Resistive Switching Memory Research Articles

Halide Perovskites for Resistive Switching Memory.

Resistive switching random access memory (RRAM), also known as memristor, is regarded as an emerging nonvolatile memory and computing-in-memory technology to address the intrinsic physical limitations of conventional memory and the bottleneck of von Neumann architecture. In particular, halide perovskite RRAMs have attracted widespread attention in recent years because of their ionic migration nature and excellent photoelectric properties. This Perspective first provides a condensed overview of halide perovskite RRAMs based on materials, device performance, switching mechanism, and potential applications. Moreover, this Perspective attempts to detail the challenges, such as the quality of halide perovskite films, the compatible processing of device fabrication, the reliability of memory performance, and clarification of the switching mechanism, and further discusses how the outstanding challenges of halide perovskite RRAMs could be met in future research.

Resistive Memory Devices Based on Reticular Materials for Electrical Information Storage.

Recently, reticular materials, such as metal-organic frameworks and covalent organic frameworks, have been proposed as an active insulating layer in resistive switching memory systems through their chemically tunable porous structure. A resistive random access memory (RRAM) cell, a digital memristor, is one of the most outstanding emergent memory devices that achieves high-density electrical information storage with variable electrical resistance states between two terminals. The overall design of the RRAM devices comprises an insulating layer sandwiched between two metal electrodes (metal/insulator/metal). RRAM devices with fast switching speeds and enhanced storage density have the potential to be manufactured with excellent scalability owing to their relatively simple device architecture. In this review, recent progress on the development of reticular material-based RRAM devices and the study of their operational mechanisms are reviewed, and new challenges and future perspectives related to reticular material-based RRAM are discussed.

Quantum conductors formation and resistive switching memory effects in zirconia nanotubes

The prospects of the development of non-volatile memory elements that involve memristive metal-dielectric-metal sandwich structures are due to the possibility of reliably implementing sustained functional states with quantized conductance. In the present paper, we have explored the properties of Zr/ZrO2/Au memristors fabricated based on an anodic zirconia layer that consists of an ordered array of vertically oriented non-stoichiometric nanotubes with an outer diameter of 30 nm. The operational stability of the designed memory devices has been analyzed in unipolar and bipolar resistive switching modes. The resistance ratio ≥105 between high-resistance (HRS) and low-resistance (LRS) states has been evaluated. It has been found that the LRS conductivity is quantized over a wide range with a fundamental minimum of 0.5G 0 = 38.74 μS due to the formation of quantum conductors based on oxygen vacancies (VO). For Zr/ZrO2/Au memristors, resistive switching mechanisms to be sensitive to the migration of VO in an applied electric field have been proposed. It has been shown that the ohmic type and space-charge-limited conductivities are realized in the LRS and HRS, respectively. Besides, we have offered a brief review of parameters for functional metal/zirconia/metal nanolayered structures to create effective memristors with multiple resistive states and a high resistance ratio.

Multilevel switching memristor by compliance current adjustment for off-chip training of neuromorphic system

Multilevel operation is one of the most essential properties for synaptic devices to realize hardware artificial neural networks. Compliance current (Icc) adjustment is a multilevel programming method that can be utilized for a large-scale one-transistor and one-resistor (1T1R) array. It protects the devices from permanent breakdown by regulating abrupt switching. However, according to the reported literature so far, the number of conductance states in the Icc control method is insufficient to implement off-chip-trained neuromorphic systems. Therefore, we experimentally explore the feasibility of a larger number of conductance states using the Icc control method. We fabricated an Al2O3/TiOx-based resistive switching memory array, observed the conductance change while increasing Icc during set operations, and 64-level conductance states were statistically demonstrated. Furthermore, we verified that the 64-level states showed recognition performance close to that of a software-based neural network through off-chip learning of the convolutional neural network (CNN) structure. The fabricated synaptic device array with the Icc-control programming method is expected to contribute to the development of hardware neural network by reducing the information loss in the transfer process.

