题名 | 纳米复合调控纳米导电细丝提升忆阻器性能研究 |
其他题名 | IMPROVING THE PERFORMANCE OF MEMRISTOR BY NANOCOMPOSITE-CONTROLLED CONDUCTIVE NANOFILAMENTS
|
姓名 | |
姓名拼音 | SONG Jiahao
|
学号 | 12132067
|
学位类型 | 硕士
|
学位专业 | 0702 物理学
|
学科门类/专业学位类别 | 07 理学
|
导师 | |
导师单位 | 材料科学与工程系
|
论文答辩日期 | 2024-05-16
|
论文提交日期 | 2024-07-05
|
学位授予单位 | 南方科技大学
|
学位授予地点 | 深圳
|
摘要 | 当今大数据时代,对计算和存储技术的要求日益提高,特别是在高密度存储和高计算能力方面。然而,传统的冯·诺依曼架构的存储与计算分离存在固有弊端,并且随着摩尔定律接近其理论极限,这些局限性进一步阻碍了计算机技术的发展。忆阻器作为一种创新的非易失性存储技术,通过模拟人类大脑神经突触的功能,为构建未来人工智能系统提供了关键的物理基础。它在实现高密度存储和低功耗操作,以及推动神经形态计算领域的发展上,展现了巨大的潜力,因此受到了科研与工业界的广泛关注。 虽然忆阻器技术在理论研究和实验探索方面已经取得了显著成就,但在推向实际系统应用时,它仍面临若干关键技术挑战,特别是在器件循环稳定性和多个器件间的一致性方面。忆阻器低阻态与高阻态受其内部纳米导电细丝动态变化的直接影响,纳米导电细丝的不稳定生成与断裂造成器件循环性能离散度大,纳米导电细丝的不均匀导致器件之间离散度大。 本研究聚焦于纳米导电细丝的可控性研究,通过纳米复合,调控优化忆阻器纳米导电细丝导通/断裂的稳定性与分布均匀性,增强忆阻器的循环稳定性和器件间的性能均一性,以推动其在高密度存储与神经形态计算中的应用。具体研究内容如下: (1)通过自组装生长方式,构建周期性分布的纳米柱结构,即钛酸钡基质中嵌入镍元素纳米柱的结构,利用镍元素纳米柱的金属镍与氧化镍相变,提升纳米导电细丝的导通/断裂的可控性以及导电细丝分布的均匀性,提升器件循环稳定特性及均一性。最终实现对忆阻器器件稳定性(低阻态与高阻态离散度分别为:13%与9%)和均一性的提升(低阻态与高阻态离散度分别为:44%与99%)。 (2)构建钛酸钡与氧化镍超晶格,实现对纳米导电细丝的多点控制,通过多点随机性生长/断裂过程增加纳米导电细丝的整体导通/断裂的可控性。纳米导电细丝生成过程中,需要跨越超晶格界面势垒,而界面处的势垒会阻碍纳米导电细丝的生成,这使部分不稳定的迁移路径被阻断,只允许部分稳定的纳米导电细丝生成并主导开关过程,实现纳米导电细丝的导通/断裂的稳定化,进一步提升器件的循环稳定性,低阻态与高阻态离散度分别为:22%与9%)。多层堆叠的方式,利用统计学中的中心极限定理,实现多个器件之间的纳米导电细丝均匀化,进而提升多个器件的均一性(低阻态与高阻态离散度分别为:39%与23%)。 本研究采用了纳米复合技术,构建纳米柱复合结构和超晶格复合结构来实现忆阻器内的纳米导电细丝的可控化,提升纳米导电细丝的稳定性以及均一性,从而有效提高了单个器件的循环稳定性以及多个器件的均一性,为忆阻器性能调控提供新的思路。 |
其他摘要 | In the information technology era, the demands for computing and storage technologies are rapidly increasingly, especially in high-density memory and computing. However, the traditional von Neumann architecture, due to the separation of storage and computation unit, exhibits inherent limitations in the advancement of computing technology, particularly as Moore’s law approaches its theoretical limit. To counter this backdrop, memristors emerge as an innovative non-volatile memory technology that emulates the functions of human brain synapses, offering a critical physical foundation for developing future artificial intelligence systems. Their potential for high-density storage and low-energy operations, alongside their role in advancing neuromorphic computing, has garnered widespread attention in the science and industry. Despite significant progress in the theoretical and experimental exploration of memristor technology, several key technical challenges persist in transitioning to practical system applications, especially concerning device cyclic stability and device-to-device variation across multiple devices. The performance of memristors is heavily influenced by the dynamics of conductive filaments, whose random growth and rupture contribute to variations in device performance and hinder the application of large-scale memristor crossbar arrays. This thesis concentrates on the controllability of conductive nanofilaments, aiming to enhance the cycle stability and uniformity of memristor devices by precisely controlling the formation and rupture