Figure 1 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Solidification sequence as simulmated with Micress 6.4 and Thermo-Calc TCFE9 and MOBFE3 databases. Red − liquid, orange − δ-ferrite, white − austenite.
Source publication
Two ingots weighing 400 g and 5 kg with nominal compositions of Fe-8Mn-4Al-2Si-0.5C-0.07V-0.05Sn were produced to investigate the effect of processing variables on microstructure development. The larger casting has a cooling rate more representative of commercial production and provides an understanding of the potential challenges arising from cast...
Contexts in source publication
Context 1
... a time step of 0.1/cooling rate was applied to ensure sufficient resolution at all cooling rates. A typical solidification sequence can be seen in Figure 1. ...
Context 2
... rolled and IA microstructures of the three tensile samples are shown in Figure 10 and the phase fractions are shown in Table 3. The fully homogenised 400 g (24 h) sample (Figure 10c) showed fine equiaxed ferrite grains, located predominantly along the austenite grain boundaries. ...
Context 3
... rolled and IA microstructures of the three tensile samples are shown in Figure 10 and the phase fractions are shown in Table 3. The fully homogenised 400 g (24 h) sample (Figure 10c) showed fine equiaxed ferrite grains, located predominantly along the austenite grain boundaries. This accurately reflects the process route where the steel was first rolled in a temperature regime where the steel was fully austenitic and the ferrite grains were newly formed α-ferrite grains which nucleated during the final passes and the intercritical annealing heat treatment. ...
Context 4
... relationship between casting condition and grain size distribution in the rolled strips can be seen in the Cumulative Distribution Function (CDF) plots in Figure 11 . Grain size was determined as the equivalent circle diameter from the EBSD data in Figure 10 but excluding grains which intersect the micrograph boundary. ...
Context 5
... relationship between casting condition and grain size distribution in the rolled strips can be seen in the Cumulative Distribution Function (CDF) plots in Figure 11 . Grain size was determined as the equivalent circle diameter from the EBSD data in Figure 10 but excluding grains which intersect the micrograph boundary. δ-ferrite grains were also excluded by first determining the percentage of δ-ferrite grains out of the total ferrite grains in Table 3, i.e. 16% and 11% of ferrite grains in the 5 kg (2 h) and 400 g (2 h) samples respectively. ...
Context 6
... ferrite grains were then sorted in decreasing size and the largest 16% and 11% of the ferrite grains were assumed to be δ-ferrite and excluded since δ-ferrite grains were consistently larger than α-ferrite grains. From Figure 11a, the austenite grain size CDFs were very similar. However, there was some spread in the α-ferrite grains size CDFs between the three rolled strips. ...
Context 7
... effect may be attributed to a lower driving force for α-ferrite grain growth during IA due to pre-existing δ-ferrite grains. Nevertheless, Figure 11 shows that the grain size CDFs, especially of the austenite phase, are largely independent of casting condition and δ-ferrite fraction. From Figure 9, the three rolled strips had nearly identical strain hardening behaviour regardless of casting condition or δ-ferrite fraction. ...
Context 8
... Figure 9, the three rolled strips had nearly identical strain hardening behaviour regardless of casting condition or δ-ferrite fraction. Since the austenite phase in all three strips had a similar grain size distribution (Figure 11a), this then strongly implies that the austenite phase across all three samples should have the composition, i.e. same Stacking Fault Energy (SFE) and stability (Md 30 ). However, the austenite phase fraction was not constant across all three samples ( Table 3), suggesting that the composition of the austenite could not have been identical under normal circumstances. ...
Context 9
... the alloying additions in medium Mn steel, C is able to significantly alter the SFE and Md 30 of austenite, even in small concentrations [15,21,30]. Therefore, in order for the austenite phase in the partially homogenised samples to maintain the same C content as the fully homogenised sample while having a lower austenite fraction, it was likely that the excess C in the partially homogenised samples precipitated in the form of carbides along the δ-ferrite interphase boundaries (Figure 10d). This phenomena was not observed in the 400 g (24 h) sample where there was no δ-ferrite. ...
Context 10
... the post mortem 5 kg (2 h) tensile sample in Fig- ure 10d, the fracture edge was observed to have sheared across the entire microstructure. Brittle transverse cleavage of δ-ferrite grains was not observed near the fracture edge nor in the gauge. ...
Context 11
... Table 3 and Figure 11, it is evident that the scale of production has little impact on the final grain size and distribution, and as such, further rolling reduction during commercial production is not likely to refine this much more. However, this work has shown that there is a strong dependency on the casting parameters on the stability of δ-ferrite and how it manifests in the final product. ...
Context 12
... this work has shown that there is a strong dependency on the casting parameters on the stability of δ-ferrite and how it manifests in the final product. Fig- ure 12 shows the influence of cast cooling rate on the area fraction of δ-ferrite after 2 h homogenisation at 1250 • C. This difference has shown to have a noticeable impact on tensile properties with the slow cooled ingot showing a significant amount of δ-ferrite that forms stringers after rolling. ...
