Memristive technologies for data storage, computation, encryption, and radio-frequency communication

Resistive switching modulation by incorporating thermally enhanced layer in HfO2-based memristor

Oxide-based memristors by incorporating thermally enhanced layer (TEL) have showed great potential in electronic devices for high-efficient and high-density neuromorphic computing owing to the improvement of multilevel resistive switching. However, research on the mechanism of resistive switching regulation is still lacking. In this work, based on the method of finite element numerical simulation analysis, a bilayer oxide-based memristor Pt/HfO2(5 nm)/Ta2O5(5 nm)/Pt with the Ta2O5TEL was proposed. The oxygen vacancy concentrates distribution shows that the fracture of conductive filaments (CF) is at the interface where the local temperature is the highest during the reset process. The multilevel resistive switching properties were also obtained by applying different stop voltages. The fracture gap of CF can be enlarged with the increase of the stopping voltage, which is attributed to the heat-gathering ability of the TEL. Moreover, it was found that the fracture position of oxygen CF is dependent on the thickness of TEL, which exhibits a modulation of device RS performance. These results provide a theoretical guidance on the suitability of memristor devices for use in high-density memory and brain-actuated computer systems.

Effect of electrode materials on resistive switching behaviour of NbOx-based memristive devices

Memristive devices that rely on redox-based resistive switching mechanism have attracted great attention for the development of next-generation memory and computing architectures. However, a detailed understanding of the relationship between involved materials, interfaces, and device functionalities still represents a challenge. In this work, we analyse the effect of electrode metals on resistive switching functionalities of NbOx-based memristive cells. For this purpose, the effect of Au, Pt, Ir, TiN, and Nb top electrodes was investigated in devices based on amorphous NbOx grown by anodic oxidation on a Nb substrate exploited also as counter electrode. It is shown that the choice of the metal electrode regulates electronic transport properties of metal-insulator interfaces, strongly influences the electroforming process, and the following resistive switching characteristics. Results show that the electronic blocking character of Schottky interfaces provided by Au and Pt metal electrodes results in better resistive switching performances. It is shown that Pt represents the best choice for the realization of memristive cells when the NbOx thickness is reduced, making possible the realization of memristive cells characterised by low variability in operating voltages, resistance states and with low device-to-device variability. These results can provide new insights towards a rational design of redox-based memristive cells.

In-situ electro-responsive through-space coupling enabling foldamers as volatile memory elements

Voltage-gated processing units are fundamental components for non-von Neumann architectures like memristor and electric synapses, on which nanoscale molecular electronics have possessed great potentials. Here, tailored foldamers with furan‒benzene stacking (f-Fu) and thiophene‒benzene stacking (f-Th) are designed to decipher electro-responsive through-space interaction, which achieve volatile memory behaviors via quantum interference switching in single-molecule junctions. f-Fu exhibits volatile turn-on feature while f-Th performs stochastic turn-off feature with low voltages as 0.2 V. The weakened orbital through-space mixing induced by electro-polarization dominates stacking malposition and quantum interference switching. f-Fu possesses higher switching probability and faster responsive time, while f-Th suffers incomplete switching and longer responsive time. High switching ratios of up to 91 for f-Fu is realized by electrochemical gating. These findings provide evidence and interpretation of the electro-responsiveness of non-covalent interaction at single-molecule level and offer design strategies of molecular non-von Neumann architectures like true random number generator.

A High-Entropy-Oxides-Based Memristor: Outstanding Resistive Switching Performance and Mechanisms in Atomic Structural Evolution

The application of high-entropy oxide (HEO) has attracted significant attention in recent years owing to their unique structural characteristics, such as excellent electrochemical properties and long-term cycling stability. However, the application of resistive random-access memory (RRAM) has not been extensively studied, and the switching mechanism of HEO-based RRAM has yet to be thoroughly investigated. In this study, HEO (Cr, Mn, Fe, Co, Ni)3 O4 with a spinel structure is epitaxially grown on a Nb:STO conductive substrate, and Pt metal is deposited as the top electrode. After the resistive-switching operation, some regions of the spinel structure are transformed into a rock-salt structure and analyzed using advanced transmission electron microscopy and scanning transmission electron microscopy. From the results of X-ray photoelectron spectroscopy and electron energy loss spectroscopy, only specific elements would change their valence state, which results in excellent resistive-switching properties with a high on/off ratio on the order of 105 , outstanding endurance (>4550 cycles), long retention time (>104 s), and high stability, which suggests that HEO is a promising RRAM material.

