I. L. Prejbeanu

Thermally assisted switching in exchange-biased storage layer magnetic tunnel junctions

A thermally assisted writing procedure is proposed in a tunnel junction based magnetic random access memory cell. The magnetic layers of the tunnel junction are both exchange-biased with antiferromagnetic layers, the reference layer having a much higher blocking temperature than the storage layer. In the operating mode, a current pulse sent through the junction generates enough heat to raise the temperature of the storage layer above its blocking temperature, without affecting the pinning of the reference layer. The concept is demonstrated here for an isolated junction using an homogeneous external magnetic field.

Probing fast heating in magnetic tunnel junction structures with exchange bias

Heat diffusion in a magnetic tunnel junction (MTJ) having a ferromagnetic/antiferromagnetic free layer is investigated. The MTJ is heated by an electric current pulse of power PHP, flowing through the junction in current perpendicular to the plane (CPP) geometry, via Joule heat dissipation in the tunnel barrier. According to a proposed one-dimensional (1D) model of heat diffusion, when an electric voltage is applied to the MTJ, the free layer experiences a transient temperature regime, characterized by an exponential increase of its temperature TAF with a time constant ?TR, followed by a steady temperature regime characterized by TAF=TRT+?PHP, where TRT is the room temperature and ? is a constant. Magnetic transport measurements of exchange bias HEX acting on the free layer allow the determination of ? and ?TR. The experimental values of ? and ?TR are in agreement with those calculated using the 1D model and an estimation of the MTJ thermodynamic parameters based on the Dulong?Petit and Widemann?Franz laws.

Dynamic heating in submicron size magnetic tunnel junctions with exchange biased storage layer

The dynamic heating of submicron size junctions developed for the thermomagnetic write scheme was investigated. The dependency of the heating power density with the voltage pulse width (Δt) was measured and shows a [1−exp(−Δt∕t0)] phenomenological variation in the investigated range of pulse widths (5ns–1s). For submicron junction areas and constant pulse width, the heating power required for writing decreases as the junction area is reduced. The thermomagnetic write scheme was demonstrated through cycling of the cell between the two low and high resistance levels. This scheme was also shown to be insensitive to stray magnetic fields even in absence of any magnetic shielding of the memory.

Crossover in heating regimes of thermally assisted magnetic memories

This work investigates the tunnel junction heating process for micrometer and submicrometer size junctions to be used in a thermally assisted magnetic random access memories write scheme. The time evolution of the heating process was obtained from experimental measurements and numerical thermal simulations. Simulation results show an initial temperature regime at very short pulse widths associated with the intrinsic heating of the junction (adiabatic regime). In this regime, for the same power density, the temperature increase is independent of the junction area. In the studied geometry, for pulse widths around 1ns and higher an additional heating occurs in the electrical leads (diffusion regime). The write power density is in this case lower for large junction areas. The use of thermal barriers is an effective mean to decrease the power density required for writing and to eliminate its junction area dependence.

Compact model of a three-terminal MRAM device based on Spin Orbit Torque switching

The combination of non-volatility, fast access time and endurance in MRAM technology paves the path toward an universal memory. Although an expanding attention is given to two-terminal Magnetic Tunnel Junctions (MTJ) based on Spin-Transfer Torque (STT) switching as the potential candidate for future memories, its reliability is significantly decreased because of the common writing/reading path. Three-terminal MTJ based on Spin-Orbit Torque (SOT) approach revitalizes the hope of an ultimate MRAM. It represents a pioneering way to triumph over current two-terminal MTJs by separating the reading and the writing path. This paper represents simulation results of the first compact model which describes the SOT-MTJ device based on recently fabricated samples. The model is developed in Verilog-A language, implemented on industrial CAD platform and validated by electrical simulations. Many experimental parameters are included in the model in order to enhance simulation accuracy. Based on simulations results, we show the capability of the model to be efficiently used to design hybrid MTJ/CMOS circuits.

Pulsewidth Dependence of Barrier Breakdown in MgO Magnetic Tunnel Junctions

Dependence of time-dependent dielectric breakdown with pulsewidth was investigated on MgO magnetic tunnel junctions. Barrier breakdown measurements were performed using different pulsewidths, from 10 ns up 1 s. The equivalence of the breakdown results between DC and pulsed voltage experiments was established using the total cumulated pulse time under electrical stress as the relevant parameter. Two different breakdown regimes were identified: the first one corresponding to pulses longer than 100 ns up to the DC limit and a second regime for pulses shorter than 100 ns. The barrier lifetime is increased for pulsewidths lower than 100 ns pulses. The breakdown process was shown to be thermally activated, but the longer lifetime at shorter pulses cannot be simply explained assuming a lower temperature during the pulse application.

MRAM with soft reference layer: In-stack combination of memory and logic functions

This paper describes an original concept of thermally assisted MRAM in which memory and logic functions are combined in the same stack. The memory cell is represented by a magnetic tunnel junction having an exchange biased storage layer and a soft reference layer (called sense layer), replacing the conventional pinned reference layer. The write of the storage layer is ensured by a combination of heating pulses and magnetic stray fields created by the soft reference layer. The read is performed in a self-referenced manner by measuring the resistance variation associated to a field variation. This makes these memories much more tolerant to cell to cell variability. In addition, this stack intrinsically performs both a storage function and a comparison (XOR) function. This device called a Magnetic Logic Unit (MLU™) is particularly suited for security applications and Content Addressable Memories.

High sensitivity magnetic field sensor for spatial applications

A high sensitivity 1D magnetic field sensor is developed for spatial applications, in order to replace the heavy search-coils currently used. This new sensor combines a flux concentrator, biasing coils for field modulation and magnetic tunnel junctions. These three elements are fabricated and independently characterized. Finally, the expected performance of a sensor combining these three elements can be estimated.

Scalability and logic functionalities of TA-MRAMs

This paper describes the working principle of thermally assisted MRAM extending the downsize scalability of MRAM and introducing new functionalities particularly useful for security applications. Ultimately, STT switching assisted by thermally induced anisotropy reorientation can be used in MTJ with perpendicular magnetic anisotropy to obtain ultimate downsize scalability with reduced power consumption.

Noise study of magnetic field sensors based on magnetic tunnel junctions

Low frequency noise has been studied for two types of magnetic field sensors based on magnetic tunnel junctions (MTJ). The first structure, composed of a few large MTJs, is designed for low noise applications; the second one, composed of hundreds of small MTJs, is designed for general purposes. At low frequency, both structures exhibit 1/f noise, but with very different amplitudes. The sensors for general purposes show a much higher noise level compared to the low-noise sensors. However, the sensitivity of the low noise sensors is much smaller compared to the other ones. Thus, the limit of detection, defined as the ratio of noise and sensitivity, turns out to be roughly the same for both technologies. Using the advantages of each sensor could help to design a sensor with an improved limit of detection.