Evolution of growth/dissolution conductive filaments (CFs) in oxide-electrolyte-based resistive switching memories are studied by in situ transmission electron microscopy. Contrary to what is commonly believed, CFs are found to start growing from the anode (Ag or Cu) rather than having to reach the cathode (Pt) and grow backwards. A new mechanism based on local redox reactions inside the oxide-electrolyte is proposed.

Resistive switching (RS) is an interesting property shown by some materials systems that, especially during the last decade, has gained a lot of interest for the fabrication of electronic devices, with electronic nonvolatile memories being those that have received the most attention. The presence and quality of the RS phenomenon in a materials system can be studied using different prototype cells, performing different experiments, displaying different figures of merit, and developing different computational analyses. Therefore, the real usefulness and impact of the findings presented in each study for the RS technology will be also different. This manuscript describes the most recommendable methodologies for the fabrication, characterization, and simulation of RS devices, as well as the proper methods to display the data obtained. The idea is to help the scientific community to evaluate the real usefulness and impact of an RS study for the development of RS technology.

Resistive memory (ReRAM) based on a solid-electrolyte insulator is a promising nanoscale device and has great potentials in nonvolatile memory, analog circuits, and neuromorphic applications. The underlying resistive switching (RS) mechanism of ReRAM is suggested to be the formation and rupture of nanoscale conductive filament (CF) inside the solid-electrolyte layer. However, the random nature of the nucleation and growth of the CF makes their formation difficult to control, which is a major obstacle for ReRAM performance improvement. Here, we report a novel approach to resolve this challenge by adopting a metal nanocrystal (NC) covered bottom electrode (BE) to replace the conventional ReRAM BE. As a demonstration vehicle, a Ag/ZrO(2)/Cu NC/Pt structure is prepared and the Cu NC covered Pt BE can control CF nucleation and growth to provide superior uniformity of RS properties. The controllable growth of nanoscale CF bridges between Cu NC and Ag top electrode has been vividly observed by transmission electron microscopy (TEM). On the basis of energy-dispersive X-ray spectroscopy (EDS) and elemental mapping analyses, we further confirm that the chemical contents of the CF are mainly Ag atoms. These testing/metrology results are consistent with the simulation results of electric-field distribution, showing that the electric field will enhance and concentrate on the NC sites and control location and orientation of Ag CFs.

Although the kinetics of CF formation/ dissolution is still unclear, it is widely accepted that the CF formation/dissolution is strongly related to the electromigration and electrochemical reaction of anion (i.e., oxygen vacancy) [13][14][15][16] or cation (i.e., Cu 2+ , Ag + or Ni 2+ ). [17][18][19][20][21][22] Generally, RS behavior can be classifi ed as two modes: nonvolatile memory switching (MS) and volatile threshold switching (TS). In the MS mode, both LRS and HRS can be maintained after removing the external voltage, while the LRS in the TS mode will be back to the HRS once the applied voltage is smaller than a critical value. [23][24][25] To avoid confusion with MS, the LRS and HRS in TS are renamed as "TS ON-state" and "TS OFFstate" in this article. The MS device can be used for the non-volatile data storage [1][2][3][4][5] while TS device can be as a selector in series with memory cell to suppress crosstalk effect in the crossbar array. [26][27][28][29][30] Recently, some groups reported that TS and MS can coexist and mutually transform in a single device at suitable external excitation. [23][24][25][26][27][28] Several models have been proposed to explain this phenomenon, including CF thermal instability, [ 23 ] strong electron correlation effect, [ 24 ] quantum-wire model, [ 25 ] interface barrier modulation, [ 26 ] and space charge effect. [ 27 ] However, the underlying mechanism of the phenomenon is still unclear, especially lacking of direct evidences to uncover when and how the two RS modes happen and what is the internal relationship between them.Here, we demonstrate that the TS and MS modes can be modulated in the Ag/SiO 2 /Pt structure by controlling the compliance current ( I CC ) in electroforming. We systematically investigate the morphologies, chemical components, and dynamic growth of the CF using scanning electron microscope (SEM), high-resolution transmission electron microscopy (HRTEM) and electron energy loss spectroscopy (EELS) analysis. The results confi rm that the TS and MS modes correspond to the CF consisting of isolated and continuous Ag nanocrystals, respectively. In addition, by Kelvin probe force microscopy (KPFM) studies, the voltage potential distribution of CF in the ON-and OFF-state further indicate that the TS mode is Volatile threshold switching (TS) and non-volatile memory switching (MS) are two typical resistive switching (RS) phenomena in oxides, which could form the basis for memory, analog circuits, and neuromorphic applications. Interestingly, TS and MS can be coexistent and converted in a single device under the suitable external excitation. However, the origin of the transition from TS to MS is still unclear due to the lack of direct experimental evidence. Here, conversion between TS and MS induced by conductive fi lament (CF) morphology in Ag/SiO 2 /Pt device is directly observed using scanning electron microscopy and high-resolution transmission electron microscopy. The MS mechanism is related to the formation and dissolution of CF consisting of continuous Ag...

