Significant Power Consumption Reduction and Speed Boosting in Phase Change Memory with Nanocurrent Channels.

The excessive power consumption is challenging for phase change memory (PCM) on its way to becoming universal memory in complex hierarchies of memory systems. Here, from the perspective of device structure, by adding a nanocurrent-channel (NCC) layer between the electrode layer and phase change layer, a RESET power consumption reduction by more than 95% and 10 times faster SET speed were realized simultaneously. Through the first principle calculations, Au and SiO2 were screened as the metal and insulating matrix material of NCC layer, respectively. Our PCM device with a Au-SiO2 NCC layer shows an ultralow RESET power consumption, down to 381 fJ, and an ultrafast SET speed (8 ns). Much higher current density near NCC in the phase change layer and thermal barrier effect of insulating matrix material were confirmed by finite element analysis (FEA), and the role of Au nanochannels was revealed by transmission electron microscopy (TEM). Our NCC layer structure provides a simple and practicable method to significantly decrease PCM power consumption.

Reconfigurable Multilevel Storage and Neuromorphic Computing Based on Multilayer Phase-Change Memory.

In the era of big data, the amount of global data is increasing exponentially, and the storage and processing of massive data put forward higher requirements for memory. To deal with this challenge, high-density memory and neuromorphic computing have been widely investigated. Here, a gradient-doped multilayer phase-change memory with two-level states, four-level states, and linear conductance evolution using different pulse operations is proposed. The mechanism of multilevel states is revealed through high-resolution transmission electron microscopy (HRTEM) and finite-element analysis (FEA), which show that the sequential phase change among different sublayers is realized due to the different physical properties of the sublayers with different doping concentrations. Taking advantage of the devices' linear conductance evolution characteristic, a handwritten digit (28 × 28 pixel) recognition task is implemented with a high learning accuracy of 93.46% by building a simulated artificial neural network made up of this gradient-doped multilayer phase-change memory. It is proved that this gradient-doped multilayer phase-change memory is capable of both binary multilevel digital storage and brain-inspired analog in-memory computing in the same device, enabling reconfigurable applications in the future.

Oxygen Vacancy Compensation-Induced Analog Resistive Switching in the SrFeO3-δ/Nb:SrTiO3 Epitaxial Heterojunction for Noise-Tolerant High-Precision Image Recognition.

Neuromorphic computing, inspired by the brain's architecture, promises to surpass the limitations of von Neumann computing. In this paradigm, synaptic devices play a crucial role, with resistive switching memory (memristors) emerging as promising candidates due to their low power consumption and scalability advantages. This study focuses on the development of metal/oxide-semiconductor heterojunctions, which offer several technological advantages and have broad potential for applications in artificial neural synapses. However, constructing high-quality epitaxial interfaces between metal and oxide semiconductors and designing modifiable contact barriers are challenging. Herein, we construct high-quality epitaxial metal/semiconductor interfaces based on the metallicity of the perovskite phase SrFeO3-δ (PV-SFO) and a small Schottky barrier in contact with Nb-doped SrTiO3 (NSTO). X-ray diffraction patterns, reciprocal space mapping results, and cross-sectional transmission electron microscopy images reveal that the prepared PV-SFO film exhibits a perfect single-crystal structure and an excellent epitaxial interface with the NSTO (111) substrate. The corresponding memristor exhibits analog-type resistive-variable characteristics with an ON/OFF ratio of ∼1000, stable data retention after 10,000 s, and no noticeable fluctuation in resistance after 10,000 pulse cycles. Electron energy loss spectroscopy, first-principles calculations, and electrical measurements reveal that compensating or restoring oxygen vacancies at the NSTO surface decreases or increases the contact barrier between PV-SFO and NSTO, respectively, thereby gradually regulating the resistance value. Furthermore, high-quality epitaxial PV-SFO/NSTO devices achieve up to 98.21% recognition accuracy for handwriting recognition tasks using LeNet-5-based network structures and 92.21% accuracy for color images using visual geometry group (VGG) network structures. This work contributes to the advancement of interface-type memristors and provides valuable insights into enhancing synaptic functionality in neuromorphic computing systems.

Crossmodal sensory neurons based on high-performance flexible memristors for human-machine in-sensor computing system.

