Conductive Bridge Random Access Memory Research Articles

Controllable multilevel quantized conduction states in vertical CBRAM using Cu-nanodot ion source

Restricting the injection of cations is crucial for implementing precise quantized conduction (QC) during the multi-level operation of conductive-bridge random-access memory (CBRAM). This study proposes a method that controls ion supply by confining the Cu ion source to 0D in a vertical structure. This confinement enables sophisticated filament control for multilevel operation. Cu nanodots are formed between the W electrodes, with W and Cu serving as the electron and ion sources for the conducting filament, respectively. When the Cu filament is confined to 0D, the controllability of the QC implementation is better than that in cases where the filament is restricted to 1D or bulk Cu. The highest number of quantized levels was observed for 0D Cu, which can be attributed to the synergistic effects of filament confinement and the decrease in the Cu electrochemical reaction rate. Furthermore, we analyzed the switching mechanism of the Cu nanodots by employing the activation energy extracted by the slope of the voltage-time dilemma. Our results demonstrate the effectiveness of confining the ion source to 0D for achieving precise filament control in CBRAM and enabling its applicability to vertical structures.

Vertical‐Switching Conductive Bridge Random Access Memory with Adjustable Tunnel Gap and Improved Switching Uniformity Using 2D Electron Gas

AbstractOwing to the high reactivity and diffusivity of Ag and Cu ions, controlling the atomic filament formation and rupture processes in conductive bridge random‐access memory (CBRAM) is challenging. In this study, it is demonstrated that by using a 2D electron gas (2DEG) as the bottom electrode (BE) in a vertical‐switching CBRAM (V‐CBRAM), filament formation and rupture can be effectively managed and the tunnel gap distance created by partial filament formation can be adjusted. The 2DEG BE induces partial filament formation by limiting the number of electrons required for this process in the V‐CBRAM device, as verified via current fitting to the quantum point contact model. Varying the electron concentration and activation energy for electrons trapped in the 2DEG, when paired with various programming voltages, leads to transitions in the device resistance state via changes in the distance of the tunnel gap. This tunnel‐gap‐tunable 2DEG V‐CBRAM device, which exhibits superior switching uniformity, can be employed for nonvolatile memory applications in the sub‐G0 conductance regime, such as 3‐bit multilevel cells and selector‐less memory.

Modulating conductive filaments via thermally stable bilayer organic memristor

The basic unit of information in conductive bridge random access memory based on the redox mechanism of metal ions is physically stored in a conductive filament (CF). Therefore, the overall performance of the device is indissolubly related to the properties of such CFs, as unreliable performance often originates from unstable CFs behavior. However, accurately controlling the dissolution of CFs during device operation can be challenging due to their non-uniformity. This paper proposes a type of memristor based on a solid polymer electrolyte with a polyvinylpyrrolidone/polyvinyl alcohol composite layer structure. The properties of the composite layer are employed to regulate the number of CFs and the growth/fracture location, while the damage to the device by Joule heat is prevented. The device exhibits low operating voltage (0.5 V), stable switching conditions (2000 cycles), and a long holdup time (>3 × 104 s).

Impact of inert electrode on the volatility and non-volatility switching behavior of SiO2-based conductive bridge random access memory devices

Impact of inert electrode on the volatility and non-volatility switching behavior of SiO2-based conductive bridge random access memory devices

Improvement of volatile switching in scaled silicon nanofin memristor for high performance and efficient reservoir computing.

Conductive-bridge random access memory can be used as a physical reservoir for temporal learning in reservoir computing owing to its volatile nature. Herein, a scaled Cu/HfOx/n+-Si memristor was fabricated and characterized for reservoir computing. The scaled, silicon nanofin bottom electrode formation is verified by scanning electron and transmission electron microscopy. The scaled device shows better cycle-to-cycle switching variability characteristics compared with those of large-sized cells. In addition, synaptic characteristics such as conductance changes due to pulses, paired-pulse facilitation, and excitatory postsynaptic currents are confirmed in the scaled memristor. High-pattern accuracy is demonstrated by deep neural networks applied in neuromorphic systems in conjunction with the use of the Modified National Institute of Standards and Technology database. Furthermore, a reservoir computing system is introduced with six different states attained by adjusting the amplitude of the input pulse. Finally, high-performance and efficient volatile reservoir computing in the scaled device is demonstrated by conductance control and system-level reservoir computing simulations.

