A5100720045- Advanced Memory and Neural Computing
- Ferroelectric and Negative Capacitance Devices
- Semiconductor materials and devices
- Neuroscience and Neural Engineering
- Transition Metal Oxide Nanomaterials
- Machine Learning and ELM
- Electronic and Structural Properties of Oxides
- ZnO doping and properties
- Thin-Film Transistor Technologies
- Neural Networks and Reservoir Computing
- Power Systems and Technologies
- CCD and CMOS Imaging Sensors
- Magnetic and transport properties of perovskites and related materials
- Smart Grid and Power Systems
- Neural dynamics and brain function
- Ferroelectric and Piezoelectric Materials
- MXene and MAX Phase Materials
- Advanced Sensor and Control Systems
- Neural Networks and Applications
- Advanced Battery Materials and Technologies
- Power Systems and Renewable Energy
- Robotics and Sensor-Based Localization
- Advancements in Semiconductor Devices and Circuit Design
- Advanced Neural Network Applications
- Conducting polymers and applications
Peking University
2016-2025
Nanjing University of Aeronautics and Astronautics
2025
Beijing Advanced Sciences and Innovation Center
2023-2024
Neusoft (China)
2022-2024
Civil Aviation Flight University of China
2024
Institute of Microelectronics
2014-2023
Nanjing University
2020
Research Institute of Petroleum Exploration and Development
2019-2020
China Southern Power Grid (China)
2020
Beijing YouAn Hospital
2019
The safety of CRISPR (clustered regularly interspaced short palindromic repeats)-based genome editing in the context human gene therapy is largely unknown. CCR5 a reasonable but not absolutely protective target for cure immunodeficiency virus type 1 (HIV-1) infection, because CCR5-null blood cells are resistant to HIV-1 entry. We transplanted CRISPR-edited CCR5-ablated hematopoietic stem and progenitor (HSPCs) into patient with infection acute lymphoblastic leukemia. leukemia was complete...
The characteristics and mechanism of conduction/set process in TiN∕ZnO∕Pt-based resistance random access memory devices with stable reproducible nanosecond bipolar switching behavior were studied. dependencies on cell area, operating temperature, frequency indicate that the conduction low-resistance states is due to electrons hopping through filament paths. We also identify set essentially equivalent a soft dielectric breakdown associated polarization effect caused by migration space charges...
Resistance switching (RS) devices have potential to offer computing and memory function. A new computer unit is built of RS array, where processing storing information occur on same devices. states stored in located arbitrary positions array can be performed various nonvolatile logic operations. Logic functions reconfigured by altering trigger signals.
The exponential growth of various complex images is putting tremendous pressure on storage systems. Here, we propose a memristor-based system with an integrated near-storage in-memory computing-based convolutional autoencoder compression network to boost the energy efficiency and speed image compression/retrieval improve density. We adopt 4-bit memristor arrays experimentally demonstrate functions system. step-by-step quantization aware training scheme equivalent transformation for transpose...
Reservoir computing is a powerful neural network-based paradigm for spatiotemporal signal processing. Recently, physical reservoirs have been explored based on various electronic devices with outstanding efficiency. However, the inflexible temporal dynamics of these posed fundamental restrictions in processing signals timescales. Here, we fabricated thin-film transistors controllable dynamics, which can be easily tuned electrical operation and showed excellent cycle-to-cycle uniformity....
An implantation doping approach is implemented to fabricate Gd-doped HfO2 resistive random access memory (RRAM) devices. The significantly enhanced performances are achieved in the RRAM devices including improved uniformity of switching parameters, enlarged ON/OFF ratio, and increased speed without obvious reliability degradation. This performance improvement clarified suppressed randomicity oxygen vacancy filaments’ formation reduced ion migration barrier induced by trivalent Gd-doping...
Neuromorphic computing is an attractive computation paradigm that complements the von Neumann architecture. The salient features of neuromorphic are massive parallelism, adaptivity to complex input information, and tolerance errors. As one most crucial components in a system, electronic synapse requires high device integration density low-energy consumption. Oxide-based resistive switching devices have been shown be promising candidate realize functions synapse. However, intrinsic variation...
A unified model is proposed to elucidate the resistive switching behavior of metal-oxide-based random access memory devices using concept electron hopping transport along filamentary conducting paths in dielectric layer. The calculation shows that a low-electron-occupied region conductive filament (CF) formed when critical electric field applied. oxygen vacancies this are recombined with ions, resulting rupture CFs. mechanism was verified by experiments and theoretical calculations. In...
Oxygen vacancy (VO) plays the critical role for resistive switching in transition metal oxide random access memory (RRAM). First principles calculation is performed to study impact of metallic ion (Al, Ti, or La) doping ZrO2 on behaviors VO, including defect energy level and formation (Evf). Trivalent dopant (Al significantly reduces Evf. Based calculated results, ZrO2-based RRAM devices are designed control improved uniformity demonstrated experiments.
