Analytic Trajectories for Mobility Edges in the Anderson Model
Wolfram T. Arnold, Roger Haydock

TL;DR
This paper introduces a new basis for analyzing the Anderson model, enabling explicit calculation of mobility edges as a function of disorder, revealing novel insights into localization transitions in disordered systems.
Contribution
It develops an analytic basis of distorted Bloch waves for the Anderson model, allowing clear identification of mobility edges and revealing new types of localization transitions.
Findings
Mobility edges are explicitly calculated as a function of disorder.
In two dimensions, mobility edges are found analytically and numerically, unlike previous perturbation approaches.
In three dimensions, the behavior of mobility edges differs from numerical scaling results.
Abstract
A basis of Bloch waves, distorted locally by the random potential, is introduced for electrons in the Anderson model. Matrix elements of the Hamiltonian between these distorted waves are averages over infinite numbers of independent site-energies, and so take definite values rather than distributions of values. The transformed Hamiltonian is ordered, and may be interpreted as an itinerant electron interacting with a spin on each site. In this new basis, the distinction between extended and localized states is clear, and edges of the bands of extended states, the mobility edges, are calculated as a function of disorder. In two dimensions these edges have been found in both analytic and numerical applications of tridiagonalization, but they have not been found in analytic approaches based on perturbation theory, or the single-parameter scaling hypothesis; nor have they been detected in…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
