This paper presents uniqueness theorems for positive solutions of the semilinear second order elliptic boundary value problem (1) below with a nonnegative nonlinearity F(x, zu) which is monotonic and sublinear (Eq. (2)) in w. The proofs of the theorems use the methods developed by Krasnosel’skii [l, Sections 6.1 and 7.2.61 and Urysohn [2] f or nonlinear operators together with the maximum principle of Hopf [3] f or elliptic difkrential operators, Since use is made of the maximum principle, we consider only classical (positive) solutions of the boundary value probiem, and the only smoothness assumptions necessary for the given functions appearing in the probIem are that the coefhcients of the differential operator be bounded in the domain J2 in which the solution is sought. In order to apply the techniques of Urysohn and Krasnosel’skii, however, it is necessary to make a careful study of the boundary of !ZJ and the behavior of the solutions near the boundary (cf. Theorem 2). The uniqueness theorem states that under certain smoothness conditions (see especially Corollary 2.2), (1) h as either at most 01x positive solution or an infinite number of positive solutions, each of which is a scalar multiple of any other. For domains whose boundaries satisfy the conditions imposed below, our results include the uniqueness results of Keller [4, 5, 61, Kdler and Cohen [7], Cohen [S], and Shampine [9], for self-adjoint elliptic operators. In contmst to the results just cited, our method of proof does not require one to assume the existence of solutions (eigenvalues and eigenfunctions) for associated linear problems or the differentiability of the nonlinearityfi: The technique used here has been applied to a closely related problem in [lo]. For ordinary differential equations, the uniqueness of the positive solutions for sublinear ~nonlinearities is proven by Pimbley [l l] (see also [12] and [l, Section 7.4.41). The existence
Read full abstract