Improvement of synaptic property of GeSe based CBRAM by engineering the deposition condition of switching matrix

A study of effect of deposition condition of germanium selenide (GeSe) switching layer to Conductive bridging random access memory (CBRAM) device property was conducted to enhance memory property and synaptic property. The devices deposited under the varied deposition conditions were analyzed by current distribution of each state (Low-resistance state [LRS], High-resistance state [HRS]), conduction mechanism analysis, chemical analysis, and synaptic analysis according to pulse conditions. When we deposited switching layer at lowest power (40 W), more Ag ions move to the switching matrix and make lower operation voltage. But there were some disadvantages, which were worst distribution and higher HRS current, less memory window, nonlinear and rapid increase of potentiation process. Conversely, although the device deposited at high voltage increased the operating voltage, but there are improvements, such as the highest memory window and improved linearity of potentiation. The best sample was conducted the synaptic property measurement, and the device successfully mimicked the biological synaptic property with the good linearity of potentiation and depression, spike-rate dependent plasticity (SRDP) and spike-timing dependent plasticity (STDP) properties.

Realizing reliable linearity and forming-free property in conductive bridging random access memory synapse by alloy electrode engineering

The linearity of conductance modulation of the artificial synapse severely restricts the recognition accuracy and the convergence rate in the learning of artificial neural networks. In this work, by alloy electrode engineering, a Ti–Ag device gained the forming-free property because Ag ions were promoted to migrate into the GeTeOx layer to form a thicker conductive filament. This facilitated a uniform change in conductance with the pulse number, and the alloy synapse achieved a significant improvement in linearity (350%), which demonstrated its enhancement in recognition accuracy. To further validate its potential as a comprehensive artificial synapse, the multi-essential synaptic behaviors, including spike-timing-dependent plasticity, spike-rate-dependent plasticity, paired-pulse facilitation, post-tetanic potentiation, and excitatory post-synaptic current, were achieved successfully. This work proposes a promising approach to enhance the performance of conductive bridging random access memory synaptic devices, which benefits the hardware implementation of neuromorphic systems.

Kinetic Monte Carlo simulations on electroforming in nanomanipulated conductive bridge random access memory devices.

Conductive bridge random access memory (CBRAM) devices exhibit great potential as the next-generation nonvolatile memory devices. However, they suffer from two major disadvantages, namely relatively high power consumption and large cycle-to-cycle and device-to-device variations, which hinder their more extensive commercialization. To learn how to enhance their device performance, kinetic Monte Carlo (KMC) simulations were employed to illustrate the variation of electroforming processes in nanomanipulated CBRAM devices by introducing an ion-blocking layer with scalable nanopores and tuning the microstructures of dielectric layers. Both the size of nanopores and the inhomogeneity of dielectric layers have significant impacts on the forming processes of conductive filaments. The dielectric layer with a high-content loose texture plus the scalable nanopore-containing ion-blocking layer leads to the formation of size-controlled and uniform filaments, which remarkably contributes to miniaturizable and stable CBRAM devices. Our study provides insights into nanomanipulation strategies to realize high-performance CBRAM devices, still awaiting future experimental confirmation.

