Domain wall-magnetic tunnel junction spin–orbit torque devices and circuits for in-memory computing

Abstract

There are pressing problems with traditional computing, especially for accomplishing data-intensive and real-time tasks, that motivate the development of in-memory computing devices to both store information and perform computation 1 . Magnetic tunnel junction (MTJ) memory elements can be used for computation by manipulating a domain wall (DW), a transition region between magnetic domains. Three leading device types that use MTJs and DWs for in-memory computing are majority logic 2-4 , mLogic 5-9 , and DW-MTJs 10-14 . But, these devices have suffered from challenges: spin transfer torque (STT) switching of a DW requires high current, and the multiple etch steps needed to create an MTJ pillar on top of a DW track has led to reduced tunnel magnetoresistance (TMR) 15-16 . These issues have limited experimental study of devices and circuits. Here, we study prototypes of three-terminal domain wall-magnetic tunnel junction (DW-MTJ) in-memory computing devices that can address data processing bottlenecks and resolve these challenges by using perpendicular magnetic anisotropy (PMA), spin-orbit torque (SOT) switching, and an optimized lithography process to produce average device tunnel magnetoresistance TMR = 164%, resistance-area product RA = 31 𝛀 − 𝝁𝒎 𝟐 , close to the RA of the unpatterned film, and lower switching current density compared to using spin transfer torque. A two-device circuit shows bit propagation between devices. Device initialization variation in switching voltage is shown to be curtailed to 7% by controlling the DW initial position, which we show corresponds to 96% accuracy in a DW-MTJ full adder simulation. These results make strides in using MTJs and DWs for in-memory and neuromorphic computing applications.Computing today faces walls when processing data-intensive and unstructured tasks. Memory access in modern computers can dominate as much as 96% of computing time 17 . SRAM idle leakage can consume over 20% of the total power of a computation 18,19 . For internet of things

Shape-Dependent Multi-Weight Magnetic Artificial Synapses for Neuromorphic Computing

“…= 3.5 𝑉 with 𝑣𝑎𝑟 = 5.7%, after which 𝑅 #$% = 67.1 Ω ± 0.3 Ω. The trapezoidal geometry and multiple notches reduce the variation in switching voltage measured here compared to our previous work of 10% 20 . See Supplementary Fig.…”

“…When used for online learning with ANNs, we have previously used micromagnetic simulations to prove that the DW-MTJ synapse is sufficiently linear, controllable, and possesses low enough device variation to make the lower on/off ratio (expressed as tunnel magnetoresistance, or 𝑇𝑀𝑅) of the sensing MTJ tolerable 7 . However, experimental work has focused mostly on binary DW-MTJs 20,21 , with only one MW MTJ paper demonstrating a multi-MTJ device 22 that could have scalability challenges. Several works have investigated theoretical and experimental stochastic functionality of magnetic devices 6,[23][24][25] , but none have also included MW switching.…”

All-Electrical Control of Compact SOT-MRAM: Toward Highly Efficient and Reliable Non-Volatile In-Memory Computing

“…In 2012, the conceptual design of spin-orbit torque magnetic random access memory (SOT-MRAM) was proposed by L. Q. Liu et al [6] with the faster magnetization switching speed (sub-ns) and the higher endurance (>10 10 ) with mitigation of barrier layer degradation by separating the writing current flow through the spin-orbit coupling layer instead of the barrier layer of magnetic tunneling junction (MTJ) [6], i.e., the novel design splits the read and write path individually and avoids the interference and damage caused by the read and write current injected into the MTJ in STT-MRAM. Experimentally, the SOT-MRAM not only enhances the reliability and endurance but also reduces the energy consumption of the device operation, facilitating the highly reliable and time-and energy-efficient applications in the field of non-volatile memory and nv-IMC, as described in our previous work [1,[7][8][9].…”

Tailoring the switching efficiency of magnetic tunnel junctions by the fieldlike spin-orbit torque

“…Current-induced spin-orbit torques (SOT) [1] offer an efficient and scalable way to control the magnetization of spintronic devices [2], including magnetic tunnel junctions (MTJ) [3][4][5][6][7] domain wall racetracks [8][9][10][11] and logic gates [12][13][14][15]. For its relevance in memory and computing applications, SOT switching has undergone much progress in terms of reliability, operation speed, energy efficiency, as well as realizing zeroexternal-field switching in systems with perpendicular magnetization [16].…”