One long-standing goal in the emerging neuromorphic field is to create a reliable neural network hardware implementation that has low energy consumption, while providing massively parallel computation. Although diverse oxide-based devices have made significant progress as artificial synaptic and neuronal components, these devices still need further optimization regarding linearity, symmetry, and stability. Here, we present a proof-of-concept experiment for integrated neuromorphic computing networks by utilizing spintronics-based synapse (spin-S) and neuron (spin-N) devices, along with linear and symmetric weight responses for spin-S using a stripe domain and activation functions for spin-N. An integrated neural network of electrically connected spin-S and spin-N successfully proves the integration function for a simple pattern classification task. We simulate a spin-N network using the extracted device characteristics and demonstrate a high classification accuracy (over 93%) for the spin-S and spin-N optimization without the assistance of additional software or circuits required in previous reports. These experimental studies provide a new path toward establishing more compact and efficient neural network systems with optimized multifunctional spintronic devices.
Highly stretchable self‐powered energy sources are promising options for powering diverse wearable smart electronics. However, commercially existing energy sources are disadvantaged by tensile strain limitations and constrained deformability. Here, 1D thread‐based highly stretchable triboelectric nanogenerators (HS‐TENGs), a crucial step toward overcoming these obstacles, are developed based on a highly stretchable coaxial‐type poly[styrene‐b‐isoprene‐b‐styrene] (SIS) elastomer tube. Carbon conductive ink is injected into the SIS tube as a core 1D electrode that remains almost unaffected even under 250% stretching because of its low Young's modulus. To further facilitate power generation by the HS‐TENG, a composite of barium titanate nanoparticles (BaTiO3 NPs) and polydimethylsiloxane (PDMS) is coated on the initial SIS tube to modulate the dielectric permittivity based on variations in the BaTiO3 NPs volume ratio. The 1D PDMS/BaTiO3 NP composite‐coated SIS and a nylon 6‐coated 2D Ni–Cu conductive fabric are selected as triboelectric bottom and top layers, respectively. Woven HS‐TENGs textiles yield consistent power output under various extreme and harsh conditions, including folded, twisted, and washed states. These experimental findings indicate that the approach may become useful for realizing stretchable multifunctional power sources for various wearable electronics.
Three-dimensional stackable memory frames involving the integration of two-terminal scalable crossbar arrays are expected to meet the demand for high-density memory storage, fast switching speed, and ultra-low power operation. However, two-terminal crossbar arrays introduce an unintended sneak path, which inevitably requires bidirectional nonlinear selectors. In this study, the advanced threshold switching (TS) features of ZnTe chalcogenide material-based selectors provide bidirectional threshold switching behavior, nonlinearity of 104, switching speed of less than 100 ns, and switching endurance of more than 107. In addition, thermally robust ZnTe selectors (up to 400 ℃) can be obtained through the use of nitrogen-annealing treatment. This process can prevent possible phase separation phenomena observed in generic chalcogenide materials during thermal annealing which occurs even at a low temperature of 250 ℃. The possible characteristics of the electrically and thermally advanced TS nature are described by diverse structural and electrical analyses through the Poole–Frankel conduction model.
In article number 1903217, Jin Pyo Hong and co‐workers demonstrate a 1D thread‐based highly stretchable triboelectric nanogenerator (HS‐TENG) that may be used in the future real textile industry, contributing to the proliferation of wearable electronic devices including stretchable electrodes, energy conversion, and storage, as 2D textile and 1D flexible‐based TENGs still provide insufficient degrees of freedom in geometric parameters such as tensile strain and deformability.
To further ensure such features, substantial efforts have been dedicated toward developing a suitable candidate for data storage media. [7][8][9] In recent years, carbon-based media has shown great potential as a reliable option owing to its low cost, simple chemical composition, affordable compatibility with complementary metal-oxidesemiconductor processes, and fast switching speed. [10][11][12][13] Thus, numerous studies have also been conducted to attain a firm understanding of the nature and to achieve improved device performance.To date, the possible underlying nature of carbon layer-based RS has been expected to the formation and rupture of existing sp 2 conductive filament (CFs) initially arising from the electroforming step: that is, the conversion between the sp 2 and sp 3 carbon complexes takes places via bias, leading to two representative RS behaviors. One class is unipolar RS frequently observed from the tetrahedral amorphous carbon (tα-C) [14,15] or diamond-like carbon (DLC) layer. [16][17][18] The corresponding unipolar RS is likely attributable to the fuse-antifuse process induced by the current-driven temperature increase. The second class is bipolar-RS, mainly observed in graphene oxides [19][20][21] or α-C:H [22] layers, where bias-dependent oxygen or hydrogen ion drift plays a crucial role in the transition between the sp 2 and sp 3 bonds through the removal or absorption of ions under bias. Recently, other promising work addressed the advancing device performance of oxygenated amorphous carbon (α-C:O x ) layers, such as the high on/off ratio and fast switching speed compared to graphene oxide. [21,23] Furthermore, this α-C:O x active layer is also expected to play a key role in providing simple wafer-scale fabrication allowing good reproducibility at room temperature, amorphous nature, high degree of tunable chemical characteristics by simply changing oxygen content, and outstanding device performance, compared with those of other resistive switching active layer. However, these α-C:O x layer-based devices exhibited higher forming voltages and insufficient stability/reliability features, even if they did possess an appreciable switching speed. [18,24] In addition, the stochastic distribution in forming voltages during consecutive sweeps led to significant variation in device performance, thus requiring high power consumption to address all individual cells. [25][26][27] Recent advances in resistive switching devices have garnered a considerable amount of interest as an alternative option for next-generation nonvolatile memories due to their distinct advantages of ultralow power consumption, fast operation, and outstanding scaling potential. Among the recently considered active media, amorphous carbon oxide (α-C:O x ) shows promise in terms of device performance, essentially due to the transition between carbon sp 2 -sp 3 complex under bias. However, widespread utilization of this media still remains a challenge due to its undesirable high forming voltage and insufficient stability issues. Her...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.