Experimental verification of the reflection matrix description in linear magneto-optics

We have experimentally studied the accuracy of the Jones-type reflection matrix Formula Presented that describes light reflection from a magnetic sample exhibiting the magneto-optical (MO) Kerr effect (MOKE). For this purpose, we performed a comprehensive experimental study, utilizing generalized MO ellipsometry, which allows for the simultaneous measurement of all complex-valued elements of Formula Presented for arbitrary magnetization vector orientations. For the collection of suitable datasets under appropriate test conditions, we also fabricated and utilized a thin film sample that exhibits a uniform magnetization state, whose vector orientation can be easily rotated by an applied magnetic field. This approach enabled us to systematically and simultaneously vary the MOKE coefficients of Formula Presented, which permitted us to verify the accuracy of Formula Presented with high precision. While we observe the widely used standard formulation of Formula Presented to be correct under most experimental conditions, we also found small systematic deviations for some specific cases. However, these deviations are not indicative of a limited correctness of Formula Presented, but instead, they are related to the material assumptions that are commonly made for the derivation of Formula Presented. For our specific sample here, the origin of the deviation is MO anisotropy, and upon considering this material property accurately, an exact description of all experimental results by Formula Presented can be reestablished.