TiN/Ti/HfO2/TiN memristive devices for neuromorphic computing: from synaptic plasticity to stochastic resonance

We characterize TiN/Ti/HfO2/TiN memristive devices for neuromorphic computing. We analyze different features that allow the devices to mimic biological synapses and present the models to reproduce analytically some of the data measured. In particular, we have measured the spike timing dependent plasticity behavior in our devices and later on we have modeled it. The spike timing dependent plasticity model was implemented as the learning rule of a spiking neural network that was trained to recognize the MNIST dataset. Variability is implemented and its influence on the network recognition accuracy is considered accounting for the number of neurons in the network and the number of training epochs. Finally, stochastic resonance is studied as another synaptic feature. It is shown that this effect is important and greatly depends on the noise statistical characteristics.

Edge learning using a fully integrated neuro-inspired memristor chip

Learning is highly important for edge intelligence devices to adapt to different application scenes and owners. Current technologies for training neural networks require moving massive amounts of data between computing and memory units, which hinders the implementation of learning on edge devices. We developed a fully integrated memristor chip with the improvement learning ability and low energy cost. The schemes in the STELLAR architecture, including its learning algorithm, hardware realization, and parallel conductance tuning scheme, are general approaches that facilitate on-chip learning by using a memristor crossbar array, regardless of the type of memristor device. Tasks executed in this study included motion control, image classification, and speech recognition.

Emerging Iontronic Neural Devices for Neuromorphic Sensory Computing

Living organisms have a very mysterious and powerful sensory computing system based on ion activity. Interestingly, studies on iontronic devices in the past few years have proposed a promising platform for simulating the sensing and computing functions of living organisms, because: 1) iontronic devices can generate, store, and transmit a variety of signals by adjusting the concentration and spatiotemporal distribution of ions, which analogs to how the brain performs intelligent functions by alternating ion flux and polarization; 2) through ionic-electronic coupling, iontronic devices can bridge the biosystem with electronics and offer profound implications for soft electronics; 3) with the diversity of ions, iontronic devices can be designed to recognize specific ions or molecules by customizing the charge selectivity, and the ionic conductivity and capacitance can be adjusted to respond to external stimuli for a variety of sensing schemes, which can be more difficult for electron-based devices. This review provides a comprehensive overview of emerging neuromorphic sensory computing by iontronic devices, highlighting representative concepts of both low-level and high-level sensory computing and introducing important material and device breakthroughs. Moreover, iontronic devices as a means of neuromorphic sensing and computing are discussed regarding the pending challenges and future directions.

Collective Control of Potential-Constrained Oxygen Vacancies in Oxide Heterostructures for Gradual Resistive Switching

Filamentary resistive switching in oxides is one of the key strategies for developing next-generation non-volatile memory devices. However, despite numerous advantages, their practical applications in neuromorphic computing are still limited due to non-uniform and indeterministic switching behavior. Given the inherent stochasticity of point defect migration, the pursuit of reliable switching likely demands an innovative approach. Herein, a collective control of oxygen vacancies is introduced in LaAlO3 /SrTiO3 (LAO/STO) heterostructures to achieve reliable and gradual resistive switching. By exploiting an electrostatic potential constraint in ultrathin LAO/STO heterostructures, the formation of conducting filaments is suppressed, but instead precisely control the concentration of oxygen vacancies. Since the conductance of the LAO/STO device is governed by the ensemble concentration of oxygen vacancies, not their individual probabilistic migrations, the resistive switching is more uniform and deterministic compared to conventional filamentary devices. It provides direct evidence for the collective control of oxygen vacancies by spectral noise analysis and modeling by Monte-Carlo simulation. As a proof of concept, the significantly-improved analog switching performance of the filament-free LAO/STO devices is demonstrated, revealing potential for neuromorphic applications. The results establish an approach to store information by point defect concentration, akin to biological ionic channels, for enhancing switching characteristics of oxide materials.

Non-volatile resistive switching mechanism in single-layer MoS2 memristors: insights from ab initio modelling of Au and MoS2 interfaces

Non-volatile memristive devices based on two-dimensional (2D) layered materials provide an attractive alternative to conventional flash memory chips. Single-layer semiconductors, such as monolayer molybdenum disulphide (ML-MoS2), enable the aggressive downscaling of devices towards greater system integration density. The "atomristor", the most compact device to date, has been shown to undergo a resistive switching between its high-resistance (HRS) and low-resistance (LRS) states of several orders of magnitude. The main hypothesis behind its working mechanism relies on the migration of sulphur vacancies in the proximity of the metal contact during device operation, thus inducing the variation of the Schottky barrier at the metal-semiconductor interface. However, the interface physics is not yet fully understood: other hypotheses were proposed, involving the migration of metal atoms from the electrode. In this work, we aim to elucidate the mechanism of the resistive switching in the atomristor. We carry out density functional theory (DFT) simulations on model Au and ML-MoS2 interfaces with and without the presence of point defects, either vacancies or substitutions. To construct realistic interfaces, we combine DFT with Green's function surface simulations. Our findings reveal that it is not the mere presence of S vacancies but rather the migration of Au atoms from the electrode to MoS2 that modulate the interface barrier. Indeed, Au atoms act as conductive "bridges", thus facilitating the flow of charge between the two materials.