Percolation theory based model of conduction mechanism and characteristic contradiction in ZnO RRAM

Resistive random access memory (RRAM) has been intensively investigated for nearly two decades. However, RRAM has not been applied widely in the market because of the poor characteristics, such as reliability and uniformity, which could be improved by the accurate comprehension of the mechanism. In this paper, a model based on percolation theory is proposed to simulate the I-V characteristics of ZnO resistive switching memory. It demonstrates that three different conductions of space charge limited current, Poole–Frenkel effect, and thermionic emission are determined by the relationship between the oxygen vacancy concentration and the bias. Furthermore, this model well explains the effect of conductive filaments' diameter and compliance current on the I-V characteristics of ZnO resistive switching memory, which demonstrates the rationality of the percolation model.

Design of Materials Configuration for Optimizing Redox-Based Resistive Switching Memories.

Redox-based resistive random access memories (ReRAMs) are based on electrochemical processes of oxidation and reduction within the devices. The selection of materials and material combinations strongly influence the related nanoscale processes, playing a crucial role in resistive switching properties and functionalities. To date, however, comprehensive studies on device design accounting for a combination of factors such as electrodes, electrolytes, and capping layer materials related to their thicknesses and interactions are scarce. In this work, the impact of materials' configuration on interfacial redox reactions in HfO2 -based electrochemical metallization memory (ECM) and valence-change memory (VCM) systems is reported. The redox processes are studied by cyclic voltammetry, and the corresponding resistive switching characteristics are investigated. In ECM cells, the overall cell resistance depends on the electrocatalytic activity of the counter electrode. Nonetheless, the capping layer material further influences the cell resistance and the SET and RESET voltages. In VCM systems, the influence of the electrode material configuration is also pronounced, and is capable of modulating the active resistive switching interface. For both types of memory cells, the switching behavior changes significantly with variation of the oxide thickness. The results present important materials selection criteria for rationale design of ReRAM cells for various memristive applications.

Electrodeposition of GeSbTe-Based Resistive Switching Memory in Crossbar Arrays

In this work, we report on the fabrication of resistive random-access memory cells based on electrodeposited GeSbTe material between TiN top and bottom electrodes in a crossbar architecture. The cells exhibit asymmetric bipolar resistive switching characteristics under the same SET and RESET compliance current (CC), showing highly uniform and reproducible switching properties. A multi-state switching behavior can be also achieved by varying the sweeping voltage and CC. Unlike phase-change switching, the switching between the high-resistance state and the low-resistance state in these cells can be attributed to the formation and rupture of conductive Te bridge(s) within the Te-rich GeSbTe matrix upon application of a high electric field. The results point toward the usage of the electrodeposition method to fabricate advanced functional device structures for application in non-volatile memory.

Switching-Modulated Phase Change Memory Realized by Si-Containing Block Copolymers.

The phase change memory (PCM) is one of the key enabling memory technologies for next-generation non-volatile memory device applications due to its high writing speed, excellent endurance, long retention time, and good scalability. However, the high power consumption of PCM devices caused by the high switching current from a high resistive state to a low resistive state is a critical obstacle to be resolved before widespread commercialization can be realized. Here, a useful approach to reduce the writing current of PCM, which depends strongly on the contact area between the heater electrode and active layer, by employing self-assembly process of Si-containing block copolymers (BCPs) is presented. Self-assembled insulative BCP pattern geometries can locally block the current path of the contact between a high resistive film (TiN) and a phase-change material (Ge2 Sb2 Te5 ), resulting in a significant reduction of the writing current. Compared to a conventional PCM cell, the BCP-modified PCM shows excellent switching power reduction up to 1/20 given its use of self-assembled hybrid SiFex Oy /SiOx dot-in-hole nanostructures. This BCP-based bottom-up process can be extended to various applications of other non-volatile memory devices, such as resistive switching memory and magnetic storage devices.

