Figure 1

(a) Experimental configuration of XFMR measurement. The RF master oscillating signals with a frequency of \(f_0 = 500.1 \hbox {MHz}\) are employed as the RF field for FMR effect. The RF signals are delivered to the sample via a coplanar waveguide (CPW) after delaying the phase and multiplying the frequency. The phase of the RF field is also modulated by \(\pi\) with a frequency of 1.0333 kHz by using a square-wave signal generated from a function generator. The square-wave signal and the detected XMCD signals were fed into a Lock-in Amplifier. A bias magnetic field is applied perpendicular to the direction of X-rays and AC magnetic field of the RF field. (b) Schematic representation of magnetization precession under FMR effect with the modulated RF field. \(\textbf{m}\) represents a magnetic moment, and IP and OOP present in-plane and out-of-plane directions to a sample surface, respectively. (c) Phase delay scans of XFMR signals for Pt(10)/Py(30) sample at the Fe \(L_3\) edge obtained by using left and right circular polarized (LCP and RCP) X-rays. (d,e) Phase delay scans for (d) Ta(2)[Pt(2)/Py(5)]\(_{6}\) and (e) Pt(10)/Py(30) samples at various bias magnetic fields across the ferromagnetic resonance field. RF field with a frequency \(f = 4.0008\) GHz was applied for the samples. Dashed lines indicate \(t_1 = 160\,\hbox {ps}\) and \(t_2 = 290\,\hbox {ps}\) for Ta(2)[Pt(2)/Py(5)]\(_{6}\), and \(t_1^{\prime } = 90\,\hbox {ps}\) and \(t_2^{\prime } = 210\,\hbox {ps}\) for Pt(10)/Py(30). (f,g) Bias magnetic field dependence of (f) the amplitude and (g) the relative phase of the magnetization precessions. Red and blue lines represent the fitting results.