Flexible, Transparent, and Wafer-Scale Artificial Synapse Array Based on TiOx /Ti3 C2 Tx Film for Neuromorphic Computing

A high-density neuromorphic computing memristor array based on 2D materials paves the way for next-generation information-processing components and in-memory computing systems. However, the traditional 2D-materials-based memristor devices suffer from poor flexibility and opacity, which hinders the application of memristors in flexible electronics. Here, a flexible artificial synapse array based on TiOx /Ti3 C2 Tx film is fabricated by a convenient and energy-efficient solution-processing technique, which realizes high transmittance (≈90%) and oxidation resistance (>30 days). The TiOx /Ti3 C2 Tx memristor shows low device-to-device variability, long memory retention and endurance, a high ON/OFF ratio, and fundamental synaptic behavior. Furthermore, satisfactory flexibility (R = 1.0 mm) and mechanical endurance (104 bending cycles) of the TiOx /Ti3 C2 Tx memristor are achieved, which is superior to other film memristors prepared by chemical vapor deposition. In addition, high-precision (>96.44%) MNIST handwritten digits recognition classification simulation indicates that the TiOx /Ti3 C2 Tx artificial synapse array holds promise for future neuromorphic computing applications, and provides excellent high-density neuron circuits for new flexible intelligent electronic equipment.

Unravelling the Data Retention Mechanisms under Thermal Stress on 2D Memristors

Memristors based on two-dimensional (2D) materials are a rapidly growing research area due to their potential in energy-efficient in-memory processing and neuromorphic computing. However, the data retention of these emerging memristors remains sparsely investigated, despite its crucial importance to device performance and reliability. In this study, we employ kinetic Monte-Carlo simulations to investigate the data retention of a 2D planar memristor. The operation of the memristor depends on field-driven on defect migration, while thermal diffusion gradually evens the defect distribution, leading to the degradation of the high resistance state (HRS) and diminishing the ON/OFF ratio. Notably, we examine the resilience of devices based on single crystals of transition metal dichalcogenides (TMDs) in harsh environments. Specifically, our simulations show that MoS2-based devices have negligible degradation after 10 years of thermal annealing at 400 K. Furthermore, the variability in data retention lifetime across different temperatures is less than 22%, indicating a relatively consistent performance over a range of thermal conditions. We also demonstrate that device miniaturization does not compromise data retention lifetime. Moreover, employing materials with higher activation energy for defect migration can significantly enhance data retention at the cost of increased switching voltage. These findings shed light on the behavior of 2D memristors and pave the way for their optimization in practical applications.

Highly Reliable Textile-Type Memristor by Designing Aligned Nanochannels

Information-processing devices are the core components of modern electronics. Integrating them into textiles is the indispensable demand for electronic textiles to form close-loop functional systems. Memristors with crossbar configuration are regarded as promising building blocks to design woven information-processing devices that seamlessly unify with textiles. However, the memristors always suffer from severe temporal and spatial variations due to the random growth of conductive filaments during filamentary switching processes. Here, inspired by the ion nanochannels across synaptic membranes, a highly reliable textile-type memristor made of Pt/CuZnS memristive fiber with aligned nanochannels, showing small set voltage variation (<5.6%) under ultralow set voltage (≈0.089 V), high on/off ratio (≈106 ), and low power consumption (0.1 nW), is reported. Experimental evidence indicate that nanochannels with abundant active S defects can anchor silver ions and confine their migrations to form orderly and efficient conductive filaments. Such memristive performances enable the resultant textile-type memristor array to have high device-to-device uniformity and process complex physiological data like brainwave signals with high recognition accuracy (95%). The textile-type memristor arrays are mechanically durable to withstand hundreds of bending and sliding deformations, and seamlessly unified with sensing, power-supplying, and displaying textiles/fibers to form all-textile integrated electronic systems for new generation human-machine interactions.

Engineering Metal-Organic Frameworks with Tunable Colors for High-Performance Wireless Communication

Metal-organic frameworks (MOFs) have emerged as excellent platforms possessing tunable and controllable optical behaviors that are essential in high-speed and multichannel data transmission in optical wireless communications (OWCs). Here, we demonstrate a novel approach to achieving a tunable wide modulation bandwidth and high net data rate by engineering a combination of organic linkers and metal clusters in MOFs. More specifically, two organic linkers of different emission colors, but equal molecular length and connectivity, are successfully coordinated by zirconium and hafnium oxy-hydroxy clusters to form the desired MOF structures. The precise change in the interactions between these different organic linkers and metal clusters enables control over fluorescence efficiency and excited state lifetime, leading to a tunable modulation bandwidth from 62.1 to 150.0 MHz and a net data rate from 303 to 363 Mb/s. The fabricated color converter MOFs display outstanding performance that competes, and in some instances surpasses, those of conventional materials commonly used in light converter devices. Moreover, these MOFs show high practicality in color-pure wavelength-division multiplexing (WDM), which significantly improved the data transmission link capacity and security by the contemporary combining of two different data signals in the same path. This work highlights the potential of engineered MOFs as a game-changer in OWCs, with significant implications for future high-speed and secure data transmission.

