A Note on a Mountain Pass Characterization of Least Energy Solutions (Q4454023)

This paper is a continuation of the study undertaken recently by the authors [Proc. Am. Math. Soc. 131, 2399--2408 (2003; Zbl 1094.35049)] for the nonlinear scalar field equations \[ -\Delta u=g(u),\qquad u(x)\in H^{1}( \mathbb R^N) ,\tag{1} \] in the case where \(N\geq2.\) Assuming now that \(N=1,\) under rather general conditions on the nonlinearity \(g,\) it is established that (a) Eq. (1) has a unique, up to translation, positive solution; (b) all nontrivial solutions of (1) are least energy solutions; (c) the natural functional, \[ I(u)=\frac{1}{2}\int_{\mathbb R^n}| \nabla u| ^{2}\,dx-\int_{\mathbb R^n}\left( \int_{0}^{u}g(\tau)\,d\tau\right) dx, \] corresponding to (1), has a mountain pass geometry. The conservation of total energy, which is a consequence of the fact that (1) is autonomous, is crucial for the proof of the main result. The authors concentrate their attention mainly on a mountain pass characterization of least energy solutions and not on the existence and uniqueness of solutions. For instance, they define a functional, \[ I(q)=\int_{-\infty}^{+\infty}\frac{1}{2}| \dot{q}| ^{2}-V(q)\,dt, \] which has a mountain pass geometry and explain how it can be used for finding homoclinic solutions of a certain Hamiltonian system by exploiting variational methods. The authors also provide an example, based on the idea from singular Hamiltonian systems, showing that the mountain pass characterization of least energy solutions is not true for systems.