Searcharxiv⌕ Search

arXiv subjects

Bruno Premoselli

Publications and source records attributed to Bruno Premoselli.

At least 19 recordsLinked to original sources
arXiv2026

A priori bounds for energy-bounded solutions of critical polyharmonic equations

We investigate critical polyharmonic equations of the type: $$ Lu = |u|^{2^\sharp-2} u \quad \text{ in } \Omega $$ with Dirichlet boundary conditions, in a smooth bounded domain $\Omega$ of $\mathbb{R}^n$. Here $L$ is an elliptic differential operator of integer order $2 \le 2k < n$ whose leading order term is $(-\Delta)^k$ and $2^\sharp = \frac{2n}{n-2k}$ is the critical Sobolev exponent. Our main result establishes, in large dimensions, uniform \emph{a priori} bounds in $C^{2k}(\overline{\Omega})$ for bounded-energy solutions of this problem, that only depend on an upper bound on the energy. We prove this under a coercivity assumption of sorts on the lower-order terms of $L$. Our results are sharp, at least when $k=1$. Our approach uses asymptotic analysis techniques and in the course of the proof we obtain in particular a new global pointwise description of bounded-energy blowing-up solutions for this problem, which is of independent interest.

math.AP↗
arXiv2025

Compactness of conformal metrics with constant $Q$-curvature of higher order

Let $k\ge1$ be a positive integer and let $P_g$ be the GJMS operator $P_{g}$ of order $2k$ on a closed Riemannian manifold $(M,g)$ of dimension $n>2k$. We investigate the compactness of the set of conformal metrics to $g$ with prescribed constant positive $Q$-curvature of order $2k$- or, equivalently, of the set of positive solutions for the $2k$-th order $Q$-curvature equation. Under a natural positivity-preserving condition on $P_{g}$ we establish compactness, for an arbitrary $1 \le k < \frac{n}{2}$, under the following assumptions: $(M,g)$ is locally conformally flat and $P_g$ has positive mass in $M$, or $2k+1 \le n \le 2k+5$ and $P_g$ has positive mass in $M$, or $n \ge 2k+4$ and $|\text{W}_g|_g >0$ in $M$. For an arbitrary $1 \le k < \frac{n}{2}$, the expression of $P_g$ is not explicit, which is an obstacle to proving compactness. We overcome this by relying on Juhl's celebrated recursive formulae for $P_g$ to perform a refined blow-up analysis for solutions of the $Q$-curvature equation and to prove a Weyl vanishing result for $P_g$. This is the first compactness result for an arbitrary $1 \le k < \frac{n}{2}$ and the first successful instance where Juhl's formulae are used to yield compactness. Our result also hints that the threshold dimension for compactness for the $2k$-th order $Q$-curvature equation diverges as $k \to + \infty$.

math.AP↗
arXiv2025

Least-energy solutions of the Br\'ezis-Nirenberg problem in the non-coercive case in dimension $3$

Let $\Omega$ be a bounded, smooth domain of $\mathbb{R}^n, n \ge 3$ and $\lambda \ge 0$. We consider the celebrated Br\'ezis-Nirenberg problem: \begin{equation}\label{eq:critlambda:abs} \tag{*} \left\{\begin{aligned} -\Delta u -\lambda u & =\left|u\right|^{2^*-2}u &\hbox{ in } \Omega, u & = 0 \quad \text{ in } \partial \Omega, \end{aligned}\right. \end{equation} where $2^* = \frac{2n}{n-2}$. When $n=3$ we investigate the existence of \emph{least-energy solutions} for this problem, that we define as having the lowest $L^{2^*}(\Omega)$ norm among all non-zero solutions. We prove that least-energy solutions of the Br\'ezis-Nirenberg problem exist when $\lambda$ belongs to a left neighbourhood of any eigenvalue of $-\Delta$ that we explicitly characterise by a positive mass assumption. We obtain in particular the first \emph{existence} result for the Br\'ezis-Nirenberg problem on a general smooth bounded domain $\Omega$ when $n=3$ and $\lambda \ge \Lambda_1$. In order to do this we introduce, for any $\lambda \ge 0$, a new variational problem inspired from spectral-theoretic considerations which is as follows: for any $u \in L^{2^*}(\Omega), u>0$ a.e., we consider the principal eigenvalue of $- \Delta-\lambda$ on the weighted space $L^2(\Omega, u^{2^*-2} dx)$, whose value we then minimise over the set of normalised weights $\Vert u \Vert_{2^*} = 1$. When $\lambda \ge \Lambda_1$ this defines a new, non-smooth variational problem for which we develop a variational theory. We prove that its minimisers exist under the aforementioned positive mass assumption and that they yield \emph{least-energy} solutions. We also obtain new results in the higher-dimensional case $n \ge 4$, where we show that the energy function of the Br\'ezis-Nirenberg problem is discontinuous exactly at the eigenvalues of $- \Delta$.

