On the characterization of the Dirichlet and Fučík spectra of the one-dimensional anisotropic p p -Laplace operator
Raul Fernandes Horta · 2025
Read On the characterization of the Dirichlet and Fučík spectra of the one-dimensional anisotropic p p -Laplace operator by Raul Fernandes Horta (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-2501-10599.txt.
Introduction
Raul Fernandes Horta and Marcos Montenegro solve a spectral puzzle for a lopsided math operator in a 2025 paper. The operator bends one way when a curve rises and another way when it falls, a twist called anisotropy. They study which tones, or eigenvalues, it can sound on a line segment with pinned ends.
The pair pin down the full tone set and the shapes of the matching modes, borrowing from the even-handed case by gluing reshaped modes together. A side prize is a sharper version of a classic inequality that bounds a curve by its slope. They extend the work to a two-sided spectrum used in stability tests.
Catalog record
| CREATOR | Raul Fernandes Horta |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-10599
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On the characterization of the Dirichlet and Fučík spectra of the one-dimensional anisotropic p p -Laplace operator Raul Fernandes Horta Abstract 2020 Mathematics Subject Classification: 34B15; 34L05; 34L10; 34L15; 34L30 Key words: Anisotropic p -Laplacian, first eigenvalue, one-sided Poincaré inequality, Dirichlet spectrum E-mail addresses : [email protected] (M. Montenegro) Abstract The paper is concerned with the Dirichlet spectrum \Lambda^{a,b}_{p}(0,L) of the anisotropic p -Laplace operator -\Delta^{a,b}_{p} on an interval (0,L) where \Delta^{a,b}_{p}u:=\left(a^{p}[(u^{\prime})^{+}]^{p-1}-b^{p}[(u^{\prime})^{-}]^{p-1}\right)^{\prime},\ \ a,b>0. The set \Lambda^{a,b}_{p}(0,L) and the respective eigenfunctions are completely characterized for a\neq b in terms of the corresponding ones within the isotropic context. As an interesting application, we derive a new optimal Poincaré inequality that is stronger than the classical counterpart. The leading ideas are based on glue arguments of conveniently modified eigenfunctions and maximum type principles. More generally, our approach allows to characterize the Fučík spectrum \Sigma^{a,b}_{p}(0,L) of -\Delta^{a,b}_{p} on (0,L) and mainly the corresponding solutions. All results are novelty even for the nonlinear operator \Delta^{a,b}_{2} . 1 Introduction and main theorems Quasilinear elliptic equations involving the p -Laplace operator arise in a variety of situations such as fluid flows through some types of porous media, singular solutions for the Emden-Fowler and Einstein-Yang-Mills equations, the Lane-Emden equations for equilibrium configurations of gaseous stellar objects, among many others, see for instance Chapter 14 of the book [ 23 ] . In this paper we carry out a complete study on the spectrum of such operators in dimension one involving asymmetric conductivity. More specifically, given numbers p>1 and a,b,L>0 with a\neq b , let \Delta^{a,b}_{p} be the anisotropic p -Laplace operator defined weakly for functions u\in W_{0}^{1,p}(0,L) as \langle\Delta^{a,b}_{p}u,\varphi\rangle=-\int_{0}^{L}\left(a^{p}[(u^{\prime}(t))^{+}]^{p-1}-b^{p}[(u^{\prime}(t))^{-}]^{p-1}\right)\varphi^{\prime}(t)\,dt for every \varphi\in W_{0}^{1,p}(0,L) , in short \Delta^{a,b}_{p}u=\left(a^{p}[(u^{\prime})^{+}]^{p-1}-b^{p}[(u^{\prime})^{-}]^{p-1}\right)^{\prime}. We consider the associated boundary value problems \left\{\begin{array}[]{lllllrlllr}-\Delta^{a,b}_{p}u=\lambda|u|^{p-2}u&{\rm in}&(0,L),\\ u(0)=u(L)=0\end{array}\right. (1) and \left\{\begin{array}[]{lllllrlllr}-\Delta^{a,b}_{p}u=\mu(u^{+})^{p-1}-\nu(u^{-})^{p-1}&{\rm in}&(0,L),\\ u(0)=u(L)=0,\end{array}\right. (2) where \lambda,\mu,\nu\in\mathbb{R} and u^{\pm}=\max\{\pm u,0\} . Corresponding to ( 1 ) and ( 2 ), we have the respective Dirichlet and Fučík spectra: \displaystyle\Lambda^{a,b}_{p}(0,L):=\{\lambda\in\mathbb{R}:\ \eqref{DS}\ \text{admits a nontrivial weak solution}\,u\in W_{0}^{1,p}(0,L)\}, \displaystyle\Sigma^{a,b}_{p}(0,L):=\{(\mu,\nu)\in\mathbb{R}^{2}:\ \eqref{FS}\ \text{admits a nontrivial weak solution}\,u\in W_{0}^{1,p}(0,L)\}. The elements of \Lambda^{a,b}_{p}(0,L) are called Dirichlet eigenvalues of -\Delta^{a,b}_{p} . When a=b , the operator \Delta^{a,a}_{p} coincides with a^{p}\Delta_{p} , where \Delta_{p}u:=\left(|u^{\prime}|^{p-2}u^{\prime}\right)^{\prime} denotes the one-dimensional isotropic p -Laplace operator. Notice also that both sets \Lambda_{p}(0,L):=a^{-p}\,\Lambda^{a,a}_{p}(0,L) and \Sigma_{p}(0,L):=a^{-p}\,\Sigma^{a,a}_{p}(0,L) are entirely known. Precisely, they are given by \displaystyle\bullet\ \ \Lambda_{p}(0,L)=\{\lambda_{k,p}(0,L):\,k\geq 1\}, \displaystyle\bullet\ \ \Sigma_{p}(0,L)=\left(\{\lambda_{1,p}(0,L)\}\times\mathbb{R}\right)\cup\left(\mathbb{R}\times\{\lambda_{1,p}(0,L)\}\right)\cup\Sigma^{e}_{p}(0,L)\cup\Sigma^{o,1}_{p}(0,L)\cup\Sigma^{o,2}_{p}(0,L), where \lambda_{k,p}(0,L)=(p-1)\left(\frac{k\pi_{p}}{L}\right)^{p}=k^{p}\lambda_{1,p}(0,L) with \pi_{p}:=2\int_{0}^{1}(1-t^{p})^{-1/p}dt and \displaystyle\Sigma^{e}_{p}(0,L):=\{(\mu,\nu)\in\mathbb{R}_{+}^{2}:\,\mu^{-1/p}+\nu^{-1/p}=2\lambda_{2k,p}(0,L)^{-1/p},\ k\geq 1\}, \displaystyle\Sigma^{o,1}_{p}(0,L):=\{(\mu,\nu)\in\mathbb{R}_{+}^{2}:\,k\mu^{-1/p}+(k+1)\nu^{-1/p}=(2k+1)\lambda_{2k+1,p}(0,L)^{-1/p},\ k\geq 1\}, \displaystyle\Sigma^{o,2}_{p}(0,L):=\{(\mu,\nu)\in\mathbb{R}_{+}^{2}:\,(k+1)\mu^{-1/p}+k\nu^{-1/p}=(2k+1)\lambda_{2k+1,p}(0,L)^{-1/p},\ k\geq 1\}. For the one-dimensional Dirichlet spectrum of -\Delta_{p} we refer to [ 14 ] and for the associated Fučík spectrum to [ 10 , 21 ] for p=2 and to [ 15 ] for p\neq 2 . As it is well known, there is a vast literature on miscellaneous problems connected to ( 1 ) and ( 2 ) in the case a=b , both in dimension 1 and higher dimensions. Most contributions can be found in a long list of old and new references. For some more closely related works, we refer to [ 2 , 3 , 4 , 5 , 6 , 8 , 9 , 11 , 12 , 13 , 16 , 17 , 18 , 19 , 20 , 22 , 24 , 25 , 26 , 27 , 28 , 29 , 30 ] . On the other hand, fairly little is known about spectra of -\Delta^{a,b}_{p} when a\neq b , even for p=2 in which \Delta^{a,b}_{2} is a nonlinear elliptic operator. Our main goal here is to determine explicitly both spectral sets as well as to characterize all corresponding solutions for any p>1 and a,b>0 . In particular, we establish an one-to-one correspondence between the structures of the anisotropic and isotropic spectra, showing that many spectral properties are not affected by the asymmetry of the operator \Delta^{a,b}_{p} . The bridge between the asymmetric and symmetric settings will be constructed by means of two fundamental ingredients: (i) The explicit value of the first Dirichlet eigenvalue of -\Delta^{a,b}_{p} as well as the classification of its eigenfunctions (Theorem 1.1 ); (ii) The simplicity of the zeroes of any nontrivial weak solution of ( 2 ) (Theorem 3.1 ). Due to the presence of asymmetry, the proofs of (i) and (ii) demand alternative arguments adapted to the current context. Particularly, the approach of (i) involves the construction of some appropriate glues of modified eigenfunctions and the recurrent use of maximum type principles. In addition, it is worth mentioning that the explicit knowledge of the first Dirichlet eigenvalue in the isotropic case, as well as of the key properties satisfied by its eigenfunctions, plays an essential role in the proof of the ingredient (i). A well-known property of the first Dirichlet eigenvalue \lambda_{1,p}(0,L) of -\Delta_{p} is its simplicity, that is, the set of all eigenfunctions associated to \lambda_{1,p}(0,L) is generated by a positive eigenfunction \varphi_{p} : \{t\varphi_{p}:\,t\neq 0,\ \varphi_{p}\in C^{1}[0,L]\ \ {\rm and}\ \ \varphi_{p}>0\ {\rm in}\ (0,L)\}. Note that this set is the union of two collinear half-lines in C^{1}[0,L] . Within the anisotropic environment, the appropriate notion of simplicity for Dirichlet eigenvalues of the operator -\Delta^{a,b}_{p} is as follows: Definition 1.1 . A Dirichlet eigenvalue of -\Delta^{a,b}_{p} is said to be simple in the generalized sense when a\neq b , if the set of corresponding eigenfunctions is the union of two non-collinear half-lines in C^{1}[0,L] , one generated by a positive function and other one by a negative function in (0,L) . For each p>1 and a,b,L>0 , consider the number \lambda^{a,b}_{1,p}(0,L):=\inf\left\{\int_{0}^{L}a^{p}[(u^{\prime}(t))^{+}]^{p}+b^{p}[(u^{\prime}(t))^{-}]^{p}\,dt:\ u\in W_{0}^{1,p}(0,L),\ \|u\|_{p}=1\right\}. From the fact that \lambda_{1,p}(0,L) is positive, it is clear that \lambda^{a,b}_{1,p}(0,L) is also positive and, in addition, it follows by direct minimization, with the aid of the Poincaré inequality, that the infimum is attained for some u\in W_{0}^{1,p}(0,L) . In particular, \lambda^{a,b}_{1,p}(0,L) is the smallest Dirichlet eigenvalue of -\Delta^{a,b}_{p} and u is a corresponding eigenfunction. Furthermore, by definition, \lambda^{a,b}_{1,p}(0,L) is the best constant of the asymmetric Poincaré inequality \lambda^{a,b}_{1,p}(0,L)\int_{0}^{L}|u(t)|^{p}\,dt\leq\int_{0}^{L}a^{p}[(u^{\prime}(t))^{+}]^{p}+b^{p}[(u^{\prime}(t))^{-}]^{p}\,dt for every u\in W_{0}^{1,p}(0,L) . Under this point of view, the heuristic intuition based on similarity of roles played by the terms (u^{\prime})^{+} and (u^{\prime})^{+} perhaps would suggest that the value of \lambda^{a,b}_{1,p}(0,L) should be \displaystyle\left(\frac{a^{p}+b^{p}}{2}\right)\lambda_{1,p}(0,L). However, its computation does not seem trivial by far and, to our surprise, it is strictly smaller than the one mentioned above. Part of our first theorem exhibits the precise value of \lambda^{a,b}_{1,p}(0,L) . Theorem 1.1 . Let p>1 and L>0 . For any a,b>0 , we have: (I) \lambda_{1,p}^{a,b}(0,L)=\displaystyle\left(\frac{a+b}{2}\right)^{p}\lambda_{1,p}(0,L) ; (II) u\in W_{0}^{1,p}(0,L) is an eigenfunction of -\Delta^{a,b}_{p} associated to \lambda^{a,b}_{1,p}(0,L) if and only if either u(t)=c\varphi_{p,L}^{a,b,+}(t):=c\varphi_{p}^{a,b,+}\left(\frac{t}{L}\right)\ \ {\rm or}\ \ u(t)=c\varphi_{p,L}^{a,b,-}(t):=c\varphi_{p}^{a,b,-}\left(\frac{t}{L}\right) for some constant c>0 , where \varphi_{p}^{a,b,+}(t)=\begin{cases}\begin{aligned} &\varphi_{p}\left(\frac{t}{2t_{0}}\right)\ \ \text{if}\ t\in(0,t_{0}),\\ &\varphi_{p}\left(\frac{t+1-2t_{0}}{2(1-t_{0})}\right)\ \ \text{if}\ t\in[t_{0},1),\end{aligned}\end{cases} \varphi_{p}^{a,b,-}(t)=\begin{cases}\begin{aligned} &-\varphi_{p}\left(\frac{t}{2(1-t_{0})}\right)\ \ \text{if}\ t\in(0,1-t_{0}),\\ &-\varphi_{p}\left(\frac{t+2t_{0}-1}{2t_{0}}\right)\ \ \text{if}\ t\in[1-t_{0},1).\end{aligned}\end{cases} Here, \varphi_{p}\in W^{1,p}_{0}(0,1) denotes the principal eigenfunction associated to \lambda_{1,p}(0,1) normalized by \|\varphi_{p}\|_{p}=1 and t_{0}=\frac{a}{a+b}. In particular, \lambda_{1,p}^{a,b}(0,L) is an simple eigenvalue in the generalized sense, in other words, the eigenspace of \lambda^{a,b}_{1,p}(0,L) is formed by two non-collinear half-lines if and only if a\neq b . This result will play an important role in the complete characterization of the Dirichlet and Fučík spectra of -\Delta^{a,b}_{p} and corresponding solutions for any a,b>0 . It also implies the following one-sided Poincaré inequalities by letting a\rightarrow 0 with b=1 (or