Suppr超能文献

在没有接触电极的情况下,在钛酸钡中形成和操控具有2纳米畴周期的畴壁。

Formation and manipulation of domain walls with 2 nm domain periodicity in BaTiO without contact electrodes.

作者信息

Barzilay Maya, Ivry Yachin

机构信息

Department of Materials Science and Engineering, Technion - Israel Institute of Technology, Haifa 3200003, Israel.

出版信息

Nanoscale. 2020 May 28;12(20):11136-11142. doi: 10.1039/d0nr01747g.

Abstract

Interfaces at the two-dimensional limit in oxide materials exhibit a wide span of functionality that differs significantly from the bulk behavior. Among such interfaces, domain walls in ferroelectrics draw special attention because they can be moved deterministically with external voltage, while they remain at place after voltage removal, paving the way to novel neuromorphic and low-power data-processing technologies. Ferroic domains arise to release strain, which depends on material thickness, following Kittel's scaling law. Hence, a major hurdle is to reduce the device footprint for a given thickness, i.e., to form and move high-density domain walls. Here, we used transmission electron microscopy to produce domain walls with periodicity as high as 2 nm without the use of contact electrodes, while observing their formation and dynamics in situ in BaTiO3. Large-area coverage of the engineered domain walls was demonstrated. The domain-wall density was found to increase with increasing effective stress, until arriving at a saturation value that reflects 150-fold effective stress enhancement. Exceeding this value resulted in strain release by domain rotation. In addition to revealing this multiscale strain-releasing mechanism, we offer a device design that allows controllable switching of domain-walls with 2 nm periodicity, reflecting a potential 144 Tb per inch2 neuromorphic network.

摘要

氧化物材料二维极限处的界面展现出广泛的功能,与体相行为有显著差异。在这类界面中,铁电体中的畴壁备受关注,因为它们可以通过外部电压确定性地移动,而在去除电压后仍保持原位,为新型神经形态和低功耗数据处理技术铺平了道路。铁电畴的出现是为了释放应变,根据基特尔标度律,应变取决于材料厚度。因此,一个主要障碍是在给定厚度下减小器件尺寸,即形成并移动高密度畴壁。在此,我们利用透射电子显微镜在不使用接触电极的情况下制备出周期高达2 nm的畴壁,同时在钛酸钡中对其形成和动力学进行原位观察。展示了工程化畴壁的大面积覆盖。发现畴壁密度随有效应力增加而增大,直至达到反映有效应力增强150倍的饱和值。超过该值会导致通过畴旋转释放应变。除了揭示这种多尺度应变释放机制外,我们还提供了一种器件设计,可实现对周期为2 nm的畴壁进行可控切换,这意味着每平方英寸有潜在的144太比特神经形态网络。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验