Constructing crossmodal in-sensor processing system based on high-performance flexible devices is of great significance for the development of wearable human-machine interfaces. A bio-inspired crossmodal in-sensor computing system can perform real-time energy-efficient processing of multimodal signals, alleviating data conversion and transmission between different modules in conventional chips. Here, we report a bio-inspired crossmodal spiking sensory neuron (CSSN) based on a flexible VO2 memristor, and demonstrate a crossmodal in-sensor encoding and computing system for wearable human-machine interfaces. We demonstrate excellent performance in the VO2 memristor including endurance (>1012), uniformity (0.72% for cycle-to-cycle variations and 3.73% for device-to-device variations), speed (<30 ns), and flexibility (bendable to a curvature radius of 1 mm). A flexible hardware processing system is implemented based on the CSSN, which can directly perceive and encode pressure and temperature bimodal information into spikes, and then enables the real-time haptic-feedback for human-machine interaction. We successfully construct a crossmodal in-sensor spiking reservoir computing system via the CSSNs, which can achieve dynamic objects identification with a high accuracy of 98.1% and real-time signal feedback. This work provides a feasible approach for constructing flexible bio-inspired crossmodal in-sensor computing systems for wearable human-machine interfaces.

High-Performance Proton Field-Effect Transistor Based on Two-Dimensional Cd Vacancy-Resided Cd0.85PS3Li0.15H0.15.

Ion transport is a critical phenomenon underpinning numerous biological, physical, and chemical systems. Proton transistors leveraging proton transport face significant limitations, such as a low on-off ratio and deficient carrier mobility, which restrict their applicability in biological and other scenarios. This study explores the use of two-dimensional (2D) vacancy-residing transition metal phosphorus trichallcogenide-based membranes as the active layer for proton field-effect transistors. The synthesized Cd0.85PS3Li0.15H0.15 membrane exhibits a well-organized layered structure and high hydrophilicity, with nanometer-sized interlayers containing interconnected water networks. These distinct features facilitate proton conduction, leading to a high proton conductivity value of 0.83 S cm-1 at 98% relative humidity and 90 °C, with an activation energy of 0.26 eV. The Cd0.85PS3Li0.15H0.15-based proton transistor demonstrates tunability via gate voltage, thereby enabling effective modulation of proton flow across source and drain electrodes. The transistor notably showcases superior switching characteristics, with an on/off ratio surpassing 5.51 and a carrier mobility of 8.84 × 10-2 cm2 V-1 s-1. The underlying mechanism for this performance enhancement is attributed to electric-field-induced switching in Cd vacancies. This research boosts the development of highly versatile ionotropic devices by introducing advanced 2D ion-conductive membranes.

Manipulating 2D Materials through Strain Engineering.

This review explores the growing interest in 2D layered materials, such as graphene, h-BN, transition metal dichalcogenides (TMDs), and black phosphorus (BP), with a specific focus on recent advances in strain engineering. Both experimental and theoretical results are delved into, highlighting the potential of strain to modulate physical properties, thereby enhancing device performance. Various strain engineering methods are summarized, and the impact of strain on the electrical, optical, magnetic, thermal, and valleytronic properties of 2D materials is thoroughly examined. Finally, the review concludes by addressing potential applications and challenges in utilizing strain engineering for functional devices, offering valuable insights for further research and applications in optoelectronics, thermionics, and spintronics.

Synapse Neurotransmitter Channel-Inspired AlOx Memristor with "V" Type Oxygen Vacancy Distribution.

Memristor possesses great potential and advantages in neuromorphic computing, while consistency and power consumption issues have been hindering its commercialization. Low cost and accuracy are the advantages of human brain, so memristors can be used to construct brain-like synaptic devices to solve these problems. In this work, a five-layer AlOx device with a V-shaped oxygen distribution is used to simulate biological synapses. The device simulates synapse structurally. Further, under electrical stimulation, O2- moves to the Ti electrode and oxygen vacancy (Vo) moves to the Pt electrode, thus forming a conductive filament (CF), which simulates the Ca2+ flow and releases neurotransmitters to the postsynaptic membrane, thus realizing the transmission of information. By controlling applied voltage, the regulation of Ca2+ gated pathway is realized to control the Ca2+ internal flow and achieve different degrees of information transmission. Long-term Potentiation (LTP)/Long-term Depression (LTD), Spike Timing Dependent Plasticity (STDP), these basic synaptic performances can be simulated. The AlOx device realizes low power consumption (56.7 pJ/392 fJ), high switching speed (25 ns/60 ns), and by adjusting the window value, the nonlinearity is improved (0.133/0.084), a high recognition accuracy (98.18%) is obtained in neuromorphic simulation. It shows a great prospect in multi-value storage and neuromorphic computing.

Detection of Biomolecules Using Solid-State Nanopores Fabricated by Controlled Dielectric Breakdown.

