Understanding disorder-induced zero-bias anomalies in systems with short-range interactions: An atomic-limit perspective

TL;DR

This paper investigates how short-range interactions and disorder in the atomic limit of the extended Anderson-Hubbard model lead to zero-bias anomalies, revealing the effects of residual charge ordering and on-site interactions.

Contribution

It provides new insights into disorder-induced zero-bias anomalies by analyzing the atomic limit with nonlocal interactions, bridging the gap between local and infinite-range interaction models.

Findings

Nearest-neighbor interactions and disorder cause zero-bias anomalies.

On-site interactions non-monotonically affect the anomaly depth.

Density of states resembles full model results for U<4V.

Abstract

Motivated by the novel electronic behaviors seen in transition metal oxides, we look for physical insight into disordered, strongly-correlated systems by exploring the atomic limit. In recent work, the atomic limit has provided a useful reference point in systems with strong {\em local} interactions. For comparison with experiments, the exploration of {\em non}local interactions is of interest. In the atomic limit, both the case of on-site interactions alone and the case of infinite-range ($1/r$) interactions are well understood; however, not so the intervening possibilities. Here we study the atomic limit of the extended Anderson-Hubbard model using classical Monte Carlo to calculate the single-particle density of states. We show that the combination of nearest-neighbor interactions and site disorder produce a zero-bias anomaly caused by residual charge ordering, and the addition of on-site interactions has a non-monotonic effect on the depth of this zero-bias anomaly. A key conclusion is that the form of the density of states in this classical system strongly resembles density of states results obtained for the full extended Anderson-Hubbard model when $U<4V$.