • Regularity: Both solutions are shown to be locally $C^{1,\kappa}$ in the interior of $\Omega$.

• Asymptotic Behavior of Ground States:
  Under the additional assumption that $a(x) \geq \sigma_a > 0$ (strictly positive), the main result characterizes the limit of the ground states as $\varepsilon \to 0^+$:

  • Pointwise Limit: The ground states $u_\varepsilon$ converge in measure to the explicit function:
    $$\bar{u}_0(x) = \left( \frac{a(x)}{b(x)} \right)^{1/(\gamma-q)}.$$
  • Convergence Modes: The convergence is strong in $L^r(\Omega)$ for $1 \leq r < \gamma$ and weak in $L^r(\Omega)$ for $1 < r \leq \gamma$.
  • Limiting Equation: The limit $\bar{u}_0$ satisfies the algebraic balance equation $a(x)|u|^{q-2}u - b(x)|u|^{\gamma-2}u = 0$ almost everywhere, effectively replacing the differential equation in the singular limit.

Significance and Claims
The paper claims that its primary contribution lies in describing the asymptotic behavior of ground states for a competing-superlinear Dirichlet problem with small diffusion. The authors emphasize that, unlike typical singular perturbation problems where solutions concentrate at isolated points (spikes) or boundaries, the solution here converges to a spatially distributed equilibrium profile determined entirely by the variable coefficients $a(x)$ and $b(x)$.

The work addresses the difficulty of the "loss of differential order" in the limit $\varepsilon \to 0$, where the boundary condition is no longer encoded in the limiting algebraic equation. The authors demonstrate that the ground state selection mechanism of the original elliptic problem uniquely selects the full positive branch $\bar{u}_0$ among the continuum of possible algebraic solutions, provided $a(x)$ is strictly positive.

The results are presented as a rigorous application of the nonlinear Rayleigh quotient method to a specific class of quasilinear parametric problems, extending the understanding of bifurcation diagrams and extremal parameters in the context of competing superlinearities ($1 < p < q < \gamma$). The paper also notes that these findings translate directly to equivalent formulations of the problem with parameters $\lambda$ and $\nu$ (as shown in Corollary 1.2), providing a complete picture of the solution branches up to the critical thresholds.