Nanopore sensor technology is widely used in biomolecular detection due to its advantages of low cost and easy operation. In a variety of nanopore manufacturing methods, controlled dielectric breakdown has the advantages of a simple manufacturing process and low cost under the premise of ensuring detection performance. In this paper, we have made enhancements to the applied pulses in controlled dielectric breakdown and utilized the improved dielectric breakdown technique to fabricate silicon nitride nanopores with diameters of 5 to 15 nm. Our improved fabrication method offers the advantage of precise control over the nanopore diameter (±0.4 nm) and enhances the symmetry of the nanopore. After fabrication, we performed electrical characterization on the nanopores, and the IV characteristics exhibited high linearity. Subsequently, we conducted detection experiments for DNA and protein using the prepared nanopores to assess the detection performance of the nanopores fabricated using our method. In addition, we also give a physical model of molecule translocation through the nanopores to give a reasonable explanation of the data processing results.

Ovonic threshold switching-based artificial afferent neurons for thermal in-sensor computing.

Artificial afferent neurons in the sensory nervous system inspired by biology have enormous potential for efficiently perceiving and processing environmental information. However, the previously reported artificial afferent neurons suffer from two prominent challenges: considerable power consumption and limited scalability efficiency. Herein, addressing these challenges, a bioinspired artificial thermal afferent neuron based on a N-doped SiTe ovonic threshold switching (OTS) device is presented for the first time. The engineered OTS device shows remarkable uniformity and robust endurance, ensuring the reliability and efficacy of the artificial afferent neurons. A substantially decreased leakage current of the SiTe OTS device by nitrogen doping results in ultra-low power consumption less than 0.3 nJ per spike for artificial afferent neurons. The inherent temperature response exhibited by N-doped SiTe OTS materials allows us to construct a highly compact artificial thermal afferent neuron over a wide temperature range. An edge detection task is performed to further verify its thermal perceptual computing function. Our work provides an insight into OTS-based artificial afferent neurons for electronic skin and sensory neurorobotics.

Structure, bonding and electronic characteristics of amorphous Se.

Ovonic threshold switching (OTS) selectors can effectively improve the storage density and suppress the leakage current of advanced phase-change memory devices. As a prototypical OTS material, amorphous GeSe is widely investigated. But the attention paid to amorphous Se (i.e., the functional constituent in amorphous GeSe) has been very limited up to now. Here we have explored the structure, bonding and electronic characteristics of amorphous Se using ab initio molecular dynamics simulations. The results reveal that the Se atoms in amorphous Se tend to form 2-coordinated configurations, and they connect with each other to form long chains. The fraction of the vibrational density of state located in the high frequency range is relatively large, and the formation energy of the Se-Se bond is as large as 4.44 eV, hinting that the Se-Se bonds in chains possess high stability. In addition, the mid-gap state related to the OTS behavior is also found in amorphous Se despite the small proportion. Our findings enrich the knowledge of amorphous Se, which aids the applications of Se-based OTS selectors.

Electrostatic Gating of Spin Dynamics of a Quasi-2D Kagome Magnet.

Electrostatic gating has emerged as a powerful technique for tailoring the magnetic properties of two-dimensional (2D) magnets, offering exciting prospects including enhancement of magnetic anisotropy, boosting Curie temperature, and strengthening exchange coupling effects. Here, we focus on electrical control of the ferromagnetic resonance of the quasi-2D Kagome magnet Cu(1,3-bdc). By harnessing an electrostatic field through ionic liquid gating, significant shifts are observed in the ferromagnetic resonance field in both out-of-plane and in-plane measurements. Moreover, the effective magnetization and gyromagnetic ratios display voltage-dependent variations. A closer examination reveals that the voltage-induced changes can modulate magnetocrystalline anisotropy by several hundred gauss, while the impact on orbital magnetization remains relatively subtle. Density functional theory (DFT) calculations reveal varying d-orbital hybridizations at different voltages. This research unveils intricate physics within the Kagome lattice magnet and further underscores the potential of electrostatic manipulation in steering magnetism with promising implications for the development of spintronic devices.

Stereo-Hindrance Engineering of A Cation toward <110>-Oriented 2D Perovskite with Minimized Tilting and High-Performance X-Ray Detection.

