Enhancing the Uniformity of a Memristor Using a Bilayer Dielectric Structure.

Resistive random access memory (RRAM) holds great promise for in-memory computing, which is considered the most promising strategy for solving the von Neumann bottleneck. However, there are still significant problems in its application due to the non-uniform performance of RRAM devices. In this work, a bilayer dielectric layer memristor was designed based on the difference in the Gibbs free energy of the oxide. We fabricated Au/Ta2O5/HfO2/Ta/Pt (S3) devices with excellent uniformity. Compared with Au/HfO2/Pt (S1) and Au/Ta2O5/Pt (S2) devices, the S3 device has a low reset voltage fluctuation of 2.44%, and the resistive coefficients of variation are 13.12% and 3.84% in HRS and LRS, respectively, over 200 cycles. Otherwise, the bilayer device has better linearity and more conductance states in multi-state regulation. At the same time, we analyze the physical mechanism of the bilayer device and provide a physical model of ion migration. This work provides a new idea for designing and fabricating resistive devices with stable performance.

An ultrasmall organic synapse for neuromorphic computing.

High-performance organic neuromorphic devices with miniaturized device size and computing capability are essential elements for developing brain-inspired humanoid intelligence technique. However, due to the structural inhomogeneity of most organic materials, downscaling of such devices to nanoscale and their high-density integration into compact matrices with reliable device performance remain challenging at the moment. Herein, based on the design of a semicrystalline polymer PBFCL10 with ordered structure to regulate dense and uniform formation of conductive nanofilaments, we realize an organic synapse with the smallest device dimension of 50 nm and highest integration size of 1 Kb reported thus far. The as-fabricated PBFCL10 synapses can switch between 32 conductance states linearly with a high cycle-to-cycle uniformity of 98.89% and device-to-device uniformity of 99.71%, which are the best results of organic devices. A mixed-signal neuromorphic hardware system based on the organic neuromatrix and FPGA controller is implemented to execute spiking-plasticity-related algorithm for decision-making tasks.

Organic Memristor with Synaptic Plasticity for Neuromorphic Computing Applications.

Memristors have been considered to be more efficient than traditional Complementary Metal Oxide Semiconductor (CMOS) devices in implementing artificial synapses, which are fundamental yet very critical components of neurons as well as neural networks. Compared with inorganic counterparts, organic memristors have many advantages, including low-cost, easy manufacture, high mechanical flexibility, and biocompatibility, making them applicable in more scenarios. Here, we present an organic memristor based on an ethyl viologen diperchlorate [EV(ClO4)]2/triphenylamine-containing polymer (BTPA-F) redox system. The device with bilayer structure organic materials as the resistive switching layer (RSL) exhibits memristive behaviors and excellent long-term synaptic plasticity. Additionally, the device's conductance states can be precisely modulated by consecutively applying voltage pulses between the top and bottom electrodes. A three-layer perception neural network with in situ computing enabled was then constructed utilizing the proposed memristor and trained on the basis of the device's synaptic plasticity characteristics and conductance modulation rules. Recognition accuracies of 97.3% and 90% were achieved, respectively, for the raw and 20% noisy handwritten digits images from the Modified National Institute of Standards and Technology (MNIST) dataset, demonstrating the feasibility and applicability of implementing neuromorphic computing applications utilizing the proposed organic memristor.

Titanium oxide-based optoelectronic synapses with visual memory synergistically adjusted by internal emotions and ambient illumination.

Brain-inspired neuromorphic computing has become one of the critical technologies to overcome the bottleneck of von Neumann architecture. It is a vital step to construct a brain-like neuromorphic computing system at the hardware level by utilizing artificial synaptic devices. Compared with electronic synaptic devices, optoelectronic synaptic devices have the advantages of low power consumption, low crosstalk, and high bandwidth. Artificial optoelectronic synapses, analogous to retinal structure, can directly respond to and process light signal information to mimic the neuromorphic visual system. As high-level nerve impulses, both generated and regulated, emotions affect the strength and persistence of memory. Ambient illumination can provide visual perception to distinguish the size, color, and other characteristics of objects as well as affect the nonvisual functions of individuals, such as emotional states, thereby affecting learning and memory function. Herein, an artificial optoelectronic synapse composed of ITO/TiO2-x /p-Si was proposed. A variety of biologically dependent synaptic plasticity relating to learning and memory, including short-term synaptic plasticity, long-term synaptic plasticity, and learning-forgetting-relearning multifunctional advanced synaptic activity, was successfully simulated. A 3 × 3 artificial optoelectronic synapse array based on 9 devices was constructed to mimic the functions of visual learning and memory affected by internal emotion and ambient illumination. The proposed artificial optoelectronic synapse will exhibit great potential in visual and image information perception and memory.

