The present investigation provides some new directions to improve our 100
understanding about the morphologic evolution of carbides. Furthermore, it helps us to clarify the conditions for the development of fibrous carbide. It has been pointed out the chemical composition of steel, the transformation temperature, the coherency of interface are the factors on determining carbide morphologies, which are summarized as follows:
a. Effects of the transformation temperature
Based on the microscopic observations and diffusion considerations, as the transformation temperature is lowered, interphase precipitation becomes dominant on the basis of minimizing interfacial energies and maximizing diffusion efficiency. It makes the carbide that the carbide platelet would be as parallel as to the terrace plane, leading to interphase precipitation. On the contrary, carbide fiber would be originated from a curved interface. Edmonds addressed that at higher transformation temperature, the habit plane of carbide does not have to be on the terrace plane because the mobility of solute atom is able to diffuse in the direction perpendicular to the interface as the transformation temperature is elevated, and carbide fiber is then developed. However, Barbacki and Honeycombe proposed that the carbide fiber is favored as the transformation temperature is lowered. According to the ledge mechanism proposed by Honeycobme, the terrace plance of interphase precipitation is supposed to be semi-coherent, followed
{
1 1 0} {
α || 1 1 1}
γ orientation relationships. As the transformation temperature is lowered, the solute atom tends to diffuse along an incoherent interface to maximum the diffusion efficiency, which is perpendicular to the ferrite/austenite interface.Cahn [85] had pointed out that the ledge mechanism of transformation would occur
101
at higher transformation temperatures because the lower transformation driving force at higher transformation temperatures is unable to make the overall interface to advance uniformly. According to this proposition, it concludes that the carbide fiber would be dominant at lower transformation temperatures because the driving force of transformation is large enough. However, this obviously deviates from the experimental observations of the carbide aggregates. The contradiction between Cahn’s proposition and microscopic observations has not been highlighted.
b. The effect of interface coherency
The deviation indicated in the previous point implies that the crystallographic and coherency of interface would not be the main factors to determine carbide morphologies.
The crystallographic aspects had been discussed by Law et al [86]. Miyamoto and Yen have shown that a ledged interface can even be developed from an incoherent interface.
However, the original ferrite/austenite interface characteristics are difficult to retain because the austenite would transform into martensite. Kitahara et al. had developed a method to re-construct the orientation of the interface by using the inverse orientation matrix [87, 88]. An area with ferrite and martensite was scanned by EBSD clarify this issue, as shown in Figure 4-6. The analysis of the misorientation of ferrite/austenite interface is according to the work of Furuhara and Miyamoto [89, 90]. The sample is a Fe-0.36C-1.35Mn-0.33V-0.33Si alloyed steel isothermally transformed at 650 oC for 3 min. It can be seen that allotriomorphic ferrite and idiomorphic ferrite were transformed from austenite with this grain. The misorientation of ferrite/austenite interface is revealed by OIM-software. SEM was then used to examine the microstructure of each ferrite grain (allotriomorphic and idiomorphic), as shown in Figure 4-7. Only the ferrite labeled as i3 was found to have carbide fibers, and no carbide was observed in the rest
102
ferrite grains. By referring to the proposition for carbide fiber development, it is expected that the misorientation of i3 is supposed to be larger than other ferrite grains.
The misorientaiton map presented in Figure 4-7(a) indeed shows that the i3 actually has a larger misorientation angle (~ 40o). However, for other grains which exhibit large misorientation angles (a3 and a5 for examples), carbide fibers did not appear. As a consequence, it can conclude that the coherency of the ferrite/austenite interface does not influence greatly on the determination of carbide morphology. This is complementary with the findings presented by Yen [14], Okamoto [91] and Miyamoto [51], which showed that a ledge interface does not have to be semi-coherent.
c. The alloying elements
Alloying elements which influence the transformation kinetics can be used to alter the carbide morphologies as well. Mn and Ni are generally recognized to slower the rate of austenite to ferrite transformation and had been found in increasing the amount of carbide fibers in alloy steels [5]. However, increasing in vanadium content did not promote the volume fraction of VC fibers. No attempts had been made on the competition between interphase precipitated carbides and fibrous carbides by varying carbon or vanadium contents. The effects of carbon and solute contents on the development of fibrous carbides are still unclear.
The conditions and features of carbide fiber proposed among studies have now are presented in Table 2-1. Nevertheless, the carbide fibers in Fe-V-C steels are less studied.
Detailed studies on the development of fibrous carbide have been few in number, and direct evidence of the propositions is still lacking, particularly in terms of how fibrous carbides nucleate and grow.
103
Figure 4-6 (a) The selected area for EBSD analysis showing ferrite allotriomorphs, idiomorphs, and martensite; (b) Orientation map of the selected area; (c) a stereographic projection showing the martensite variants transformed from this prior austenite grain, and (d) coupling with measured martensite (M1) and ferrite (a1) orientations.
Figure 4-7 (a) A micrograph showing the calculated deviation angle of ferrite grains; the ferrite with carbide fiber is labeled, (b) a SEM image of the labeled ferrite grain, and (c) a larger magnification of (b).
104