Progress in Nonlinear Differential Equations and Their Applications

Volume 82
Editor Haim Brezis Universit Pierre et Marie Curie Paris and Rutgers University New Brunswick, N.J. Editorial Board Antonio Ambrosetti, Scuola Internationale Superiore di Studi Avanzati, Trieste A. Bahri, Rutgers University, New Brunswick Felix Browder, Rutgers University, New Brunswick Luis Caffarelli, The University of Texas, Austin Lawrence C. Evans, University of California, Berkeley Mariano Giaquinta, University of Pisa David Kinderlehrer, Carnegie-Mellon University, Pittsburgh Sergiu Klainerman, Princeton University Robert Kohn, New York University P. L. Lions, University of Paris IX Jean Mawhin, Universite Catholique de Louvain Louis Nirenberg, New York University Lambertus Peletier, University of Leiden Paul Rabinowitz, University of Wisconsin, Madison John Toland, University of Bath

For other titles published in this series, go to http://www.springer.com/series/4889

Antonio Ambrosetti David Arcoya

An Introduction to Nonlinear Functional Analysis and Elliptic Problems

Birkhuser Boston Basel Berlin

Antonio Ambrosetti SISSA-International School for Advanced Studies 34136 Trieste Italy ambr@sissa.it

David Arcoya Departamento de Anlisis Matemtico Universidad de Granada 18071 Granada Spain darcoya@ugr.es

ISBN 978-0-8176-8113-5 e-ISBN 978-0-8176-8114-2 DOI 10.1007/978-0-8176-8114-2 Springer New York Dordrecht Heidelberg London
Library of Congress Control Number: 2011930398 Mathematics Subject Classication (2011): 35-01, 46-01, 49-01 Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Birkhuser Boston, c/o Springer Science+Business Media, LLC, 233 Spring Street, NewYork, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identied as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Birkhuser is part of Springer Science+Business Media (www.springer.com)

Preface

The purpose of this book is to introduce the reader to some of the main abstract tools of nonlinear functional analysis and their applications to semilinear elliptic Dirichlet boundary value problems. In the rst chapter we outline some general results on Frchet differentiability, Nemitski operators, weak and strong solutions of the linear Laplace equation, linear compact operators and their eigenvalues and Sobolev spaces. This last topic is discussed in greater generality in Appendix A. Chapter 2 deals with the Banach contraction principle and with a xed point theorem for increasing operators. In Chap. 3 we study the local inversion theorem, the HadamardCaccioppoli global inversion theorem and the case in which the map to be inverted has fold singularities. Chapter 4 is concerned with the LeraySchauder topological degree. Variational methods are discussed in Chap. 5. Minima, the mountain pass theorem and the linking theorem are stated and proved. Chapter 6 deals with local and global bifurcation theory. The abstract results collected in rst part of the book are applied in the second part to prove existence and multiplicity results for semilinear elliptic Dirichlet boundary value problems on bounded domains in RN . We emphasize that the choice of the appropriate abstract tool depends on the behavior of the nonlinearity f as well as on the kind of results one expects. First, in Chap. 7, we outline how a semilinear elliptic boundary value problem can be transformed into an operator equation in an appropriate Banach or Hilbert function space. In Chap. 8 we consider the case in which, roughly, f is sublinear at innity and one can prove a priori estimates for possible solutions. In this case one can use degree theory or variational methods or the global inversion theorem. Chapter 9 deals with asymptotically linear problems, for which one can also use several different approaches such as global bifurcation or variational methods. In Chap. 10 we study problems with asymmetric nonlinearities, when the behaviors at + and are different. If one aims to nd the precise number of multiple solutions, the most appropriate approach turns out to be the global inversion theorem in the presence of fold singularities. But one can also use sub- and super-solutions jointly with degree theoretical arguments. Nonlinearities that are superlinear at innity are considered in Chap. 11 by means of the mountain pass or linking theorems.
v

vi

Preface

In all of the preceding chapters we do not consider more general sophisticated versions of problems, but prefer to study model cases containing the main features of the arguments without unnecessary technical details. The last two chapters of the book are concerned with slightly more advanced topics of current research. In Chap. 12 a class of quasilinear elliptic problems is discussed using critical point theory. Here the corresponding Euler functional is not C 1 , and hence a new form of the mountain pass theorem has to be proved. Chapter 13 deals with nonlinear Schrdinger equations on R N . We prove the existence of ground and bound states as well as semiclassical states. The book is addressed to senior undergraduate and graduate students of mathematics as well as to students of applied sciences, who wish to utilize a modern approach to the fascinating topic of nonlinear elliptic partial differential equations.

Contents

Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Sobolev Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1.1 Embedding Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Linear Elliptic Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.1 Frchet Differentiability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.2 Nemitski Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.3 Dirichlet Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.4 Regularity of the Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.5 The Inverse of the Laplace Operator . . . . . . . . . . . . . . . . . . . . . 1.3 Linear Elliptic Eigenvalue Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.1 Linear Compact Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.2 Variational Characterization of The Eigenvalues . . . . . . . . . . . Some Fixed Point Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 The Banach Contraction Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Increasing Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Local and Global Inversion Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 The Local Inversion Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 The Implicit Function Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 The LyapunovSchmidt Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 The Global Inversion Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 A Global Inversion Theorem with Singularities . . . . . . . . . . . . . . . . . . LeraySchauder Topological Degree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 The Brouwer Degree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 The LeraySchauder Topological Degree . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Index of an Isolated Zero and Computation by Linearization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 3 4 4 5 6 10 10 10 11 13 17 17 20 23 23 25 26 27 28 33 33 36 39