Low‐Frequency‐Noise Spectroscopy of TaOx‐based Resistive Switching Memory

AbstractResearch and development into resistive switching memories, such as resistive random access memory (ReRAM) and memristors, are being actively promoted toward the realization of new computing techniques. To improve the reliability of these devices, it is important to clarify their resistance characteristics. Low‐frequency‐noise spectroscopy (LFNS) is a powerful method for investigating the nature of traps in conduction paths, even at the nanoscale. In this article, the results of LFNS measurements on a TaOx‐based ReRAM device are presented. The temperature dependence of the noise spectra at given frequencies reveals the existence of multiple trap levels, which are observed over a wide range of resistance values. These experimental results show that the ReRAM device is particularly advantageous when used in analog resistive switching applications.

A Physically Transient Self-Rectifying and Analogue Switching Memristor Synapse

A physical transient memristor synapse with self-rectifying and analogue switching behaviors based on Mo/MgO/AZO (Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> 2 wt %, ZnO 98 wt %)/W is presented. By modulating thickness of MgO layer, the device with 5 nm MgO insulator layer shows stable resistive switching memory with a rectification ratio up to 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> , and precise tuning synaptic functions. Meanwhile, the formation of Schottky barrier and oxygen vacancy conductive filament is proposed to further explain self-rectifying and analogue switching behaviors. Additionally, the electrical characteristics of the Mo/MgO/AZO/W devices degraded after being immersed in deionized water (DI) for 3 min, which demonstrated the physically transient features successfully. The physical transient self-rectifying memristors have great potential for applications in secure neuromorphic computing systems, and bio-integrated electronics.

Logic-in-memory application of CMOS compatible silicon nitride memristor

Resistive switching memory (RRAM) performs its role as a next-generation memory that overcomes the limitations of von Neumann computing. RRAM can be stated as set or reset state by applying a certain voltage and has advantages in maintaining the states with non-volatile characteristics. That is, both the change of state and storage of operations occur in a single RRAM device. Both bipolar RRAM which operates on two opposite voltage polarity, and unipolar RRAM, which operates on a single polarity, can be combined with the switch to effectively implement the Boolean function. In this paper, both unipolar and bipolar characteristics are conducted in CMOS compatible Ni/SiN/Si RRAM devices. We succeed in demonstrating 14 logical operators out of 16 by applying pulses to implement the Boolean function. In addition, we prove that the other two logical operators cannot be represented by the existing system, and we also suggest ideas to solve this problem.

Resistance Switching Behavior of a Perhydropolysilazane-Derived SiOx-Based Memristor.

SiOx is an important dielectric material layer for resistive switching memory due to its compatibility with complementary metal-oxide semiconductor (CMOS) technology. Here we propose a solution process for a SiOx dielectric layer based on perhydropolysilazane (PHPS). A series of SiOx layers with different compositions are prepared by controlling the conversion process from PHPS, then the resistance switching behaviors of typical Ag/SiOx/Au memristors are analyzed. The effect of oxygen vacancies and Si-OH groups on the formation and rupture of Ag conducting filaments (CFs) in the SiOx layer was thoroughly investigated. Ultimately, we achieved a high-performance memristor with a coefficient of variation (σ/μ) as low as 0.16 ± 0.08 and an on/off ratio as high as 106, which can rival the performance of the SiOx memristors based on the high-vacuum and high-cost vapor deposition methods. These findings demonstrate the high promise of the PHPS-derived SiOx dielectric layer in the field of memristors.

Combinatorial Screening of Anodic Memristors Based on Hafnium-tantalum Mixed Oxides