2D <100>-oriented Dion-Jacobson or Ruddlesden-Popper perovskites are widely recognized as promising candidates for optoelectronic applications. However, the large interlayer spacing significantly hinders the carrier transport. <110>-oriented 2D perovskites naturally exhibit reduced interlayer spacings, but the tilting of metal halide octahedra is typically serious and leads to poor charge transport. Herein, a <110>-oriented 2D perovskite EPZPbBr4 (EPZ = 1-ethylpiperazine) with minimized tilting is designed through A-site stereo-hindrance engineering. The piperazine functional group enters the space enclosed by the three [PbBr6]4- octahedra, pushing Pb─Br─Pb closer to a straight line (maximum Pb─Br─Pb angle ≈180°), suppressing the tilting as well as electron-phonon coupling. Meanwhile, the ethyl group is located between layers and contributes an extremely reduced effective interlayer distance (2.22 Å), further facilitating the carrier transport. As a result, EPZPbBr4 simultaneously demonstrates high µτ product (1.8 × 10-3 cm2 V-1) and large resistivity (2.17 × 1010 Ω cm). The assembled X-ray detector achieves low dark current of 1.02 × 10-10 A cm-2 and high sensitivity of 1240 µC Gy-1 cm-2 under the same bias voltage. The realized specific detectivity (ratio of sensitivity to noise current density, 1.23 × 108 µC Gy-1 cm-1 A-1/2) is the highest among all reported perovskite X-ray detectors.

Resistive Memory Devices at the Thinnest Limit: Progress and Challenges.

The Si-based integrated circuits industry has been developing for more than half a century, by focusing on the scaling-down of transistor. However, the miniaturization of transistors will soon reach its physical limits, thereby requiring novel material and device technologies. Resistive memory is a promising candidate for in-memory computing and energy-efficient synaptic devices that can satisfy the computational demands of the future applications. However, poor cycle-to-cycle and device-to-device uniformities hinder its mass production. 2D materials, as a new type of semiconductor, is successfully employed in various micro/nanoelectronic devices and have the potential to drive future innovation in resistive memory technology. This review evaluates the potential of using the thinnest advanced materials, that is, monolayer 2D materials, for memristor or memtransistor applications, including resistive switching behavior and atomic mechanism, high-frequency device performances, and in-memory computing/neuromorphic computing applications. The scaling-down advantages of promising monolayer 2D materials including graphene, transition metal dichalcogenides, and hexagonal boron nitride are presented. Finally, the technical challenges of these atomic devices for practical applications are elaborately discussed. The study of monolayer-2D-material-based resistive memory is expected to play a positive role in the exploration of beyond-Si electronic technologies.

Self-Rectifying Memristors for Three-Dimensional In-Memory Computing.

Costly data movement in terms of time and energy in traditional von Neumann systems is exacerbated by emerging information technologies related to artificial intelligence. In-memory computing (IMC) architecture aims to address this problem. Although the IMC hardware prototype represented by a memristor is developed rapidly and performs well, the sneak path issue is a critical and unavoidable challenge prevalent in large-scale and high-density crossbar arrays, particularly in three-dimensional (3D) integration. As a perfect solution to the sneak-path issue, a self-rectifying memristor (SRM) is proposed for 3D integration because of its superior integration density. To date, SRMs have performed well in terms of power consumption (aJ level) and scalability (>102  Mbit). Moreover, SRM-configured 3D integration is considered an ideal hardware platform for 3D IMC. This review focuses on the progress in SRMs and their applications in 3D memory, IMC, neuromorphic computing, and hardware security. The advantages, disadvantages, and optimization strategies of SRMs in diverse application scenarios are illustrated. Challenges posed by physical mechanisms, fabrication processes, and peripheral circuits, as well as potential solutions at the device and system levels, are also discussed.

Magnetic-ferroelectric synergic control of multilevel conducting states in van der Waals multiferroic tunnel junctions towards in-memory computing.

van der Waals (vdW) multiferroic tunnel junctions (MFTJs) based on two-dimensional materials have gained significant interest due to their potential applications in next-generation data storage and in-memory computing devices. In this study, we construct vdW MFTJs by employing monolayer Mn2Se3 as the spin-filter tunnel barrier, TiTe2 as the electrodes and In2S3 as the tunnel barrier to investigate the spin transport properties based on first-principles quantum transport calculations. It is highlighted that apparent tunneling magnetoresistance (TMR) and tunneling electroresistance (TER) effects with a maximum TMR ratio of 6237% and TER ratio of 1771% can be realized by using bilayer In2S3 as the tunnel barrier under finite bias. Furthermore, the physical origin of the distinguished TMR and TER effects is unraveled from the k||-resolved transmission spectra and spin-dependent projected local density of states analysis. Interestingly, four distinguishable conductance states reveal the implementation of four-state nonvolatile data storage using one MFTJ unit. More importantly, in-memory logic computing and multilevel data storage can be achieved at the same time by magnetic switching and electrical control, respectively. These results shed light on vdW MFTJs in the applications of in-memory computing as well as multilevel data storage devices.

