A review of advantages of high-efficiency X-ray spectrum imaging for analysis of nanostructured ferritic alloys

Abstract

a b s t r a c tNanostructured ferritic alloys (NFAs) exhibit complex microstructures consisting of 100-500 nm ferrite grains, grain boundary solute enrichment, and multiple populations of precipitates and nanoclusters (NCs). Understanding these materials' excellent creep and radiation-tolerance properties requires a combination of multiple atomic-scale experimental techniques. Recent advances in scanning transmission electron microscopy (STEM) hardware and data analysis methods have the potential to revolutionize nanometer-to micrometer-scale materials analysis. Modern high-brightness, high-X-ray collection STEM instruments are capable of enabling advanced experiments, such as simultaneous energy dispersive X-ray spectroscopy and electron energy loss spectroscopy spectrum imaging at nm to sub-nm resolution, that are now well-established for the study of nuclear materials. In this paper, we review past results and present new results illustrating the effectiveness of latest-generation STEM instrumentation and data analysis.

Microstructure of HIPed and SPSed 9Cr-ODS steel and its effect on helium bubble formation

“…Interestingly, higher concentrations of Cr, Mn, Ti and O are observed at the surface of oxide than internal by in APT atom distribution maps. Such surface enrichment was not observed in the EDS maps due to the inherent limits to EDS spatial resolution [27]. Fig.…”

Microstructure and hydrothermal corrosion behavior of NITE-SiC with various sintering additives in LWR coolant environments

“…A modern STEM like the F200X produces copious X-ray data, which then must be interrogated. Traditional analyst-dependent methods such as generating count maps from X-rays of interest becomes cumbersome and unlikely to yield subtle insights as the datasets become very large [42,43], so we employed multivariate statistical analysis methods to mine the X-ray spectrum image (SI) datacubes for additional insight and correlations. Specifically, we used the Sandia National Laboratories AXSIA code [44].…”

When will Low-Contrast Features be Visible in a STEM X-Ray Spectrum Image?

“…This is within the visibility range of M i >10 3 inferred from the results of Figure 9b. Other maps (Parish & Miller, 2014), which required significant MVSA processing and binning of the pixels by factors of 8×8 to find >rank-1 MVSA solutions, showed M i ~0.9–1×10 3 nC-sr. In previous CM200 data (Parish et al, 2011), in which NCs were visible by high-angle annular dark-field (HAADF)-STEM imaging, but barely visible in X-ray maps, M i ~2×10 3 nC-sr. With an FEI Tecnai Osiris (Ω=0.9 sr, E 0 =200 keV, uncorrected probe), previous work (Miller & Parish, 2011) showed obvious visibility of NCs in the Ti and Y lines at M i ~10 2 –10 3 nC-sr.…”

“…As described elsewhere (Parish and Miller, in press), STEM-EDS spatial resolution is calculated assuming a Gaussian beam profile at the incident surface of the sample and a Gaussian beam-broadening effect, added in quadrature. The beam-broadening term is as per Reed (1982), assuming single-scattering and no crystallographic effects: Z is the atomic number of the target, A the atomic mass, ρ the density (g/cm 3 ), t the thickness of the specimen (cm), E 0 the primary beam energy (eV), and b has units cm.…”

“…In 14YWT, the NCs are Y- and Ti-rich and are either Ti-O-Y-vacancy complexes or ceramic-like particles (e.g., Y 2 Ti 2 O 7 ) (Bhattacharyya et al, 2012; Brandes et al, 2012), although Al-bearing and large TiN-related precipitates are also found (at a lower-number density) (Parish et al, 2011, 2014). NFAs also damage under intense STEM-beam irradiation (Parish & Miller, 2014). The combination of small particles (NCs) with multiple compositions, combined with a limit to the allowable electron dose, coupled with experimental data from multiple microscopes, makes this an ideal test case for a STEM X-ray mapping model.…”