High-Performance Edge-Line Contact Memristors with In-Plane Solid–Liquid–Solid Grown Silicon Nanowires for Probabilistic Neuromorphic Computing DOI
Lei Yan, Yifei Zhang,

Zhiyan Hu

и другие.

ACS Nano, Год журнала: 2025, Номер unknown

Опубликована: Март 11, 2025

Memristors have garnered increasing attention in neuromorphic computing hardware due to their resistive switching characteristics. However, achieving uniformity across devices and further miniaturization for large-scale arrays remain critical challenges. In this study, we demonstrate the scalable production of highly uniform, quasi-one-dimensional diffusive memristors based on heavily doped n-type silicon nanowires (SiNWs) with diameters as small ∼50 nm, fabricated via in-plane solid–liquid–solid (IPSLS) growth technology. The edge-line contact structural design improves control nucleation sites size conductive filaments (CFs) Ag/SiO2/n-SiNW memristors. These exhibit excellent self-compliance threshold characteristics, including a low operating voltage (∼0.8 V) standard deviation 0.073 V, leakage current (1 pA), high ratio (>107), ultrafast speed (∼8 ns), extremely energy (47.2 fJ per operation). Additionally, developed neurons tunable sigmoidal probabilistic activation functions, demonstrating different devices. achieved an accuracy 96.2% binary tumor classification tasks, underscoring potential IPSLS-fabricated SiNWs advanced hardware. This work highlights effectiveness SiNW-based addressing challenges integration.

Язык: Английский

High-Performance Edge-Line Contact Memristors with In-Plane Solid–Liquid–Solid Grown Silicon Nanowires for Probabilistic Neuromorphic Computing DOI
Lei Yan, Yifei Zhang,

Zhiyan Hu

и другие.

ACS Nano, Год журнала: 2025, Номер unknown

Опубликована: Март 11, 2025

Memristors have garnered increasing attention in neuromorphic computing hardware due to their resistive switching characteristics. However, achieving uniformity across devices and further miniaturization for large-scale arrays remain critical challenges. In this study, we demonstrate the scalable production of highly uniform, quasi-one-dimensional diffusive memristors based on heavily doped n-type silicon nanowires (SiNWs) with diameters as small ∼50 nm, fabricated via in-plane solid–liquid–solid (IPSLS) growth technology. The edge-line contact structural design improves control nucleation sites size conductive filaments (CFs) Ag/SiO2/n-SiNW memristors. These exhibit excellent self-compliance threshold characteristics, including a low operating voltage (∼0.8 V) standard deviation 0.073 V, leakage current (1 pA), high ratio (>107), ultrafast speed (∼8 ns), extremely energy (47.2 fJ per operation). Additionally, developed neurons tunable sigmoidal probabilistic activation functions, demonstrating different devices. achieved an accuracy 96.2% binary tumor classification tasks, underscoring potential IPSLS-fabricated SiNWs advanced hardware. This work highlights effectiveness SiNW-based addressing challenges integration.

Язык: Английский

Процитировано

0