Probing deeper: Determining the 3D gamma-gamma'-microstructure in a Ni-base superalloy by FIB-SEM tomography

Kurzfassung

Ni-based superalloys consist of gamma- and gamma'-phases which control the high-temperature mechanical properties of these alloys [1]. Their microstructure is typically investigated by 2D imaging methods. Quantitative investigation of the 3D microstructure using 2D image data relies on assumptions and statistical modelling to construct a virtual 3D model [2,3]. Serial sectioning of the material in a Scanning Electron Microscope (SEM) equipped with a Focused Ion Beam (FIB) offers an approach to reconstruct the actual microstructure of a given sample. The morphology is interpolated from a FIB tomography consisting of a large and therefore statistically relevant set of serially acquired images. Preprocessing methods such as selective etching yield contrast enhanced images (Figure 1 a) but are impractical for serial imaging. Sufficient image quality in serial imaging requires the SEM to be operated with retractable detectors between electron source and sample, such as a Circular BackScatter (CBS) detector (Figure 1 b). However, FIB ablation and subsequent SEM imaging require the sample to be tilted facing the ion column, which in turn inhibits retractable detectors to be used for image capturing. In this work, a FIB-SEM setup is implemented that allows for serial sectioning of a Ni-based superalloy with a CBS detector. Here, the sample faces the electron column instead of the ion column. This setup accommodates for a detector to be inserted between electron column and sample, resulting in an angular ablation of the sample. The thus obtained images have high resolution and little noise but are distorted (Figure 1 b). Postprocessing of the image series therefore requires proper alignment and virtual tilt correction. The corrected images (Figure 1 c) are then segmented, and the gamma-gamma'-microstructure can be visualized in 3D representation (Figure 1 d). The 3D microstructural data is then evaluated for morphological features such as phase volume fractions, precipitate sizes, and shape factors. The results can be used to estimate the reliability of microstructural data obtained from 3D stereographic reconstructions from 2D images. References [1] R. C. Reed The Superalloys, Cambridge University Press, 2009 [2] A. Brahme et al. Scripta Materialia, 2006, 55, 75-80. [3] C. Zhang et al. Metallurgical and Materials Transactions A, 2004, 35(7), 1927-1933