Morse index, topological degree and local uniqueness of multi-spikes solutions to the Lane-Emden problem in dimension two

Isabella Ianni, Peng Luo, Shusen Yan · 2025

Read Morse index, topological degree and local uniqueness of multi-spikes solutions to the Lane-Emden problem in dimension two by Isabella Ianni, Peng Luo, Shusen Yan (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-2506-12905.txt.

Introduction

Isabella Ianni, Peng Luo, and Shusen Yan count the instability of spiky solutions to a classic flat-domain equation, in this 2025 mathematics paper. The Lane-Emden problem, born in star theory, admits positive solutions that pile up into sharp spikes at chosen points. For domains in three or more dimensions, older theorems measured exactly how unstable each spike pattern is. The flat, two-dimensional case resisted.

The authors compute the Morse index, that instability count, for multi-spike solutions in any smooth bounded planar region. Digging into how the spikes concentrate also yields the total topological degree, and together the two give a new proof that such solutions are locally unique. Counting formulas like these organize whole solution landscapes for nonlinear equations.

Catalog record

CREATORIsabella Ianni, Peng Luo, Shusen Yan
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE174 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-12905

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Morse index, topological degree and local uniqueness of multi-spikes solutions to the Lane-Emden problem in dimension two Isabella Ianni, Peng Luo, Shusen Yan Abstract . We consider multi-spike positive solutions to the Lane-Emden problem in any bounded smooth planar domain and compute their Morse index, extending to the dimension N = 2 N=2 classical theorems due to Bahri-Li-Rey (1995) [ 2 ] and Rey (1999) [ 29 ] when N ≥ 4 N\geq 4 and N = 3 N=3 , respectively. Furthermore, by deeply investigating their concentration behavior, we also derive the total topological degree. The Morse index and the degree counting formula yield a new local uniqueness result. 1991 Mathematics Subject Classification 2020 Mathematics Subject Classification: 35A01 \cdot 35B25 \cdot 35J20 \cdot 35J60 1. Introduction and main results We consider the Lane-Emden problem \begin{cases}-\Delta u=u^{p}&\text{in}~\Omega,\\[1.42262pt] u>0&\text{in}~\Omega,\\[1.42262pt] u=0&\text{on}~\partial\Omega,\end{cases} (1.1) where \Omega\subset\mathbb{R}^{2} is a smooth bounded domain and p>1 . It is well known that problem ( 1.1 ) admits at least one solution, while the question of uniqueness or multiplicity is significantly more complex, depending on both the topology and geometry of the domain \Omega , as well as on the value of the exponent p (see for instance [ 10 ] ). A complete understanding of this issue is still far from being reached both in dimension N=2 than in the higher dimensions. For example, the solution is unique in any domain \Omega provided that p is sufficiently close to 1 ( [ 9 , 11 , 25 ] ). Moreover, for any value of p , uniqueness is known to hold when \Omega is a ball ( [ 17 ] ) or, more generally, when it is a symmetric and convex domain with respect to 2 orthogonal directions ( [ 10 , 9 ] ). More recently, uniqueness for problem ( 1.1 ) has also been established in general convex domains, without any symmetry assumption, for sufficiently large p (see [ 13 , 21 ] ), thus partially solving a longstanding conjecture proposed in [ 10 ] ). Conversely, multiple solutions exist in suitable non-convex domains \Omega (see [ 10 , 16 ] for examples involving dumbbell shaped domains, or [ 16 , 3 ] for the case of non-simply connected domains). In particular all the solutions found in [ 16 ] concentrate at a finite number of distinct points in \Omega , and vanish elsewhere as p\rightarrow+\infty . Note that all solutions of problem ( 1.1 ) are uniformly bounded in p (see [ 24 ] ). Notably, for the concentrating solutions built in [ 16 ] , the local maximum values around each concentration point converge all to \sqrt{e} . The location of these concentration points is determined by the Green function G and the Robin function R associated with -\Delta in \Omega under Dirichlet boundary conditions. Especially, in [ 16 ] , the existence of a solution concentrating at k\in\mathbb{N} distinct points x_{\infty,1},\cdots,x_{\infty,k}\in\Omega is obtained whenever \boldsymbol{x_{\infty}}:=(x_{\infty,1},\cdots,x_{\infty,k}) is a non-degenerate critical point of the Kirchhoff-Routh function \Psi_{k}:\Omega^{k}\rightarrow\mathbb{R} , which is defined as \Psi_{k}(\boldsymbol{a}):=\sum^{k}_{j=1}\Psi_{k,j}(\boldsymbol{a}),\hskip 