An annulus multiplier and applications to the limiting absorption principle for Helmholtz equations with a step potential (Q2225633)

From MaRDI portal





scientific article
Language Label Description Also known as
default for all languages
No label defined
    English
    An annulus multiplier and applications to the limiting absorption principle for Helmholtz equations with a step potential
    scientific article

      Statements

      An annulus multiplier and applications to the limiting absorption principle for Helmholtz equations with a step potential (English)
      0 references
      0 references
      0 references
      8 February 2021
      0 references
      The authors are interested in the equation \[ -\Delta u + Vu - \lambda u = f \tag{1}\] in \(\mathbb{R}^n\), \(n\geq 2\), as well as a nonlinear pendant, where \(V\) is assumed throughout to be a model step potential taking the constant value \(V_1\) in the upper half-space and \(V_2 < V_1\) in the lower half space. Denote by \(\mathcal{R}(\mu) := (-\Delta + V - \mu)^{-1}\) the resolvent operator associated with (1), \(\mu \in \mathbb{C} \setminus \sigma (-\Delta + V)\). The main result (Theorem 1) is the \(L^p\)--\(L^q\) resolvent estimate \[ \sup_{0 < |\varepsilon| \leq 1} \|\mathcal{R}(\lambda + i\varepsilon)\|_{L^p(\mathbb{R}^n) \to L^q(\mathbb{R}^n)} < \infty \] for all \(\lambda > V_1\) and all pairs \((p,q)\) in a somewhat involved region related to, and almost as large as, the corresponding region in the corresponding result for constant potential \(V \in \mathbb{R}\) (the precise region here is also related to \textit{T. Tao}'s restriction problem in Harmonic Analysis, cf. [in: Fourier analysis and convexity. Boston, MA: Birkhäuser. 217--243 (2004; Zbl 1083.42008)]). In this case, the authors also obtain convergence, as \(\varepsilon \to 0\), of \(\mathcal{R} (\lambda + i\varepsilon)\) to \(\mathcal{R} (\lambda \pm i0)\) in the weak topology of bounded linear operators from \(L^p (\mathbb{R}^n)\) into \(L^q (\mathbb{R}^n)\). Two other results are obtained as corollaries: firstly, for the problem (1), with \(f \in L^p(\mathbb{R}^n)\), for the ``outgoing'' and ``incoming'' solutions \(u_+ := \mathbb{R} (\lambda + i0)f\) and \(u_- := \mathbb{R} (\lambda - i0)f\), respectively, under the same conditions as the main theorem we have \[ u_\pm \in L^q (\mathbb{R}^n) \cap W^{2,p}_{\textrm{loc}}(\mathbb{R}^n) \] together with the estimate \[ \|u_+\|_{L^q(\mathbb{R}^n)} + \|u_-\|_{L^q(\mathbb{R}^n)} \leq C(V_1,V_2,n,p,q,\lambda) \|f\|_{L^p (\mathbb{R})}. \] The second application is to prove the existence of a nontrivial solution \(u \in L^q (\mathbb{R}^n) \cap W^{2,r}_{\textrm{loc}}(\mathbb{R}^n)\) for all \(r<\infty\), of the nonlinear equation \[ -\Delta u + Vu - \lambda u = \Gamma |u|^{q-2}u \] in \(\mathbb{R}^n\), \(n\geq 2\), where \(\Gamma\) is assumed to be a strictly positive \(L^\infty\) function decaying at \(\infty\), and \(\frac{2(n+1)}{n-1} \leq q < \frac{2n}{n-2}\). The proof of the main theorem is based on Fourier restriction theory as well as the analysis of, including obtaining \(L^p\)--\(L^q\) estimates for, a certain singular Fourier multiplier supported in an annular region (Theorems 2 and 3).
      0 references
      Helmholtz equation
      0 references
      resolvent estimates
      0 references
      limiting absorption principle
      0 references
      restriction conjecture
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references