Impact of Reference-Layer Stray Field on the Write-Error Rate of Perpendicular Spin-Transfer-Torque Random-Access Memory

Arshid Nisar, Tanmoy Pramanik, and Brajesh Kumar Kaushik

Phys. Rev. Applied 19, 024016 – Published 6 February 2023

Abstract

A finite-temperature micromagnetic study of magnetization switching and write-error rates in a perpendicular magnetic tunnel junction with and without synthetic antiferromagnetic layer (SAF) is presented. In the absence of SAF, magnetization switching is induced by domain-wall nucleation and propagation. Although the various modes of domain-wall propagation are observed to delay switching, it does not show an appreciable impact on the overall write-error-rate slopes. In the presence of the nonuniform stray field from the SAF assembly, the domain-wall-based switching modes turn on more complex magnetization dynamics that impedes the switching process. In cases where the SAF layers fail to balance each other contributing to a stronger stray field, incoherent switching modes give rise to metastable states with significantly longer lifetimes, and a dramatic change in the write-error slopes is observed. Simulation results are compared to recent experimental findings from time-domain measurements of spin-transfer-torque switching and measurements of anomalous write-error rates. These results directly prove the long-predicted relation of the SAF stray field to write-error anomaly in perpendicular spin-transfer-torque magnetic random-access memory and could be useful to solve the anomalous write-error problems.

Received 5 August 2022
Revised 5 December 2022
Accepted 12 January 2023

DOI:https://doi.org/10.1103/PhysRevApplied.19.024016

Physics Subject Headings (PhySH)

Magnetic anisotropy Magnetization switching Micromagnetism Spin transfer torque

Devices for digital logic, storage & processing Magnetic tunnel junctions Synthetic antiferromagnetic multilayers

Micromagnetic modeling

Condensed Matter, Materials & Applied Physics

Authors & Affiliations

Arshid Nisar , Tanmoy Pramanik ^*, and Brajesh Kumar Kaushik ^†

Department of Electronics and Communication Engineering, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand 247667, India

^*pramanik.tanmoy@ece.iitr.ac.in
^†bkk23fec@iitr.ac.in

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 19, Iss. 2 — February 2023

Subject Areas

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
Device structure, dimension, and layer thicknesses considered in this work. (a) Standalone FL—the RL is assumed to have no stray field on the FL and (b) MTJ stack with SAF layers included. Layer thicknesses are also marked. Other material parameters are listed in Table 1. Three different SAF designs for (b) are considered—“SAF I,” “SAF II,” and “SAF III”. For all three SAF configurations, the stray field inside the FL along the diameter is plotted in (c) and (d) for perpendicular and in-plane directions, respectively.
Reuse & Permissions
Figure 2
(a)–(c) Typical switching processes observed for the standalone FL at V = 3V_C₀. Domain-wall-mediated switching is observed in all cases. (a) Ballistic domain wall motion through the device achieving fast switching. (b) oscillatory and (c) multiple swing-like behaviors when the domain wall passes through the center of the FL. Inset in (a) shows the colormap used for the magnetization snapshots. (d) WER slopes calculated for varying applied voltages.
Reuse & Permissions
Figure 3
(a) WER slopes at V = 3V_C₀ showing the impact of SAF-induced stray field. Compared to standalone FL, one of the switching directions is aided and the other direction is hindered by the stray field. The WER slopes change dramatically for the SAF II P → AP switching and SAF III AP → P switching configurations as the metastable oscillatory states prevail for these two configurations. (b) WER plotted versus normalized write voltage for selected cases. SAF II P → AP and SAF III AP → P switching show distinct plateau in the WER tails—similar to the ballooning effect reported in experimental studies.
Reuse & Permissions
Figure 4
Typical switching events and metastable states for varying SAF configurations—(a) for SAF I, P → AP switching starts by a nucleation of a reversed domain at the center followed by domain-wall motion as in standalone FL. AP → P starts by edge nucleation; (b) for SAF II with stronger stray fields, P → AP switching is hindered by the oscillation of unswitched bubble around the edges while AP → P switching becomes mostly ballistic aided by the stray field; (c) for SAF III, nucleation of reverse domain starts from the edge for both direction of switching, but the AP → P direction now shows similar metastable states as in P → AP direction for SAF II. (d) Trapped bubble of unswitched domain evolves during P → AP switching in SAF II. After the write pulse is turned off at 30 ns, the FL may return to initial state (path 1) or switch to the desired state (path 2) causing increased write-error rates. Magnetization snapshots are shown on the right corresponding to the time instants marked by the red dots. Inset in (a) shows the colormap used for the magnetization snapshots.
Reuse & Permissions
Figure 5
Impact of SAF-induced stray field on the WER slopes for varying FL diameter: (a) 120, (b) 90, (c) 80, and (d) 56 nm. For each case the WER slope of standalone FL (red) and WER including SAF III (black) is plotted. Additionally, WER slopes including an average SAF field (blue) is also plotted for comparison. At 56 nm, the effect of SAF is the same as the average out-of-plane field. For larger diameter, SAF field causes additional features. All simulations are done for 300 K. The temperature dependence of WER is shown in (a) for the 120-nm device with 233 K (dotted lines) and 300 K (solid lines) simulation results. Higher temperature lowers WER. The same SAF parameters (Table 1) are used for all, except the diameter. Applied voltage is twice the critical voltage for each diameter.
Reuse & Permissions

Physical Review Applied