IoT Sensor Challenges for Geothermal Energy Installations Monitoring: A Survey

Geothermal energy installations are becoming increasingly common in new city developments and renovations. With a broad range of technological applications and improvements in this field, the demand for suitable monitoring technologies and control processes for geothermal energy installations is also growing. This article identifies opportunities for the future development and deployment of IoT sensors applied to geothermal energy installations. The first part of the survey describes the technologies and applications of various sensor types. Sensors that monitor temperature, flow rate and other mechanical parameters are presented with a technological background and their potential applications. The second part of the article surveys Internet-of-Things (IoT), communication technology and cloud solutions applicable to geothermal energy monitoring, with a focus on IoT node designs, data transmission technologies and cloud services. Energy harvesting technologies and edge computing methods are also reviewed. The survey concludes with a discussion of research challenges and an outline of new areas of application for monitoring geothermal installations and innovating technologies to produce IoT sensor solutions.

Highly Trustworthy In-Sensor Cryptography for Image Encryption and Authentication

The prevailing transmission of image information over the Internet of Things demands trustworthy cryptography for high security and privacy. State-of-the-art security modules are usually physically separated from the sensory terminals that capture images, which unavoidably exposes image information to various attacks during the transmission process. Here we develop in-sensor cryptography that enables capturing images and producing security keys in the same hardware devices. The generated key inherently binds to the captured images, which gives rise to highly trustworthy cryptography. Using the intrinsic electronic and optoelectronic characteristics of the 256 molybdenum disulfide phototransistor array, we can harvest electronic and optoelectronic binary keys with a physically unclonable function and further upgrade them into multiple-state ternary and double-binary keys, exhibiting high uniformity, uniqueness, randomness, and coding capacity. This in-sensor cryptography enables highly trustworthy image encryption to avoid passive attacks and image authentication to prevent unauthorized editions.

Wurtzite and fluorite ferroelectric materials for electronic memory

Ferroelectric materials, the charge equivalent of magnets, have been the subject of continued research interest since their discovery more than 100 years ago. The spontaneous electric polarization in these crystals, which is non-volatile and programmable, is appealing for a range of information technologies. However, while magnets have found their way into various types of modern information technology hardware, applications of ferroelectric materials that use their ferroelectric properties are still limited. Recent advances in ferroelectric materials with wurtzite and fluorite structure have renewed enthusiasm and offered new opportunities for their deployment in commercial-scale devices in microelectronics hardware. This Review focuses on the most recent and emerging wurtzite-structured ferroelectric materials and emphasizes their applications in memory and storage-based microelectronic hardware. Relevant comparisons with existing fluorite-structured ferroelectric materials are made and a detailed outlook on ferroelectric materials and devices applications is provided.

Hardware and Information Security Primitives Based on 2D Materials and Devices

Hardware security is a major concern for the entire semiconductor ecosystem that accounts for billions of dollars in annual losses. Similarly, information security is a critical need for the rapidly proliferating edge devices that continuously collect and communicate a massive volume of data. While silicon-based complementary metal-oxide-semiconductor technology offers security solutions, these are largely inadequate, inefficient, and often inconclusive, as well as resource intensive in time, energy, and cost, leading to tremendous room for innovation in this field. Furthermore, silicon-based security primitives have shown vulnerability to machine learning (ML) attacks. In recent years, 2D materials such as graphene and transition metal dichalcogenides have been intensely explored to mitigate these security challenges. In this review, 2D-materials-based hardware security solutions such as camouflaging, true random number generation, watermarking, anticounterfeiting, physically unclonable functions, and logic locking of integrated circuits (ICs) are summarized with accompanying discussion on their reliability and resilience to ML attacks. In addition, the role of native defects in 2D materials in developing high entropy hardware security primitives is also examined. Finally, the existing challenges for 2D materials, which must be overcome for large-scale deployment of 2D ICs to meet the security needs of the semiconductor industry, are discussed.