Among valve metals, Ta and Hf are the most studied high-κ dielectric materials for development of the next-generation non-volatile electronic components, commonly known as memristors. Redox-based resistive switching memories (ReRAMs) and artificial neural networks are applications examples of metal-insulator-metal memory types. An insulating layer, interposed between the bottom and the top conducting electrodes, is the determining factor in memristive performance. From this aspect, oxidized forms of Hf and Ta are promising candidates to be used as solid-state electrolytes, allowing conductive filaments formation (CFs) inside the oxide layer. There is a strongly motivated scientific interest for the implementation of anodic oxides in memristive devices due to its simple, cost-efficient and controllable fabrication. However, properties of anodic memristors based on Hf and Ta are not yet fully investigated and several parameters still have to be examined in detail. Herein, influence of the electrolyte selection on Hf and Ta anodic memristors was studied. Consequently, anodization parameters were optimized and applied for the fabrication of memristors based on Hf-Ta alloys spread across a thin film combinatorial library. Parent metals and their mixtures served as bottom electrodes and, identical to the case of Pt top electrode, they were deposited by sputtering. The total compositional spread of Hf-Ta library was ranging from 5 to 92 at.% Hf and was distributed across three Si wafers. The memristive properties of devices based on mixed anodic oxides were screened with a resolution of 1 at.%, thus offering a comparative investigation of a high number of compositionally different devices. Generally, the ratios between resistance states for Ta and Hf memristors were approximately in the same range, as well as their number of switching voltage levels. Nevertheless, in both cases memory and electrical characteristics were dictated by the incorporation of electrolyte species, which was confirmed by XPS. Therefore, for devices based on pure metals, the memristive behavior varied only as depending on the electrochemical conditions used for device fabrication. However, no electrolyte incorporation was observed in case of devices based on Hf-Ta library. Memristive properties could have been related only to different compositions which were grouped in three zones containing alloys showing similar behaviors. An improvement of resistance states ratios in some zones by orders of magnitude was achieved, as compared to ratios obtained for devices based on pure metals. Additionally, variations between the zones of unipolar and bipolar switching behavior were observed. HRTEM and hard X-ray XPS were done in order to compare the CFs within the mixed oxide layers for memristors from different zones. Clearly, the novelty of this work is the found spectra of different memristive properties for anodic Hf and Ta memristors and anodic memristors based on numerous Hf-Ta alloys. Such approach is offering the possibility of better understanding whether the switching mechanism is based on electrolyte species, oxygen vacancies, metallic species movement or formation of additional vacancies by redox reactions or thermal field for each specific composition or electrochemical parameters. Conclusively, this may be an important step towards defect engineered memristors fabrication, as intrinsically offered by the anodic process of memristive formation.

