“…The linearly fitted curves in Figure c,d indicate that Joule heating in this experiment does not play an important role in the final results. In addition, if we assume that all spin currents, exerted on the FM layer, are generated from the HM layer, eq can be used to calculate the spin–torque efficiency θ SH . ,, In eq , ℏ is the Planck constant, M s is the saturation magnetization, e is the electronic charge, t FM is the thickness of the FM layer, and J is the current density. In this work, the current density is assumed to flow through the multilayer uniformly.…”