In-memory factorization of holographic perceptual representations

Disentangling the attributes of a sensory signal is central to sensory perception and cognition and hence is a critical task for future artificial intelligence systems. Here we present a compute engine capable of efficiently factorizing high-dimensional holographic representations of combinations of such attributes, by exploiting the computation-in-superposition capability of brain-inspired hyperdimensional computing, and the intrinsic stochasticity associated with analogue in-memory computing based on nanoscale memristive devices. Such an iterative in-memory factorizer is shown to solve at least five orders of magnitude larger problems that cannot be solved otherwise, as well as substantially lowering the computational time and space complexity. We present a large-scale experimental demonstration of the factorizer by employing two in-memory compute chips based on phase-change memristive devices. The dominant matrix-vector multiplication operations take a constant time, irrespective of the size of the matrix, thus reducing the computational time complexity to merely the number of iterations. Moreover, we experimentally demonstrate the ability to reliably and efficiently factorize visual perceptual representations.

On the switching mechanism and optimisation of ion irradiation enabled 2D MoS2 memristors

Memristors are prominent passive circuit elements with promising futures for energy-efficient in-memory processing and revolutionary neuromorphic computation. State-of-the-art memristors based on two-dimensional (2D) materials exhibit enhanced tunability, scalability and electrical reliability. However, the fundamental of the switching is yet to be clarified before they can meet industrial standards in terms of endurance, variability, resistance ratio, and scalability. This new physical simulator based on the kinetic Monte Carlo (kMC) algorithm reproduces the defect migration process in 2D materials and sheds light on the operation of 2D memristors. The present work employs the simulator to study a two-dimensional 2H-MoS₂ planar resistive switching (RS) device with an asymmetric defect concentration introduced by ion irradiation. The simulations unveil the non-filamentary RS process and propose routes to optimize the device's performance. For instance, the resistance ratio can be increased by 53% by controlling the concentration and distribution of defects, while the variability can be reduced by 55% by increasing 5-fold the device size from 10 to 50 nm. Our simulator also explains the trade-offs between the resistance ratio and variability, resistance ratio and scalability, and variability and scalability. Overall, the simulator may enable an understanding and optimization of devices to expedite cutting-edge applications.

Evaluating charge-type of polyelectrolyte as dielectric layer in memristor and synapse emulation

Based on credible advantages, organic materials have received more and more attention in memristor and synapse emulation. In particular, an implementation of the ionic pathway as a dielectric layer is essential for organic materials used as building blocks of memristor and artificial synaptic devices. Herein, we describe an evaluation of the use of positive and negative polyelectrolytes as dielectric layers for a memristor with calcium ion (Ca²⁺) doping. The device based on a negative polyelectrolyte shows the potential to obtain an excellent resistive switching performance and synapse functionality, especially in the transformation behaviours from short-term plasticity (STP) to long-term plasticity (LTP) in both the potentiation and depression processes, which were comparable to the perfomrmance obtained with a positive polyelectrolyte. The mechanism of electrical resistance transition and synaptic function can be attributed to the migration of the doped Ca²⁺ and the ionic functional groups of polyelectrolyte, which result in the formation and vanishing filament-like Ca²⁺ flux.

Learning from the Brain: Bioinspired Nanofluidics

The human brain completes intelligent behaviors such as the generation, transmission, and storage of neural signals by regulating the ionic conductivity of ion channels in neuron cells, which provides new inspiration for the development of ion-based brain-like intelligence. Against the backdrop of the gradual maturity of neuroscience, computer science, and micronano materials science, bioinspired nanofluidic iontronics, as an emerging interdisciplinary subject that focuses on the regulation of ionic conductivity of nanofluidic systems to realize brain-like functionalities, has attracted the attention of many researchers. This Perspective provides brief background information and the state-of-the-art progress of nanofluidic intelligent systems. Two main categories are included: nanofluidic transistors and nanofluidic memristors. The prospects of nanofluidic iontronics' interdisciplinary progress in future artificial intelligence fields such as neuromorphic computing or brain-computer interfaces are discussed. This Perspective aims to give readers a clear understanding of the concepts and prospects of this emerging interdisciplinary field.

Thermal Characterization of Conductive Filaments in Unipolar Resistive Memories

A methodology to estimate the device temperature in resistive random access memories (RRAMs) is presented. Unipolar devices, which are known to be highly influenced by thermal effects in their resistive switching operation, are employed to develop the technique. A 3D RRAM simulator is used to fit experimental data and obtain the maximum and average temperatures of the conductive filaments (CFs) that are responsible for the switching behavior. It is found that the experimental CFs temperature corresponds to the maximum simulated temperatures obtained at the narrowest sections of the CFs. These temperature values can be used to improve compact models for circuit simulation purposes.

