Comprehensive Scaling Analysis of Current Induced Switching in Magnetic Memories Based on In-Plane and Perpendicular Anisotropies

Abstract

Spin transfer torque based magnetic memories (STT-MRAMs) are leading contender for the replacement of SRAM caches. However, STT-MRAMs suffer from high write current, read/write stability conflicts and other failure mechanisms. In this paper, we present a comprehensive scaling analysis for STT-MRAMs based on in-plane and perpendicular anisotropy magnets in context to different failure mechanisms. Write failures are taken into consideration by the write current, read disturb failures by the critical current and read decision failure by the tunnel magneto-resistance scaling trends. Bit-cells comprising three different device structures-the conventional magnetic tunnel junctions (MTJs), the dual pillar MTJs (DP-MTJs) and the spin-orbit-torque based MTJs (SOT-MTJs) are investigated. We analyze the robustness of the aforementioned devices within the voltage constraints specified by ITRS. We also report predictive analysis results with futuristic material parameters. Through a coupled simulation framework consisting of spin transport and magnetization dynamics, we show that conventional MTJs would require higher voltages at scaled technology nodes. DP-MTJs, within ITRS voltage specifications, show better scalability (with larger bit-cell area). SOT-MTJs provide attractive power savings ( improvement) at a larger bit-cell area. Furthermore, our analysis indicates that among various possible improved material parameters, high interface perpendicular anisotropy shows the most promising way of achieving scalable memory cells at assumed ITRS voltages.Index Terms-Dual-pillar magnetic tunnel junction (MTJ), Landau-Lifshitz-Gilbert (LLG), readability, scaling, spin-orbittorque magnetic tunnel junction (SOT-MTJ), spin transfer torque based magnetic memories (STT-MRAMs), writability.

Practical Experiments to Evaluate Quality Metrics of MRAM-Based Physical Unclonable Functions

“…1b, one must heat AFML2 up to 160℃. For this purpose, a parameter called "heating voltage" is employed [21].…”

In-situ, In-Memory Stateful Vector Logic Operations based on Voltage Controlled Magnetic Anisotropy

“…In general, higher the EB, higher is the current required to switch the MTJ. One of the key challenges associated with the STT phenomenon is the high switching current requirement 32 . In order to reduce the current requirement for switching the nano-magnet various voltage driven switching phenomenon are under intense research investigation 33 , 34 .…”

“…1 relates to magnetization precession along H EFF while the second and last terms describe the damping torque and STT, respectively. H EFF includes an external field ( H ext ), demagnetization field due to shape anisotropy 44 ( H demag ), the interface perpendicular anisotropy field 45 ( H ani ) and stochastic field due to thermal noise ( H thermal ), as described in Eq. 2 .…”

Energy-Efficient Memory Using Magneto-Electric Switching of Ferromagnets

“…where H EF F is the effective magnetic field. H EF F is the sum of the demagnetization field [18], [19], the interface anisotropy field [3] and any other external field.m is the unit magnetization vector, γ is the gyromagnetic ratio and α is the Gilbert damping constant. The thermal noise is modeled using the Brown's model [20] and is accounted for by expressing a contributing field to H EF F as H thermal = ζ 2αkT |γ|M S V dt , where ζ is a vector with components that are zero mean Gaussian random variables with standard deviation of 1.…”

“…M AGNETO-RESISTIVE memories based on current driven Spin Transfer Torque (STT) [1], have attracted immense research interest due to their non-volatility, almost unlimited endurance and area-efficiency [2]. However, STT based memories suffer from inherent low switching speed and high write-energy consumption [3]. Recently, voltage induced Magneto-Electric (ME) effect, has shown potential for fast and energy-efficient switching of ferromagnets [4].…”