Enhanced electrical performances with HZO/β-Ga2O3 3D FinFET toward highly perceptual synaptic device application

IF 4.6 3区 工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC
Seohyeon Park , Jaewook Yoo , Seokjin Oh , Hongseung Lee , Minah Park , Seongbin Lim , Soyeon Kim , Sojin Jung , Bongjoong Kim , Keun Heo , Taehwan Moon , TaeWan Kim , Mengwei Si , Peide D. Ye , Hagyoul Bae
{"title":"Enhanced electrical performances with HZO/β-Ga2O3 3D FinFET toward highly perceptual synaptic device application","authors":"Seohyeon Park ,&nbsp;Jaewook Yoo ,&nbsp;Seokjin Oh ,&nbsp;Hongseung Lee ,&nbsp;Minah Park ,&nbsp;Seongbin Lim ,&nbsp;Soyeon Kim ,&nbsp;Sojin Jung ,&nbsp;Bongjoong Kim ,&nbsp;Keun Heo ,&nbsp;Taehwan Moon ,&nbsp;TaeWan Kim ,&nbsp;Mengwei Si ,&nbsp;Peide D. Ye ,&nbsp;Hagyoul Bae","doi":"10.1016/j.mssp.2025.110104","DOIUrl":null,"url":null,"abstract":"<div><div>In this study, we have demonstrated a novel high-performance ferroelectric FinFET (Fe-FinFET) that integrates ultra-wide-bandgap (UWBG) beta-gallium oxide (<em>β</em>-Ga<sub>2</sub>O<sub>3</sub>) channel with an atomic layer deposited (ALD) hafnium zirconium oxide (HZO) ferroelectric layer for the first time, and experimentally validated the effectiveness of the robust <em>β</em>-Ga<sub>2</sub>O<sub>3</sub> platform as a memory application. Compared with conventional planar ferroelectric FET (FeFET), the Fe-FinFET exhibits a markedly wider counter-clockwise hysteresis loop in its transfer characteristics, achieving a large memory window (MW) of 13.9 V with a single HZO layer. When normalized to the actual channel width, the Fe-FinFETs show an improved <em>I</em><sub>ON</sub>/<em>I</em><sub>OFF</sub> ratio of 2.3 × 10<sup>7</sup> and a subthreshold swing value of 110 mV/dec; Y-function method further indicates that the intrinsic mobility is improved to 4.25 × 10<sup>2</sup> cm<sup>2</sup>/Vs. The enhanced polarization due to larger electric fields across the ferroelectric layer in Fe-FinFET is validated using Sentaurus TCAD, and this result is further supported by the energy-dependent distribution of interface trap density (<em>D</em><sub>it</sub>) extracted in both forward and reverse sweep directions. After 5 × 10<sup>6</sup> program/erase (PGM/ERS) cycles, the MW was maintained at 9.2 V, and the retention time was measured up to 3 × 10<sup>4</sup> s with low degradation. To verify its potential as an artificial synapse, we trained a convolutional neural network (CNN) and achieved an accuracy of 91.7 %. These results establish the HZO/<em>β</em>-Ga<sub>2</sub>O<sub>3</sub> Fe-FinFET as a promising candidate for high voltage-enduring, non-volatile memory devices that also offer synaptic functionality for neuromorphic applications.</div></div>","PeriodicalId":18240,"journal":{"name":"Materials Science in Semiconductor Processing","volume":"201 ","pages":"Article 110104"},"PeriodicalIF":4.6000,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Materials Science in Semiconductor Processing","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S136980012500842X","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
引用次数: 0