Toward the Speed Limit of Phase-Change Memory

Phase-change memory (PCM) is one of the most promising candidates for next-generation data-storage technology, the programming speed of which has enhanced within a timescale from milliseconds to sub-nanosecond (≈500 ps) through decades of effort. As the potential applications of PCM strongly depend on the switching speed, namely, the time required for the recrystallization of amorphous chalcogenide media, the finding of the ultimate crystallization speed is of great importance both theoretically and practically. In this work, through systematic analysis of discovered phase-change materials and ab initio molecular dynamics simulations, elemental Sb-based PCM is predicted to have a superfast crystallization speed. Indeed, such cells experimentally present extremely fast crystallization speeds within 360 ps. Remarkably, the recrystallization process is further sped up as the device shrinks, and a record-fast crystallization speed of only 242 ps is achieved in 60 nm-size devices. These findings open opportunities for dynamic random-access memory (DRAM)-like and even cache-like PCM using appropriate storage materials.

Performance estimation for the memristor-based computing-in-memory implementation of extremely factorized network for real-time and low-power semantic segmentation

Nowadays many semantic segmentation algorithms have achieved satisfactory accuracy on von Neumann platforms (e.g., GPU), but the speed and energy consumption have not meet the high requirements of certain edge applications like autonomous driving. To tackle this issue, it is of necessity to design an efficient lightweight semantic segmentation algorithm and then implement it on emerging hardware platforms with high speed and energy efficiency. Here, we first propose an extremely factorized network (EFNet) which can learn multi-scale context information while preserving rich spatial information with reduced model complexity. Experimental results on the Cityscapes dataset show that EFNet achieves an accuracy of 68.0% mean intersection over union (mIoU) with only 0.18M parameters, at a speed of 99 frames per second (FPS) on a single RTX 3090 GPU. Then, to further improve the speed and energy efficiency, we design a memristor-based computing-in-memory (CIM) accelerator for the hardware implementation of EFNet. It is shown by the simulation in DNN+NeuroSim V2.0 that the memristor-based CIM accelerator is ∼63× (∼4.6×) smaller in area, at most ∼9.2× (∼1000×) faster, and ∼470× (∼2400×) more energy-efficient than the RTX 3090 GPU (the Jetson Nano embedded development board), although its accuracy slightly decreases by 1.7% mIoU. Therefore, the memristor-based CIM accelerator has great potential to be deployed at the edge to implement lightweight semantic segmentation models like EFNet. This study showcases an algorithm-hardware co-design to realize real-time and low-power semantic segmentation at the edge.

Thousands of conductance levels in memristors integrated on CMOS

Neural networks based on memristive devices^1-3 have the ability to improve throughput and energy efficiency for machine learning^4,5 and artificial intelligence⁶, especially in edge applications^7-21. Because training a neural network model from scratch is costly in terms of hardware resources, time and energy, it is impractical to do it individually on billions of memristive neural networks distributed at the edge. A practical approach would be to download the synaptic weights obtained from the cloud training and program them directly into memristors for the commercialization of edge applications. Some post-tuning in memristor conductance could be done afterwards or during applications to adapt to specific situations. Therefore, in neural network applications, memristors require high-precision programmability to guarantee uniform and accurate performance across a large number of memristive networks^22-28. This requires many distinguishable conductance levels on each memristive device, not only laboratory-made devices but also devices fabricated in factories. Analog memristors with many conductance states also benefit other applications, such as neural network training, scientific computing and even 'mortal computing'^25,29,30. Here we report 2,048 conductance levels achieved with memristors in fully integrated chips with 256 × 256 memristor arrays monolithically integrated on complementary metal-oxide-semiconductor (CMOS) circuits in a commercial foundry. We have identified the underlying physics that previously limited the number of conductance levels that could be achieved in memristors and developed electrical operation protocols to avoid such limitations. These results provide insights into the fundamental understanding of the microscopic picture of memristive switching as well as approaches to enable high-precision memristors for various applications. Fig. 1 HIGH-PRECISION MEMRISTOR FOR NEUROMORPHIC COMPUTING.: a, Proposed scheme of the large-scale application of memristive neural networks for edge computing. Neural network training is performed in the cloud. The obtained weights are downloaded and accurately programmed into a massive number of memristor arrays distributed at the edge, which imposes high-precision requirements on memristive devices. b, An eight-inch wafer with memristors fabricated by a commercial semiconductor manufacturer. c, High-resolution transmission electron microscopy image of the cross-section view of a memristor. Pt and Ta serve as the bottom electrode (BE) and top electrode (TE), respectively. Scale bars, 1 μm and 100 nm (inset). d, Magnification of the memristor material stack. Scale bar, 5 nm. e, As-programmed (blue) and after-denoising (red) currents of a memristor are read by a constant voltage (0.2 V). The denoising process eliminated the large-amplitude RTN observed in the as-programmed state (see Methods). f, Magnification of three nearest-neighbour states after denoising. The current of each state was read by a constant voltage (0.2 V). No large-amplitude RTN was observed, and all of the states can be clearly distinguished. g, An individual memristor on the chip was tuned into 2,048 resistance levels by high-resolution off-chip driving circuitry, and each resistance level was read by a d.c. voltage sweeping from 0 to 0.2 V. The target resistance was set from 50 µS to 4,144 µS with a 2-µS interval between neighbouring levels. All readings at 0.2 V are less than 1 µS from the target conductance. Bottom inset, magnification of the resistance levels. Top inset, experimental results of an entire 256 × 256 array programmed by its 6-bit on-chip circuitry into 64 32 × 32 blocks, and each block is programmed into one of the 64 conductance levels. Each of the 256 × 256 memristors has been previously switched over one million cycles, demonstrating the high endurance and robustness of the devices.

