Validity of the thermal activation model for spin-transfer torque switching in magnetic tunnel junctions

Abstract

We have performed spin-transfer torque switching experiments with a large number of trials (up to 107 switching events) on nanoscale MgO magnetic tunnel junctions in order to test the validity and the limits of the thermal activation model for spin-torque-assisted switching. Three different methods derived from the model (“read disturb rate,” “switching voltage versus pulse duration,” and “switching voltage distribution” measurements) are used to determine the thermal stability factor and the intrinsic switching voltage. The results obtained from the first two methods agree well with each other as well as with values obtained from quasistatic measurements, if we use only the data for which the voltage is smaller than approximately 0.8 of the intrinsic switching voltage. This agreement also shows that, in our samples, in the low voltage region, the influence from other factors contributing to the switching (such as current-induced heating and field-like torque) is negligible. The third method (switching voltage distribution measurements) yields incorrect values for the time-scales (<1μs) at which the experiments are performed. Macrospin simulations confirm our findings that the model must be applied only in the low voltage limit, and that in certain devices this limit can extend up to about 0.9 of the intrinsic switching voltage.

Historical Background and Introductory Concepts

“…These are maintained via the steady state magnetization precession due to STT representing the conversion of dc input current into an a.c. output voltage. Now, in STT devices, the thermal fluctuations cannot be ignored at finite temperatures because they lead to mainly noise-induced switching at currents far less than the critical switching current without noise as well as introducing randomness into the precessional orbits [138]. This phenomenon has been corroborated by many experiments (e.g., [139]) demonstrating that STT near room temperature alters thermally activated switching processes, which then exhibit a pronounced dependence on both material and geometrical parameters.…”

Estimation of thermal stability factor and intrinsic switching current from switching distributions in spin-transfer-torque devices with out-of-plane magnetic anisotropy

“…The current values that we used for estimating ∆ and I c0 span a range ∼ 0.6-0.8 I c0 . The reason for this is that currents > 0.8 I c0 would violate the model used to fit the data (as explained in reference Heindl, et al 4 ), while going lower than 0.6 I c0 in the RDR analysis is not practical -it is very time consuming to simulate (or experimentally acquire) low switching probabilities. As mentioned by Taniguchi et al 11 for the case of in-plane anisotropy, the value of n could be a function of current for the case of out-of-plane anisotropy too.…”

“…These two methods (read-disturb and long-times) for obtaining the values of ∆ and Ic 0 using n = 1 for the case of in-plane magnetic anisotropy have been validated with the restriction that I < 0.8Ic 0 . 4 The value of the exponent n, for both the in-and out-of-plane anisotropy, is a debated issue. As discussed above, Li, et al 7 and Koch et al 8 used n = 1 for the case of in-plane anisotropy in the limit of small currents.…”

“…The simulations were performed with the same code as in reference Heindl, et al, 4 using a macrospin model that includes finite temperature effects. 5 The simulation parameters for the outof-plane structure were: µ 0 M s = 1.38 T (free layer magnetization), α = 0.01 (magnetic damping parameter), µ 0 H z k = 1.64 T (out-of-plane magnetic anisotropy), P = 0.5 (polarization of spin injection current), T = 300 K, and Λ = 1 (spin-torque asymmetry).…”

Spin-transfer torque effects in the dynamic forced response of the magnetization of nanoscale ferromagnets in superimposed ac and dc bias fields in the presence of thermal agitation

“…Consequently, the magnetization M of the ferromagnet may be altered by spinpolarized currents, which underpin the novel subject of spintronics [8], i.e., current-induced control over magnetic nanostructures. Applications include (a) very high speed current-induced magnetization switching by reversing the orientation of magnetic bits [5,9] and (b) using spinpolarized currents to manipulate steady state microwave oscillations [9] via the steady state magnetization precession due to STT representing the conversion of dc input current into an ac output voltage [5]. Now due to thermal fluctuations [5,9], STT devices invariably represent an open system on the nanoscale in an out-of-equilibrium steady state quite unlike conventional nanostructures characterized by the Boltzmann equilibrium distribution.…”