math.AP↗
arXiv2025

Extremising eigenvalues of the GJMS operators in a fixed conformal class

Let $(M,g)$ be a closed Riemannian manifold of dimension $n\geq 3$. If $s$ is a positive integer satisfying $2s<n$, we let $P_g^s$ be the GJMS operator of order $2s$ in $M$. We investigate in this paper the extremal values taken by fixed eigenvalues of $P_h^s$ as $h$ runs through the whole conformal class $[g]$. These extremal values -- that we call throughout the paper \emph{conformal eigenvalues} -- are conformal invariants of $(M,g)$ and optimisers for these problems, when they exist, are known to not be smooth metrics in general. In this paper we develop a general framework that allows us to address the the existence theory for extremals of conformal eigenvalues. We define and investigate eigenvalues for singular conformal metrics, that we call \emph{generalised eigenvalues}. We develop a new variational framework for renormalised eigenvalues of any index over the set of admissible (singular) conformal factors: we obtain semi-continuity results and Euler-Lagrange equations for local extremals. Using this framework we prove, under mild assumptions on $(M,g)$ and $s$, several new (non)-existence results for extremals of renormalised eigenvalues over $[g]$. These include, among other results, a maximisation result for negative eigenvalues, the minimisation of the principal eigenvalue of $P_g^s$ and the analysis of the conformal eigenvalues of the round sphere $(\mathbb{S}^n, g_0)$. We also establish a strong connection between the existence of optimisers and (nodal) solutions of prescribed $Q$-curvature equations. Our analysis allows any order $s \ge 1$ and allows $P_g^s$ to have kernel. Previous results only covered the cases $s=1,2$ and $k=1,2$. Our work strongly generalises these results to any $s \ge 1$ and to eigenvalues of any order.

math.DG↗
arXiv2024

Compactness results for Sign-Changing Solutions of critical nonlinear elliptic equations of low energy

Let $\Omega$ be a bounded, smooth connected open domain in $\mathbb{R}^n$ with $n\geq 3$. We investigate in this paper compactness properties for the set of sign-changing solutions $v \in H^1_0(\Omega)$ of \begin{equation} \tag{*} -\Delta v+h v =\left|v\right|^{2^*-2}v \hbox{ in } \Omega, \quad v = 0 \hbox{ on } \partial \Omega \end{equation} where $h\in C^1(\overline{\Omega})$ and $2^*:=2n/(n-2)$. Our main result establishes that the set of sign-changing solutions of $(*)$ at the lowest sign-changing energy level is unconditionally compact in $C^2(\overline{\Omega})$ when $3 \le n \le 5$, and is compact in $C^2(\overline{\Omega})$ when $n \ge 7$ provided $h$ never vanishes in $\overline{\Omega}$. In dimensions $n \ge 7$ our results apply when $h >0$ in $\overline{\Omega}$ and thus complement the compactness result of Devillanova-Solimini, Adv. Diff. Eqs. 7 (2002). Our proof is based on a new, global pointwise description of blowing-up sequences of solutions of $(*)$ that holds up to the boundary. We also prove more general compactness results under perturbations of $h$.

math.AP↗
arXiv2024

Nonexistence of minimizers for the second conformal eigenvalue near the round sphere in low dimensions

We consider the problem of minimizing the second conformal eigenvalue of the conformal Laplacian in a conformal class of metrics with renormalized volume. We prove, in dimensions $n\in\left\{3,\dotsc,10\right\}$, that a minimizer for this problem does not exist for metrics sufficiently close to the round metric on the sphere. This is in striking contrast with the situation in dimensions $n \ge 11$, where Ammann and Humbert obtained the existence of minimizers for the second conformal eigenvalue on any smooth closed non-locally conformally flat manifold. As a byproduct of our techniques, we also obtain a lower bound on the energy of sign-changing solutions of the \nobreak Yamabe equation in dimensions 3, 4 and 5, which extends a result obtained by Weth in the case of the round sphere.

math.DG↗
arXiv2024

One-bubble nodal blow-up for asymptotically critical stationary Schr\"odinger-type equations