vii

viii

Contents

4.3 Continuation Theorem of LeraySchauder . . . . . . . . . . . . . . . . . . . . . . 4.3.1 A Topological Lemma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2 A Theorem by Leray and Schauder . . . . . . . . . . . . . . . . . . . . . . 4.4 Other Continuation Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 An Outline of Critical Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Denitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Minima . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 The Mountain Pass Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 The Ekeland Variational Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5 Another MinMax Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6 Some Perturbation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Bifurcation Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Local Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1.1 Bifurcation from a Simple Eigenvalue . . . . . . . . . . . . . . . . . . . 6.1.2 Bifurcation from an Odd Eigenvalue . . . . . . . . . . . . . . . . . . . . 6.2 Bifurcation for Variational Operators . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.1 A Krasnoselskii Theorem for Variational Operators . . . . . . . . 6.2.2 Branching Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Global Bifurcation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Elliptic Problems and Functional Analysis . . . . . . . . . . . . . . . . . . . . . . . . 7.1 Nonlinear Elliptic Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.1 Classical Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.2 Weak Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Sub- and Super-Solutions and Increasing Operators . . . . . . . . . . . . . . Problems with A Priori Bounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1 An Elementary Nonexistence Result . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 Existence of A Priori Bounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3 Existence of Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.1 Using the Global Inversion Theorem . . . . . . . . . . . . . . . . . . . . 8.3.2 Using Degree Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.3 Using Critical Point Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4 Positive Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

40 40 42 44 47 47 48 48 53 54 56 61 61 62 63 64 65 68 70 73 73 74 74 76 83 83 84 85 85 86 87 90 97 97 99 101 104 105 107

9 Asymptotically Linear Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1 Existence of Positive Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 Bifurcation from Innity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3 On the Behavior of the Bifurcations from Innity . . . . . . . . . . . . . . . . 9.4 The Local Anti-Maximum Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5 The LandesmanLazer Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5.1 A Variational Proof of the LandesmanLazer Result . . . . . . .

Contents

ix

10 Asymmetric Nonlinearities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1 The Approach by Ambrosetti and Prodi . . . . . . . . . . . . . . . . . . . . . . . . 10.2 The Approach by AmannHess . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3 Variational Approach by Mountain Pass and Suband Super-Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.4 Approach by Degree Giving a Continuum of Solutions . . . . . . . . . . .

111 111 113 117 119

11 Superlinear Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 11.1 Using MinMax Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 11.2 Superlinear AmbrosettiProdi Problem . . . . . . . . . . . . . . . . . . . . . . . . . 124 12 Quasilinear Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.1 First Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.2 Mountain Pass Theorem for Nondifferentiable Functionals and Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.3 Application to Quasilinear Variational Problems . . . . . . . . . . . . . . . . . 12.4 Some Nonvariational Quasilinear Problems . . . . . . . . . . . . . . . . . . . . . 13 Stationary States of Evolution Equations . . . . . . . . . . . . . . . . . . . . . . . . . 13.1 Soliton States to Stationary NLS Equations . . . . . . . . . . . . . . . . . . . . . 13.2 Semiclassical States of NLS Equations with Potentials . . . . . . . . . . . . 13.3 Systems of NLS Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4 Nonautonomous Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.5 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix A: Sobolev Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.1 Weak Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.2 Sobolev Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.3 Boundary Values in Sobolev Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . A.4 Embedding Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 131 133 136 142 149 149 155 158 163 167 169 169 175 177 179

Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

Notation

For every s R we consider the positive and negative parts given by s + = max{s , 0} and s = min{s , 0}. C , C1 , C2 , . . . denote possibly different positive constants. If is a measurable set in RN , we denote by | | the Lebesgue measure of and by u the integral of a function u in . Hence, unless it is explicitly stated, the integrals are always understood to be on . denotes that is compactly embedded in , that is, the closure of is a compact subset of . If is an open set in RN and is a multi-index, namely = (1 , 2 , . . . , N ), with | | u i a non-negative integer, we denote by D u the partial derivative x 1 x , 2 ...x N
N where | | = i =1 i is the order of . The set of all innitely-differentiable functions of compact support in is represented by C0 ( ). For a non-negative integer k and 0 < 1, we denote by C k, ( ) the space of the functions whose derivatives up to order k are -Hlder continuous in . In 1 particular, we write C k ( ) if = 0. Moreover, C0 ( ) is the space of all functions 1 of class C in an open neighborhood of such that they vanish at the boundary of . For 1 p +, u p is the usual norm of a function u Lp ( ). 1 We have equipped the standard Sobolev space H0 ( ) with the norm u = 1/2
1 2 N

| u|2

. 2N N 2 , if N 3 +, if N 2

We also denote by 2 =

the critical Sobolev exponent and by S = sup{ u 2 : u = 1} the Sobolev embedding constant. 2 = 2N/(N + 2) is the Hlder conjugate exponent of 2 . The truncature functions Tk and Gk are given by Tk (s ) = max{min{s , k }, k } and Gk (s ) = s Tk (s ), for every s R.
xi

xii

Notation

We denote a Banach (resp. Hilbert) space with the letter X (resp. E ). The identity operator is denoted by I . The functionals, i.e., (nonlinear) operators from a Banach space X to R, are denoted by letters J , H, I , . . . In general, the operators between different Banach spaces X and Y are denoted by letters F , G, . . . , while letters T , S , . . . are used for operators from a Banach space into itself. The weak convergence of a sequence wn in a Banach space to w will be denoted wn w. If F : X Y is an operator between Banach spaces, we denote Ker F = {u X : F (u) = 0} and Range F = {F (x ) : x X }. 0 < 1 < 2 3 , . . . denote the eigenvalues of u = u, u E and i satises i = i i with i = 1 and (i | j ) = 0 for i = j . We take 1 > 0.