Zentralblatt MATH - REVIEW SUBMISSION FORM |
Zentralblatt MATH
HOME
|
Name: | ||
Miloslav Znojil | ||
Reviewer number: | ||
9689 | ||
Email: | ||
znojil@ujf.cas.cz | ||
Item's zbl-Number: | ||
DE 018 207 684 | ||
Author(s): | ||
Remling, Christian: | ||
Shorttitle: | ||
Schroedinger operators with decaying potentials: some counterexamples | ||
Source: | ||
Duke Math. J. 105, No. 3, 463 - 496 (2000) | ||
Classification: | ||
| ||
Primary Classification: | ||
| ||
Secondary Classification: | ||
| ||
Keywords: | ||
Schroedinger operator on half-line; slowly decaying potentials; absolutely continuous spectrum; embedded singular spectrum; Hausdorff dimension of the exceptional set; contructive proof of optimality of its estimate; sum rules for asymptoticaly Coulomb potentials | ||
Review: | ||
Schropedinger equation is studied on half-axis, with an asymptoticaly slowly decreasing potential V(x) which is asymptotically bounded by a negative power (between minus one and - not included - minus one half) of the coordinate. A mixed boundary condition in the origin is admitted. A review is offered of the results concerning the essential spectrum (on the positive energy half-line) and its possible (i.e., point or absolutely continuous or singular continuous) more detailed character. The purpose is to construct examples which show that the results available in the literature are, in fact, optimal. An extremely well readable presentation of rigorous results (well motivated and complemented by heuristic arguments, e.g., in secton 3) constructs its``counterexample" sample potentials, mostly, by pasting together carefully chosen periodic pieces. Then, the bound on the embedded singular spectrum can only be show optimal (in Theorem 1.3) for the asymptotic decay of V(x) betwen two thirds and one, which is a not entirely pleasant natural limitation of the constructive approach. Its strength, on the other hand, manifests itself at the Coulombic end (especially via sum rules of Theorem 1.5 and its Corollary 1.6). The proofs pay due attention to the effects of pasting (non-smoothness) and make ample use of periodicity of the pasted pieces, say, for related asymptotic estimates of the so called Ljapunov function (= trace of the transfer matrix), with considerations closely related to some recent Kiselev's results [1]. | ||
Remarks to the editors: | ||