We report fabrication and characterization of double magnetic tunnel junction (DMTJ) magneto-resistive random access memory cells that exhibit characteristic about 2X reduction of switching current compared to single reference layer junctions, but maintain high tunneling magnetoresistance ratio exceeding 120 %, high coercive fields of the free layer of more than 2 kOe for 65 nm cells, and magnetically stable reference layers with pinning fields above 6 kOe. Switching analysis performed for two different relative magnetization orientations of the reference layers shows that the net switching current is the result of combined spin transfer torque effects of the individual reference layers, with tunneling and spin-valve-like contributions adding constructively. Our work shows that efficient reduction of switching current can be achieved in double magnetic tunnel junctions with dual MgO layers where one of the layers has significantly lower resistance-area product to enable high magnetoresistance ratio.

We report device-level damping measurements using spin-torque driven ferromagnetic resonance on perpendicular magnetic random-access memory cells. It is shown that thermal agitation enhances the apparent damping for cells smaller than about 55 nm. The effect is fundamental and does not reflect a true damping increase. In addition to the thermal effect, it is still found that device-level damping is higher than film-level damping and increases with decreasing cell size. This is attributed to edge damage caused by device patterning.

Implementation of spin transfer torque magneto-resistive random access memory (STT-MRAM) in memory chips requires that the write margin of the MRAM cell, defined as the difference between breakdown voltage and write voltage for the specified endurance, write error rate, and write speed, is sufficiently large in order to accommodate resistance variations arising from external chip circuitry. We show that by increasing only the thickness of the MgO cap layer to make its resistance-area product close to that of the main MgO barrier, the write margin can be increased substantially without affecting the thermal stability of the cell.

In this paper, we present a new compact model of threshold switching devices which is suitable for efficient circuit-level simulations. First, a macro model, based on a compact transistor based circuit, was implemented in LTSPICE. Then, a descriptive model was extracted and implemented in MATLAB, which is based on the macro model. This macro model was extended to develop a physical model that describes the processes that occur during the threshold switching. The physical model derived comprises a delay structure with few electrical components adjacent to the second junction. The delay model incorporates an internal state variable, which is crucial to transform the descriptive model into a compact model and to parameterize it in terms of electrical parameters that represent the component's behavior. Finally, we applied our model by fitting measured i-v data of an OTS device manufactured by Western Digital Research.

We investigate experimentally the time evolution of the magnetization reversal in spin-transfer torque driven perpendicular magnetoresistive memory cells in the diameter range of 20 to 65nm. The switching process is characterized by two metrics: the incubation time and the reversal time, where both are stochastic in nature. We find that the average reversal time increases with cell size. The reversal time is dominated by domain wall pinning effects that can last several tens of nanoseconds for the bigger devices. However, even for the smallest cells, we still observe similar behavior albeit with shorter pinning durations of the order of 2 to 3 nanoseconds, indicating that the magnetization reversal is incoherent.

Accurate modeling of locally active dynamics plays a critical role in memristor based neuromorphic circuit design. This paper introduces a physically meaningful locally active device model which is suitable for modeling locally active memristors or threshold switching devices. The proposed model is derived through a 2-transistor-1-resistor (2T1R) circuit which is originally composed of 2 BJTs and a linear bias resistor and is exhibiting negative differential resistance characteristics. To obtain a robust mathematical description with minimal complexity, we introduce an external capacitor into the 2T1R circuit such that it physically governs the respective dynamics. Adopting the Ebers-Moll model for the BJTs and utilizing the external capacitor voltage as the state variable, we precisely derive the differential algebraic set of equations for the 2T1R circuit. The numerical simulation results of the proposed model match very well with the Spice simulation results of the 2T1R circuit. The value of the external capacitor can be further tuned to dominate internal parasitics and control the switching speed of the 2T1R circuit accurately, which is an essential requirement in the design of neuromorphic circuits.

Spin transfer torque magnetoresistive random access memory (STT MRAM) is an alternative to SRAM [1] and Flash [2] , [3] in various embedded applications, as it can provide non-volatility concurrently with low power, high speed operation, and high endurance. In order to become a viable option for replacing DRAM as a stand-alone memory at relevant areal densities, further optimization of MRAM performance is needed, as continued technology scaling and advanced computing systems impose challenging specifications on MRAM in terms of its retention (thermal stability), read-out latencies, and write margins [4] . This talk will present our recent experimental, theoretical, and modeling results on understanding physical mechanisms that affect performance of perpendicular STT-MRAM cells under electrical, magnetic and thermal excitations. In particular, we will report experimental and modeling results that extend of our recent work on electrical self-heating in STT-MRAM [5] to smaller device size, describe our work on optimization of FL materials for improving STT efficiency and thermal stability [6] , present analytical model for calculating energy barrier for domain-wall-mediated magnetization reversal of the perpendicular FL [7] , and show our experimental results which suggest that, contrary to common understanding, fitting the magnetic-field switching probabilities P(H) to a macrospin-reversal model provides approximately correct values of perpendicular magnetic anisotropy field H k on device level.

