Jing Zhang

Nature Communications, Published online: 08 December 2022; doi:10.1038/s41467-022-35366-3

Information encryption technology calls for versatile multi-mode luminescent materials. Here, the authors develop phosphors with five integrated luminescence modes by exploiting the interplay of defect levels and rare-earth emission centers.

Nature Materials, Published online: 08 December 2022; doi:10.1038/s41563-022-01409-9

Both bosonic and fermionic collective states can emerge in two-dimensional semiconductor lattices, and mixing these species can further expand the landscape of quantum phases. Here, the authors report Bose–Fermi mixtures of neutral and charged excitons and the emergence of dual-density waves in an electrostatic lattice in a GaAs bilayer.

Nature Electronics, Published online: 08 December 2022; doi:10.1038/s41928-022-00881-0

By combining p-type transistors made with silicon-on-insulator technology and n-type transistors made with two-dimensional molybdenum disulfide, heterogeneous complementary field-effect transistors can be fabricated on the wafer scale.

Nature Electronics, Published online: 08 December 2022; doi:10.1038/s41928-022-00901-z

By controlling ion-dynamic capacitance, electrolyte-gated transistors can be switched between different operating modes, providing flexible neural network implementations.

Nature Nanotechnology, Published online: 08 December 2022; doi:10.1038/s41565-022-01299-7

2D materials for fast flash memory devices

Nature Electronics, Published online: 05 December 2022; doi:10.1038/s41928-022-00877-w

A van der Waals gap of 5.3 Å can be formed between a hafnium oxide dielectric and molybdenum disulfide channel through oxygen accumulation, which weakens the influence of dielectric defects on the channel material and results in transistors with low hysteresis and steep subthreshold slopes.

Silicon photonics has developed as a mature technology for communications, while laser integration on silicon remains challenging since silicon is an indirect bandgap material. In article number 2201008, Nanxi Li and co-workers present a comprehensive review on the recent advances of lasers integrated on silicon with different gain materials including III–V semiconductors, germanium/germanium tin, silicon with Raman effect, and rare-earth-doped thin film. Also, the future outlook for integrated lasers on silicon is included. The review provides valuable reference for readers interested in lasers and integrated photonics.

A monolithic neuromorphic machine vision system (NMVS), consisting of front-end retinomorphic sensors and a back-end neuromorphic convolutional neural network, is constructed based on a ferroelectric-semiconductor-transistor (FST) device structure. The front-end FST-based sensors have a broadband visual adaption capability with a large dynamic range. Combing with a linear weight programming, the NMVS shows a high image recognition accuracy.

Abstract

Bio-inspired machine visions have caused wide attentions due to the higher time/power efficiencies over the conventional architectures. Although bio-mimic photo-sensors and neuromorphic computing have been individually demonstrated, a complete monolithic vision system has rarely been studied. Here, a neuromorphic machine vision system (NMVS) integrating front-end retinomorphic sensors and a back-end convolutional neural network (CNN) based on a single ferroelectric-semiconductor-transistor (FST) device structure is reported. As a photo-sensor, the FST shows a broadband (275–808 nm) retina-like light adaption function with a large dynamic range of 20.3 stops, and as a unit of the CNN, the FST's weight can be linearly programmed. In total, the NMVS has a high recognition accuracy of 93.0% on a broadband-dim-image classification task, which is 20% higher than that of an incomplete system without the retinomorphic sensors. Because of the monolithic unit, the NVMS shows high feasibility for integrated bio-inspired machine vision systems.

An integration-type optoelectronic synapse based on ReS₂/hBN/monolayer graphene floating structure is proposed. The device mimics the visual attention behavior of the human visual system, benefiting from its excellent photosensitive memory characteristics and charge-trapping capability. The attention-based neuromorphic vision systems composed of this synapse successfully perform perceptual learning and multi-target recognition tasks.

