Fig. 1. Schematic illustration of the experimental HPSR setup: (a) side view and the coordinate system for sectioned specimens, where the red frame indicates location examined using SEM; (b) locations of steel rolls on the tool.

Fig. 2. SEM images showing regions near the surface: (a) secondary electron image taken after etching. The framed area in this image is magnified in (b) to indicate sublayers and vortex-like structures in more detail; (c, d) ECC images (taken after electropolishing) showing branched cracks in (c) and nanoscale lamellae in (d). Note that the upper region in (a, c) is a thin Cu layer electrodeposited on the top surface to protect it during metallographic preparations.

Fig. 3. Orientation maps and {111} pole figures showing the microstructure and texture at different depths: (a, e) ≤130 µm; (b, f) 370-500 µm; (c, g) 600-1100 µm; (d, h) 1500-3200 µm. Clusters of black pixels near the sample surface in (a) correspond to non-indexed EBSD patterns. The inset in (d) shows the color code for crystallographic orientations, where TD is the radial (transverse) direction of the sample. White and black lines show LABs and HABs, respectively.

Fig. 4. Parameters of the deformed microstructure in HPSR-processed copper as a function of depth: (a) average boundary spacing dND measured using ECC images for the topmost layer and using EBSD data for the other depths; (b) fraction of HABs determined using EBSD; (c) stored energy calculated from the EBSD data. Note that each data point represents the value for a layer centered at the position indicated along the X-axis.

Fig. 5. Orientation maps demonstrating the annealed microstructure (with recrystallized grains shown separately next to each orientation map) within the first 300 µm below the surface: (a) 50 min at 150 °C; (b) 130 min at 150 °C; (c) 200 min at 150 °C. Different colors in the maps correspond to different crystallographic directions as shown in the inset in Fig. 3(d). White, black and purple lines show LABs, HABs and twin boundaries, respectively.

Fig. 6. Microstructure in the subsurface of the sample annealed for 100 min at 150 °C. The microstructure is seen as a result of local differences in the quality of EBSD patterns, where dark pixels correspond to poor pattern quality, while lighter pixels correspond to better pattern quality.

Fig. 7. Quantitative parameters of recrystallized microstructure analyzed within the first 300 µm below the surface after annealing at 150 °C: (a) area fraction; (b) average recrystallized grain size measured ignoring twin boundaries.

Fig. 8. Orientation maps from different depths in the sample annealed for 960 min at 150 °C: (a) first 300 µm below the surface; (b) 600-900 µm below the surface. Different colors in the maps correspond to different crystallographic directions as shown in the inset in Fig. 3(d). White, black and purple lines show LABs, HABs and twin boundaries, respectively.

Fig. 9. Fraction recrystallized as a function of depth in the sample annealed for 960 min at 150 °C.

Fig. 10. Mechanical properties measured in the as-deformed sample and the sample annealed for 960 min at 150 °C: (a) Vickers hardness at different depths; (b) stress-strain curves obtained by tensile testing of 1-mm-thick specimens (80 µm to 1080 µm from the immediate surface of the disc).

Fig. 11. Comparison of hardness profiles in copper samples processed using different surface deformation techniques. For consistency with the literature data, all HV values are presented here in GPa.

Fig. 12. JMAK plot describing the recrystallization kinetics within the first 300 µm below the surface of the HPSR copper sample annealed at 150 °C. The Avrami exponent (n = 1.35) was determined for annealing durations in the range 50-320 min.