Determining Medium Range Atomic Ordering in Metallic Glasses Using 4D-STEM

Soohyun Im, Pengyang Zhao, Geunhee Yoo, Zhen Chen, Gabriel Calderon Ortiz, Mehrdad Abbasi Gharacheh, Olivia Licata, Baishakhi Mazumder, David Muller, Eun Soo Park, Yunzhi Wang, Jinwoo Hwang
2020 Microscopy and Microanalysis  
Medium range atomic ordering (MRO) attributes to nanometer scale structural heterogeneity in metallic glasses (MGs), and it has been under extensive interest due to its potential connection to the important properties of the material, including mechanical properties and glass forming ability. However, the precise characterization of MRO has been challenging. Conventional large area diffraction methods, such as using X-ray, neutron, or electron, inherently average the structure within the
more » ... ated volume, through which the MRO information tend to be lost. MRO domains in MGs is typically smaller and much less ordered as compared to nanocrystals, and therefore observing them using high-resolution TEM is also very difficult. The difficulties have made the understanding of the exact relationship between MRO and important properties of MGs very challenging. For example, it has been speculated that MRO may be related to the mechanical properties of MGs (e.g. [2]), especially their ductility, but no clear experimental evidence for this suggested relationship has been shown so far. MRO may also be related to the glass forming ability of MGs [3], although spatially resolved information of MRO must be required to confirm the hypothesis. In this work, we determine the structural parameters of MRO with unprecedented details using 4dimensional scanning transmission electron microscopy (4D-STEM) [4, 5] . Our 4D-STEM is enabled by the quantitative analysis of the data acquired using the new-generation Electron Microscopy Pixel Array Detector (EMPAD) [6] , which provides high dynamic range essential for the quantification. About 250,000 nanodiffraction patterns were acquired per sample using STEM probes with 1 nm in diameter, continuously throughout the sample area with over sampling (Fig. 1a) . The resulting 4D data were then reconstructed into dark field images in the real space for all scattering vector magnitude, k, and the inplane azimuthal angle of ( Fig. 1b and 1c) . These images show bright speckles of the MRO domains, which we use to quantitatively determine the MRO parameters, including their type, size, distribution, and volume fraction. The determined MRO parameters are shown to be directly related to the important properties of the Zr-Cu-Co-Al MGs. The data shows that the smaller and more diverse MRO types (Fig. 1d) lead to higher ductility. Certain MRO types are more structurally frustrated, as revealed by the angular correlation function analysis of the 4D-data (Fig. 2a) , which correlates well with the glass forming ability of the MGs. To understand the mechanism of the observed correlation, we have used mesoscale deformation
doi:10.1017/s1431927620013896 fatcat:xkjju2eztzhlfo7lirttrpu6om