Global multiplicity results in a Moore–Nehari type problem with a spectral parameter
Julián López-Gómez, Eduardo Muñoz-Hernández · 2025
Read Global multiplicity results in a Moore–Nehari type problem with a spectral parameter by Julián López-Gómez, Eduardo Muñoz-Hernández (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-00431.txt.
Introduction
Julián López-Gómez, Eduardo Muñoz-Hernández, and Fabio Zanolin charted solution families for a nonlinear equation in a 2025 paper. Their Moore-Nehari type problem has a coefficient that jumps between values in steps, and a spectral parameter tunes the equation's strength. Positive solutions can multiply or vanish as the step position shifts the problem between its autonomous and linear extremes, behavior earlier numerics had only suggested.
The authors prove several of those multiplicity patterns rigorously. They also analyze the delicate limit as the step approaches the endpoint, where solutions blow up pointwise and numerics go blind. That singular limit, handled here for the first time, shows analysis succeeding exactly where computation fails.
Catalog record
| CREATOR | Julián López-Gómez, Eduardo Muñoz-Hernández |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-00431
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Global multiplicity results in a Moore–Nehari type problem with a spectral parameter Thanks: This paper has been written under the auspices of the Ministry of Science and Innnovation of Spain under Reserach Grant PID2021-123343NB-I00, and the Institute of Interdisciplinary Mathematics of Complutense University of Madrid. Julián López-Gómez, Eduardo Muñoz-Hernández Abstract This paper analyzes the structure of the set of positive solutions of ( 1.1 ), where a ≡ a h a\equiv a_{h} is the piece-wise constant function defined in ( 1.3 ) for some h ∈ ( 0 , 1 ) h\in(0,1) . In our analysis, λ \lambda is regarded as a bifurcation parameter, whereas h h is viewed as a deformation parameter between the autonomous case when a = 1 a=1 and the linear case when a = 0 a=0 . In this paper, besides establishing some of the multiplicity results suggested by the numerical experiments of [ 2 ] , we have analyzed the asymptotic behavior of the positive solutions of ( 1.1 ) as h ↑ 1 h\uparrow 1 , when the shadow system of ( 1.1 ) is the linear equation − u ′′ = π 2 u -u^{\prime\prime}=\pi^{2}u . This is the first paper where such a problem has been addressed. Numerics is of no help in analyzing this singular perturbation problem because the positive solutions blow-up point-wise in ( 0 , 1 ) (0,1) as h ↑ 1 h\uparrow 1 if λ < π 2 \lambda<\pi^{2} . Keywords : Moore–Nehari equation. Multiplicity of positive solutions. Point-wise blow-up to a metasolution. Spectral parameter. MSC 2020 : 34B08, 34B16, 34B18. 1 Introduction This paper studies the one-dimensional second-order superlinear problem \left\{\begin{array}[]{ll}-u^{\prime\prime}=\lambda u+a(x)u^{p},\quad x\in(0,1),\\[1.0pt] u(0)=u(1)=0,\end{array}\right. (1.1) where \lambda\in\mathbb{R} is regarded as a bifurcation parameter, p>1 is fixed, and a:[0,1]\to[0,\infty) is a non-zero piece-wise constant function. Thus, a\gneq 0 in the sense that a\geq 0 , but a\neq 0 . The main goal of this paper is to study the global structure of the set of positive solutions of ( 1.1 ) in terms of the parameter \lambda and of the nature of the weight function a(x) . As p>1 , by the uniqueness of solution for the Cauchy problem associated to the differential equation of ( 1.1 ), any positive solution, u , of ( 1.1 ) must be strongly positive, in the sense that u(x)>0\;\;\hbox{for all}\;x\in(0,1),\;\;u^{\prime}(0)>0\;\;\hbox{and}\;\;u^{\prime}(1)<0. (1.2) In the special case when \lambda=0 and p is an odd integer, the problem ( 1.1 ) has been already analyzed by Moore and Nehari [ 16 ] for a symmetric piecewise weight function a(x) defined, after a spatial scaling, as a(x)=\;\begin{cases}1\qquad\hbox{if}\;x\in[0,\tfrac{1-h}{2}]\cup[\tfrac{1+h}{2},1],\\[4.30554pt] 0\qquad\hbox{if}\;x\in(\tfrac{1-h}{2},\tfrac{1+h}{2}),\end{cases} (1.3) where h\in(0,1) is fixed. The main result of [ 16 ] shows that, for some values of h\in(0,1) , the problem ( 1.1 ) possesses, at least, three positive solutions: a first solution which is symmetric (even) with respect to x=0.5 , and two additional asymmetric solutions, reflected from each other about x=0.5 . In recent years, also in the special case when \lambda=0 , some additional work has been done by Kajikiya et al. [ 9 ] by considering h as a bifurcation parameter and establishing the existence of a symmetry-breaking from the symmetric solution to a continuum of asymmetric solutions at some critical value h_{*}\in(0,1) (see also [ 7 ] for nodal solutions if p\in(0,1) , and [ 8 ] for nodal solutions if p>1 ). Some related problems