Proton Transport in Hydrated GdOx Measured with GdCo Magnetic Sensing Layer

Solid state ionic materials such as metal oxide ceramics are of interest due to the wide variety of properties that can be controlled through voltage control of ion concentration and transport in these materials, e.g. electrical resistivity, optical properties, and magnetic properties. Ion movement through a solid oxide electrolyte film can be controlled very simply by electrical gating of the oxide, i.e. applying a voltage across the film. The ease of processing and CMOS compatibility of metal oxides makes them strong candidate materials for next-generation devices for information storage and computing, sensing, energy generation (fuel cells), and energy storage (batteries).Hydrogen ions (protons) take the form of (OH)O • defects in oxides and move by hopping between oxygen sites (Grotthuss mechanism); their small size allows for high conductivity.8 Protonic conduction in oxides has been studied in solid oxide fuel cells (SOFCs)1; proton incorporation has also been shown to change the optical4, electrical2,3, and magnetic properties of the oxide5, or a combination of multiple properties e.g. through inducing phase changes in strontium cobalt oxide (SCO)6. However, current work focuses on materials that require high-temperature processing and operation (>200C, or up to 600-1000C for SOFCs), and bulk or microscale films. This work aims to take solid oxide protonic devices down to the ultrathin (10-100nm) scale, to introduce simpler fabrication techniques (sputtering), and to achieve room-temperature operation of ionic devices. We investigate gadolinium oxide (GdOx) as a model material due to its hygroscopic nature, which allows for fast room-temperature proton conduction.7,8 Using variations of a Pt/Co/GdOx/Au device structure, a gate voltage of 2-5V is applied, initiating a hydrolysis reaction at the top electrode and allowing water to be incorporated into the film. The electric field within the GdOx drives protons to the bottom interface, where it can diffuse into the adjacent magnetic layer and affect a variety of properties. Thin film ferromagnetic materials such as cobalt have been shown to undergo dramatic changes in their magnetic anisotropy with voltage gating, due to the incorporation and removal of hydrogen at the interface with the electrolyte and within the magnetic layer.6 Control of the compensation temperature of a thin film ferrimagnet and control of exchange bias in an antiferromagnet have also been demonstrated with this magneto-ionic proton pump mechanism.9,10 Using a gadolinium cobalt (GdCo) magnetic layer inserted into the MOM device stack to detect the presence of hydrogen, we have developed a sensor for use in a novel time-domain proton transport measurement. We apply short voltage pulses to the device and use magneto-optic Kerr effect (MOKE) to observe the resulting changes in the compensation temperature and magnetic state of the sensor layer. This can be used as a more direct measure of the presence of protons at the bottom interface, as compared to frequency domain measurements such as electrochemical impedance spectroscopy (EIS). We also show that humidity affects the proton transport in the oxide layer, using neutron and x-ray reflectometry to observe the growth of a gadolinium hydroxide (Gd(OH)3) layer with high hydrogen concentration upon exposing the film to a high humidity environment.11 References Q. Minh, M. B. Mogensen. Reversible Solid Oxide Fuel Cell Technology for Green Fuel and Power Production. Electrochem. Soc. Interface, 55–62 (2013).Huang et al. Three-terminal resistive switch based on metal/metal oxide redox reactions, Scientific Reports, 7 (2017).Messerschmitt, F. Kubicek, M., Rupp, J. L. M. How Does Moisture Affect the Physical Property of Memristance for Anionic–Electronic Resistive Switching Memories?, Advanced Functional Materials, 25, 5117-5125 (2015).Huang, M., Jun Tan, A., Büttner, F. et al. Voltage-gated optics and plasmonics enabled by solid-state proton pumping. Nat Commun 10, 5030 (2019).Tan et al. Magneto-ionic control of magnetism using a solid-state proton pump, Nature Materials, 17 (2018).Lu, N. et al. Electric-field control of tri-state phase transformation with a selective dual-ion switch, Nature 546, 124–128, (2017)Norbya, T. Proton conduction in oxides, Solid State lonics, 40/41, 857-862 (1990).Larring and Norby. Protons in rare earth oxides, Solid State Ionics, 77 (1995).Huang, M., Hasan, M.U., Klyukin, K. et al. Voltage control of ferrimagnetic order and voltage-assisted writing of ferrimagnetic spin textures. Nanotechnol. 16, 981–988 (2021).Zehner, J. et al. Magnetoionic control of perpendicular exchange bias. Phys. Rev. Materials 5, L061401 (2021).Tan, A.J., Huang, M., Sheffels, S., et al. Hydration of gadolinium oxide (GdOx) and its effect on voltage-induced Co oxidation in a Pt/Co/GdOx/Au heterostructure. Phys. Rev. Materials 3, 064408 (2019). Figure 1

Nonvolatile resistive switching memory based on monosaccharide fructose film

In this paper, we report resistive random access memory (RRAM) based on a monosaccharide—fructose for nonvolatile memory in biocompatible and “green” electronics. Fructose thin film acts as the resistive switching layer with Al and Ag top electrodes for comparison. Both devices demonstrated highly reproducible nonvolatile bipolar resistive switching behaviors with a large on/off ratio of ∼106 for the Al electrode and ∼105 for the Ag electrode. The forming voltage, set voltage, and memory window are also larger for the Al electrode than the Ag electrode, but the reset voltages are comparable. Dominant conduction mechanisms of fructose films were proposed. At a high resistance state, both electrodes reveal space charge limited conduction, while at a low resistance state, the governing mechanism is Ohm's law, and in addition, the Ag electrode also shows trap-fill limited conduction when approaching the reset voltage. This observation has yet to be reported in RRAM based on natural bio-organic materials.

