We discuss a simple and effective technique for obtaining nonexistence results for nonnegative solutions of the following problem:

\begin{equation}\label{eq:D} -\Delta_p u = g_\lambda(x, u, \nabla u), \quad x \in \Omega, \end{equation} where $p>1$, $\lambda \in \mathbb{R}$ is a parameter, and $\Omega \subset \mathbb{R}^N$ ($N \geq 1$) is a bounded domain whose boundary $\partial \Omega$ is of class $C^{1}$.

Our main assumption on the nonlinearity $g_\lambda$ is the following:

$(\ast)$ There exists $M \subset \mathbb{R}$ such that $g_\lambda(x, t, \eta) > \lambda_1 t^{p-1}$ for all $\lambda \in M$, $x \in \Omega$, $t > 0$, and $z \in \mathbb{R}^N$.

Here $\lambda_1$ is the first eigenvalue of the $p$-Laplacian with the corresponding eigenfunction $\varphi_1$, i.e.,

It is known [1] that $\lambda_1$ is positive, simple and isolated, and $\varphi_1 > 0$ in $\Omega$. Moreover by the regularity result of [2] we have $\varphi_1 \in C^{1,\beta}(\overline{\Omega})$ for some $\beta \in (0,1)$.

Under a weak solution $u$ of \eqref{eq:D}, we mean a function $u \in W^{1,p}(\Omega)$ where $W^{1,p}(\Omega)$ is the standart Sobolev space, which satisfies \begin{equation} \label{eq:weak} \int_{\Omega} |\nabla u|^{p-2} \left( \nabla u, \nabla \phi \right) dx = \int_{\Omega} g_\lambda(x, u, \nabla u) \phi \, dx, \quad \forall \phi \in W_0^{1,p}(\Omega) \backslash { 0}. \end{equation}

Our main result is the following.

Theorem. Assume $(\ast)$ is satisfied. Then for every $\lambda \in M$ problem \eqref{eq:D} has no weak nonnegative solution $u \in C^{1}(\overline{\Omega})$.


Particular examples of equation \eqref{eq:D} with $g_\lambda(x, t, \eta)$ satisfying $(\ast)$ are:

1) equations with concave nonlinearity [3]

2) equations with convex-concave nonlinearity [4]

3) example from [5]

4) equations with singular nonlinearity [6]

5) the Liouville-Bratu-Gelfand problem [7]

6) equations with polynomial-type reaction term [8]

For each example of (1)-(6) there exist $\lambda^* > 0$ such that there are no weak positive $C^1$-solutions of \eqref{eq:D} for any $\lambda > \lambda^*$.


The main tool for the proof of Theorem is the following Picone inequality, see Theorem 1.1 in [9].

Lemma. Let $\phi \in W_0^{1,p}(\Omega) \cap C^1(\overline{\Omega})$ and $u \in C^1(\overline{\Omega})$ be such that $u > 0$, $\phi \geq 0$, and $\partial u(x) /\partial \nu < 0$ whenever $u(x) = 0$, $x \in \partial \Omega$, where $\nu$ is an outward normal to the boundary $\partial \Omega$. Then $\phi^p/u^{p-1} \in W_0^{1,p}(\Omega)$ and


Let us now prove our main result.

Proof of Theorem. Let $\lambda \in M$ where $M$ is defined by $(\ast)$. Assume that there exists a weak nonnegative solution $u \in C^1(\overline{\Omega})$ of \eqref{eq:D}. It follows from $(\ast)$ that $g_\lambda(x, t, \eta) > \lambda_1 t^{p-1} > 0$ for all $t>0$ and $x \in \Omega$. Hence $-\Delta_p u \geq 0$, and we apply the Hopf maximum prinicple from [10] to deduce that $u > 0$ in $\Omega$ and $\partial u(x) /\partial \nu < 0$ if $u(x)=0$ for $x \in \partial \Omega$.

Therefore, by Lemma,

\begin{equation} \label{eq:Il} \int_{\Omega} |\nabla u|^{p-2} \left( \nabla u, \nabla \left( \frac{\varphi_1^{p}}{u^{p - 1}} \right) \right) dx \leq \int_{\Omega} |\nabla \varphi_1|^{p} \,dx = \lambda_1 \int_{\Omega} |\varphi_1|^{p} \, dx. \end{equation}

On the other hand, since $u$ is the weak solution of \eqref{eq:D}, we test \eqref{eq:weak} by $\varphi_1^{p} / u^{p-1}$ and get

\begin{equation} \label{eq:u} \int_{\Omega} |\nabla u|^{p-2} \left( \nabla u, \nabla \left( \frac{\varphi_1^{p}}{u^{p - 1}} \right) \right) dx = \int_{\Omega} g_\lambda(x, u, \nabla u) \frac{\varphi_1^{p}}{u^{p - 1}} \, dx. \end{equation}

From \eqref{eq:Il} and \eqref{eq:u} it follow that

which reads as

However, since $\lambda \in M$, the assumption $(\ast)$ gives us a contradiction. $\square$


Remark. Essentially, Theorem is a corollary of the fact that the eigenvalue problem

does not possess any nonnegative supersolution whenever $\lambda > \lambda_1$. In particular, there are no positive eigenfunctions other than $\varphi_1$. Or, in other words, any higher eigenfunction have to be sign-changing.


Bibliography

  1. Lindqvist, P. (1990). On the equation $\text{div}(|\nabla u|^{p-2} \nabla u) + \lambda |u|^{p-2} u = 0$. Proceedings of the American Mathematical Society, 109(1), 157-164. 

  2. Lieberman, G. M. (1988). Boundary regularity for solutions of degenerate elliptic equations. Nonlinear Analysis: Theory, Methods & Applications, 12(11), 1203-1219. 

  3. Brezis, H., & Nirenberg, L. (1983). Positive solutions of nonlinear elliptic equations involving critical Sobolev exponents. Communications on Pure and Applied Mathematics, 36(4), 437-477. 

  4. Ambrosetti, A., Brezis, H., & Cerami, G. (1994). Combined effects of concave and convex nonlinearities in some elliptic problems. Journal of Functional Analysis, 122(2), 519-543. 

  5. Boccardo, L., Escobedo, M., & Peral, I. (1995). A Dirichlet problem involving critical exponents. Nonlinear Analysis: Theory, Methods & Applications, 24(11), 1639-1648. 

  6. Coclite, M. M., & Palmieri, G. (1989). On a singular nonlinear Dirichlet problem. Communications in Partial Differential Equations, 14(10), 1315-1327. 

  7. Jacobsen, J., & Schmitt, K. (2002). The Liouville–Bratu–Gelfand problem for radial operators. Journal of Differential Equations, 184(1), 283-298. 

  8. Ferrero, A. (2004). On the solutions of quasilinear elliptic equations with a polynomial-type reaction term. Advances in Differential Equations, 9(11-12), 1201-1234. 

  9. Allegretto, W., & Huang, X. Y. (1998). A Picone’s identity for the $p$-Laplacian and applications. Nonlinear Analysis: Theory, Methods & Applications, 32(7), 819-830. 

  10. Vazquez, J. L. (1984). A strong maximum principle for some quasilinear elliptic equations. Applied mathematics & optimization, 12(1), 191-202.