9.24994pt~\mbox{ with }~\Psi_{k,j}(\boldsymbol{a}):=R\big(a_{j}\big)-\sum^{k}_{m=1,m\neq j}G\big(a_{j},a_{m}\big), (1.2) for \boldsymbol{a}=(a_{1},\cdots,a_{k}) , with a_{j}\in\Omega, j=1,\cdots,k . Recently, a rather complete characterization of the asymptotic behavior as p\rightarrow+\infty of any family u_{p} of solutions to problem ( 1.1 ) has been obtained in [ 14 ] , with sharp constants later provided in [ 12 , 30 ] , under the following energy uniform bound assumption: \sup_{p}\ p\|\nabla u_{p}\|^{2}_{2}\leq C. It was shown that solutions u_{p} necessarily concentrate, up to a subsequence, at a critical point (x_{\infty,1},\cdots,x_{\infty,k}) of \Psi_{k} , for some k\in\mathbb{N} , and vanish elsewhere, in the same way as the solutions found in [ 16 ] . Moreover, their local maximum values converge to \sqrt{e} , as p\rightarrow+\infty , and the total energy is asymptotically quantized as an integer multiple of 8\pi e : \lim_{p\rightarrow+\infty}p\|\nabla u_{p}\|^{2}_{2}=k\cdot 8\pi e. Furthermore, the asymptotic local profile of the solutions (suitably rescaled around each concentration point) was also described, showing that it is given by a radial solution U of the Liouville equation -\Delta U=e^{U} in the whole \mathbb{R}^{2} . We refer to solutions having these properties as spike solutions . It is worth noting that least energy solutions of problem ( 1.1 ) are single-spike solutions (i.e. k=1 ). For previous characterizations of this simpler case we refer to [ 27 , 28 , 1 ] . Observe that when k=1 , the Kirchhoff-Routh function \Psi_{1} reduces to the Robin function R . Furthermore, by [ 23 ] , when \Omega is convex it must be k=1 for any spike-solution. The sharp asymptotic description provided in [ 14 ] plays a crucial role in the proof of the uniqueness result for convex domains developed in [ 13 ] . Indeed, a key ingredient to derive uniqueness in [ 13 ] is the computation of the Morse index for single-spike solutions ( k=1 ). A central observation is that for the Morse index m(u_{p}) (resp. the augmented Morse index m_{0}(u_{p}) ) of a solution u_{p} to problem ( 1.1 ), the following holds: m(u_{p})=\sharp\big\{m\in\mathbb{N}:\lambda_{p,m}<1\big\}\hskip 9.24994pt(\text{resp. }m_{0}(u_{p})=\sharp\big\{m\in\mathbb{N}:\lambda_{p,m}\leq 1\big\}), (1.3) where \lambda_{p,1}<\lambda_{p,2}\leq\lambda_{p,3}\leq\cdots denotes the sequence of eigenvalues for the problem \begin{cases}-\Delta v=\lambda p(u_{p})^{p-1}v~&\mbox{in}~\Omega,\\ v=0~&\mbox{on}~\partial\Omega.\end{cases} (1.4) Thus, it becomes evident that the asymptotic information about the eigenvalues \lambda_{p,j} can be derived from the asymptotic information on u_{p} , leading to the computation of m(u_{p}) . For k -spike solutions, this computation has been rigorously carried out in [ 13 ] in the case k=1 , while when k\geq 2 , determining the exact Morse index is still an open question, whose answer requires a deeper understanding of the asymptotic behavior of u_{p} . More recently, the characterization in [ 14 ] has been refined in [ 21 ] , where higher order asymptotic expansions have been established (see Section 2.1 for more details). One of the aims of this paper is to compute the Morse index of any multi-spike solution of problem ( 1.1 ), making use of this sharpened analysis. Furthermore, the improved characterization in [ 21 ] enabled the proofs in [ 21 ] of the nondegeneracy of multi-spike solutions, and of a local uniqueness result for single-spike solutions. Here, local uniqueness means that if u and \widetilde{u} are two solutions of problem ( 1.1 ) concentrating at the same points, then u\equiv\widetilde{u} , for sufficiently large p . This property is particularly significant in non-convex domains, where uniqueness tout-court is not expected for problem ( 1.1 ), and the focus instead shifts to counting the number of solutions. However, for k -spike solutions, with k\geq 2 , local uniqueness remains an open problem. Indeed, dealing with multi-spike solutions is considerably more intricate, as the interactions between the spikes must be carefully taken into account. In this work, we also explore the local uniqueness property for multi-spike solutions, which, as we will see, is closely linked to the computation of the Morse index. Our first result concerns the computation of the Morse index for any multi-spike solution of problem ( 1.1 ). To state it precisely, let us denote by m(\boldsymbol{a},f) and m_{0}(\boldsymbol{a},f) the Morse and augmented Morse index , respectively, of \boldsymbol{a}=(a_{1},\cdots,a_{k})\in\Omega^{k} as a critical point of a

Read the complete text