HSLA 강 내의 TiC 조대화에 미치는 Mo 와 W 효과

Abstract

There is a new vigour in the development of low-alloy steels containinga fine dispersion of substitutionally alloyed carbides. The steels have amicrostructure which is essentially ferritic, but with TiC or (Ti,Mo)C particleswhich are less than 10 nm in size, generated at the austenite-ferriteinterface during the course of phase transformation. This interphase precipitationmechanism has been known for a long time, but its applicationto automotive steels which compete with dual phase and transformationplasticity based alloys is much more recent.The steels are mass produced and in the final stages are coiled at temperaturesin the vicinity of 600C. A key feature of alloy design, therefore,is the use of the complex (Ti,Mo)C carbide, which is found to coarsen ata much slower rate than the pure TiC, during the cooling of the coil toambient temperature. The mechanism for the effect of molybdenum isnot understood, but it is the fine dispersion of carbides that permits theotherwise weak ferrite to gain sufficient strength to be of use in a varietyof engineering applications.Many of the variants that have been developed also contain niobiumas a microalloying addition. High-resolution transmission electron microscopyhas been used to characterise the precipitates in Ti-Nb andTi-Nb-Mo bearing steels, using both thin foil and extracted carbides. In this way, precipitation and coarsening kinetics have been characterisedand form the basis for comparison against mathematical models later inthe thesis.The role of molybdenum has been investigated using wide rangingfirst-principles calculations for a variety of (Ti,M)C precipitates, where'M' stands for niobium, vanadium, molybdenum or tungsten. The purposeof this was to see whether the molybdenum acts to reduce coarsening by athermodynamic effect or whether the phenomenon is principally kinetic.In fact, molybdenum has the effect of reducing the stability of TiC, butit at the same time reduces the crystallographic misfit between ferriteand the carbide, and as a consequence, the interfacial energy per unitarea. It is this latter parameter which controls coarsening and explainswhy molybdenum leads to a more stable dispersion. Furthermore, it isfound that molybdenum incorporated into the carbide at the early stagesof precipitation, is rejected as the carbide grows beyond the nucleationstage, confirming the first principles estimates that its presence in theTiC is not favoured.In an interesting the results from the ab-initio calculations suggesta new alloy system based on (Ti,W)C precipitates which should be aseffective as (Ti,Mo)C by the same mechanism, in resisting coarsening.Finally, a detailed analysis is reported on three different models forrepresenting the observed coarsening behaviours. The first is based onclassical Ostwald ripening theory due to Lifshitz, Slyozov and Wagner,which essentially assigns the problem to the diffusion of a 'controlling'solute (i.e., a binary alloy), and leads to a result in which the normalisedsize distribution is invariant with time, even though the small particlesdissolve and larger ones grow. A model due to Kampmann and Wagneravoids the assumption of a particular form of particle size distribution, butstill treats the problem as if the system concerned is binary. A computationalmodel based on the LSW particle size distribution, but which properlytreats multicomponent diffusion has also been studied

this model also has the advantage of revealing concentration profies within the matrixas the particles evolve. Naturally, the different models give similar resultsexcept for the Kampmann-Wagner method, where the particle sizedistribution is not invariant with time.