Dependence of the density of states outer measure on the potential for deterministic Schr\"odinger operators on graphs with applications to ergodic and random models

TL;DR

This paper investigates how the density of states measure and related spectral functions depend on the potential in Schr"odinger operators on graphs, extending results to deterministic and ergodic potentials and improving understanding of their continuity properties.

Contribution

It introduces a deterministic analysis of the density of states outer measure and establishes its Lipschitz and log-H"older continuity on various graphs, extending prior random operator results.

Findings

Lipschitz continuity of DOSm on the lattice Z^d

Log-H"older continuity of DOSoM on the Bethe lattice

Extension of continuity results to deterministic and ergodic potentials

Abstract

We continue our study of the dependence of the density of states measure and related spectral functions of Schr\"odinger operators on the potential. Whereas our earlier work focused on random Schr\"odinger operators, we extend these results to Schr\"odinger operators on infinite graphs with deterministic potentials and ergodic potentials, and improve our results for random potentials. In particular, we prove the Lipschitz continuity of the DOSm for random Schr\"odinger operators on the lattice, recovering results of \cite{kachkovskiy, shamis}. For our treatment of deterministic potentials, we first study the density of states outer measure (DOSoM), defined for all Schr\"odinger operators, and prove a deterministic result of the modulus of continuity of the DOSoM with respect to the potential. We apply these results to Schr\"odinger operators on the lattice $Z^d$ and the Bethe lattice. In the former case, we prove the Lipschitz continuity of the DOSoM, and in the latter case, we prove that the DOSoM is $\frac{1}{2}$-log-H\"older continuous. Our technique combines the abstract Lipschitz property of one-parameter families of self-adjoint operators with a new finite-range reduction that allows us to study the dependency of the DOSoM and related functions on only finitely-many variables and captures the geometry of the graph at infinity.