A Wideband Digital Attenuator Based on Conductive Bridging Random Access Memory Switches for RF System-in-Package

A Wideband Digital Attenuator Based on Conductive Bridging Random Access Memory Switches for RF System-in-Package

Bi2O2Se-based CBRAM integrated artificial synapse

Integrating two-dimensional (2D) semiconducting materials into memristor structures has paved the way for emerging 2D materials to be employed in a vast field of memory applications. Bismuth oxyselenide (Bi2O2Se), a 2D material with high electron mobility, has attracted significant research interest owing to its great potential in various fields of advanced applications. Here, we explore the out-of-plane intrinsic switching behavior of few-layered Bi2O2Se via a cross point device for application in conductive bridge random access memory (CBRAM) and artificial synapses for neuromorphic computing. Via state-of-the-art methods, CVD-grown Bi2O2Se nanoplate is applied as a switching material (SM) in an Al/Cu/Bi2O2Se/Pd CBRAM structure. The device exhibits ∼90 consecutive DC cycles with a tight distribution of the SET/RESET voltages under a compliance current (CC) of 1 mA, a retention of over 10 ks, and multilevel switching characteristics showing four distinct states at Vread values of 0.1, 0.2, 0.25, and 0.3 V. Moreover, an artificial synapse is realized with potentiation and depression by modulating the conductance. The switching mechanism is explained via Cu migration through Bi2O2Se based on HRTEM analysis. The present structure shows potential for future integrated memory applications.

Self-Assembled Au Nanoelectrodes: Enabling Low-Threshold-Voltage HfO2-Based Artificial Neurons.

Filamentary-type resistive switching devices, such as conductive bridge random-access memory and valence change memory, have diverse applications in memory and neuromorphic computing. However, the randomness in filament formation poses challenges to device reliability and uniformity. To overcome this issue, various defect engineering methods have been explored, including doping, metal nanoparticle embedding, and extended defect utilization. In this study, we present a simple and effective approach using self-assembled uniform Au nanoelectrodes to controll filament formation in HfO2 resistive switching devices. By concentrating the electric field near the Au nanoelectrodes within the BaTiO3 matrix, we significantly enhanced the device stability and reduced the threshold voltage by up to 45% in HfO2-based artificial neurons compared to the control devices. The threshold voltage reduction is attributed to the uniformly distributed Au nanoelectrodes in the insulating matrix, as confirmed by COMSOL simulation. Our findings highlight the potential of nanostructure design for precise control of filamentary-type resistive switching devices.

Enhanced resistive switching characteristics of conductive bridging memory device by a Co–Cu alloy electrode

One of the main challenges in the development of conductive bridging random access memory (CBRAM) is the large stochastic nature of ion movement that ultimately leads to large parameter variability. In this study, the resistive switching variability of CBRAM devices is significantly improved by employing Co–Cu alloy as the active electrode. By comparing with Pt/Ta2O5/Co devices, the Co70Cu30 alloy exhibited lower forming voltage (<2 V), lower SET voltage (<0.70 V), and faster response time (∼70 ns). The filament stability indicated by the distribution of SET/RESET voltage and high resistance state/low resistance state variation was significantly improved. Our experimental results suggest the formation of Co filaments, and the proposed mechanism is governed by the galvanic effect. In addition, a comparison between Co70Cu30 and Co30Cu70 alloys highlights that the relative proportion between Co and Cu plays an essential role in the device performance. A physical model based on different electrochemical activities of the alloys has been proposed to explain the filament formation and the improved switching uniformity in the Co70Cu30 alloy. This study not only develops a CBRAM with enhanced performance but also advances the implementation of suitable alloy systems for the application of such devices.

Transition of short-term to long-term memory of Cu/TaOx/CNT conductive bridge random access memory for neuromorphic engineering

This work presents the resistive switching characteristics of the TaOx-based conductive-bridge random-access memory (CBRAM) for neuromorphic engineering. Controlling the Cu filament inside the TaOx film allows the device to operate as both volatile and nonvolatile memory. For volatile switching induced by a lower compliance current (Icc), a threshold switching operation is observed. Upon completion of the set process, the retention and current decay were observed, suggesting that the device has the potential for short-term memory applications. Increasing Icc enables the CBRAM to act as a memory-switching device, as confirmed by the lengthy retention time of up to 104 s. Additionally, short-term memory (STM) and long-term memory (LTM) of the device were demonstrated by time-dependent memory decay, where the various magnitude differences of the time-dependent operations. STM was identified by applying two identical pulses to the device to mimic the paired-pulse facilitation (PPF) of the neural system. Furthermore, long-term potentiation and depression were accomplished via consequent identical pulse stimuli under different switching modes to demonstrate stable LTM properties.

