Full text loading...
(a) Measurement schematic showing coplanar stripline on left with microwave current, , generating a perpendicular microwave field over the bar area. A microwave current, , is coupled into the bar. The voltage is measured across the bar contacts with a lock-in amplifier. (b) The bar consists of a Pt layer deposited on top of a cobalt layer. The in-plane angle θ is defined as the angle between the bar direction and the magnetisation.
(a) Detected voltage for a 2 nm device for a single field sweep. The FMR peak is fitted (solid green line) by a combination of symmetric (dotted red line) and antisymmetric (dashed blue line) Lorentzian curves. (b) The angular dependences of the symmetric (full red circles) and antisymmetric (open blue circles) voltages are each fitted by a linear combination of and terms.
(a) and (b) Cobalt thickness dependence of the fitted symmetric (redcircles) and antisymmetric (blue diamonds) and voltage components.
(a) Measured values of (red circles) and (blue diamonds) are fitted well by Eqs. (7) (dotted line) and (6) (dashed line), respectively. (b) Cobalt thickness dependence of the spin-pumping charge current is plotted (red circles). The relative size of the spin-current (solid blue line), which is calculated using the fits to the measured values of and , decreases in the thinner layers. In contrast, the charge current increases in the thinner layers. (c) The relative size of (red circles) is enhanced in the 1 nm Co layer. The error bars show the standard error from fitting the parameter to the angular-dependent symmetric voltage data. The small variance between the data points of the same thickness could also be from a small difference in the size of the microwave field in each device. The inhomogeneous part of the linewidth (blue diamonds) also shows an increase in thinner Co layers.
Article metrics loading...