A global compactness result and multiplicity of solutions for a class of critical exponent problems in the hyperbolic space

Abstract

This paper deals with the global compactness and multiplicity of positive solutions to problems of the type-Delta(N)(B) u-lambda u = a(x)|u|(2*-2)u + f(x) in B-N, u is an element of H-1(B-N),where B-N denotes the ball model of the hyperbolic space of dimension N >= 4, 2* = 2N/ N-2, N(N-2)/4 < lambda < (N-1)(2)/4 and f is an element of H-1(B-N) ( f not equivalent to 0 ) is a non-negative functional in the dual space of H-1(B-N). The potential a is an element of L-infinity(B-N) is assumed to be strictly positive, such that lim(d(x,0)->infinity) a(x)=1, where d(x,0) denotes the geodesic distance. We establish profile decomposition of the associated functional. We show that concentration takes place along two different profiles, namely along hyperbolic bubbles and localized Aubin-Talenti bubbles. For f=0 and a equivalent to 1, profile decomposition was studied by Bhakta and Sandeep [Calc. Var. PDE, 2012]. However, due to the presence of a(.), an extension of profile decomposition to the present set-up is highly nontrivial and requires several delicate estimates and geometric arguments concerning the isometry group (Mobius group) of the hyperbolic space. Further, using the decomposition result, we derive various energy estimates involving the interacting hyperbolic bubbles and hyperbolic bubbles with localized Aubin-Talenti bubbles. Finally, combining these estimates with topological and variational arguments, we establish a multiplicity of positive solutions in the cases: a >= 1 and a<1 separately. The equation studied in this article can be thought of as a variant of a scalar-field equation with a critical exponent in the hyperbolic space, although such a critical exponent problem in the Euclidean space RN has only a trivial solution when f equivalent to 0, a(x)equivalent to 1 and lambda<0.