Low‐operation voltage conductive‐bridge random access memory based on amorphous NbS2

AbstractAmorphous NbS2 was proposed as the resistive switching (RS) layer for conductive‐bridge random access memory (CBRAM) for the first time, with Cu and Au as the top and bottom electrodes, respectively. NbS2 films were prepared at room temperature, which exhibited an amorphous structure and did not crystalize even annealed at 500°C, showing good thermal stability. The amorphous NbS2 CBRAM devices present stable bipolar non‐volatile RS characteristics. Repetitive RS behavior is demonstrated in amorphous NbS2 CBRAMs. The operating voltage during all RS cycles is less than 1 V, demonstrating that the NbS2 CBRAM is a low‐operation voltage memory device. The distribution of the high and low resistive state resistance is relatively concentrated, and the on‐off ratio has been kept above 100, offering a sufficient data read/write window. The formation and fracture of the Cu metal conductive filament is considered to be the RS mechanism by analyzing the dependence of current and voltage in logarithmic coordinates. Our study demonstrated that amorphous NbS2 is a promising material for low‐operation voltage CBRAM.

Integration of Ag-CBRAM crossbars and Mott ReLU neurons for efficient implementation of deep neural networks in hardware

In-memory computing with emerging non-volatile memory devices (eNVMs) has shown promising results in accelerating matrix-vector multiplications. However, activation function calculations are still being implemented with general processors or large and complex neuron peripheral circuits. Here, we present the integration of Ag-based conductive bridge random access memory (Ag-CBRAM) crossbar arrays with Mott rectified linear unit (ReLU) activation neurons for scalable, energy and area-efficient hardware (HW) implementation of deep neural networks. We develop Ag-CBRAM devices that can achieve a high ON/OFF ratio and multi-level programmability. Compact and energy-efficient Mott ReLU neuron devices implementing ReLU activation function are directly connected to the columns of Ag-CBRAM crossbars to compute the output from the weighted sum current. We implement convolution filters and activations for VGG-16 using our integrated HW and demonstrate the successful generation of feature maps for CIFAR-10 images in HW. Our approach paves a new way toward building a highly compact and energy-efficient eNVMs-based in-memory computing system.

Controlling Electrodeposition Via Electrokinetic Phenomena and Phase Separation

This talk will review the basic physics of dendritic instabilities in electrodeposition and how they may be suppressed, in some cases, by controlling electrokinetic phenomena in the electrolyte or phase separation in the electrode: 1) It is shown both theoretically and experimentally that, if electrodeposition occurs in charged porous media above the diffusion-limited current, then dendritic growth can be suppressed by surface conduction and electro-osmotic flows through the phenomenon of “shock electrodeposition”, as in many examples with copper. 2) It is also shown theoretically and experimentally that dendritic electrodeposition on a multiphase electrode can be avoided by avoiding the separation of a resistive phase, as illustrated by lithium plating on graphite. Examples include composite metal materials manufacturing, metal nanotube growth in templates, rechargeable metal batteries, and conductive-bridge random access memory (CB-RAM). These phenomena may also be leveraged to enhance the selectivity of electrodeposition for metal separations.