Amorphous chalcogenide alloys are key materials for data storage and energy scavenging applications due to their large non-linearities in optical and electrical properties as well as low vibrational thermal conductivities. Here, we report on a mechanism to suppress the thermal transport in a representative amorphous chalcogenide system, silicon telluride (SiTe), by nearly an order of magnitude via systematically tailoring the cross-linking network among the atoms. As such, we experimentally demonstrate that in fully dense amorphous SiTe the thermal conductivity can be reduced to as low as 0.10 ± 0.01 W m −1 K −1 for high tellurium content with a density nearly twice that of amorphous silicon. Using ab-initio simulations integrated with lattice dynamics, we attribute the ultralow thermal conductivity of SiTe to the suppressed contribution of extended modes of vibration, namely propagons and diffusons. This leads to a large shift in the mobility edge - a factor of five - towards lower frequency and localization of nearly 42% of the modes. This localization is the result of reductions in coordination number and a transition from over-constrained to under-constrained atomic network.

The hemodynamic, renal, and hematologic responses to fluid resuscitation with four different hydrating solutions (lactated Ringer's and hypertonic salt solutions, with and without albumin) administered in equal quantities were compared in an ovine burn model. Forty-five animals, including a sham group, were studied. The burn (40%, flame) was inflicted under anesthesia, but the animals were then studied while in the awakened state. Fluid resuscitation was begun one hour after the burn. While all animals survived the burn and disclosed reasonable hemodynamic stability throughout the experiment, those that received lactated Ringer's with albumin (LRA) restored their cardiac output to preburn values, by 24 h postinjury demonstrated higher serum albumin and colloid osmotic pressure levels, experienced no electrolyte or acid-base imbalances, and maintained serum osmolality within normal limits. In contrast to the other solutions, LRA did not induce edema in unburned tissues, and seemed optimal for burn resuscitation.

In phase change memory cells, the majority of heat is lost through the electrodes during the programming process, which leads to significant drops in the performance of the memory device. In this Letter, we report on the thermal properties of thin film carbon nitride with a modest electrical resistivity of 5–10 mΩ cm, a low thermal conductivity of 1.47 ± 0.09 W m−1 K–1, and a low interfacial thermal conductance between carbon nitride and phase change material for length scales below 40 nm. The thermally insulating property of carbon nitride makes it a suitable thermal barrier, allowing for less heat loss during Joule heating within the memory unit. We compare the thermal properties of carbon nitride against the commonly used electrodes and insulators such as tungsten and silicon nitride, respectively, to demonstrate the promise of carbon nitride as a potential material candidate for electrode applications in phase change memory devices.

The switching speed of a threshold switching Cr-doped V 2 O 3 device as function of voltage and resistive load is understood by means of a thermal device model. It is found that conditions that result in switching trajectories traveling further from equilibrium result in reduced switching time and energy, providing interesting guidelines for the optimization of the 1S1R memory cell. Secondly, the addition of capacitive loads is investigated. Voltage-controlled oscillatory behavior in the frequency range of 10-30 MHz has been obtained and also properly simulated by our device model. Such nanodevice oscillators are of high interest for new neuromorphic computing paradigms.

In this paper, we study diverse device and chip- level parameters of MRAM and ReRAM to make their storage capacity and power consumption comparable to DDR4 SDRAM under different system read/write workload conditions. Using a parameterized NVM power calculator, our analysis results show that both MRAM and ReRAM become beneficial at high- stress workload cases (Read:Write=60%:20%, 40%:40%) at the cost of downsizing the page size to 2,048 or 1,024 bits. For both low and high- stress workload, the suggested retention time of each 16Gb MRAM and 32Gb ReRAM is >1s and >10s, respectively to compete with 8Gb DDR4 SDRAM in power aspect. The 16Gb MRAM and 32Gb ReRAM show negligible refresh power (<;1mW) when retention time is longer than 100s.

In this work, thin film (down to 10 nm) (V1-xCrx)(2)O-3 Mott-oxide based nano-devices (electrode width down to 120 nm) are fabricated for the first time. The devices show volatile threshold switching and NDR caused by thermal feedback. Fast (< 10 ns) and very stable (< 5% variation) cycle to cycle threshold switching is obtained over 10(12) cycles. Thickness and area dependence of the NDR curves are consistent with uniform volume switching and are explained with a thermal feedback model calibrated to the temperature dependent conductance of the (V1-xCrx)(2)O-3 films, enabling predictions for further scaled device geometries.

