Mimicking biological synapses with a-HfSiOx-based memristor: implications for artificial intelligence and memory applications

IF 13.4 2区材料科学 Q1 MATERIALS SCIENCE, MULTIDISCIPLINARY

Nano Convergence Pub Date : 2023-07-10 DOI:10.1186/s40580-023-00380-8

Muhammad Ismail, Maria Rasheed, Chandreswar Mahata, Myounggon Kang, Sungjun Kim

{"title":"Mimicking biological synapses with a-HfSiOx-based memristor: implications for artificial intelligence and memory applications","authors":"Muhammad Ismail, Maria Rasheed, Chandreswar Mahata, Myounggon Kang, Sungjun Kim","doi":"10.1186/s40580-023-00380-8","DOIUrl":null,"url":null,"abstract":"<div><p>Memristors, owing to their uncomplicated structure and resemblance to biological synapses, are predicted to see increased usage in the domain of artificial intelligence. Additionally, to augment the capacity for multilayer data storage in high-density memory applications, meticulous regulation of quantized conduction with an extremely low transition energy is required. In this work, an a-HfSiO<sub>x</sub>-based memristor was grown through atomic layer deposition (ALD) and investigated for its electrical and biological properties for use in multilevel switching memory and neuromorphic computing systems. The crystal structure and chemical distribution of the HfSiOx/TaN layers were analyzed using X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS), respectively. The Pt/a-HfSiO<sub>x</sub>/TaN memristor was confirmed by transmission electron microscopy (TEM) and showed analog bipolar switching behavior with high endurance stability (1000 cycles), long data retention performance (10<sup>4</sup> s), and uniform voltage distribution. Its multilevel capability was demonstrated by restricting current compliance (CC) and stopping the reset voltage. The memristor exhibited synaptic properties, such as short-term plasticity, excitatory postsynaptic current (EPSC), spiking-rate-dependent plasticity (SRDP), post-tetanic potentiation (PTP), and paired-pulse facilitation (PPF). Furthermore, it demonstrated 94.6% pattern accuracy in neural network simulations. Thus, a-HfSiO<sub>x</sub>-based memristors have great potential for use in multilevel memory and neuromorphic computing systems.</p><h3>Graphical Abstract</h3>\n <figure><div><div><div><picture><source><img></source></picture></div></div></div></figure>\n </div>","PeriodicalId":712,"journal":{"name":"Nano Convergence","volume":"10 1","pages":""},"PeriodicalIF":13.4000,"publicationDate":"2023-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://nanoconvergencejournal.springeropen.com/counter/pdf/10.1186/s40580-023-00380-8","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nano Convergence","FirstCategoryId":"88","ListUrlMain":"https://link.springer.com/article/10.1186/s40580-023-00380-8","RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATERIALS SCIENCE, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 5

Abstract

Memristors, owing to their uncomplicated structure and resemblance to biological synapses, are predicted to see increased usage in the domain of artificial intelligence. Additionally, to augment the capacity for multilayer data storage in high-density memory applications, meticulous regulation of quantized conduction with an extremely low transition energy is required. In this work, an a-HfSiO_x-based memristor was grown through atomic layer deposition (ALD) and investigated for its electrical and biological properties for use in multilevel switching memory and neuromorphic computing systems. The crystal structure and chemical distribution of the HfSiOx/TaN layers were analyzed using X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS), respectively. The Pt/a-HfSiO_x/TaN memristor was confirmed by transmission electron microscopy (TEM) and showed analog bipolar switching behavior with high endurance stability (1000 cycles), long data retention performance (10⁴ s), and uniform voltage distribution. Its multilevel capability was demonstrated by restricting current compliance (CC) and stopping the reset voltage. The memristor exhibited synaptic properties, such as short-term plasticity, excitatory postsynaptic current (EPSC), spiking-rate-dependent plasticity (SRDP), post-tetanic potentiation (PTP), and paired-pulse facilitation (PPF). Furthermore, it demonstrated 94.6% pattern accuracy in neural network simulations. Thus, a-HfSiO_x-based memristors have great potential for use in multilevel memory and neuromorphic computing systems.

Graphical Abstract

查看原文本刊更多论文

用基于a- hfsiox的忆阻器模拟生物突触:对人工智能和记忆应用的影响

忆阻器由于其结构简单且与生物突触相似，预计将在人工智能领域得到越来越多的应用。此外，为了增加高密度存储器应用中多层数据存储的容量，需要以极低的跃迁能量对量子化传导进行细致的调节。在这项工作中，通过原子层沉积(ALD)生长了基于a- hfsiox的忆阻器，并研究了其在多电平开关存储器和神经形态计算系统中的电学和生物学特性。利用x射线衍射(XRD)和x射线光电子能谱(XPS)分析了HfSiOx/TaN层的晶体结构和化学分布。Pt/a-HfSiOx/TaN忆阻器经透射电子显微镜(TEM)验证，具有高耐久稳定性(1000次循环)、长数据保持性能(104 s)和均匀电压分布的模拟双极开关行为。通过限制电流顺应性(CC)和停止复位电压来证明其多电平能力。记忆电阻器表现出短期可塑性、兴奋性突触后电流(EPSC)、spike -rate依赖性可塑性(SRDP)、强直后增强(PTP)和成对脉冲促进(PPF)等突触特性。此外，该方法在神经网络模拟中显示出94.6%的模式准确率。因此，基于a- hfsiox的记忆电阻器在多电平存储和神经形态计算系统中具有很大的应用潜力。图形抽象

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Nano Convergence Engineering-General Engineering

CiteScore

15.90

自引率

2.60%

发文量

审稿时长

13 weeks

期刊介绍： Nano Convergence is an internationally recognized, peer-reviewed, and interdisciplinary journal designed to foster effective communication among scientists spanning diverse research areas closely aligned with nanoscience and nanotechnology. Dedicated to encouraging the convergence of technologies across the nano- to microscopic scale, the journal aims to unveil novel scientific domains and cultivate fresh research prospects. Operating on a single-blind peer-review system, Nano Convergence ensures transparency in the review process, with reviewers cognizant of authors' names and affiliations while maintaining anonymity in the feedback provided to authors.