Abstract

The human visual attention mechanism enables them to rapidly perceive important information and objects in a complex external scene; this effectively solves the problems of data redundancy, low-resolution images, and substantial computing resources. The process by which the attention system reconstructs the visual information can be considered as integrating internal attention signals with external visual details in the postsynaptic neuron. However, electronic devices that simulate visual attention modulation by incorporating device characteristics into neuromorphic vision systems (NVSs) to achieve visual attention behavior are rarely reported. Herein, a synapse device that integrates optical and electrical stimulation is designed using ReS₂/hBN/monolayer graphene heterojunction to mimic attention regulation and integrate multiple neuron signals successfully. The synapse array can imitate perceptual learning of the human visual system (HVS) to realize visual preprocessing, such as image contrast improvement and weak signal enhancement at the sensory terminal, and overcome data redundancy. Moreover, by applying gate voltage pulses, electric-tunable synaptic plasticity is successfully observed, attributed to the carrier trapping and de-trapping mechanism in the floating layer. Attention stabilization, fluctuation, distraction, and reinforcement are exhibited, simulating the attention behaviors of the HVS. Thus, an NVS with attention mechanism is established depending on the optoelectronic hybrid synaptic plasticity of the device, which successfully mimics the HVS to perform a multi-target recognition task. Furthermore, the effect of device defects on the NVS is rarely evaluated, in which a method is provided to analyze the application results of the NVS when considering uniformity and fault rate. This study may provide new inspiration for developing neuromorphic vision systems for autonomous driving and brainwave control in the future.

A ferroelectric domain patterning enabled approach is developed to achieve nanoscale reconfigurable modulation of the amplitude and polarization of SHG light in MoS₂ interfaced with PbZr_0.2Ti_0.8O₃ thin films and membranes. The light polarization exhibits threefold symmetry at polar domains and twofold symmetry at domain walls. This study opens up new opportunities for developing smart nano-photonics and optical computing systems.

Abstract

Two-dimensional (2D) transition metal dichalcogenides (TMDCs) such as MoS₂ exhibit exceptionally strong nonlinear optical responses, while nanoscale control of the amplitude, polar orientation, and phase of the nonlinear light in TMDCs remains challenging. In this work, by interfacing monolayer MoS₂ with epitaxial PbZr_0.2Ti_0.8O₃ (PZT) thin films and free-standing PZT membranes, the amplitude and polarization of the second harmonic generation (SHG) signal are modulated via ferroelectric domain patterning, which demonstrates that PZT membranes can lead to in-operando programming of nonlinear light polarization. The interfacial coupling of the MoS₂ polar axis with either the out-of-plane polar domains of PZT or the in-plane polarization of domain walls tailors the SHG light polarization into different patterns with distinct symmetries, which are modeled via nonlinear electromagnetic theory. This study provides a new material platform that enables reconfigurable design of light polarization at the nanoscale, paving the path for developing novel optical information processing, smart light modulators, and integrated photonic circuits.

Nature Physics, Published online: 01 December 2022; doi:10.1038/s41567-022-01829-z

The Luttinger liquid is a theoretical concept used to describe interacting fermions in a 1D system. Now it is shown that the model also describes electron physics in η-Mo4O11, a quasi-2D material in which 1D chains cross each other.

Memtransistors

In article number 2108025, Vinod K. Sangwan, Mark C. Hersam, and co-workers review recent advances in memtransistor devices and circuits to highlight their unique gate-tunable attributes in the context of neuromorphic computing. This work also outlines remaining challenges and future research directions, such as wafer-scale and 3D integration, that will benefit the progression of memtransistors and other multiterminal nonvolatile memory devices for use in practical neuromorphic circuits. The image depicts a crossbar array of memtransistors, which is one of the leading architectures for neuromorphic hardware.

2D Semiconductors

In article number 2106886, Peng Zhou and co-workers review and discuss the two main development paths for 2D semiconductors in the silicon age: mitigating the challenges of silicon-based devices and creating technologies that go beyond the von Neumann architecture; claiming that achieving heterogeneous integration of 2D semiconductors with silicon is the most promising option.

