EELS Spectrum Imaging of Ca Segregation at Grain Boundaries in Magnesium Aluminate Spinel
EELS Spectrum Imaging of Ca Segregation at Grain Boundaries in Magnesium Aluminate Spinel
复制标题
镁铝尖晶石晶界处 Ca 偏析的 EELS 光谱成像
DOI：
10.1093/micmic/ozad067.191
复制
发表时间：
2023
影响因子：
2.8
通讯作者：
Watanabe, Masashi
中科院分区：
文献类型：
--
作者：
Campos-Quiros, Alexander;Kundu, Animesh;Watanabe, Masashi
晶界（GB）化学和结构已被证明在决定镁铝尖晶石（MAS，MgAl 2 O 4）的微观结构和机械性能方面起着至关重要的作用[1，2]。出于这个原因，它是非常感兴趣的，以证实在GB的掺杂剂原子的存在，以获得更好地理解之间的关系偏析行为和体性能。高角度环形暗场扫描透射电子显微镜（HAADF-STEM）成像是检测GB偏析的有用技术，因为图像强度与Δ Z 2成比例[3]。然而，如果掺杂剂元素和主体材料之间的原子序数差异不是非常显著并且偏析水平相当低，例如在MAS中的Ca偏析的情况下，则通过Z对比成像确认GB偏析不再是简单的。作为替代方案，可以采用通过X射线能量色散光谱法（XEDS）的元素绘图来确认GB偏析，从而提供化学和空间分布信息。然而，微量元素的XEDS映射仍然是一项困难的任务，主要是由于低信号生成和收集[4]。此外，需要更高的探针电流和更长的采集时间来补偿低信号可能会导致光束损坏和空间漂移，这会导致GB化学变化，并阻碍在高放大倍数下进行元素映射，以进行偏析表征。因此，为了研究MAS中Ca原子的GB偏析行为，与XEDS相比，在较低的探针电流下使用具有较高信号收集效率的电子能量损失谱（EELS）。使用高纯度MAS粉末（Baikowski，S25 CR）通过在1250 C和40 MPa下热压1 h来制备Ca掺杂的MAS（500 ppm Ca）样品。随后，样品在空气气氛中在1400 ℃下热处理48小时，并通过用50-nm金刚石精加工进行机械抛光。使用在30 kV下操作的聚焦离子束（FIB）仪器（Thermo/Fisher Scientific Scios）制备电子透明的薄样品。FIB制备的薄样品的表面上的损坏层通过Fischione 1040纳米研磨仪在900 eV下去除。使用HAADF-STEM成像在200 kV下操作的像差校正扫描TEM JEOL JEM-ARM 200 CF仪器中进行详细的原子尺度表征以评估GB中的Ca偏析行为。EELS分析通过配备有混合像素电子检测器Dectris ELA的CEOS能量过滤和成像装置（CEFID）进行。使用100× 100像素采集几个不同GB附近的光谱成像（SI）数据集，停留时间为30 ms/像素（总采集时间约为5 min），能量色散为0.75 eV/通道，范围为770 eV。对于EELS分析，使用34 mrad的会聚半角和60 mrad的收集半角，相机长度为3 cm。此外，使用CEOS PantaTM软件进行SI数据集的采集和分析。
Grain boundary (GB) chemistry and structure have been demonstrated to play a crucial role in determining microstructural and mechanical properties of magnesium aluminate spinel (MAS, MgAl2O4)[1, 2]. For this reason, it is of great interest to corroborate the presence of dopant atoms at GBs to gain a better understanding of the relationship between segregation behavior and bulk properties. High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) imaging is a useful technique to detect GB segregation since the image intensity is proportional to∼ Z2 [3]. However, if the atomic number difference between the dopant element and the host material is not very significant and the segregation levels are rather low, such as in the case of Ca-segregation in MAS, confirmation of GB segregation by Z-contrast imaging is no longer straightforward. As an alternative, elemental mapping by X-ray energy dispersive spectrometry (XEDS) can be employed to confirm the GB segregation providing chemical and spatial distribution information. However, XEDS mapping of trace elements remains a difficult task, mainly due to the low signal generation and collection [4]. In addition, the requirement for higher probe currents and longer acquisition times to compensate for the low signals could cause beam damage and spatial drifts, which lead to changes in the GB chemistry and impede elemental mapping at high magnifications needed for segregation characterization. Therefore, to study the GB segregation behavior of Ca atoms in MAS, electron energy-loss spectrometry (EELS) with a higher signal collection efficiency was used at lower probe currents compared to XEDS.High-purity MAS powders (Baikowski, S25CR) were used to fabricate a Ca-doped MAS (500 ppm of Ca) sample using hotpressing at 1250 C and 40 MPa for 1 h. Subsequently, the sample was heat-treated at 1400 C for 48 h in an air atmosphere and mechanically polished by finishing with 50-nm diamond. Electron transparent thin specimens were prepared using a focused-ion beam (FIB) instrument (Thermo/Fisher Scientific Scios) operated at 30 kV. Damaged layers on the surface of FIB-prepared thin specimens were removed by a Fischione 1040 nanomill instrument at 900 eV. Detailed atomic-scale characterization to evaluate the Ca segregation behavior in GB was performed using HAADF-STEM imaging in an aberration-corrected scanning TEM JEOL JEM-ARM200CF instrument operated at 200 kV. EELS analysis was performed by a CEOS energy filtering and imaging device (CEFID) equipped with a hybrid-pixel electron detector Dectris ELA. Spectrum-imaging (SI) datasets in the vicinity of several different GBs were acquired with 100× 100 pixels for a dwell time of 30 ms per pixel (total acquisition time∼ 5 min) with an energy dispersion of 0.75 eV/channel over a range of 770 eV. A convergence semi-angle of 34 mrad and a collection semi-angle of 60 mrad were used with a camera length of 3 cm for the EELS analysis. Furthermore, the acquisition and analysis of the SI datasets were performed using CEOS Panta Rhei software.