On the origin of systematic errors in VSM torque curves

Abstract

Magnetic anisotropy (MA) is a basic and crucial magnetic property which describes how the magnetic properties change with field angle in a magnetic composite. Torque measurements are normally used to measure a sample’s anisotropy constant and can be determined directly using a torque magnetometer (TM) or indirectly using a vibrating sample magnetometer (VSM). Most TM only operate at room temperature. However, a biaxial VSM allows straightforward torque measurements at varying temperatures. VSM torque curves comprise a 2θ background caused by small misalignments of the sample on the sample rod, and/or a non-azimuthal symmetric sample shape. These errors originate from the position dependence of the pickup coil sensitivity matrix. Here, the magnetic torque of a strontium ferrite/PA-12 3D printed samples was explored using biaxial VSM and compared to measurements done with a true torque magnetometer. Torque curves were taken from 16-22kOe every 2kOe on a 4.8 mm wire and a 1.23 mm dot sample using a MicroSense EZ9 VSM. Fourier analysis was used to determine the second harmonics from each torque curve. The biaxial torque curves of the wire samples have a much larger torque amplitude which is mainly due to the crosstalk of the y-coil set. Crosstalk originates from the non-azimuthal symmetric shape of the sample and due to the wobble originating from small sample misalignments on the sample rod. Background can be minimized by using samples with near cylindrical symmetry and using a microscope when mounting the sample on the sample rod.