Memory Cells

In article number 2106321, Yanqing Wu and co-workers report the smaller footprint of nonconventional computing-in-memory devices based on black phosphorus and rhenium disulfide transistors. By adopting the charge-trapping mechanism, four-transistor nonvolatile ternary content-addressable memory cells are realized for parallel search operations with reduced complexity and thermal budget.

Logic Circuits

Are 2D semiconductors ready for next-generation IC application? In article number 2202472, Yufeng Xie, Lifeng Bian, Wenzhong Bao, and co-workers present a wafer-scale demonstration by circuit-level fabrication on a 4-inch MoS₂ wafer. While pass-transistor configuration is more like an expedient to build logic circuits based on n-type MoS₂, future development should add complementary p-type 2D semiconductors to realize more complex ICs.

Reservoir Computing

Physically implemented reservoir computing systems have aroused wide interest, due to their capability of efficiently handling temporally complex information. In article number 2108826, Ru Huang, Yuchao Yang, and co-workers present a multilayer reservoir system using ferroelectric In₂Se₃ as the key function material, which enables hierarchical information-processing capability that is suitable for time-series data processing.

Advanced Materials, Volume 34, Issue 48, December 1, 2022.

2D Transistors

2D transition metal chalcogenide (TMDC) materials have recently attracted great interest in ultra-scaled chips. The current device structures, contact engineering, and doping methods of 2D TMDC for scaling down and performance optimization are summarized by Xing Wu, He Tian, Tian-Ling Ren, and co-workers in article number 2201916, who propose the Moore's law of 2D materials and present a state-of-the-art compilation of the most sophisticated strategies developed to date.

The unique organic/inorganic alternating chain structure of 2D Ruddlesden–Popper perovskite enables the combination of the advantages of floating gate and charge trap memories to achieve fast, multi-bit, and vis-infrared broadband nonvolatile optoelectronic memory.

Abstract

Nonvolatile optoelectronic memory (NVOM) integrating the functions of optical sensing and long-term memory can efficiently process and store a large amount of visual scene information, which has become the core requirement of multiple intelligence scenarios. However, realizing NVOM with vis-infrared broadband response is still challenging. Herein, the room temperature vis-infrared broadband NVOM based on few-layer MoS₂/2D Ruddlesden–Popper perovskite (2D-RPP) van der Waals heterojunction is realized. It is found that the 2D-RPP converts the initial n-type MoS₂ into p-type and facilitates hole transfer between them. Furthermore, the 2D-RPP rich in interband states serves as an effective electron trapping layer as well as broadband photoresponsive layer. As a result, the dielectric-free MoS₂/2D-RPP heterojunction enables the charge to transfer quickly under external field, which enables a large memory window (104 V), fast write speed of 20 µs, and optical programmable characteristics from visible light (405 nm) to telecommunication wavelengths (i.e., 1550 nm) at room temperature. Trapezoidal optical programming can produce up to 100 recognizable states (>6 bits), with operating energy as low as 5.1 pJ per optical program. These results provide a route to realize fast, low power, multi-bit optoelectronic memory from visible to the infrared wavelength.

Nature Communications, Published online: 30 November 2022; doi:10.1038/s41467-022-35150-3

Author Correction: Emergence of distinct electronic states in epitaxially-fused PbSe quantum dot superlattices

Nature, Published online: 30 November 2022; doi:10.1038/s41586-022-05329-1

By using Si3N4 photonic integrated circuits on a silicon chip, a continuous-travelling-wave parametric amplifier is shown to yield a parametric gain exceeding both on-chip propagation loss as well as fibre–chip–fibre coupling losses.

Nature Electronics, Published online: 29 November 2022; doi:10.1038/s41928-022-00907-7

Publisher Correction: An anomalous Hall effect in altermagnetic ruthenium dioxide