Search

As the basic storage unit of spin transfer torque magnetic random-access memory (STT-MRAM), the perpendicular magnetic anisotropy (PMA) magnetic tunnel junction (MTJ) has been extensively studied in recent years. Lowering the critical switching current and improving the data retention are two crucial pathways to optimize the performance of STT-MRAM. However, the conventional MTJ can merely achieve both. In this paper, we present a physics-based compact model of Ta/CoFeB/MgO PMA double-barrier MTJ (DMTJ) with enhanced thermal stability recording structure. Combination of double-barrier and synthetic double-free layers can heighten the STT effect and enhance the thermal stability simultaneously. A larger STT switching efficiency, compared with conventional MTJ, can thus be realized. The modeling results show great agreement with experimental results. A 1-bit magnetic full adder (MFA) based on DMTJ, as a hybrid logic-in-memory circuit example, has been designed and simulated to validate its functionality. This SPICE-compatible compact model will be useful for high-performance hybrid MTJ/CMOS circuit and system designs. [ABSTRACT FROM AUTHOR]

Magnetic tunnel junction nanopillar with interfacial perpendicular magnetic anisotropy (PMA-MTJ) becomes a promising candidate to build up spin transfer torque magnetic random access memory (STT-MRAM) for the next generation of non-volatile memory as it features low spin transfer switching current, fast speed, high scalability, and easy integration into conventional complementary metal oxide semiconductor (CMOS) circuits. However, this device suffers from a number of failure issues, such as large process variation and tunneling barrier breakdown. The large process variation is an intrinsic issue for PMA-MTJ as it is based on the interfacial effects between ultra-thin films with few layers of atoms; the tunneling barrier breakdown is due to the requirement of an ultra-thin tunneling barrier (e.g., <1 nm) to reduce the resistance area for the spin transfer torque switching in the nanopillar. These failure issues limit the research and development of STT-MRAM to widely achieve commercial products. In this paper, we give a full analysis of failure mechanisms for PMA-MTJ and present some eventual solutions from device fabrication to system level integration to optimize the failure issues. [ABSTRACT FROM AUTHOR]

A multilevel cell (MLC) allows gigabit high-density integration of a spin-transfer torque magnetic random-access memory (STT-MRAM). In this paper, we present an MLC structure based on MgO/CoFeB interfacial perpendicular magnetic anisotropy (PMA) and capping layer adjustment. The effect of capping layer toward interfacial PMA modulation has been investigated by first-principles calculations, which is consistent with the previous experimental results. Two methods have been proposed to realize this STT-MRAM MLC: 1) capping layer thickness and 2) material composition design. Using a physics-based compact model of perpendicular magnetic tunnel junctions, resistance–current curves with four-level resistance states are demonstrated, for which the threshold switching currents with clear separations prove the feasibility of this concept. [ABSTRACT FROM AUTHOR]

Recently, spin-transfer torque magnetic random access memory (STT-MRAM) has been considered as a promising universal memory candidate for future memory and computing systems, thanks to its nonvolatility, high speed, low power, good endurance, and scalability. However, as technology scales down, STT-MRAM suffers from serious process variations and thermal fluctuations, which greatly degrade the performance and stability of STT-MRAM. In general, the optimization and robustness of STT-MRAM under process variations often require a hybrid design flow and multilevel codesign strategies. In this paper, we quantitatively analyze the impacts of process variations and thermal fluctuations on the STT-MRAM performances from physics, technology, and circuit design point of views. Based on the analyses, we found that readability is becoming the newest challenge for deeply scaled STT-MRAM due to the conflict between sensing margin and read disturbance. To deal with this problem, a novel reconfigurable design strategy from device, circuit, and architecture codesign perspective is then presented. Finally, a conceptual hybrid magnetic/CMOS design flow is also proposed for STT-MRAM in deeply scaled technology nodes. [ABSTRACT FROM AUTHOR]

Spin-transfer torque magnetic random access memory (STT-MRAM) has emerged as a promising candidate for the next-generation high-speed, low-power, and scalable nonvolatile memory technology. Its advantageous features have attracted much attention in the area of research and development, and a number of preindustrial prototypes and small-scale products have been demonstrated. One expects to widely commercialize it in the next few years. One of the critical issues that blocks STT-MRAM’s wide commercialization is its low sensing reliability due to the relatively small tunnel magnetoresistance ratio of the magnetic tunnel junction (MTJ) and the continuously increasing process variations, especially as technology process scales down to the deep submicrometer nodes (e.g., 40 nm). In this paper, we present a variation-tolerant high-reliability sensing circuit for deep submicrometer STT-MRAM. This circuit, using triple-stage sensing and charge transfer amplification, is able to tolerate significantly the process variations and improve greatly the sensing margin by sacrificing some sensing time. Using a CMOS 40 nm design kit and a precise STT-MTJ compact model, transient and Monte Carlo simulations have been carried out to demonstrate its sensing performance. [ABSTRACT FROM AUTHOR]

Spin transfer torque-based magnetic random access memory (STT-MRAM) is considered as one of the most promising candidates for the next generation of nonvolatile memories; however, its storage density and reliability are currently uncompetitive compared with other nonvolatile memories (e.g., NAND flash). In this paper, a dual-functional memory cell structure, named DFSTT-MRAM, is proposed by stacking multiple magnetic tunnel junctions (MTJs) on top of the CMOS access transistor. In such a structure, the cell can be dynamically configured between two possible functional modes, i.e., high-reliability mode (HR-mode) and multilevel cell mode (MLC-mode), based on the data requirements of the addressed applications. The DFSTT-MRAM cell was electrically modeled based on the perpendicular magnetic anisotropy CoFeB/MgO/CoFeB MTJ integrating the STT stochastic switching behaviors. Transient and Monte Carlo simulations were then performed to evaluate its MLC functionality and reliability performance. Our evaluation results show that the DFSTT-MRAM cell can indeed realize MLC capability providing proper write/read control at the MLC-mode and enhance the intrinsic cell reliability by several orders of magnitude when operating at the HR-mode. This cell structure provides a flexible memory cell design for future advanced applications, such as high-density nonvolatile memory and neuromorphic circuits. [ABSTRACT FROM AUTHOR]

Magnetic tunnel junctions (MTJ) are considered as one of the most promising candidates for the next generation of nonvolatile memories and programmable logic chips. Spin transfer torque (STT) in CoFeB/MgO/CoFeB MTJs with perpendicular magnetic anisotropy (PMA) exhibits noticeable performance enhancements compared to that with In-plane magnetic anisotropy, particularly in terms of thermal stability, critical current for switching, access speed and power consumption. However, the STT switching of MTJ has been revealed stochastic, which results from unavoidable thermal fluctuations of magnetization. This leads to the occurrence of write errors which deeply affects the reliability of hybrid CMOS/MTJ circuits. In this paper, we present the first spice-compact model of CoFeB/MgO/CoFeB structure PMA-MTJ integrating STT stochastic behaviors. Depending on the relative magnitude between the switching current (I) and the critical current (Ico), the STT stochastic behaviors of this PMA-MTJ can be categorized into two regions: Sun model (I>Ico) and Neel-Brown model (I<0.8Ico). The Monte-Carlo simulations for single cell and hybrid CMOS/MTJ circuits show the stochastic behaviors in both writing and sensing operations. This model can be very useful for investigating the reliability issues during the design and simulation before process fabrication. [ABSTRACT FROM AUTHOR]