We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Condensed Matter > Materials Science

Title: Mapping Structural Heterogeneity at the Nanoscale with Scanning Nano-structure Electron Microscopy (SNEM)

Abstract: Scanning electron diffraction in a transmission electron microscope (TEM), is often used to resolve spatial information on structures at the nano-scale. By coupling a data reduction and QoI extraction data analysis pipelines to a relational precession-assisted electron diffraction probing, we discuss the `scanning nano-structure electron microscopy' (SNEM) approach. SNEM reveals structural and related compositional information in materials at high spatial and, potentially, temporal resolution.
SNEM provides new horizons for understanding disordered materials. By choose scalar quantities of interest (QoIs) and mapping them to their positional indexes, SNEM allows following spatially variations in order. We mainly use electron pair distribution functions (ePDF), which are reduced from diffraction patterns, from which we extract structurally relevant QoIs. Besides some straightforward QoIs, we take advantage of the large relational data set and use non-negative matrix factorization (NMF) to generate `bond type distribution maps' and the related chemical short-range order distribution maps.
We use a composite of a hot-rolled Ni-encapsulated Zr$_{65}$Cu$_{17.5}$Ni$_{10}$Al$_{7.5}$ bulk metallic glass (BMG). We expose structural heterogeneity within the amorphous BMG layer that is both structural and compositional. We find regions of enhanced crystallinity, compositional gradients, and tendencies towards chemical clustering or ordering. These are all resolved at the nm length scale. More specifically, we discover that the crystalline Ni undergoes a non-trivial amorphization as it mixes with the BMG, which might be the origin for the successful binding between the BMG and the hosting Ni. We find that Zr-rich regions result in a tendency towards chemical clustering, while Cu-depleted regions are suspected as effective nucleation centers for crystallization.
Subjects: Materials Science (cond-mat.mtrl-sci); Disordered Systems and Neural Networks (cond-mat.dis-nn)
Cite as: arXiv:2110.03589 [cond-mat.mtrl-sci]
  (or arXiv:2110.03589v1 [cond-mat.mtrl-sci] for this version)

Submission history

From: Yevgeny Rakita [view email]
[v1] Thu, 7 Oct 2021 16:14:52 GMT (40103kb,D)
[v2] Mon, 18 Oct 2021 02:02:32 GMT (40100kb,D)
[v3] Sun, 7 Nov 2021 03:40:59 GMT (36015kb,D)

Link back to: arXiv, form interface, contact.