On Brezis-Nirenberg problems: open questions and new results in dimension six

Fengliu Li, Giusi Vaira, Juncheng Wei, Yuanze Wu · 2025

Read On Brezis-Nirenberg problems: open questions and new results in dimension six by Fengliu Li, Giusi Vaira, Juncheng Wei, Yuanze Wu (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-2509-19863.txt.

Introduction

A classic partial differential equation problem, named for Brezis and Nirenberg, sits at the critical exponent where standard methods break. Fengliu Li and three co-authors survey its history and open questions, then add new results in dimension six. The equation couples a linear term with a critical power term on a bounded domain, with zero boundary values. Existence of solutions hinges on the parameter and the dimension in delicate ways.

Dimension six has resisted several standard attacks, which makes progress there notable. The paper mixes review, new theorems, and a list of open problems for the community. Critical problems act as benchmarks for nonlinear analysis as a whole. A clear map of what is known, plus fresh ground in a hard dimension, serves both specialists and newcomers.

Catalog record

CREATORFengliu Li, Giusi Vaira, Juncheng Wei, Yuanze Wu
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE79 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-19863

Browse the full library · About the Tower of Babel

Full text

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

On Brezis-Nirenberg problems: open questions and new results in dimension six Thanks: G. Vaira is partially supported by the MUR-PRIN-P2022YFAJH “Linear and Nonlinear PDE’s: New directions and Applications” and by the INdAM-GNAMPA project “Fenomeni non lineari: problemi locali e non locali e loro applicazioni”, CUP E5324001950001. J. Wei is partially supported by GRF of Hong Kong entitled ”New frontiers in singularity formations in nonlinear partial differential equations”. Y. Wu is supported by NSFC (No. 12171470). F. Li is partially supported by NSFC (No. 12171470) and the Graduate Innovation Program of China University of Mining and Technology 2025WLKXJ133 Fengliu Li, Giusi Vaira, Juncheng Wei, Yuanze Wu Abstract . In this paper, we consider the Brezis-Nirenberg problem { − Δ u = λ u + | u | 2 ∗ − 2 u , in ​ Ω , u = 0 , on ​ ∂ Ω , \left\{\begin{aligned} &-\Delta u=\lambda u+|u|^{2^{*}-2}u,\quad&\mbox{in}\,\Omega,\\ &u=0,\quad&\mbox{on}\,\partial\Omega,\end{aligned}\right. where Ω \Omega is a smoothly bounded domain of ℝ N \mathbb{R}^{N} with N ≥ 3 N\geq 3 , λ > 0 \lambda>0 is a parameter and 2 ∗ = 2 ​ N N − 2 2^{*}=\frac{2N}{N-2} is the critical Sobolev exponent. We first recall the history of the Brezis-Nirenberg problem and then provide new results of it in dimension six. Finally, we also list some open questions on the Brezis-Nirenberg problem. Keywords: Brezis-Nirenberg problem; Multi-bump solution; Bubbling phenomenon; Sign-changing solution. AMS Subject Classification 2020: 35B33; 35B40; 35B44; 35J15. To the memory of Louis Nirenberg with admiration 1. Introduction In this paper we consider the Brezis-Nirenberg problem, namely we let (1.1) \left\{\begin{aligned} &-\Delta u=\lambda u+|u|^{2^{*}-2}u,\quad&\mbox{in}\,\Omega,\\ &u=0,\quad&\mbox{on}\,\partial\Omega,\end{aligned}\right. where \Omega is a smoothly bounded domain of \mathbb{R}^{N} with N\geq 3 , \lambda>0 is a parameter and 2^{*}=\frac{2N}{N-2} is the critical Sobolev exponent. 1.1. History of ( 1.1 ) Problem ( 1.1 ) is introduced by Brezis and Nirenberg in their celebrated paper [ 13 ] , where they gave a detailed study on the existence of a positive solution of it. By applying the Pohozaev identity [ 49 ] , Brezis and Nirenberg first pointed out in [ 13 ] that problem ( 1.1 ) only has the trivial solution when \Omega is star-shaped and \lambda\leq 0 , and by multiplying ( 1.1 ) with the unique eigenfunction of \lambda_{1}(\Omega) , where \lambda_{1}(\Omega) is the first eigenvalue of -\Delta with Dirichlet boundary condition, Brezis and Nirenberg next pointed out in [ 13 ] that problem ( 1.1 ) has no positive solutions for \lambda\geq\lambda_{1}(\Omega) . Moreover, they also showed in [ 13 ] that for N\geq 4 the problem ( 1.1 ) admits a positive solution for every \lambda\in(0,\lambda_{1}(\Omega)) . The situation in dimension N=3 is delicate and in [ 13 ] Brezis and Nirenberg are able to prove the existence of \lambda_{*}(\Omega)>0 