Boundedness and monotonicity of principal eigenvalues for boundary value problems with indefinite weight functions英文文献资料.docVIP

PII.S0161171202007780 BOUNDEDNESSANDMONOTONICITYOFPRINCIPAL EIGENVALUESFORBOUNDARYVALUEPROBLEMS WITHINDEFINITEWEIGHTFUNCTIONS G.A.AFROUZI Received 24June 2001 We study the principal eigenvalues (i.e., eigenvalues corresponding topositive eigenfunc- tions) for the boundary value problem: ??u(x) = λg(x)u(x), x ∈ D; (?u/?n)(x) + αu(x) =0, x ∈?D, where ? is the standard Laplace operator, D is a bounded domain with smooth boundary, g : D → R is a smooth function which changes sign on D and α∈R.Wediscuss therelation between αand theprincipal eigenvalues. 2000 Mathematics Subject Classi?cation: 35J15, 35J25. 1. Introduction. Weinvestigatethepropertyofprincipaleigenvaluesforthebound- aryvalueproblem ??u(x)=λg(x)u(x), x∈D, ?u?n(x)+αu(x)=0, x∈?D, (1.1) where D is a bounded region in RN with smooth boundary, g :D →R is a smooth function whichchangessignonDandα∈R. Suchproblems havebeenstudiedinrecentyearssinceFleming[4]studiedthefol- lowingassociated nonlinear problems arisinginthestudyofpopulation genetics: ut(x,t)=?u+λg(x)f(u), x∈D, (1.2) wheref issomefunction ofclassC1 suchthatf(0)=0=f(1). Fleming’s results suggested thatnontrivial steady-state solutions werebifurcating the trivial solutions u≡0 and u≡1. In order to investigate these bifurcation phe- nomena, it was necessary to understand the eigenvalues and eigenfunctions of the corresponding linearized problem ??u(x)=λg(x)u(x), x∈D. (1.3) The study of the linear ordinary di?erential equation case, however, goes back to Bocher [3]. Attention has been con?ned mainly tothe cases ofDirichlet (α=∞ )and Neumann boundary conditions. In the case of Dirichlet boundary conditions, it is well known (see [5]) that there existsadoublesequence ofeigenvalues for(1.1) ···λ?2 λ+1 0λ1+λ+2 ···, (1.4)  26 G. A. AFROUZI λ+1(λ?1) being the unique positive (negative) principal eigenvalue, that is, (1.1) has solutionu(v) whichispositiveinD.Itisalsowellknownthatthecasewhere0α ∞issimilartotheDirichlet case. In the case