processes through nano-composite techniques. The main research content includes: (1) Through self-assembly growth to construct aperiodic array of nanocolumn structure, the structure of nickel element nanocolumns embedded in barium titanate matrix, Utilizing the phase transition between metallic nickel and nickel oxide within the nickel element nanocolumns to enhance the controllability of filament formation/breakdown ,as well as the uniformity of filament distribution, ultimately enhancing the stability and uniformity of memristor devices (Low resistance state and High resistance state cyclic stability dispersion reduced to 13% and 9%, and device to device variation reduced to 44% and 99%). (2) Constructing a barium titanate and nickel oxide superlattice enables multi-point control over the formation/breakdown of conductive filaments. This control is realized through the random growth/rupture processes at multiple points, enhancing the controllability of filament formation/breakdown. As filaments attempt to form across superlattice interfaces, they encounter barriers that impede the migration of oxygen vacancies, blocking unstable migration paths and allowing only stable conductive pathways to emerge. This mechanism further enhances device cyclic stability (Low resistance state and High resistance state cyclic stability dispersion reduced to 22% and 9%). The multi-layer stacking approach, aligned with the Central Limit Theorem from statistics, ensures uniformity among device conductive filaments, thereby improving device-to-device uniformity (Low resistance state and High resistance state device uniformity dispersion reduced to 39% and 23%). This study employed nano-composite technology to construct composite structures of nanorods and superlattice, we achieve controlled conductive nanofilament formation/rupture and uniform distribution within memristors, significantly enhancing filament stability and uniformity. Through this strategy, we not only improve the cyclic stability of individual devices but also the uniformity across multiple devices, offering new approach to improve memristor performance. |
关键词 | |
其他关键词 | |
语种 | 中文
|
培养类别 | 独立培养
|
入学年份 | 2021
|
学位授予年份 | 2024-06
|
参考文献列表 | [1] Zidan M A, Strachan J P, Lu W D. The future of electronics based on memristive systems[J]. Nature Electronics, 2018, 1(1): 22-29. |
所在学位评定分委会 | 材料科学与工程
|
国内图书分类号 | 0484
|
来源库 | 人工提交
|
成果类型 | 学位论文 |
条目标识符 | http://sustech.caswiz.com/handle/2SGJ60CL/779027 |
专题 | 工学院_材料科学与工程系 |
推荐引用方式 GB/T 7714 |
宋家豪. 纳米复合调控纳米导电细丝提升忆阻器性能研究[D]. 深圳. 南方科技大学,2024.
|
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | 操作 | |
12132067-宋家豪-材料科学与工程(8087KB) | -- | -- | 限制开放 | -- | 请求全文 |
个性服务 |
原文链接 |
推荐该条目 |
保存到收藏夹 |
查看访问统计 |
导出为Endnote文件 |
导出为Excel格式 |
导出为Csv格式 |
Altmetrics Score |
谷歌学术 |
谷歌学术中相似的文章 |
[宋家豪]的文章 |
百度学术 |
百度学术中相似的文章 |
[宋家豪]的文章 |
必应学术 |
必应学术中相似的文章 |
[宋家豪]的文章 |
相关权益政策 |
暂无数据 |
收藏/分享 |
|
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论