We investigate in this work families $(u_\epsilon)_{\epsilon >0}$ of sign-changing blowing-up solutions of asymptotically critical stationary nonlinear Schr\"odinger equations of the following type: $$\Delta_g u_\epsilon + h_\epsilon u_\epsilon = |u_{\epsilon}|^{p_\epsilon-2} u_\epsilon $$ in a closed manifold $(M,g)$, where $h_\epsilon$ converges to $h$ in $C^1(M)$. Assuming that $(u_\epsilon)_{\epsilon >0}$ blows-up as \emph{a single sign-changing bubble}, we obtain necessary conditions for blow-up that constrain the localisation of blow-up points and exhibit a strong interaction between $h$, the geometry of $(M,g)$ and the bubble itself. These conditions are new and are a consequence of the sign-changing nature of $u_\epsilon$.

math.AP↗
arXiv2022

Classification of radial blow-up at the first critical exponent for the Lin-Ni-Takagi problem in the ball

We investigate the behaviour of radial solutions to the Lin-Ni-Takagi problem in the ball $B_R \subset \mathbb{R}^N$ for $N \ge 3$: \begin{equation*} \left \{ \begin{aligned} - \triangle u_p + u_p & = |u_p|^{p-2}u_p & \textrm{ in } B_R, \\ \partial_\nu u_p & = 0 & \textrm{ on } \partial B_R, \end{aligned} \right. \end{equation*} when $p $ is close to the first critical Sobolev exponent $2^* = \frac{2N}{N-2}$. We obtain a complete classification of finite energy radial smooth blowing up solutions to this problem. We describe the conditions preventing blow-up as $p \to 2^*$, we give the necessary conditions in order for blow-up to occur and we establish their sharpness by constructing examples of blowing up sequences. Our approach allows for asymptotically supercritical values of $p$. We show in particular that, if $p \geq 2^\ast$, finite-energy radial solutions are precompact in $C^2(\bar{B_R})$ provided that $N\geq 7$. Sufficient conditions are also given in smaller dimensions if $p=2^\ast$. Finally we compare and interpret our results to the bifurcation analysis of Bonheure, Grumiau and Troestler in Nonlinear Anal. 147 (2016).

math.AP↗
arXiv2022

Sign-changing blow-up for the Yamabe equation at the lowest energy level

We investigate the blow-up behavior of sequences of sign-changing solutions for the Yamabe equation on a Riemannian manifold $(M,g)$ of positive Yamabe type. For each dimension $n\ge11$, we describe the value of the minimal energy threshold at which blow-up occurs. In dimensions $11 \le n \le 24$, where the set of positive solutions is known to be compact, we show that the set of sign-changing solutions is not compact and that blow-up already occurs at the lowest possible energy level. We prove this result by constructing a smooth, non-locally conformally flat metric on space forms $\mathbb{S}^n/\Gamma$, $\Gamma \neq \{1\}$, whose Yamabe equation admits a family of sign-changing blowing-up solutions. As a counterpart of this result, we also prove a sharp compactness result for sign-changing solutions at the lowest energy level, in small dimensions or under strong geometric assumptions.

math.AP↗
arXiv2022

Stability and instability results for sign-changing solutions to second-order critical elliptic equations

On a smooth, closed Riemannian manifold $\left(M,g\right)$ of dimension $n\ge3$, we consider the stationary Schr\"odinger equation $\Delta_gu+h_0u=\left|u\right|^{2^*-2}u$, where $\Delta_g:=-\text{div}_g\nabla$, $h_0\in C^1\left(M\right)$ and $2^* :=\frac{2n}{n-2}$. We prove that, up to perturbations of the potential function $h_0$ in $C^1\left(M\right)$, the sets of sign-changing solutions that are bounded in $H^1\left(M\right)$ are precompact in the $C^2$ topology. We obtain this result under the assumptions that $\left(M,g\right)$ is locally conformally flat, $n\ge7$ and $h_0\ne\frac{n-2}{4\left(n-1\right)}\text{Scal}_g$ at all points in $M$, where $\text{Scal}_g$ is the scalar curvature of the manifold. We then provide counterexamples in every dimension $n\ge3$ showing the optimality of these assumptions.

math.AP↗
arXiv2021

A priori estimates for finite-energy sign-changing blowing-up solutions of critical elliptic equations

We prove sharp pointwise blow-up estimates for finite-energy sign-changing solutions of critical equations of Schrödinger-Yamabe type on a closed Riemannian manifold $(M,g)$ of dimension $n \ge 3$. This is a generalisation of the so-called $C^0$-theory for positive solutions of Schrödinger-Yamabe type equations. To deal with the sign-changing case we develop a method of proof that combines an \emph{a priori} bubble-tree analysis with a finite-dimensional reduction, and reduces the proof to obtaining sharp \emph{a priori} blow-up estimates for a linear problem.

math.AP↗
arXiv2020

Towers of bubbles for Yamabe-type equations and for the Brézis-Nirenberg problem in dimensions $n \ge 7$