Resistive switching processes in HfO2 are studied by electron holography and situ energy-filtered imaging. The results show that oxygen vacancies gradually generated the oxide layer under ramped electrical bias, finally form several conductive channels connecting two electrodes. It also shows process occurs at top interface of hafnia layer. As a service to our authors readers, this journal provides supporting information supplied authors. Such materials peer reviewed may be re-organized for...
Resistive random-access memory devices with atomic layer deposition HfO2 and radio frequency sputtering TiOx as resistive switching layers were fabricated successfully. Low-power characteristic 1.52 μW set power (1 μA@1.52 V) 1.12 reset μA@1.12 was obtained in the HfO2/TiOx (RRAM) by controlling oxygen content of layer. Besides, influence during process on properties would be discussed detail. The investigations indicated that "soft breakdown" occurred easily electrical forming/set RRAM high...
A technical solution is presented to improve the uniformity of -based resistive switching memory by embedding thin Al layers between and electrode layers. Compared with those pure devices, a remarkably improved parameters such as forming voltages, set resistances in high/low states was demonstrated devices embedded atoms are assumed diffuse into films intended localize oxygen vacancies due reduced vacancy formation energy, thus stabilizing generation conductive filaments, which helps uniformity.
<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <formula formulatype="inline"><tex Notation="TeX">$\hbox{Al}/\hbox{CeO}_{x}/\hbox{Pt}$</tex></formula> devices with nonstoichiometric Notation="TeX">$\hbox{CeO}_{x}\ (\hbox{1.5} ≪ x \hbox{2})$</tex> </formula> films were fabricated. The unique resistive switching (RS) behaviors for random access memory applications, including stable and sharp bipolar RS processes a multilevel self-stop set...
The retention failure of bipolar oxide-based resistive switching memory is investigated. A new physical model proposed to elucidate the typical behavior with a sudden resistance transition, which quite different from that traditional memories. In model, temperature- and bias-dependent probability time devices can be quantified. voltage-acceleration method developed evaluate
An analog neuromorphic system is developed based on the fabricated resistive switching memory array. A novel training scheme proposed to optimize performance of by utilizing segmented synaptic behavior. The demonstrated a grayscale image recognition. According experiment results, optimized one improves learning accuracy from 77.83% 91.32%, decreases energy consumption more than two orders, and substantially boosts efficiency compared traditional scheme.
HfOx-based resistive switching device has been explored as one of the promising candidates for electronic synapses neuromorphic computing systems due to its high performance, low cost, and compatibility with CMOS technology. To meet codesign requirement neurons in systems, a compact model that can capture synaptic futures is developed. The developed accurately describe multilevel conductance transition behaviors during RESET process depression learning well binary stochastic behavior SET...
In this paper, a hardware-realized neuromorphic system for pattern recognition is presented. The directly captures images from the environment, and then conducts classification using single layer neural network. Metal-oxide resistive random access memory (RRAM) used as electronic synapses, threshold-controlled neurons are proposed postsynaptic to save area simplify operation. neuron, no capacitor utilized, which contributes higher integration density. total energy consumption of RRAM...
Nonvolatile and cascadable stateful logic operations are experimentally demonstrated within a 1 k-bit one-transistor-one-resistor (1T1R) resistive random access memory (RRAM) array, where NAND gates serve as the building blocks. A robust dual-gate-voltage operation scheme is proposed. The effects of transistor ON robustness to device parameter variations discussed. parallel 4-bit bitwise XOR implemented in 1T1R array by cascading gates. This letter presents feasible approach in-memory...
In this work, a new operation scheme is developed to improve the state stability of multi-level resistive random-access memory (RRAM) array. We found that instability after programming mainly derived from excessive oxygen vacancies generated by abrupt SET process. Based on understanding operation, triangular pulse adopted for RRAM, which can effectively suppress short-term relaxation and long-time retention. Comparison results show degradation recognition accuracy Cifar-10 over time in...
The development of efficient and durable electrocatalysts for the oxygen evolution reaction (OER) is critical to advancing anion exchange membrane water electrolysis (AEMWE) technology sustainable hydrogen production. Herein, we...
A novel strategy based on defect engineering is proposed for high-performance multibit data storage in oxide-based resistive random access memory (RRAM). Key innovations are: 1) material-oriented cell desired modification of physical locations generated oxygen vacancies switching layer; and 2) operation scheme to control the amount vacancy conducting filament regions during switching. Proper doping approach applied suppress formation clusters due avalanching effect forming set processes....
In this paper, we devise and optimize schemes for the resistive random-access memory (RRAM)-based hardware implementation of convolutional neural networks (CNNs). The key achievements are as follows: 1) a specific circuit CNN corresponding operation methods is presented; 2) quantization utilizing binary RRAM or with multilevel resistances synapses proposed, simulations show that our system percentage accuracy MNIST data set using 97% synapses; 3) influence factors such number size kernels,...
Abstract The traditional Boolean computing paradigm based on the von Neumann architecture is facing great challenges for future information technology applications such as big data, Internet of Things (IoT), and wearable devices, due to limited processing capability issues binary data storage computing, non-parallel processing, buses requirement between memory units logic units. brain-inspired neuromorphic believed be one promising solutions realizing more complex functions with a lower...