Generalized magneto-optical ellipsometry in the presence of strong second-order magneto-optical effects

Abstract Generalized magneto-optical ellipsometry (GME) is designed to suppress second-order magneto-optical contributions in its data analysis, yet previously, this robustness has not been tested under challenging conditions, in which large second-order magneto-optical effects occur. Here, we investigate the reliability of GME using epitaxial bismuth-substituted yttrium iron garnet films that exhibit pronounced second-order magneto-optical effects, which can even become dominant under certain experimental conditions. Despite the presence of these strong effects, the GME data sets and associated fits show excellent agreement for all tested samples, applied field values, and corresponding magnetization states. The extracted magneto-optical parameters are also in excellent agreement with independent magnetometry measurements of the field-dependent magnetization properties of the samples. Further analysis of the intensity modifications caused by second-order magneto-optical effects indicates that they originate from the intrinsic quadratic magneto-optical response of the film materials. These results validate the reliability of GME under challenging conditions and further suggest that combining conventional GME data analysis with polarization-dependent intensity data sets could enable the simultaneous determination of both linear and second-order magneto-optical contributions.