A review of memristor: material and structure design, device performance, applications and prospects

With the booming growth of artificial intelligence (AI), the traditional von Neumann computing architecture based on complementary metal oxide semiconductor devices are facing memory wall and power wall. Memristor based in-memory computing can potentially overcome the current bottleneck of computer and achieve hardware breakthrough. In this review, the recent progress of memory devices in material and structure design, device performance and applications are summarized. Various resistive switching materials, including electrodes, binary oxides, perovskites, organics, and two-dimensional materials, are presented and their role in the memristor are discussed. Subsequently, the construction of shaped electrodes, the design of functional layer and other factors influencing the device performance are analyzed. We focus on the modulation of the resistances and the effective methods to enhance the performance. Furthermore, synaptic plasticity, optical-electrical properties, the fashionable applications in logic operation and analog calculation are introduced. Finally, some critical issues such as the resistive switching mechanism, multi-sensory fusion, system-level optimization are discussed.

Laser printed microelectronics

Printed organic and inorganic electronics continue to be of large interest for sensors, bioelectronics, and security applications. Many printing techniques have been investigated, albeit often with typical minimum feature sizes in the tens of micrometer range and requiring post-processing procedures at elevated temperatures to enhance the performance of functional materials. Herein, we introduce laser printing with three different inks, for the semiconductor ZnO and the metals Pt and Ag, as a facile process for fabricating printed functional electronic devices with minimum feature sizes below 1 µm. The ZnO printing is based on laser-induced hydrothermal synthesis. Importantly, no sintering of any sort needs to be performed after laser printing for any of the three materials. To demonstrate the versatility of our approach, we show functional diodes, memristors, and a physically unclonable function based on a 6 × 6 memristor crossbar architecture. In addition, we realize functional transistors by combining laser printing and inkjet printing.

Printed Electronics Based on 2D Material Inks: Preparation, Properties, and Applications toward Memristors

Printed electronics, which fabricate electrical components and circuits on various substrates by leveraging functional inks and advanced printing technologies, have recently attracted tremendous attention due to their capability of large-scale, high-speed, and cost-effective manufacturing and also their great potential in flexible and wearable devices. To further achieve multifunctional, practical, and commercial applications, various printing technologies toward smarter pattern-design, higher resolution, greater production flexibility, and novel ink formulations toward multi-functionalities and high quality have been insensitively investigated. 2D materials, possessing atomically thin thickness, unique properties and excellent solution-processable ability, hold great potential for high-quality inks. Besides, the great variety of 2D materials ranging from metals, semiconductors to insulators offers great freedom to formulate versatile inks to construct various printed electronics. Here, a detailed review of the progress on 2D material inks formulation and its printed applications has been provided, specifically with an emphasis on emerging printed memristors. Finally, the challenges facing the field and prospects of 2D material inks and printed electronics are discussed.

Electro-mechano responsive elastomers with self-tunable conductivity and stiffness

Materials with programmable conductivity and stiffness offer new design opportunities for next-generation engineered systems in soft robotics and electronic devices. However, existing approaches fail to harness variable electrical and mechanical properties synergistically and lack the ability to self-respond to environmental changes. We report an electro-mechano responsive Field's metal hybrid elastomer exhibiting variable and tunable conductivity, strain sensitivity, and stiffness. By synergistically harnessing these properties, we demonstrate two applications with over an order of magnitude performance improvement compared to state-of-the-art, including a self-triggered multiaxis compliance compensator for robotic manipulators, and a resettable, highly compact, and fast current-limiting fuse with an adjustable fusing current. We envisage that the extraordinary electromechanical properties of our hybrid elastomer will bring substantial advancements in resilient robotic systems, intelligent instruments, and flexible electronics.

Emerging memristive neurons for neuromorphic computing and sensing

Inspired by the principles of the biological nervous system, neuromorphic engineering has brought a promising alternative approach to intelligence computing with high energy efficiency and low consumption. As pivotal components of neuromorphic system, artificial spiking neurons are powerful information processing units and can achieve highly complex nonlinear computations. By leveraging the switching dynamic characteristics of memristive device, memristive neurons show rich spiking behaviors with simple circuit. This report reviews the memristive neurons and their applications in neuromorphic sensing and computing systems. The switching mechanisms that endow memristive devices with rich dynamics and nonlinearity are highlighted, and subsequently various nonlinear spiking neuron behaviors emulated in these memristive devices are reviewed. Then, recent development is introduced on neuromorphic system with memristive neurons for sensing and computing. Finally, we discuss challenges and outlooks of the memristive neurons toward high-performance neuromorphic hardware systems and provide an insightful perspective for the development of interactive neuromorphic electronic systems.

