Unlocking nanoscale microstructural detail in aluminium alloys through differential phase contrast segmentation in STEM

Abstract

Differential phase contrast (DPC) imaging in scanning transmission electron microscopy (STEM) maps projected electric fields through the phase sensitivity of segmented low-angle detectors. Although typically applied to atomic-resolution imaging at low beam currents, STEM-DPC is here demonstrated as a rapid micro- and nanoscale image-segmentation tool for materials characterization in advanced aluminium alloys. Decomposition of false-colour DPC micrographs in hue-saturation-value space enables simultaneous identification and quantification of nanoclusters, GP zones, intermediate precipitate phases, dislocation cores, and associated strain fields within a single field of view. The method is demonstrated across multiple alloy systems, including clustering and strain-field mapping in a deformed AlMgZn(Cu) crossover alloy, precipitate identification in a paint-baked automotive sheet alloy, phase-variant segmentation in overaged AA7075-T7, and nanopore and nanoparticle detection in an anodic coating on AA2024-T3. Coupling DPC with neural-network segmentation further enables automated grain-boundary delineation and quantification in nanocrystalline aluminium thin films. Combined with STEM-EDX, DPC-based segmentation enables correlative microstructural analysis, establishing DPC as a rapid complement to techniques such as SPED and 4D-STEM.

Figures (6)

• Figure 1: Imaging microstructures with DPC | (A) A convergent electron beam passes through an electron-transparent aluminium alloy sample, where different microstructural sites can rotate the diffraction pattern which is then projected onto a segmented LAADF detector. Before acquiring DPC images, the LAADF-STEM detector (DF4) segments must be both fully covered by the projected beam as well as calibrated to equal intensities, as shown in the actual micrographs in (B). (C) The deflected beam will cause intensity variations across the four LAADF-STEM detector segments. (D) DPC processes the variations between opposing segments and calculate the electron beam direction (0--360°) and magnitude (visualised as a gradient from black to one RGB colour in an HSV colour wheel), generating a false-colour micrograph of the alloy microstructure: this "DPC micrograph" can be used for nanoscale image segmentation.
• Figure 2: Revealing clusters, GPI Zones, dislocations and strain fields in a AlMgZn(Cu) crossover alloy | The DPC micrograph in (A) shows the microstructure of a pre-aged AlMgZn crossalloy alloy after 2% plastic deformation. From (A), the segments (B) and (C) were extracted and show in detail, respectively, the alloy matrix with GPI Zones and dislocations with strain fields. Similarly, the DPC micrograph in (D) shows the microstructure of a long-term-aged AlMgZn crossover alloy after 5% plastic deformation. From (D), the segments (E) and (F) were extracted and show in detail, respectively, the alloy matrix mainly with small clusters and dislocations. As expanded in section \ref{['sec:resndis:clustering']}, direct quantitative and qualitative comparisons between both alloys' microstructures are permitted with the DPC technique. The insets in the segmented images (B), (C), (E), and (F) show enlarged areas of their respective micrographs.
• Figure 3: Mosaic of intermediate T$^{\prime}$-phase precipitates and dislocations in a paint-baked alloy | The AlMgZn(Cu) crossover alloy was subjected to paint-bake heat treatment and 2% deformation. The LAADF micrographs from (A) to (D) show the segmented LAADF signal on the DF4 detector segments. With the signals from (A) to (D), a DPC micrograph is reconstructed and shown in (E) revealing dislocation loops and precipitates. With nanoscale image segmentation, the DPC micrograph in (E) can be decomposed into matrix (F), dislocation and precipitates (G). Notes: The scale bar in (A) applies to (B-D) and the scale bar in (E) also applies to (F) and (G). The inset in (F) shows the SAED pattern indexed to be the Al [110] zone axis.
• Figure 4: Correlative microscopy with STEM-EDX and -DPC of overaged microstructures | AA7075-T7 is a high-strength aerospace alloy in the overaged state. (A) Presents a composite LAADF micrograph with all the segments of the DF4 detector, STEM-EDX elemental maps for Al, Zr, Mg, Cr, Cu, and the SAED pattern of the grain taken along the [001] zone axis. (B) Shows the complementary DPC analysis from the same region in (A) where 4 segmented images of distinct precipitates could be extracted from the DPC micrograph along with the alloy's matrix.
• Figure 5: Advancing aluminium alloys' corrosion protection research with DPC | BF-STEM micrograph in (A) shows the sample's cross-section of an AA2024-T3 alloy substrate with a grown anodic layer. Micrographs (B) and (C) show both the iDPC and DPC micrographs from the anodic layer, where anodic walls, pores and Ce nanoparticles are distinguishable. Complementary STEM-EDX elemental mapping in (D-F) reveals aluminium oxide nature of the anodic layer as well as the presence of Ce nanoparticles. DPC micrographs (G) and (H) taken at the interface between the alloy and the anodic layer revealing in nanoscale detail the "nanotubular" morphological aspect of the latter, matching with the overlaid artistic conception by Prada-Ramirez et al.ramirez2020tartaric.