where a symmetry-breaking occurs leading to multiplicity of positive solutions can be found in [ 5 , 6 , 17 , 18 ] . The main goal of this paper is to ascertain the global topological structure of the set of positive solutions of ( 1.1 ), (\lambda,u) , regarding \lambda\in\mathbb{R} as the primary bifurcation parameter. Then, we will ascertain how changes the underlying global bifurcation diagram in \lambda as h varies in the interval (0,1) . Thus, h is regarded as the secondary parameter of ( 1.1 ) for the special choice ( 1.3 ). Throughout this paper, to emphasize the dependence of a(x) on h\in(0,1) , we will eventually set a\equiv a_{h}. According to, e.g., Amann and López-Gómez [ 1 ] , under these assumptions, for every h\in(0,1) , the set of non-trivial non-negative solutions, (\lambda,u) , of problem ( 1.1 ) has a component, \mathscr{C}_{h}^{+} , such that (\pi^{2},0)\in\mathscr{C}_{h}^{+} and \mathcal{P}_{\lambda}(\mathscr{C}_{h}^{+})=(-\infty,\pi^{2}], where \mathcal{P}_{\lambda} stands for the \lambda -projection operator, \mathcal{P}_{\lambda}(\lambda,u)=\lambda\;\;\hbox{for all}\;\;(\lambda,u)\in\mathbb{R}\times\mathcal{C}([0,1];\mathbb{R}). In particular, ( 1.1 ) admits a positive solution if, and only if, \lambda<\pi^{2} . Therefore, the condition \lambda<\pi^{2} will be imposed throughout this paper. Moreover, according to the numerical experiments of Cubillos et al. [ 2 ] , for every h\in(0,1) , there is a value of the parameter \lambda , denoted by \lambda_{h,b} , which is a subcritical pitchfork bifurcation value for \mathscr{C}_{h}^{+} . These numerical experiments also suggest that ( 1.1 ) has three positive solutions for every \lambda<\lambda_{h,b} , one symmetric and two asymmetric, and that \lim_{h\downarrow 0}\lambda_{h,b}=-\infty\quad\hbox{and}\quad\lim_{h\uparrow 1}\lambda_{h,b}=\pi^{2}. (1.4) Although the first limit of ( 1.4 ) is rather natural, because the limiting shadow problem as h\downarrow 0 , \left\{\begin{array}[]{ll}-u^{\prime\prime}=\lambda u+u^{p}\quad\hbox{in}\;\;(0,1),\\[1.0pt] u(0)=u(1)=0,\end{array}\right. (1.5) has a unique positive solution for each \lambda\in(-\infty,\pi^{2}) , which follows by a direct analysis of the phase portrait of the autonomous differential equation -u^{\prime\prime}=\lambda u+u^{p}, the second limit of ( 1.4 ) is far from expected, because it entails that, for every \varepsilon>0 and sufficiently small 1-h , ( 1.1 ) has three positive solutions for every \lambda\in(-\infty,\pi^{2}-\varepsilon] , while the limiting shadow system of ( 1.1 ) as h\uparrow 1 is the linear problem \left\{\begin{array}[]{ll}-u^{\prime\prime}=\lambda u\quad\hbox{in}\;\;(0,1),\\[1.0pt] u(0)=u(1)=0,\end{array}\right. (1.6) which has a positive solution if, and only if, \lambda=\pi^{2} . Thus, the global multiplicity result as h\uparrow 1 , for every \lambda\in(-\infty,\pi^{2}) , is attributable to the nonlinear character of a_{h}u^{p}=u^{p} in [0,\frac{1-h}{2})\cup(\frac{1+h}{2},1] , which is a very tiny set as h\uparrow 1 . Another multiplicity result in a similar vein, caused by spatial heterogeneities, has been documented by López-Gómez and Rabinowitz in [ 15 ] . The main goal of this paper is to deliver some analytical proofs of some of these features and analyze the point-wise behavior of the three positive solutions of ( 1.1 ) as h\uparrow 1 . Precisely, our main findings are the following ones: (F1) For every \lambda<\pi^{2} , ( 1.1 ) has a unique positive symmetric solution, u_{\lambda} , and \mathscr{C}_{h,s}^{+}=\{(\lambda,u_{\lambda})\;:\;\lambda\in(-\infty,\pi^{2}]\} is a continuous curve such that \mathscr{C}_{h,s}^{+}\subset\mathscr{C}_{h}^{+} . (F2) For every h\in(0,1) , there exists \tilde{\lambda}=\tilde{\lambda}(h)<\pi^{2} such that, for every \lambda\leq\tilde{\lambda}(h) , ( 1.1 ) has, at least, three positive solutions. (F3) For any given \lambda\in\left(\pi^{2}/4,\pi^{2}\right) , there exists R_{\lambda}>0 such that, for every R>R_{\lambda} , the problem ( 1.1 ) has, at least, two asymmetric positive solutions for a suitable h=h(R)\in(0,1) . Moreover, \lim_{R\uparrow+\infty}h(R)=1. (F4) For every \lambda<\pi^{2} and h\in(0,1) , let u_{h} be any positive solution of ( 1.1 ). Then, \lim_{h\uparrow 1}u_{h}(x)=+\infty\;\;\hbox{for all}\;\;x\in(0,1). (1.7) In particular, the components \mathscr{C}_{h,s}^{+} and \mathscr{C}_{h}^{+} blow-up, point-wise in (0,1) , as h\uparrow 1 . (F5) For every \alpha>0 , there are two increasing sequences, \{h_{n}\}_{n\geq 1} in (0,1) and \{\lambda_{n}\}_{n\geq 1} in (0,\pi^{2}) , such that \lim_{n\to+\infty}h_{n}=1,\quad\lim_{n\to+\infty}\lambda_{n}=\pi^{2},\quad\hbox{and}\;\;\|u_{\lambda_{n}}\|_{\infty}=\alpha\;\;\hbox{for all}\;\;n\geq 1, where u_{\lambda_{n}} is the unique symmetric positive solution of ( 1.1 ) for (\lambda,h)=(\lambda_{n},h_{n})