Demonstration of Artificial Afferent Nerve Properties With Forming-Free and SiO2-Based Memristive Synapses

The development of artificial nerve systems is considered of great significance for the fabrication of artificial appendages or intelligent robotics with environmental communication and awareness. Along these lines, in this work, the synaptic properties of a SiO <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{\text{2}}$</tex-math> </inline-formula> -based conductive bridge random access memory (CBRAM) were thoroughly investigated. Initially, a detailed study was conducted to understand and catalog the short-and long-term properties of the device, and their underlying physical mechanisms. To that end, a numerical model was introduced for interpreting the synaptic pattern within the device and its results were validated with the acquired experimental data. In addition, an artificial afferent nerve concept was demonstrated, comprising a piezoelectric sensory receptor, to generate electrical signals in response to either impulsive (“touch and go”) or continuous external stimuli (“touch and stay”), a microcontroller-emulated leaky-integrate-and-fire neural node converting the signals into spiking action potentials, and the CBRAM element as a terminal synapse, responsible for processing/attenuating the action potentials Throughout these procedures, the synaptic adaptation and the memory consolidation processes were recorded and analyzed.

Real-time study of imaging electron current density on metal filament evolution in SiO2 during in situ TEM

Conductive-bridging random access memory devices are a candidate for artificial synapses for neuromorphic computing. However, there is still an incomplete understanding of the fundamentals of the filament evolution process. In this work, we study the effect of three imaging electron current densities on nanoscale filament dynamics in a model Cu/SiO2/Cu structure during in situ TEM electroforming of the device. We find that the filaments grow from the anode to the cathode in the form of discontinuous precipitates for all the imaging electron current densities. However, increasing the imaging electron current density results in a larger injection of Cu into SiO2. Comparing the results of voltage ramp tests in air, in the TEM vacuum without electron irradiation and, in the TEM vacuum with electron irradiation, we suggest a possible mechanism of filament evolution in vacuum. Specifically, we postulate a vacancy defect generation enabled injection of Cu ions into the dielectric as the mechanism behind filament evolution in vacuum that reconciles differing observations found in the literature.

HfOx-Based Conductive Bridge Random Access Memory with Al2O3 Sandglass Nanostructures via Glancing Angle Deposition Technology toward Neuromorphic Applications

Conductive bridge random access memory (CBRAM) is one of the promising nonvolatile memories for next-generation technology owing to its high density, low power consumption, and fast switching speed, which is also a potential candidate for implementation of neuromorphic computing. However, CBRAMs suffer from stochastically growing conducting filaments in the insulator layer. Herein, we demonstrated Al2O3 sandglass nanostructures (SNGSs) embedded into HfOx-based CBRAMs via glancing angle deposition technology with the AlN thermal enhanced layer to prevent the overinjection of cations and localize the growth of conducting filaments in the HfOx switching layer. With the assistance of Al2O3 SNGSs and the AlN layer, the Cu/Al2O3 SNGSs/HfOx/AlN/TiN device exhibited a stable on/off ratio of >10 for more than 6000 cycles. Furthermore, with a Te top electrode, the Te/Al2O3 SNGSs/HfOx/AlN/TiN device shows a multilevel cell characteristic by controlling compliance currents. In addition, it possesses excellent potentiation and depression nonlinearities of 1.28 and 0.4, respectively, which is beneficial for future applications in neuromorphic computing.

GeS conducting-bridge resistive memory device with IGZO buffer layer for highly uniform and repeatable switching

A double stacked monochalcogenide GeS-based conducting-bridge random access memory (CBRAM) device with a IGZO buffer layer is investigated for highly improved resistive memory characteristics. The IGZO/GeS double layer is found to provide the CBRAM with a markedly improved sub-1V DC set/reset-voltage distributions (&amp;lt;±0.1 V variation). High endurance (&amp;gt;107 cycles) and retention (&amp;gt;105 s at 85 °C) performance are also achieved. The metal ion diffusion and migration rates in the solid electrolytes along with the redox reaction rates at the electrodes determine the respective resistive switching (RS) mechanism in the CBRAM device. Considering this fact, it is proposed that Ag diffusion into IGZO creates a virtual electrode, when coupled with strong ionic transport in GeS, consistently mediate the formation/dissolution of Ag filament there, thus reducing switching variation. Understanding the RS mechanism based on the materials' physical and chemical properties and tailoring the device structure allow an optimal control over cycle to cycle and device to device variability. The findings show that this material combination or similar oxide/chalcogenide stacks may offer a facile means for mitigating CBRAM variability.