Structural sensitivity of the spin Hall magnetoresistance in antiferromagnetic thin films

Reading the magnetic state of antiferromagnetic (AFM) thin films is key for AFM spintronic devices. We investigate the underlying physics behind the spin Hall magnetoresistance (SMR) of bilayers of platinum and insulating AFM hematite $\left(\alpha -{\mathrm{Fe}}_2{\mathrm{O}}_3\right)$ and find an SMR efficiency of up to 0.1%, comparable to ferromagnetic-based structures. To understand the observed complex SMR field dependence, we analyze the effect of misalignments of the magnetic axis that arise during growth of thin films, by electrical measurements and direct magnetic imaging, and find that a small deviation can result in significant signatures in the SMR response. This highlights the care that must be taken when interpreting SMR measurements on AFM spin textures.

Figure 1

(a) Schematic of the measurement geometry employed of a Pt Hall bar atop hematite films. The charge current and longitudinal voltage contacts are indicated. The sample $xy$ plane is the antiferromagnetic easy plane (indicated in orange) above the Morin transition temperature. Normalized longitudinal resistance ($\mathrm{\Delta }{R}_L=R_L-R_0$, where $R_0$ is the zero-field resistance) of a Pt Hall bar atop (0001)-orientated hematite at 300 K in the easy-plane antiferromagnetic phase. The resistance is measured as a function of the magnetic field applied along the (b) $\mathit{x}$ axis, (c) $\mathit{y}$ axis, and (d) $\mathit{z}$ axis. (e) Normalized longitudinal (black) and transverse (red) resistance for an in-plane rotation of the magnetic field ${\mu }_{0}H=5\mathrm{T}$ ($\mathrm{\Delta }{R}_L=R_L-R_{\alpha =0^{\circ }}$). Error bars, where visible, represent the standard deviation of the measurement points.

Figure 2

Longitudinal resistance of Pt/hematite at 175 K in the easy-axis phase for a magnetic field applied along the (a) $\mathit{x}$ axis (b) $\mathit{y}$ axis, and (c) $\mathit{z}$ axis. $\mathrm{\Delta }{R}_L=R_L-R_0,$ where $R_0$ is the zero-field resistance. Error bars where visible represent the standard deviation of the data point. We indicate in (c) the critical magnetic field $H_c^{\left|\right|}$ for an increasing magnetic field. (d), (e) Expected effect on the SMR ratio for a misalignment of the easy axis in the $xz$ plane (d) or the $yz$ plane (e) for a magnetic field applied in plane along either $x$ or $y$. The critical field of the magnetic field induced second-order transition of the Néel vector is $H_c=\left(2J{K}_z-D^2\right)/D$, where $J$ is the strength of the exchange interaction, $K_z$ is the uniaxial anisotropy, and $D$ is the strength of the antisymmetric exchange interaction (DMI). This transition describes a rotation of the Néel vector to a state perpendicular to both the easy axis and the magnetic field.

Figure 3

(a) Temperature dependence of the normalized longitudinal resistance for a magnetic field $\mathit{H}//\mathit{x}$. The arrows indicate the critical magnetic field $H_{\mathrm{c}}^{\perp }$ for each temperature. (b) Temperature dependence of the critical fields for $\mathit{H}//\mathit{x}$ ($\mathit{H}//\mathit{y}$) (black) and $\mathit{H}//\mathit{z}$ (gray). (c) Temperature dependence of the SMR ratio for $\mathit{H}//\mathit{x}$ (black) and $\mathit{H}//\mathit{y}$ (blue). (d) Normalized magnetic susceptibility of 100-nm-thick (0001)-orientated hematite for a field parallel to the $c$ axis. The Morin transition is indicated.