Tunable Microstructures in Entropy-Stabilized Oxide Thin Films Studied with Unsupervised Machine Learning Assisted Electron Microscopy
Tunable Microstructures in Entropy-Stabilized Oxide Thin Films Studied with Unsupervised Machine Learning Assisted Electron Microscopy
复制标题
DOI：
10.1017/s1431927622009862
复制
发表时间：
2022-07
影响因子：
2.8
通讯作者：
L. Miao;George N. Kotsonis;J. Maria;N. Alem
中科院分区：
文献类型：
--
作者：
L. Miao;George N. Kotsonis;J. Maria;N. Alem
源源不断的新型先进材料对于科学发现和技术机会至关重要。高熵材料已成为材料发现前沿丰富的晶体材料新来源。高熵 (HE) 材料利用增加的构型熵来降低整体自由能，将多个原子种类稳定到单个晶格中。[1] HE 材料已在电化学[2]、热电[3]、催化[4] 和许多其他应用中展现出有前景的功能特性。熵稳定首先在高熵合金（HEA）中实现。这一概念最近才随着 2015 年岩盐结构高熵氧化物 (R-HEO) 的合成而扩展到非金属晶体。[1,5] HEO 在室温下表现出固有的亚稳定性。我们之前对脉冲激光沉积（PLD）方法合成的HEO薄膜的研究表明，晶体在室温下的亚稳定性可用于调节阳离子氧化态随生长温度的变化。[6,7]在这项研究中，我们利用先进的透射电子显微镜进一步研究了成分为（Co 0.2 Cu 0.2 Mg 0.2 Ni 0.2 Zn 0.2 ）O的PLD生长的R-HEO薄膜的微观结构与合成条件之间的相关性。 （TEM）由无监督机器学习辅助。我们特别关注生长速率和薄膜厚度对微观结构的影响。研究的样品由四个在 400°C 下沉积的外延 HEO 薄膜组成，激光脉冲速率和总厚度为 1) 5Hz 和 80 nm，2) 5Hz 和 400 nm，3)
A steady stream of new and advanced materials is crucial for scientific discovery and technology opportunities. High-entropy materials have become a rich new source of crystalline materials at the frontier of materials discovery. High entropy (HE) materials leverage increased configurational entropy to lower the overall free energy, stabilizing multiple atomic species into a single lattice.[1] HE materials have demonstrated promising functional properties for electrochemical[2], thermoelectric[3], catalytic[4], and many more applications. Entropy stabilization was first realized in high entropy alloys (HEAs). This concept was only recently expanded into the non-metallic crystals with the synthesis of the rocksalt structure high entropy oxides (R-HEOs) in 2015.[1,5] HEOs show inherent metastability at room temperature. Our previous studies on the HEO thin film synthesized with pulsed laser deposition (PLD) method demonstrated the metastability of the crystal at room temperature can be exploited for tuning the cation oxidation states with growth temperature.[6,7] In this study, we further investigated the correlation between microstructure and the synthesis conditions in PLD grown R-HEO thin film with the composition (Co 0.2 Cu 0.2 Mg 0.2 Ni 0.2 Zn 0.2 )O with advanced transmission electron microscopy (TEM) assisted by unsupervised machine learning. In particular, we focused on the impact of the growth rate and the film thickness on the microstructures. The samples for the study consisted of four epitaxial HEO thin films deposited at 400°C, with the laser pulse rate and the total thickness of 1) 5Hz & 80 nm, 2) 5Hz and 400 nm, 3)