Perovskite-Based Memristor with 50-Fold Switchable Photosensitivity for In-Sensor Computing Neural Network.

In-sensor computing can simultaneously output image information and recognition results through in-situ visual signal processing, which can greatly improve the efficiency of machine vision. However, in-sensor computing is challenging due to the requirement to controllably adjust the sensor's photosensitivity. Herein, it is demonstrated a ternary cationic halide Cs0.05FA0.81MA0.14 Pb(I0.85Br0.15)3 (CsFAMA) perovskite, whose External quantum efficiency (EQE) value is above 80% in the entire visible region (400-750 nm), and peak responsibility value at 750 nm reaches 0.45 A/W. In addition, the device can achieve a 50-fold enhancement of the photoresponsibility under the same illumination by adjusting the internal ion migration and readout voltage. A proof-of-concept visually enhanced neural network system is demonstrated through the switchable photosensitivity of the perovskite sensor array, which can simultaneously optimize imaging and recognition results and improve object recognition accuracy by 17% in low-light environments.

A Quantized Convolutional Neural Network Implemented With Memristor for Image Denoising and Recognition.

The interference of noise will cause the degradation of image quality, which can have a negative impact on the subsequent image processing and visual effect. Although the existing image denoising algorithms are relatively perfect, their computational efficiency is restricted by the performance of the computer, and the computational process consumes a lot of energy. In this paper, we propose a method for image denoising and recognition based on multi-conductance states of memristor devices. By regulating the evolution of Pt/ZnO/Pt memristor wires, 26 continuous conductance states were obtained. The image feature preservation and noise reduction are realized via the mapping between the conductance state and the image pixel. Furthermore, weight quantization of convolutional neural network is realized based on multi-conductance states. The simulation results show the feasibility of CNN for image denoising and recognition based on multi-conductance states. This method has a certain guiding significance for the construction of high-performance image noise reduction hardware system.

Improving the Recognition Accuracy of Memristive Neural Networks via Homogenized Analog Type Conductance Quantization.

Conductance quantization (QC) phenomena occurring in metal oxide based memristors demonstrate great potential for high-density data storage through multilevel switching, and analog synaptic weight update for effective training of the artificial neural networks. Continuous, linear and symmetrical modulation of the device conductance is a critical issue in QC behavior of memristors. In this contribution, we employ the scanning probe microscope (SPM) assisted electrode engineering strategy to control the ion migration process to construct single conductive filaments in Pt/HfOx/Pt devices. Upon deliberate tuning and evolution of the filament, 32 half integer quantized conductance states in the 16 G0 to 0.5 G0 range with enhanced distribution uniformity was achieved. Simulation results revealed that the numbers of the available QC states and fluctuation of the conductance at each state play an important role in determining the overall performance of the neural networks. The 32-state QC behavior of the hafnium oxide device shows improved recognition accuracy approaching 90% for handwritten digits, based on analog type operation of the multilayer perception (MLP) neural network.

Direct Observation of Structural Deformation Immunity for Understanding Oxygen Plasma Treatment-Enhanced Resistive Switching in HfOx-Based Memristive Devices.

