Suppr超能文献

基于全氧化物的高透明光子突触用于神经形态计算。

All-Oxide-Based Highly Transparent Photonic Synapse for Neuromorphic Computing.

出版信息

ACS Appl Mater Interfaces. 2018 Oct 10;10(40):34370-34376. doi: 10.1021/acsami.8b10870. Epub 2018 Sep 25.

Abstract

The neuromorphic system processes enormous information even with very low energy consumption, which practically can be achieved with photonic artificial synapse. Herein, a photonic artificial synapse is demonstrated based on an all-oxide highly transparent device. The device consists of conformally grown InO/ZnO thin films on a fluorine-doped tin oxide/glass substrate. The device showed a loop opening in current-voltage characteristics, which was attributed to charge trapping/detrapping. Ultraviolet illumination-induced versatile features such as short-term/long-term plasticity and paired-pulse facilitation were truly confirmed. Further, photonic potentiation and electrical habituation were implemented. This study paves the way to develop a device in which current can be modulated under the action of optical stimuli, serving as a fundamental step toward the realization of low-cost synaptic behavior.

摘要

神经形态系统可以在极低的能耗下处理大量信息,这实际上可以通过光子人工突触来实现。在此,展示了一种基于全氧化物高透明器件的光子人工突触。该器件由在掺氟氧化锡/玻璃基底上外延生长的 InO/ZnO 薄膜组成。器件的电流-电压特性呈现出开环,这归因于电荷俘获/释放。实验证实了紫外光诱导的多种特性,如短期/长期可塑性和成对脉冲易化。此外,还实现了光子增强和电习惯化。本研究为开发一种可以在光刺激作用下调节电流的器件铺平了道路,为实现低成本突触行为迈出了重要的一步。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验