Time-domain study of frequency-power correlation in spin-torque oscillators

Abstract

This paper describes a numerical experiment, based on full micromagnetic simulations of current-driven magnetization dynamics in nanoscale spin valves, to identify the origins of spectral linewidth broadening in spin torque oscillators. Our numerical results show two qualitatively different regimes of magnetization dynamics at zero temperature: regular (single-mode precessional dynamics) and chaotic. In the regular regime, the dependence of the oscillator integrated power on frequency is linear, and consequently the dynamics is well described by the analytical theory of current-driven magnetization dynamics for moderate amplitudes of oscillations. We observe that for higher oscillator amplitudes, the functional dependence of the oscillator integrated power as a function of frequency is not a single-valued function and can be described numerically via introduction of nonlinear oscillator power. For a range of currents in the regular regime, the oscillator spectral linewidth is a linear function of temperature. In the chaotic regime found at large current values, the linewidth is not described by the analytical theory. In this regime we observe the oscillator linewidth broadening, which originates from sudden jumps of frequency of the oscillator arising from random domain wall nucleation and propagation through the sample. This intermittent behavior is revealed through a wavelet analysis that gives superior description of the frequency jumps compared to several other techniques.

Spin transfer nano-oscillators

“…In general, the analysis of the time-domain measurements allows to separate the linear and nonlinear contributions to the phase noise, being the nonlinear contribution due to the power-phase coupling. 87 For high current regime, the linewidth is not only related to the thermal fluctuations, but an additional intrinsic linewidth is present due to the chaotization of the magnetization trajectory 20,46 as also observed in recent experimental works. 75,[88][89][90] In addition, for real STNO devices the micromagnetic effects caused by devices edges, defects or inhomogeneous current injections are sources of nonuniform precession and thereby of linewidth broadening.…”

Micromagnetic understanding of the skyrmion Hall angle current dependence in perpendicularly magnetized ferromagnets

“…1, the most relevant IMFs are IMF4 and IMF5, although there are other modes with high CC, but with relatively low frequency fluctuations (such as the sub-GHz modes indicated with IMF11 and IMF15), which are far from the frequencies obtained by deterministic micromagnetic simulations (see below). [51] that clearly brings the conclusion of the non-stationary feature of those two thermal modes. 7 To show that those excitations are linked to the skyrmion breathing, we carried out deterministic micromagnetic simulations by using the value of the physical parameters as calculated by scaling relations [21,52] (see Supplementary note 3 [43] and Supplemental Material [43], Movie 2).…”

Amplitude Noise in Spin-Torque Oscillators

“…The oscillation is, however, again intermittent above 0.8 mA, which may be ascribed to the chaotic oscillation in the highcurrent region. 9) These changes of the oscillation states are continuous. The authors examine whether the amplitude fluctuation in the steady oscillation can be described by a recent nonlinear theory 6,7) based essentially on a macrospin model under thermal fluctuation.…”

“…The relative power fluctuation estimated from the distribution in the steady oscillation state is in good agreement with that of a recent nonlinear theory, 6,7) which is essentially based on a macrospin model under the thermal fluctuation, indicating that the fluctuation arises mainly from the thermal magnetization fluctuation rather than from (nonthermal) mode instabilities. 8,9) A tunnel magnetoresistance (TMR) film is deposited on a sapphire wafer using high-vacuum magnetron sputtering. The TMR multilayer consists of underlayer/IrMn (10)/ Co 50 Fe 50 (2.5)/Ru (0.85)/(Co 50 Fe 50 ) 80 B 20 (3) (pinned layer: PL)/MgO (0.65)/(Co 50 Fe 50 ) 80 B 20 (3) (free layer: FL)/cap layer, where the thicknesses in nm are given in parentheses.…”