3.2.2 Evolution of proeutectoid ferrite and eutectoid ferrite
Figure 4 showed the microstructure at 400 µm-200 µm(γ = 0.012–0.21) from the surface. The comparison of the original microstructure showed that the pearlite was basically unchanged (Fig. 3), while the contrast of proeutectoid ferrite was significantly changed (Fig. 4(a), Fig. 4(b), Fig. 4(c)). At 400 µm from the surface (γ = 0.012), some of the sub-grain in the proeutectoid ferrite began to change to the LABs (Fig. 4(a1), 4(a2)). At 300 µm(γ = 0.051), the number of LABs in the proeutectoid ferrite increased significantly, and a small amount of LABs changed to HABs, but there was no obvious change in the eutectoid ferrite (Fig. 4(b1), Fig. 4(b2)). When the distance from the surface was 200 µm(γ = 0.051), the number of LABs changed to HABs in the proeutectoid ferrite gradually increased (Fig. 4(c1), Fig. 4(c2)), forming different shape grains with an average size of 710 nm. The misorientation between some eutectoid ferrite lamellars reached 2 ~ 10° (Fig. 4(c2)).
The microstructure at 100 µm (γ = 0.84) from the surface as shown in Fig. 5. In the proeutectoid ferrite, the number of LABs changed to HABs increased significantly (Fig. 5(c)), and the grains were further refined, forming smaller equiaxial grains with an average size of 550 nm. Although the microstructure of pearlite colony did not change significantly under SEM, the contrast difference was more and more obvious, indicating that the spatial misorientation of eutectoid ferrite changed significantly (Fig. 5(a)). The number of eutectoid ferrite lamellar with low-angles misorientation also increased gradually, and even a few changed to high-angle misorientation (Fig. 5(c)). Meanwhile, a low-angle boundary was formed inside the eutectoid ferrite lamellar, and then began to be divided into bamboo-like (Fig. 5(a)-(c), 1).
Figure 6 showed the microstructure at 50µm (γ = 1.692) from the surface. The pearlite colony has undergone obvious plastic deformation, cementite suffered significant bending, and the interlamellar spacing decreased significantly (Fig. 6(a) and 6(b)). The proeutectoid ferrite grains after refining had an average size of 428 nm, and some of which were obviously deformed to form bamboo-like grains (Fig. 6(d), A), and the number of HABs further increased (Fig. 6(c), Fig. 6(d)). The number of high-angle between eutectoid ferrite lamellars increased markedly, and the number of LABs changed to HABs inside the eutectoid ferrite lamellar also increased remarkably, resulting in the formation of bamboo-like grains, and the eutectoid ferrite was further divided and refined (Fig. 6(d), B).
Figure 7 showed the microstructure within 15 − 0 µm (2.763 ≤ γ < 3.45). Figure 8 showed the grain size changes of two kinds of ferrite at different distances from the surface. It was known from Fig. 7(a) that within 15 − 2 µm, the plastic deformation and refinement of the surface microstructure were becoming more and more severe. The lamellar pearlite has lost its original lamellar characteristics, cementite was severely fragmented, and its size was dramatically reduced, and the fine carbides existed in the form of particles at the ferrite boundaries. Therefore, according to the distribution of cementite particles, the areas of proeutectoid ferrite and eutectoid ferrite could still be distinguished under SEM. Within 2 − 0 µm (3.314 ≤ γ < 3.45), the two kinds of ferrite continued to be refined and the carbides were largely dissolved, and the entire microstructure became fibrous parallel to the surface, thence the difference between the two could not be resolved under the SEM.
As could be seen from the EBSD analysis results, within 15 − 2 µm (2.763 ≤ γ < 3.314), proeutectoid ferrite and eutectoid ferrite grains continued to be divided and refined, and the number of HABs increased significantly, and fine bamboo-like grains were formed continuously (Fig. 7(b)-(c)). According to Fig. 8, the average grain size of the two measured at 11 µm (γ = 2.763) was 365 nm and 270 nm, 287 nm and 195 nm at 8µm (γ = 3.047), 190 nm and 153 nm at 4 µm (γ = 3.223), and the difference between the two gradually narrowed. However, within 2 − 0 µm (3.223 < γ ≤ 3.409), the ultrafine grains of about 110 nm were formed. It should be noted that due to the analytical capability of EBSD, the information of the most superficial finer grains is limited (Fig. 7(b), Fig. 7(c)), thence the actual average grain size should be less than the measured 110 nm (Fig. 9(d)).
According to TEM analysis, it could be seen from the bright field that the size of some grains was less than 100 nm (Fig. 9 (a)), and selected-are electron diffraction of ferrite was basically ring (Fig. 9 (b)), which also indicated that nanocrystalline was formed on the surface. This was basically consistent with the EBSD results.
The results of comprehensive statistics on the relative amounts of proeutectoid ferrite and eutectoid ferrite low-angle boundaries and high-angle boundaries at different distances from the surface as shown in Fig. 10. At 400 µm(γ = 0.012), the shear strain was relatively small, and the boundaries were mainly dominated by LABs, however, the HABs distribution were basically the same as the original microstructure, mainly distributed at 30–62°(Fig. 10(a)). At 300 µm(γ = 0.051), the number of LABs formed increased significantly, resulting in a significant increase in the proportion of LABs (Fig. 10(b)). With the decrease of the distance from the surface, the shear strain gradually increased, and the LABs gradually changed to the HABs, leading to the proportion of LABs decreased while the proportion of HABs increased significantly, and it was mainly distributed between 10–56° (Fig. 10(c)-(f)).