Low-Frequency 1/f Noise Characteristics of Ultra-Thin AlOx-Based Resistive Switching Memory Devices with Magneto-Resistive Responses

Low-frequency 1/f voltage noise has been employed to probe stochastic charge dynamics in AlOx-based non-volatile resistive memory devices exhibiting both resistive switching (RS) and magneto-resistive (MR) effects. A 1/fγ noise power spectral density is observed in a wide range of applied voltage biases. By analyzing the experimental data within the framework of Hooge’s empirical relation, we found that the Hooge’s parameter α and the exponent γ exhibit a distinct variation upon the resistance transition from the low resistance state (LRS) to the high resistance state (HRS), providing strong evidence that the electron trapping/de-trapping process, along with the electric field-driven oxygen vacancy migration in the AlOx barrier, plays an essential role in the charge transport dynamics of AlOx-based RS memory devices.

Influence of aging on the resistive switching behavior of epitaxial strontium titanate based heterostructures

In this work, we investigated the impact of aging on the resistive switching behavior of high quality epitaxial heterostructure resistive switching devices [Pt/SrTiO3/LaNiO3/LaAlO3 (001) with SrTiO3 (STO) as switching layer, Pt as top electrode and LaNiO3 as bottom electrode]. For this study, the resistive switching memory devices of two different ultrathin switching layer (STO) thicknesses (5 and 10 nm) were fabricated with different shape and size of top electrode (Pt) i.e. 20 µm (circular shape) and 50 µm (square shaped). Subsequently, their resistive switching behavior at different instant of time i.e. freshly prepared, 4 months and 16 months were studied. With long aging, we find a decrease in the positive bais voltages for which current start rising exponentially (set voltages) for most of the devices besides emergence of various features such as multiple crossovers, noisy/spike region during set process and openness in the switching hysteresis loop. The existence of these behaviors is attributed to the fact that the moisture permeates through the interfacial region between top electrode (Pt) and switching layer (STO) irrespective of the thickness of switching layer and dimensions of top electrode.

In materia reservoir computing with a fully memristive architecture based on self-organizing nanowire networks.

Neuromorphic computing aims at the realization of intelligent systems able to process information similarly to our brain. Brain-inspired computing paradigms have been implemented in crossbar arrays of memristive devices; however, this approach does not emulate the topology and the emergent behaviour of biological neuronal circuits, where the principle of self-organization regulates both structure and function. Here, we report on in materia reservoir computing in a fully memristive architecture based on self-organized nanowire networks. Thanks to the functional synaptic connectivity with nonlinear dynamics and fading memory properties, the designless nanowire complex network acts as a network-wide physical reservoir able to map spatio-temporal inputs into a feature space that can be analysed by a memristive resistive switching memory read-out layer. Computing capabilities, including recognition of spatio-temporal patterns and time-series prediction, show that the emergent memristive behaviour of nanowire networks allows in materia implementation of brain-inspired computing paradigms characterized by a reduced training cost.

(Invited) Nanoscale Probing of Field-Driven Ion Migration in TaOx for Neuromorphic Memristor Applications

Stable operations as resistive switching memory were demonstrated in amorphous TaOx (a-TaOx) thin films with very flat surfaces by conductive atomic force microscopy (c-AFM). The a-TaOx thin films fabricated on glass and Nb-doped SrTiO3 substrates by pulsed laser deposition showed high surface flatness with root-mean-square roughness of less than 0.2 nm. The c-AFM observations on the surfaces revealed that the resistive switching in a-TaOx causes almost no change in the topographic structures, and the significant structural deformation appears after the electrical breakdown by longer-range migration of the constituent ions. The possible mechanisms of the resistive switching phenomena were discussed based on the changes in the topographic structures and conduction states.

Effect of Graphene Composite on Biological Egg Protein Resistance Switching Memory Properties

In this work, a resistive switching memory device was fabricated based on egg protein, a natural biomaterial. The effect of graphene composite on the resistive switching characteristics of the device was investigated. The experimental results show that both pure egg protein and graphene composite devices exhibit bipolar nonvolatile resistive conversion properties. Both devices have good data retention capability. Furthermore, the composite of graphene can effectively improve the device endurance and the consistency of the on-state current distribution of the device. Based on the theory of capture and de-capture of charge carrier, the mechanism of resistive switching is analyzed.