Structural and Magnetic Characterization of Epitaxial Co(10.0)/Pt(110) Multi-layers for Future Anisotropic DMI Systems Based on C2v Symmetry

Abstract

The interfacial Dzyaloshinskii-Moriya Interaction has had profound consequences on the possible magnetic configurations that can be stabilized in multi-layer thin films, most notably chiral Neel skyrmions and domain walls.[1] More recently, there have been theoretical calculations and experimental work confirming the existence of an appreciable anisotropic DMI in epitaxial thin films with further reduced symmetry – namely the C2v (or mm2) point group.[2] However, so far the systems under consideration have been based on cubic crystals where the magnetocrystalline anisotropy is nearly isotropic.Here, we identify a new epitaxial relationship with C2v symmetry based on HCP Co(10.0)/FCC Pt(110). Although it is quite common to grow FCC Co(111)/Pt(111), for the case of a Pt(110) underlayer, the lattice mismatch is actually smaller for the HCP Co(10.0) orientation [as opposed to FCC Co(110)] where the c-axis lies along a single direction in the plane of the film (figure 1). Such an orientation is expected to offer some degree of anisotropic DMI based on the heavy metal Pt and reduced symmetry as well as the added feature of significant uniaxial in-plane magnetocrystalline anisotropy.Samples were prepared by DC magnetron sputtering on HF-etched Si(110) single crystal substrates. Following the work of Yang et al., we use a Ag (10nm) buffer layer, which is known to grow with (110) orientation on Si(110) at room temperature, thereby avoiding further unwanted interface diffusion.[3] This is followed by deposition of the Pt (10nm) /Co (10-50nm) bilayers, which are the focus of this work. All depositions were done in an Ar atmosphere with the working pressure fixed at 2.5 mTorr.The films were characterized structurally by standard θ/2θ x-ray diffraction (XRD) measurements and phi-scans to characterize the epitaxial relationship. The XRD scan shows primarily Ag(220), Pt(220), and Co(10.0) peaks with no identifiable Co(220) peak suggesting the film is predominantly HCP. Phi scans (figure 2) on Pt(111), Ag(111), and Co(10.1) confirm the epitaxial relationship shown in figure 1. Moreover, reciprocal space q-scans on the HCP Co spots were leveraged to characterize directional peak broadening due to stacking faults. As it is known that stacking faults cause broadening only of certain peaks (h-k ≠ 3m) along the c* direction (with breadth also determined by the parity of l), it is possible to extract the density of growth and deformation faults in the Co layer, which are found to be approximately 10% and 3% respectively in 50nm Co films. Details on how these stacking fault densities are calculated will presented in detail.Alternating gradient field magnetometry was used to measure M-H loops along different crystallographic directions in the plane of the film. A clear in-plane easy axis is identified along Si[001] (parallel to Co[0001]). The hard axis saturation field of 1.5 kOe suggests a notably smaller magnetocrystalline anisotropy than would be expected for bulk cobalt, likely due to the aforementioned stacking faults.Although the relatively thick Co layer in these films makes any interface induced DMI unlikely, the epitaxial relationship is expected to be preserved in thinner samples along with the reported uniaxial anisotropy. Together with the expected interfacial anisotropic DMI, this in-plane anisotropy makes the proposed material system an interesting playground for exploring new chiral magnetic ground states. **