Oxygen ions' migration is the fundamental resistive switching (RS) mechanism of the binary metal oxides-based memristive devices, and recent studies have found that the RS performance can be enhanced through appropriate oxygen plasma treatment (OPT). However, the lack of experimental evidence observed directly from the microscopic level of materials and applicable understanding of how OPT improves the RS properties will cause significant difficulties in its further application. In this work, we apply scanning probe microscope (SPM)-based techniques to study the OPT-enhanced RS performance in prototypical HfOx based memristive devices through in situ morphology and electrical measurements. It is first found that the structural deformations in HfOx nanofilm induced by migration of oxygen ions and interfacial electrochemical reactions can be recovered by OPT effectively. More importantly, such structural deformations no longer occur after OPT due to the strengthening in lattice structure, which directly illustrates the enhanced quantity of HfOx nanofilm and the nature of enhanced RS properties after OPT. Finally, the underlying mechanisms of OPT-enhanced RS performance are analyzed by the results of X-ray photoelectron spectroscopic (XPS) surface analysis. In the OPT-enhanced HfOx nanofilm, oxygen vacancies in crystalline regions can be remarkably reduced by active oxygen ions' implantation. The oxygen ions transport will depend considerably on the grain boundaries and OPT-enhanced lattice structure will further guarantee the stability of conductive filaments, both of which ensure the uniformity and repeatability in RS processes. This study could provide a scientific basis for improving RS performance of oxides-based memristive devices by utilizing OPT.

An Oxide Schottky Junction Artificial Optoelectronic Synapse.

The rapid development of artificial intelligence techniques and future advanced robot systems sparks emergent demand on the accurate perception and understanding of the external environments via visual sensing systems that can co-locate the self-adaptive detecting, processing, and memorizing of optical signals. In this contribution, a simple indium-tin oxide/Nb-doped SrTiO3 (ITO/Nb:SrTiO3) heterojunction artificial optoelectronic synapse is proposed and demonstrated. Through the light and electric field co-modulation of the Schottky barrier profile at the ITO/Nb:SrTiO3 interface, the oxide heterojunction device can respond to the entire visible light region in a neuromorphic manner, allowing synaptic paired-pulse facilitation, short/long-term memory, and "learning-experience" behavior for optical information manipulation. More importantly, the photoplasticity of the artificial synapse has been modulated by heterosynaptic means with a sub-1 V external voltage, not only enabling an optoelectronic analog of the mechanical aperture device showing adaptive and stable optical perception capability under different illuminating conditions but also making the artificial synapse suitable for the mimicry of interest-modulated human visual memories.

A univariate ternary logic and three-valued multiplier implemented in a nano-columnar crystalline zinc oxide memristor.

Memristors, which feature small sizes, fast speeds, low power, CMOS compatibility and nonvolatile modulation of device resistance, are promising candidates for next-generation data storage and in-memory computing paradigms. Compared to the binary logics enabled by memristor devices, ternary logics with larger information-carrying capacity can provide higher computation efficiency with simple operation schemes, reduced circuit complexity and smaller chip areas. In this study, we report the fabrication of memristor devices based on nano-columnar crystalline ZnO thin films; they show symmetric and reliable multi-level resistive switching characteristics over three hundred cycles, which benefits the implementation of univariate ternary logic operations. Experimental results demonstrate that a three-valued logic complete set can be realized by the univariate operations of the present ZnO memristor device, and a ternary multiplier unit circuit is designed for potential applications. The present methodology can be beneficial for constructing future high-performance computation architectures.

Improving Unipolar Resistive Switching Uniformity with Cone-Shaped Conducting Filaments and Its Logic-In-Memory Application.

Resistive random access memory (RRAM) with inherent logic-in-memory capability exhibits great potential to construct beyond von-Neumann computers. Particularly, unipolar RRAM is more promising because its single polarity operation enables large-scale crossbar logic-in-memory circuits with the highest integration density and simpler peripheral control circuits. However, unipolar RRAM usually exhibits poor switching uniformity because of random activation of conducting filaments and consequently cannot meet the strict uniformity requirement for logic-in-memory application. In this contribution, a new methodology that constructs cone-shaped conducting filaments by using chemically a active metal cathode is proposed to improve unipolar switching uniformity. Such a peculiar metal cathode will react spontaneously with the oxide switching layer to form an interfacial layer, which together with the metal cathode itself can act as a load resistor to prevent the overgrowth of conducting filaments and thus make them more cone-like. In this way, the rupture of conducting filaments can be strictly limited to the tip region, making their residual parts favorable locations for subsequent filament growth and thus suppressing their random regeneration. As such, a novel "one switch + one unipolar RRAM cell" hybrid structure is capable to realize all 16 Boolean logic functions for large-scale logic-in-memory circuits.