Experimental and Modeling Study of Metal-Insulator Interfaces to Control the Electronic Transport in Single Nanowire Memristive Devices

Memristive devices relying on redox-based resistive switching mechanisms represent promising candidates for the development of novel computing paradigms beyond von Neumann architecture. Recent advancements in understanding physicochemical phenomena underlying resistive switching have shed new light on the importance of an appropriate selection of material properties required to optimize the performance of devices. However, despite great attention has been devoted to unveiling the role of doping concentration, impurity type, adsorbed moisture, and catalytic activity at the interfaces, specific studies concerning the effect of the counter electrode in regulating the electronic flow in memristive cells are scarce. In this work, the influence of the metal-insulator Schottky interfaces in electrochemical metallization memory (ECM) memristive cell model systems based on single-crystalline ZnO nanowires (NWs) is investigated following a combined experimental and modeling approach. By comparing and simulating the electrical characteristics of single NW devices with different contact configurations and by considering Ag and Pt electrodes as representative of electrochemically active and inert electrodes, respectively, we highlight the importance of an appropriate choice of electrode materials by taking into account the Schottky barrier height and interface chemistry at the metal-insulator interfaces. In particular, we show that a clever choice of metal-insulator interfaces allows to reshape the hysteretic conduction characteristics of the device and to increase the device performance by tuning its resistance window. These results obtained from single NW-based devices provide new insights into the selection criteria for materials and interfaces in connection with the design of advanced ECM cells.

Compact artificial neuron based on anti-ferroelectric transistor

Neuromorphic machines are intriguing for building energy-efficient intelligent systems, where spiking neurons are pivotal components. Recently, memristive neurons with promising bio-plausibility have been developed, but with limited reliability, bulky capacitors or additional reset circuits. Here, we propose an anti-ferroelectric field-effect transistor neuron based on the inherent polarization and depolarization of Hf_0.2Zr_0.8O₂ anti-ferroelectric film to meet these challenges. The intrinsic accumulated polarization/spontaneous depolarization of Hf_0.2Zr_0.8O₂ films implements the integration/leaky behavior of neurons, avoiding external capacitors and reset circuits. Moreover, the anti-ferroelectric neuron exhibits low energy consumption (37 fJ/spike), high endurance (>10¹²), high uniformity and high stability. We further construct a two-layer fully ferroelectric spiking neural networks that combines anti-ferroelectric neurons and ferroelectric synapses, achieving 96.8% recognition accuracy on the Modified National Institute of Standards and Technology dataset. This work opens the way to emulate neurons with anti-ferroelectric materials and provides a promising approach to building high-efficient neuromorphic hardware.

Roles of Low-Dimensional Nanomaterials in Pursuing Human-Machine-Thing Natural Interaction

A wide variety of low-dimensional nanomaterials with excellent properties can meet almost all the requirements of functional materials for information sensing, processing, and feedback devices. Low-dimensional nanomaterials are becoming the star of hope on the road to pursuing human-machine-thing natural interactions, benefiting from the breakthroughs in precise preparation, performance regulation, structural design, and device construction in recent years. This review summarizes several types of low-dimensional nanomaterials commonly used in human-machine-thing natural interactions and outlines the differences in properties and application areas of different materials. According to the sequence of information flow in the human-machine-thing interaction process, the representative research progress of low-dimensional nanomaterials-based information sensing, processing, and feedback devices is reviewed and the key roles played by low-dimensional nanomaterials are discussed. Finally, the development trends and existing challenges of low-dimensional nanomaterials in the field of human-machine-thing natural interaction technology are discussed.

Neuromorphic-computing-based adaptive learning using ion dynamics in flexible energy storage devices

High-accuracy neuromorphic devices with adaptive weight adjustment are crucial for high-performance computing. However, limited studies have been conducted on achieving selective and linear synaptic weight updates without changing electrical pulses. Herein, we propose high-accuracy and self-adaptive artificial synapses based on tunable and flexible MXene energy storage devices. These synapses can be adjusted adaptively depending on the stored weight value to mitigate time and energy loss resulting from recalculation. The resistance can be used to effectively regulate the accumulation and dissipation of ions in single devices, without changing the external pulse stimulation or preprogramming, to ensure selective and linear synaptic weight updates. The feasibility of the proposed neural network based on the synapses of flexible energy devices was investigated through training and machine learning. The results indicated that the device achieved a recognition accuracy of ∼95% for various neural network calculation tasks such as numeric classification.