Date of Graduation


Document Type



All conducting materials have a magnetoresistive response (MR). In bulk ferromagnets the MR is due to the anisotropic magnetoresistance (AMR). AMR results from the spin-orbit interaction and s-d electron scattering. The change in resistance due to AMR is only a few percent. In contrast, the giant magnetoresistance (GMR) can be over 100%. The primary mechanism responsible for GMR is spin-dependent scattering. But where does this scattering occur? Some experiments show that the magnitude of the GMR depends on the thickness of the ferromagnetic layer, concluding that the scattering occurs in the bulk of the ferromagnet. Other studies that vary the interface roughness and composition conclude that the interface scattering is important. To study this problem, we chose to investigate Co/Re superlattices. These superlattices are hcp with the c-axis, the magnetic easy axis, in the film plane, and have GMR and AMR contributions of comparable size. The basic idea is to use the AMR, as a probe to determine whether the scattering responsible for GMR occurs primarily at the interface or inside the Co layers. To do this, neutron reflectivity was used to find the magnetization vector in adjacent layers of Co and the MR was measured as a function of temperature. We found that in some geometries the GMR behaves like the AMR. Here the scattering responsible for GMR occurs in the Co layer. In other geometries, the GMR and the AMR behave differently as a function of temperature, so interface scattering is more important. This demonstrates that a fundamental understanding of the GMR must take into account the direction of current flow and the band structure of the materials. In addition to the expected in-plane anisotropy, we surprisingly observed an interface induced out-of-plane anisotropy. This type of anisotropy to our knowledge has never been measured in Co systems with an in-plane c-axis before. Co/Re superlattices can be used to test theories on ideal bulk antiferromagnets. Other magnetic superlattices have already been used in this capacity. In this dissertation we tested a surface spin-flop theory using our superlattices and found excellent agreement between the theoretical predictions and the experimental results.