# ARPES study of orbital characters, symmetry breakings and pseudogaps in doped and pure Sr2IrO4

## TL;DR

This study uses ARPES to analyze the orbital characters, symmetry properties, and pseudogap phenomena in Sr2IrO4, revealing detailed orbital modulations, symmetry considerations, and disorder effects in doped and pure samples.

## Contribution

It provides the first detailed polarization-dependent ARPES analysis of orbital characters in Sr2IrO4, combining experimental data with theoretical calculations to clarify symmetry and pseudogap features.

## Key findings

- Orbital character of bands varies with momentum, showing strong modulation.
- Symmetry breaking effects are minimal once intensity variations are corrected.
- A consistent energy scale of 30 meV for spectral weight suppression suggests disorder effects.

## Abstract

Sr2IrO4 is characterized by a large spin-orbit coupling, which gives rise to bands with strongly entangled spin and orbital characters, called J=1/2 and J=3/2. We use light-polarization dependent ARPES to study directly the orbital character of these bands and fully map out their dispersion. We observe bands in very good agreement with our cluster dynamical mean-field theory calculations. We show that the J=1/2 band, the closest to the Fermi level Ef, is dominated by dxz character along kx and dyz along ky. This is actually in agreement with an isotropic J=1/2 character on average, but this large orbital dependence in k-space was mostly overlooked before. It gives rise to strong modulations of the ARPES intensity that we explain and carefully take into account to compare dispersions in equivalent directions of the Brillouin zone. Although the latter dispersions look different at first, suggesting possible symmetry breakings, they are found essentially similar, once corrected for these intensity variations. In particular, the pseudogap-like features close to the $X$ point appearing in the nearly metallic 15% Rh-doped Sr2IrO4 strongly depend on experimental conditions. We reveal that there is nevertheless an energy scale of 30meV below which spectral weight is suppressed, independent of the experimental conditions, which gives a reliable basis to analyze this behavior. We suggest it is caused by disorder.