Interface engineering of phase separation in SrRuO3/SrTiO3 hybrid superlattices

Abstract

Most of the observed electronic phase separation (EPS) phenomena in complex oxides occur in systems with chemical doping, and theories have established the strong correlation between EPS and chemical-doping induced quenched disorder. Recent experiments on fabricated oxide superlattices with exclusion of chemical disorder also confirmed that the EPS disappears in the clean systems. Thus far, the existence of EPS in strongly correlated oxides without the presence of chemical doping or chemical disorder has not been proved. Here, we have built fully chemically ordered hybrid superlattices using prototypical SrRuO3 and SrTiO3 perovskite oxides. Contrary to previous theoretical and experimental results, we observe an EPS of two magnetic phases with perpendicular and cubic magnetic anisotropies, indicating the coexisting metallic and insulating phases in the superlattices. Our results provide an alternative pathway other than the chemical doping to introduce EPS in correlated oxides and strongly suggest that EPS can exist in clean systems without the need of chemical disorder.3 Electronic phase separation (EPS) in complex oxides is often manifested by the competition between multiple phases with distinct electronic/magnetic properties [1]. This phenomenon is most prominent in superconducting cuprates and colossal magnetoresistance manganites which typically show the coexistence of metallic and insulating phases on the micron to nanometer length scales [2,3]. In most cases, EPS phenomena are induced by chemical doping through ionic replacements, interstitials, or vacancies in the parent compounds [2][3][4][5]. Occasionally, EPS appears in some oxides with no apparent chemical doping, such as CaFe3O5 and LuFe2O4; however, cation nonstoichiometry still exists in these oxides [6,7] such that they can be viewed as solid solutions of Fe 2+ and Fe 3+ . The chemical dopants inevitably distribute randomly in the lattice, leading to a quenched disorder, and the role of such randomness in phase transition has been one of the main research subjects in condensed matter physics [8,9].