No abstract

Conductive‐bridge random access memory (CBRAM) is considered a strong contender of the next‐generation nonvolatile memory technology. Resistive switching (RS) behavior in CBRAM is decided by the formation/dissolution of nanoscale conductive filament (CF) inside RS layer based on the cation injection from active electrode and their electrochemical reactions. Remarkably, RS is actually a localized behavior, however, cation injects from the whole area of active electrode into RS layer supplying excessive cation beyond the requirement of CF formation, leading to deterioration of device uniformity and reliability. Here, an effective method is proposed to localize cation injection into RS layer through the nanohole of inserted ion barrier between active electrode and RS layer. Taking an impermeable monolayer graphene as ion barrier, conductive atomic force microscopy results directly confirm that CF formation is confined through the nanohole of graphene due to the localized cation injection. Compared with the typical Cu/HfO2/Pt CBRAM device, the novel Cu/nanohole‐graphene/HfO2/Pt device shows improvement of uniformity, endurance, and retention characteristics, because the cation injection is limited by the nanohole graphene. Scaling the nanohole of ion barrier down to several nanometers, the single‐CF‐based CBRAM device with high performance is expected to achieve by confining the cation injection at the atomic scale.

systems, which can be integrated with terminal sensors to form intelligent sensory systems. [3-7] Information processing in biological sensory nervous systems involves billions of neurons interconnected through trillions of synapses, constituting immense neural networks. [8] Compared with traditional von Neumannbased computing architecture, neural networks greatly reduce time-and energy consumption by taking the advantage of co-location of logic and memory, hyperconnectivity, robustness, and massively parallel processing. [9] As the third-generation artificial neural network, spiking neural networks (SNNs) are inspired by biological nervous systems. They employ spiking neurons as computational units that process information with timing of spikes. [10] Therefore, SNNs provide the potential for spatiotemporal information processing with high time-and energy-efficiency. The learning and recognition functions aiming at spatiotemporal patterns have been demonstrated in SNN formed by resistive switching synaptic devices. [11] However, this synaptic device suffers from several performance limitations such as the discrete weight update, write variation, and high energy programing related to the filamentary switching mechanism. [12-17] Recently, He et al. use capacitively coupled multi-terminal neurotransistors for spatiotemporal information processing, where the neurotransistors mimic the dendritic discriminability of different spatiotemporal input sequences. [18] To physically emulate neural networks in hardware, large-scale crossbar Spiking neural networks (SNNs) sharing large similarity with biological nervous systems are promising to process spatiotemporal information and can provide highly time-and energy-efficient computational paradigms for the Internet-of-Things and edge computing. Nonvolatile electrolyte-gated transistors (EGTs) provide prominent analog switching performance, the most critical feature of synaptic element, and have been recently demonstrated as a promising synaptic device. However, high performance, large-scale EGT arrays, and EGT application for spatiotemporal information processing in an SNN are yet to be demonstrated. Here, an oxide-based EGT employing amorphous Nb 2 O 5 and Li x SiO 2 is introduced as the channel and electrolyte gate materials, respectively, and integrated into a 32 × 32 EGT array. The engineered EGTs show a quasi-linear update, good endurance (10 6) and retention, a high switching speed of 100 ns, ultralow readout conductance (<100 nS), and ultralow areal switching energy density (20 fJ µm −2). The prominent analog switching performance is leveraged for hardware implementation of an SNN with the capability of spatiotemporal information processing, where spike sequences with different timings are able to be efficiently learned and recognized by the EGT array. Finally, this EGT-based spatiotemporal information processing is deployed to detect moving orientation in a tactile sensing system. These results provide an insight into oxide-based EGT devices for energy-efficient neuro...

Negative-SET behavior is observed in various cation-based memories, which degrades the device reliability. Transmission electron microscopy results demonstrate the behavior is caused by the overgrowth of the conductive filament (CF) into the Pt electrode. The CF overgrowth phenomenon is suppressed and the negative-SET behavior is eliminated by inserting an impermeable graphene layer. The graphene-based devices show high reliability and satisfying performance.