Is Exchange Bias at the Hybrid Organic/Ferromagnet Interface an Intrinsic Effect?

Abstract

Exchange bias (EB) plays an important role in spintronic devices. It has been thoroughly studied between inorganic anti-ferromagnets and metallic ferromagnetic (FM) layers but rather surprisingly, similar EB effects have been reported between metallic ferromagnets and organic molecular (OM) layers. In this case, EB is considered “spinterface” induced magnetic effect and could play big role in the next generation of environment-friendly electronic devices. Molecular EB effect has been first observed in 2015 [1] between cobalt films and manganese phthalocyanine, then expanded and reproduced with different metal (Zn, Fe, Co) phthalocyanine [2] and metal octa-ethyl porphyrin molecules [3]. However, fundamentals of molecular EB and its origins is still not understood and stays controversial. In this work, we investigate the possibility of molecular EB in cobalt/metal tetra-phenyl porphyrin (M=Zn and Ni) hybrid structures. Careful SQUID magnetometer measurements show the absence of EB in “freshly deposited” hybrid samples. Surprisingly, EB appears within few hours and gradually increases with time. Additionally, new, and similar “freshly deposited” samples kept under ultra-high vacuum (UHV) for the same time span did not show EB when measured directly after removal from the UHV chamber. This indicates that EB effect has its origin from the exposure of samples to air. In the light of these unexpected results, atomic force microscopy and X-ray photoelectron spectroscopy have been performed to identify and understand the reasons of our observations. The experimental evidences strongly suggest that EB in FM/OM systems is not an intrinsic effect and originates from air-driven oxidation of cobalt films transforming part of the metallic cobalt into CoOx that is well known to produce EB effects.