Search

Mo-based perpendicular magnetic tunnel junctions (Mo-pMTJs) can outperform mainstream Ta-pMTJs in terms of perpendicular magnetic anisotropy (PMA) and thermal tolerance. However, studies on the ultrafast switching of Mo-pMTJ devices remain limited. In addition, although pMTJ devices have potential to function as cryogenic memory cells, there has been no report on the performance of Mo-pMTJs at low temperatures until now. In this Letter, Mo-pMTJs were prepared with strong PMA and patterned into nanoscale devices. Scanning transmission electron microscopy was employed to characterize the device lateral dimension and structure integrity. Systematic probability measurements were conducted under various pulse widths and current densities. On the ultrafast timescale down to sub-ns, the switching is confirmed to enter the precessional regime. The optimization of the switching energy is discussed. Moreover, we investigate the magneto-transport properties and switching of Mo-pMTJs at low temperatures down to 2 K. The feasibility of utilizing Mo-pMTJ devices in cryogenic memory is verified through this work. [ABSTRACT FROM AUTHOR]

In several recent studies, the probabilistic nature of magnetic tunnel junction (MTJ) switching has been utilized rather than suppressed for proposals and demonstrations of specific or novel scenarios or applications, such as true random number generation, neural spike generation, stochastic computing (SC), and probabilistic spin logic. Among all schemes of operations for generating tunable random signals, dual-biasing is very simple but also robust against device variations. When an MTJ is connected to a voltage source with polarity encouraging parallel (P)-to-antiparallel (AP) switching by spin-transfer torque (STT), and under a static bias field that favors P-state, the MTJ can switch back and forth between two states randomly, due to the state-dependent modulation of current by the tunneling magnetoresistance (TMR) of the MTJ. In this work, we demonstrate bipolar random signal generation by connecting two MTJs in series with a voltage source. [ABSTRACT FROM AUTHOR]

Spin-transfer torque magnetic random access memory (STT-MRAM) is a promising solution for onboard, radiation-tolerant memory for space environments. This is because magnetic tunnel junctions (MTJs) have demonstrated outstanding resilience to ionizing radiation. In this study, we examine the effect of total ionizing dose (TID) on the MTJ’s thermal stability factor, intrinsic critical switching voltage, and write energy. These parameters were calculated from switching probability distribution curves taken at three different pulsewidths. Additionally, dc current hysteresis plots were obtained to measure the MTJ’s tunnel magnetoresistance ratio, its dependence on voltage, and the dc current required to switch states. Tests were carried out at two TID exposure levels up to 1 Mrad(Si) on a total of 24 MTJs of varying cross section dimensions in order to identify the influence of scaling as well as to obtain statistical variation for any changes observed. Results indicate that TID exposure had small effects on the MTJ’s thermal stability and critical switching voltage, which affected the write energy, especially for switching from the antiparallel (AP) to parallel (P) direction. For switching from P to AP, these changes fluctuated, suggesting that TID has multiple effects, which have competing influences on the MTJ’s switching properties. The work presented here shows that, while MTJs are highly resistant to ionizing radiation, a subtle influence on some properties for the STT-MRAM write operation should be considered in radiation environment assessments for both terrestrial and space applications. [ABSTRACT FROM AUTHOR]

Recently, magnetic tunnel junctions (MTJs) with superparamagnetic free layers (sMTJs) have been proposed as key components in probabilistic bits (p-bits). However, the average fluctuation rates of sMTJs are very sensitive to slight variations in the device dimensions, which creates a major obstacle to their realization in large-scale networks. One possible solution is to implement dual-biasing on p-bits to offset the effects of variations through onboard corrections. In our previous work, we demonstrated that dual-biasing has the unique capability of separate control over the high- and low-state dwell times, which adds an extra degree of tunability in the signals generated. However, these studies investigated dual-biasing on thermally stable MTJs; therefore, the maximum switching rates were lower than the gigahertz rates desired for sMTJ-based p-bit circuits. While dual-biasing on sMTJs rather than thermally stable MTJs would improve the switching rates, some of the flexibility and robustness of dual-biasing may be sacrificed due to the sensitivity of sMTJs. In this letter, we applied the dual-biasing method on 10 MTJs with varying thermal stability factors to test if sMTJs can achieve the same degree of separation between high- and low-state dwell-time tunability as thermally stable MTJs. Our results show that two degrees of tunability was achieved on all MTJs tested, thus demonstrating that dual-biased p-bits can achieve switching rates equal to or greater than single-biased sMTJs. [ABSTRACT FROM AUTHOR]

Magnetic tunnel junctions (MTJs) with low thermal stability at room temperature have been proposed as read units in beyond CMOS computing architectures including stochastic computing unit and probabilistic-bit (p-bit). Networks of multiple interconnected MTJs may face challenges due to potential device-to-device variations in thermal stability from design targets. Recently, we generated tunable telegraphic signals using a thermally stable MTJ through proper control over an external bias field and a dc voltage bias, where we showed that the average dwell times in the antiparallel (AP) and parallel states could be tuned separately. The implication for this method for p-bit designs is that it allows for p-bits to be compatible with the state-of-the-art magnetoresistive random-access memory (MRAM) technology and introduces a second degree of tunability to the input–output characteristics of the device. In this article, we expand on this method in two important ways. First, we demonstrate the applicability of our method to p-bit designs by modeling the transfer function using the existing p-bit models. Our results indicate that the transfer function can be adjusted with slight modifications to the bias field, which allows for the possibility of p-bit circuits capable of on-chip corrections against device-to-device variations in their thermal stabilities. Second, we identify the physical mechanisms that allow for two degrees of tunability in the output signal, which is explained through the Néel–Brown model. This article provides both applicability and predictability to the dual-biasing method. [ABSTRACT FROM AUTHOR]