An optimization study of heat-assisted magnetic recording (HAMR) media optical and thermal designs is presented. A simplified stack with five component layers is used, which allows a systematic exploration of their optical and thermal properties parameter space. The optimum HAMR media design maximizes the downtrack temperature gradient at a set target write width and for a laser power limited by the near-field transducer (NFT) lifetime. Best absorption results are obtained for plasmonic-like underlayer/heat sink and a recording layer with low refractive index and low extinction coefficient. This combination of optical properties leads to the optimum confinement of the electromagnetic field between the NFT and the lower interface of the recording layer. The optimal optical design is not sensitive to the three NFT designs considered and to the thermal optimization. For the thermal properties optimization, weighted gradient and power are introduced that account for the variations of recording velocity across the disk surface in 7200 rpm hard disk drives. The ideal thermal design contains a strong heat sink with an optimal effective thermal conductance heff between the recording layer and the heat sink. Sufficient thermal anisotropy is required in the recording layer to minimize the lateral heat flow in that layer. The optimum heff also depend on disk velocity and in-plane thermal conductivity in the recording layer. This paper also present the quantitative comparisons of different optical and thermal tradeoffs/variations that might be necessary in practical HAMR media designs.

Recording experiments and decoding algorithms are presented for evaluating the bit error rate (BER) of state-of-the-art magnetic bit-patterned media (BPM). The recording experiments are performed with a static tester and conventional hard-disk drive heads. As the reader dimensions are larger than the bit dimensions in both the down-track and the cross-track directions, a 2-D bit-decoding algorithm is required. Two such algorithms are presented in detail together with the methodology implemented to accurately retrieve island positions during recording. Using these techniques, a 1.6 Td/in 2 magnetic BPM is demonstrated to support 2-D BER below 1e-2 under shingled magnetic recording conditions.

A new technique for characterizing the thermal properties of heat-assisted magnetic recording media (HAMR) using pulsed near-field heating recording with a static tester is reported. It has the advantage over traditional recording techniques of enabling studies of the 1-100 ns transient thermal configurations rather than only thermal steady states. The technique is applied to HAMR media with different heatsink thicknesses. While each media has a different steady-state temperature profile, the heatsinking differences are eliminated or enhanced by tuning the pulse duration for each media. The technique is used to extract thermal properties of the media.

An advanced scanning magnetoresistive microscopy (SMRM) — a robust magnetic imaging and probing technique — will be presented, which utilizes state-of-the-art recording heads of a hard disk drive as sensors. The spatial resolution of modern tunneling magnetoresistive sensors is nowadays comparable to the more commonly used magnetic force microscopes. Important advantages of SMRM are the ability to detect pure magnetic signals directly proportional to the out-of-plane magnetic stray field, negligible sensor stray fields, and the ability to apply local bipolar magnetic field pulses up to 10 kOe with bandwidths from DC up to 1 GHz. Moreover, the SMRM can be further equipped with a heating stage and external magnetic field units. The performance of this method and corresponding best practices are demonstrated by presenting various examples, including a temperature dependent recording study on hard magnetic L10 FeCuPt thin films, imaging of magnetic vortex states in an in-plane magnetic field, and their controlled manipulation by applying local field pulses.

We present a method for growing bit patterned magnetic recording media using directed growth of sputtered granular perpendicular magnetic recording media. The grain nucleation is templated using an epitaxial seed layer, which contains Pt pillars separated by amorphous metal oxide. The scheme enables the creation of both templated data and servo regions suitable for high density hard disk drive operation. We illustrate the importance of using a process that is both topographically and chemically driven to achieve high quality media.

Bit-patterned media (BPM) for magnetic recording provides a route to thermally stable data recording at >1 Tb/in2 and circumvents many of the challenges associated with extending conventional granular media technology. Instead of recording a bit on an ensemble of random grains, BPM comprises a well-ordered array of lithographically patterned isolated magnetic islands, each of which stores 1 bit. Fabrication of BPM is viewed as the greatest challenge for its commercialization. In this paper, we describe a BPM fabrication method that combines rotary-stage e-beam lithography, directed self-assembly of block copolymers, self-aligned double patterning, nanoimprint lithography, and ion milling to generate BPM based on CoCrPt alloy materials at densities up to 1.6 Td/in2. This combination of novel fabrication technologies achieves feature sizes of <;10 nm, which is significantly smaller than what conventional nanofabrication methods used in semiconductor manufacturing can achieve. In contrast to earlier work that used hexagonal arrays of round islands, our latest approach creates BPM with rectangular bit cells, which are advantageous for the integration of BPM with existing hard disk drive technology. The advantages of rectangular bits are analyzed from a theoretical and modeling point of view, and system integration requirements, such as provision of servo patterns, implementation of write synchronization, and providing for a stable head-disk interface, are addressed in the context of experimental results. Optimization of magnetic alloy materials for thermal stability, writeability, and tight switching field distribution is discussed, and a new method for growing BPM islands from a specially patterned underlayer-referred to as templated growth-is presented. New recording results at 1.6 Td/in2 (roughly equivalent to 1.3 Tb/in2) demonstrate a raw error rate <;10-2, which is consistent with the recording system requirements of modern hard drives. Extendibility of BPM to higher densities and its eventual combination with energy-assisted recording are explored.