Adaptive Neural Activation and Neuromorphic Processing via Drain-Injection Threshold-Switching Float Gate Transistor Memory.

Hetero-modulated neural activation is vital for adaptive information processing and learning that occurs in brain. To implement brain-inspired adaptive processing, previously various neurotransistors oriented for synaptic functions are extensively explored, however, the emulation of nonlinear neural activation and hetero-modulated behaviors are not possible due to the lack of threshold switching behavior in a conventional transistor structure. Here, a 2D van der Waals float gate transistor (FGT) that exhibits steep threshold switching behavior, and the emulation of hetero-modulated neuron functions (integrate-and-fire, sigmoid type activation) for adaptive sensory processing, are reported. Unlike conventional FGTs, the threshold switching behavior stems from impact ionization in channel and the coupled charge injection to float gate. When a threshold is met, a sub-30 mV dec-1 increase of transistor conductance by more than four orders is triggered with a typical switch time of approximately milliseconds. Essentially, by feeding light sensing signal as the modulation input, it is demonstrated that two typical tasks that rely on adaptive neural activation, including collision avoidance and adaptive visual perception, can be realized. These results may shed light on the emulation of rich hetero-modulating behaviors in biological neurons and the realization of biomimetic neuromorphic processing at low hardware cost.

Research Progress on the Application of Topological Phase Transition Materials in the Field of Memristor and Neuromorphic Computing.

Topological phase transition materials have strong coupling between their charge, spin orbitals, and lattice structure, which makes them have good electrical and magnetic properties, leading to promising applications in the fields of memristive devices. The smaller Gibbs free energy difference between the topological phases, the stable oxygen vacancy ordered structure, and the reversible topological phase transition promote the memristive effect, which is more conducive to its application in information storage, information processing, information calculation, and other related fields. In particular, extracting the current resistance or conductance of the two-terminal memristor to convert to the weight of the synapse in the neural network can simulate the behavior of biological synapses in their structure and function. In addition, in order to improve the performance of memristors and better apply them to neuromorphic computing, methods such as ion doping, electrode selection, interface modulation, and preparation process control have been demonstrated in memristors based on topological phase transition materials. At present, it is considered an effective method to obtain a unique resistive switching behavior by improving the process of preparing functional layers, regulating the crystal phase of topological phase transition materials, and constructing interface barrier-dependent devices. In this review, we systematically expound the resistance switching mechanism, resistance switching performance regulation, and neuromorphic computing of topological phase transition memristors, and provide some suggestions for the challenges faced by the development of the next generation of non-volatile memory and brain-like neuromorphic devices based on topological phase transition materials.

Erratum: Roadmap for phase change materials in photonics and beyond.

[This corrects the article DOI: 10.1016/j.isci.2023.107946.].

2D van der Waals Vertical Heterojunction Transistors for Ternary Neural Networks.

Compared with binary systems, ternary computing systems can utilize fewer devices to realize the same information density. However, most ternary computing systems based on binary CMOS circuits require additional devices to bridge binary processing and ternary computing. Exploring new device architectures for direct ternary processing and computing becomes the key to promoting ternary computing systems. Here, we demonstrated a 2D van der Waals vertical heterojunction transistor (V-HTR) with three flat conductance states, which can be the basic cell in ternary circuits to perform ternary processing and computing, without additional devices. A ternary neural network (TNN) and a ternary inverter were demonstrated based on the V-HTRs. The TNN can eliminate fuzzy data and output only clear data by building a ternary quantization function. By demonstrating both ternary logic and a TNN on the same device architecture, the 2D V-HTR shows potential as a basic hardware unit for future ternary computing systems.

Landauer-QFLPS Model for Mixed Schottky-Ohmic Contact Two-Dimensional Transistors.

Two-dimensional material-based field-effect transistors (2DM-FETs) are playing a revolutionary role in electronic devices. However, before electronic design automation (EDA) for 2DM-FETs can be achieved, it remains necessary to determine how to incorporate contact transports into model. Reported methods compromise between physical intelligibility and model compactness due to the heterojunction nature. To address this, quasi-Fermi-level phase space theory (QFLPS) is generalized to incorporate contact transports using the Landauer formula. It turns out that the Landauer-QFLPS model effectively overcomes the issue of concern. The proposed new formula can describe 2DM-FETs with Schottky or Ohmic contacts with superior accuracy and efficiency over previous methods, especially when describing non-monotonic drain conductance characteristics. A three-bit threshold inverter quantizer (TIQ) circuit is fabricated using ambipolar black phosphorus and it is demonstrated that the model accurately predicts circuit performance. The model could be very effective and valuable in the development of 2DM-FET-based integrated circuits.