Abstract

In this study, we have demonstrated a novel high-performance ferroelectric FinFET (Fe-FinFET) that integrates ultra-wide-bandgap (UWBG) beta-gallium oxide (β-Ga2O3) channel with an atomic layer deposited (ALD) hafnium zirconium oxide (HZO) ferroelectric layer for the first time, and experimentally validated the effectiveness of the robust β-Ga2O3 platform as a memory application. Compared with conventional planar ferroelectric FET (FeFET), the Fe-FinFET exhibits a markedly wider counter-clockwise hysteresis loop in its transfer characteristics, achieving a large memory window (MW) of 13.9 V with a single HZO layer. When normalized to the actual channel width, the Fe-FinFETs show an improved ION/IOFF ratio of 2.3 × 107 and a subthreshold swing value of 110 mV/dec; Y-function method further indicates that the intrinsic mobility is improved to 4.25 × 102 cm2/Vs. The enhanced polarization due to larger electric fields across the ferroelectric layer in Fe-FinFET is validated using Sentaurus TCAD, and this result is further supported by the energy-dependent distribution of interface trap density (Dit) extracted in both forward and reverse sweep directions. After 5 × 106 program/erase (PGM/ERS) cycles, the MW was maintained at 9.2 V, and the retention time was measured up to 3 × 104 s with low degradation. To verify its potential as an artificial synapse, we trained a convolutional neural network (CNN) and achieved an accuracy of 91.7 %. These results establish the HZO/β-Ga2O3 Fe-FinFET as a promising candidate for high voltage-enduring, non-volatile memory devices that also offer synaptic functionality for neuromorphic applications.
HZO/β-Ga2O3 3D FinFET在高感知突触器件中的应用提高了电性能
在这项研究中,我们首次展示了一种新型的高性能铁电FinFET (Fe-FinFET),它将超宽带隙(UWBG) β-氧化镓(β-Ga2O3)通道与原子层沉积(ALD)氧化铪锆(HZO)铁电层集成在一起,并通过实验验证了稳健的β-Ga2O3平台作为存储应用的有效性。与传统平面铁电场效应管(FeFET)相比,Fe-FinFET在传输特性上具有明显更宽的逆时针磁滞回线,单HZO层实现了13.9 V的大存储窗口(MW)。当归一化到实际通道宽度时,fe - finfet的ION/IOFF比提高到2.3 × 107,亚阈值摆幅值为110 mV/dec;y函数法进一步表明,本征迁移率提高到4.25 × 102 cm2/Vs。利用Sentaurus TCAD验证了Fe-FinFET中铁电层上较大电场导致的极化增强,并且在正反扫描方向上提取的界面陷阱密度(Dit)的能量依赖分布进一步支持了这一结果。经过5 × 106次编程/擦除(PGM/ERS)循环后,MW维持在9.2 V,保留时间达3 × 104 s,降解率低。为了验证其作为人工突触的潜力,我们训练了一个卷积神经网络(CNN),并达到了91.7%的准确率。这些结果表明,HZO/β-Ga2O3 Fe-FinFET是耐高压、非易失性存储器件的有希望的候选者,也为神经形态应用提供突触功能。
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
Materials Science in Semiconductor Processing
Materials Science in Semiconductor Processing 工程技术-材料科学:综合
CiteScore
8.00
自引率
4.90%
发文量
780
审稿时长
42 days
期刊介绍: Materials Science in Semiconductor Processing provides a unique forum for the discussion of novel processing, applications and theoretical studies of functional materials and devices for (opto)electronics, sensors, detectors, biotechnology and green energy. Each issue will aim to provide a snapshot of current insights, new achievements, breakthroughs and future trends in such diverse fields as microelectronics, energy conversion and storage, communications, biotechnology, (photo)catalysis, nano- and thin-film technology, hybrid and composite materials, chemical processing, vapor-phase deposition, device fabrication, and modelling, which are the backbone of advanced semiconductor processing and applications. Coverage will include: advanced lithography for submicron devices; etching and related topics; ion implantation; damage evolution and related issues; plasma and thermal CVD; rapid thermal processing; advanced metallization and interconnect schemes; thin dielectric layers, oxidation; sol-gel processing; chemical bath and (electro)chemical deposition; compound semiconductor processing; new non-oxide materials and their applications; (macro)molecular and hybrid materials; molecular dynamics, ab-initio methods, Monte Carlo, etc.; new materials and processes for discrete and integrated circuits; magnetic materials and spintronics; heterostructures and quantum devices; engineering of the electrical and optical properties of semiconductors; crystal growth mechanisms; reliability, defect density, intrinsic impurities and defects.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信