Optical Properties of Correlated Materials -- or Why Intelligent Windows may look Dirty

TL;DR

This paper reviews recent advances in calculating optical properties of correlated materials, especially vanadium dioxide, using dynamical mean field theory, and discusses their potential for designing intelligent windows with optimized optical performance.

Contribution

It introduces a novel, efficient scheme for calculating optical transition matrix elements in correlated oxides and discusses how these methods can guide experimental design of smart window materials.

Findings

Dynamical mean field theory can describe optical properties of correlated materials.

A new scheme simplifies calculations of optical transition matrix elements.

Optimized geometries can improve intelligent window performance.

Abstract

Materials with strong electronic Coulomb correlations play an increasing role in modern materials applications. "Thermochromic" systems, which exhibit thermally induced changes in their optical response, provide a particularly interesting case. The optical switching associated with the metal-insulator transition of vanadium dioxide, for example, has been proposed for use in numerous applications, ranging from anti-laser shields to "intelligent" windows, which selectively filter radiative heat in hot weather conditions. Are present-day electronic structure techniques able to describe, or -- eventually even predict -- such a kind of behavior ? How far are we from materials design using correlated oxides ? These are the central questions we try to address in this article. We review recent attempts of calculating optical properties of correlated materials within dynamical mean field theory, and summarize results for vanadium dioxide obtained within a novel scheme aiming at particularly simple and efficient calculations of optical transition matrix elements within localized basis sets. Finally, by optimizing the geometry of "intelligent windows", we argue that this kind of technique can in principle be used to provide guidance for experiments, thus giving a rather optimistic answer to the above questions.