Let $(M,g)$ be a closed locally conformally flat Riemannian manifold of dimension $n \ge 7$ and of positive Yamabe type. If $ξ_0$ denotes a non-degenerate critical point of the mass function we prove the existence, for any $ k \ge 1$ and $\varepsilon >0$, of a positive blowing-up solution $u_{\varepsilon}$ of $$\triangle_g u_{\varepsilon} +\big( c_n S_g +\varepsilon h\big) u_{\varepsilon} = u_{\varepsilon}^{2^*-1},$$ that blows up like the superposition of $k$ positive bubbles concentrating at different speeds at $ξ_0$. The method of proof combines a finite-dimensional reduction with the sharp pointwise analysis of solutions of a linear problem. As another application of this method of proof we construct sign-changing blowing-up solutions $u_{\varepsilon}$ for the Brézis-Nirenberg problem $$ \triangle_ξ u_{\varepsilon} - \varepsilon u_{\varepsilon} = |u_{\varepsilon}|^{\frac{4}{n-2}} u_{\varepsilon} \textrm{ in } Ω, \quad u_{\varepsilon} = 0 \textrm{ on } \partial Ω$$ on a smooth bounded open set $Ω\subset \mathbb{R}^n$, $n \ge 7$, that look like the superposition of $k$ positive bubbles of alternating sign.

math.AP↗
arXiv2020

Examples of compact Einstein four-manifolds with negative curvature

We give new examples of compact, negatively curved Einstein manifolds of dimension $4$. These are seemingly the first such examples which are not locally homogeneous. Our metrics are carried by a sequence of 4-manifolds $(X_k)$ previously considered by Gromov and Thurston. The construction begins with a certain sequence $(M_k)$ of hyperbolic 4-manifolds, each containing a totally geodesic surface $Σ_k$ which is nullhomologous and whose normal injectivity radius tends to infinity with $k$. For a fixed choice of natural number $l$, we consider the $l$-fold cover $X_k \to M_k$ branched along $Σ_k$. We prove that for any choice of $l$ and all large enough $k$ (depending on $l$), $X_k$ carries an Einstein metric of negative sectional curvature. The first step in the proof is to find an approximate Einstein metric on $X_k$, which is done by interpolating between a model Einstein metric near the branch locus and the pull-back of the hyperbolic metric from $M_k$. The second step in the proof is to perturb this to a genuine solution to Einstein's equations, by a parameter dependent version of the inverse function theorem. The analysis relies on a delicate bootstrap procedure based on $L^2$ coercivity estimates.

math.DG↗
arXiv2016

A pointwise finite-dimensional reduction method for a fully coupled system of Einstein-Lichnerowicz type

We construct non-compactness examples for the fully coupled Einstein-Lichnerowicz constraint system in the focusing case. The construction is obtained by combining pointwise a priori asymptotic analysis techniques, finite-dimensional reductions and a fixed-point argument. More precisely, we perform a fixed-point procedure on the remainders of the expected blow-up decomposition. The argument consists of an involved finite-dimensional reduction coupled with a ping-pong method. To overcome the non-variational structure of the system, we work with remainders which belong to strong function spaces and not merely to energy spaces. Performing both the ping-pong argument and the finite-dimensional reduction therefore heavily relies on a priori pointwise asymptotic techniques.

math.AP↗
arXiv2015

Stability and instability of the Einstein-Lichnerowicz constraint system

We investigate the relevance of the conformal method by investigating stability issues for the Einstein-Lichnerowicz conformal constraint system in a nonlinear scalar-field setting. We prove the stability of the system with respect to arbitrary perturbations of generic focusing physics data on closed locally conformally flat manifolds, in any dimension. We also show that our stability result is sharp by constructing explicit instability examples when its assumptions are not satisfied. Our results apply to a more general class of constraint-like systems.

math.AP↗
arXiv2014

Stability of the Einstein-Lichnerowicz constraints system

We study the Einstein-Lichnerowicz constraints system, obtained through the conformal method when addressing the initial data problem for the Einstein equations in a scalar field theory. We prove that this system is stable with respect to the physics data when posed on the standard $3$-sphere.

math.AP↗
arXiv2013

Effective multiplicity for the Einstein-scalar field Lichnerowicz equation

We prove the stability of the Einstein-scalar field Lichnerowicz equation under subcritical perturbations of the critical nonlinearity in dimensions 3, 4 and 5. As a consequence, we obtain the existence of a second solution to the equation in several cases. In particular, in the positive case, including the CMC positive cosmological constant case, we show that each time a solution exists, the equation produces a second solution with the exception of one critical value for which the solution is unique.

math.AP↗