Quantitative limiting absorption principle in the semiclassical limit (Q2510438): Difference between revisions
From MaRDI portal
ReferenceBot (talk | contribs) Changed an Item |
Normalize DOI. |
||
Property / DOI | |||
Property / DOI: 10.1007/s00039-014-0273-8 / rank | |||
Property / DOI | |||
Property / DOI: 10.1007/S00039-014-0273-8 / rank | |||
Normal rank |
Latest revision as of 03:09, 19 December 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Quantitative limiting absorption principle in the semiclassical limit |
scientific article |
Statements
Quantitative limiting absorption principle in the semiclassical limit (English)
0 references
1 August 2014
0 references
This work gives an elementary proof of Burq's resolvent bounds for the following range semiclassical Schrö\-din\-ger operators \(P_h\). Let \(\Delta\leq 0\) be the Laplacian on \(\mathbb R^n\), \(n\not= 2\), and let \(E>0\). Then \[ P=P_h:=-h^2\Delta+ V-E,\qquad h>0, \eqno(1.1) \] where, using polar coordinates \((r,\omega)\in (0,\infty) \times S^{n-1}\), we suppose that \(V=V_h(r,\omega)\) and its distributional derivative \(\partial_r V\) are in \(L^{\infty}((0,\infty)\times S^{n-1})\). We suppose further that \[ V\leq (1+r)^{-\delta_0},\qquad \partial_rV\leq (1+r)^{-1-\delta_0}, \eqno(1.2) \] for some \(\delta_0>0\). The resolvent \((P-i\varepsilon)^{-1}\) is defined \(L^2(\mathbb R^n)\to H^2(\mathbb R^n)\) for \(\varepsilon >0\) by the Kato-Rellich theorem. The author proves the following weighted resolvent bounds: Theorem. For any \(s>1/2\) there are \(C, R_0, h_0>0\) such that \[ \| (1+r)^{-s} (P-i\varepsilon)^{-1} (1+r)^{-s}\|_{L^2(\mathbb R^n)\to L^2(\mathbb R^n)}\leq e^{C/h}, \eqno(1.3) \] \[ \| (1+r)^{-s} \mathbf{1}_{\geq R_0}(P-i\varepsilon)^{-1}\text\textbf{1}_{\geq R_0} (1+r)^{-s}\|_{L^2(\mathbb R^n)\to L^2(\mathbb R^n)}\leq C/h, \eqno(1.4) \] for all \(\varepsilon>0, h\in(0,h_0]\), where \textbf{1}\(_{\geq R_0}\) is the characteristic function of \(\{ x\in \mathbb R^n: |x|\geq R_0\}\). The proof of this theorem is based on a new global Carleman estimate of the form \[ \| (1+r)^{-s} e^{\varphi/h} v\|^2_{L^2(\mathbb R^n)}\leq \frac{C}{h^2} \| (1+r)^{s} e^{\varphi/h} (P-i\varepsilon)v\|^2_{L^2(\mathbb R^n)}+ \frac{C\varepsilon}{h}\| e^{\varphi/h}v\|^2_{L^2(\mathbb R^n)}, \] with \(C\) independent of the support of \(v\), and with \(\varphi=\varphi(r)\) nondecreasing and constant outside of a compact set.
0 references
Burq's resolvent bounds
0 references
Schrödinger operator
0 references
Carleman estimate
0 references
0 references
0 references