such that a positive solution of ( 1.1 ) exists for every \lambda\in(\lambda^{*}(\Omega),\lambda_{1}(\Omega)) . The existence of the positive value \lambda_{*}(\Omega) in the existence theory of positive solutions of ( 1.1 ) is necessary in general bounded domains in dimension three, since they also showed that if \Omega\subset\mathbb{R}^{3} is strictly star-shaped about the origin, then there are no positive solutions of ( 1.1 ) for \lambda close to zero. When \Omega is the unit ball B_{1} in dimension three, the existence theory of positive solutions of ( 1.1 ) and the value of \lambda_{*}(B_{1}) are completely known, i.e. \lambda_{*}(B_{1})=\frac{\lambda_{1}(B_{1})}{4} and a positive solution of ( 1.1 ) exists if and only if \lambda\in\left(\frac{\lambda_{1}(B_{1})}{4},\lambda_{1}(B_{1})\right) . Conversely, Bahri and Coron [ 7 ] presented an existence result for a positive solution to problem ( 1.1 ) for \Omega with a nontrivial topology and \lambda=0 . In the last decades problem ( 1.1 ) has attracted the attention of many researchers due to the critical behavior and the lack of compactness of the Brezis-Nirenberg problem ( 1.1 ) that prevents to use standard variational methods. A first question is about the asymptotic behavior of the solution founded by Brezis and Nirenberg. This was studied by Han [ 30 ] and Rey [ 52 ] and they showed, with different proofs, that such solutions blow up at exactly one point. They also determined the exact blow-up rate as well as the location of the limit concentration point which is a critical point of the Robin function (see also [ 62 ] in which it was shown that this concentration point is a global minimum of the Robin function). Bahri, Li and Rey [ 8 ] determined the asymptotic behavior and locations of multi-bubbling solutions to Brezis-Nirenberg problem. Most recently, a fine multi-bubbles analysis of positive solutions has been object of two papers by König and Laurin. Indeed, in [ 37 ] the authors give an accurate description of the blow-up profile of a solution with multiple blow-up points when N\geq 4 while in [ 36 ] they consider the case N=3 . The 3-dimensional case was firstly faced by Druet [ 25 ] . In particular, the asymptotic analysis of blowing-up solutions ensures that the blow-up points are nothing but the critical points of a suitable function which involves the Green function of -\Delta in \Omega with Dirichlet boundary condition and the Robin function. Furthermore, a counterpart of these asymptotic results are studied. The first result in this direction is due to Rey that built in [ 51 ] a solution that blows up at one non-degenerate critical point of the Robin function when N\geq 5 and \lambda\to 0^{+} . In [ 42 ] , Musso and Pistoia constructed positive solutions with multiple blow-up points when N\geq 5 as \lambda\to 0^{+} whereas, Musso and Salazar [ 44 ] considered the case of multiple blow-up points in dimension N=3 as \lambda goes to the strictly positive number \lambda^{*} predicted by Druet in [ 25 ] (see also [ 21 ] for the construction of one-peak solution as \lambda\to\lambda^{*} in dimension N=3 ). It is only very recently that the first true bubbling solutions have been constructed in dimension N=4 as \lambda\to 0^{+} , see [ 47 ] for the case of one peak solution and [ 46 ] for the case of multiple peak solutions. For what concern sign-changing solutions of ( 1.1 ), several existence results have been obtained for N\geq 4 . Indeed one can get sign-changing solutions for every \lambda>0 , \lambda\not\in\sigma(-\Delta) if N=4 and for every \lambda>0 if N\geq 5 (see [ 16 , 59 ] ). The case N=3 presents even more difficulties (as for positive solutions) and the question on what is the range of \lambda for which a sign-changing solution of ( 1.1 ) exists in dimension N=3 has been raised by Brezis in a collection of his favorite open problem [ 12 ] , which, as said by Brezis, is open for 40 years. We point out that, very recently, by using the del Pino-Musso-Pacard-Pistoia sign-changing solutions of the Yamabe problem in [ 22 ] as the building blocks, Sun, Wei and Yang constructed infinitely many non-radial sign-changing solutions of the 3-dimensional Brezis-Nirenberg problem ( 1.1 ) in \Omega=B_{1} the unit ball in \mathbb{R}^{3} and for any \lambda>0 in [ 57 ] , which gives a complete answer to the Brezis’ first open problem proposed in

Read the complete text