| Metallic glasses form a new category of materials with unique properties. However, theirformation is realized by controlling the cooling process of alloy melt to overpass crystallization,which in practice is not easy to manage, and no unified theory for metallic glass formation existsas yet. Some empirical rules to increase the chance to form bulk metallic glasses (BMG) areknown but the realization of BMG is not guaranteed. In addition these rules act differently fordifferent alloy systems. So it has been very difficult to know the compositions with both goodglass forming ability (GFA) and good material properties. It is crucial to understand and be ableto use criteria proposed by various researchers for GFA characterization, taking into account bothkinetics and thermodynamics of glass formation, to predict the GFA in different alloy systems.This good understanding is especially needed for alloy systems with poor or strange GFAbehavior, because for these systems, some of the criteria have shown a deviation fromexperimentally observed results. The aim of this thesis is to clarify the simple but reliable criteriathat we can use, to study the reason of unusual GFA in Fe-based system, and then to optimizeGFA for a specific application.Ones of the simplest criteria of GFA are those based on characteristic temperatures of thealloy. In this thesis, the efficiency of the most recent criteria including β, θ, γ, γm, γcand ω wasintensively analyzed through criteria-GFA correlations. In addition, because the critical coolingrate (Rc) is the most accurate measure of GFA but the maximum obtainable diameter (Zmax) themost crucial parameter in applications, nine different alloy systems were analyzed separately andthe correlations between the criteria and Rcand/or Zmaxwere evaluated intensively. Also, theparameters defining kinetics and thermodynamic behavior of crystallization were calculated forFe-based alloys using calorimetric measurements and physical properties of constituent elements.A nonlinear optimization of GFA for hard magnetic alloys with Rcas objective function waspresented in this thesis as well.We found that β-criterion has a weak correlation whereas θ, γ, γm, γcand ω criteria have almostthe same good correlation and same stability to database changes. These five criteria have almost the same R2-values for correlation with Rcand the same but small R2-values for correlation withZmax. The results revealed that the criteria-Zmaxdistribution strongly depends on the alloy systemstudied and for some alloy systems the distribution is totally random, which implies that the useof CTs-based criteria to predict GFA, as defined by the maximum diameter of obtainable totallyamorphous rod, is still questionable for some alloy systems including Fe-, and Zr-based alloys.For the Fe-based alloys, Rcestimated by combining kinetic and thermodynamic parametershighly correlates with the measured Rcfound in literatures. It is also found that for this group ofalloys, the compositions with high melting enthalpy ΔHmcan easily form glass even without highundercooling. High values of the β-parameter from Turnbull’s theory and high values of ΔHmobserved for high glass former alloys revealed that the glass formation in this group of alloys ismostly controlled by growth limitation lather than nucleation limitation.The results of GFA optimization in FeMoYBNd and FeNbYBNd hard magnetic alloysystems are consistent with the existing experimental results. It is found that the role of elementsbelieved to be GFA enhancers like Y is determined by the content of other elements in the systemand some times their addition can act negatively to the GFA improvement. For a given content ofB and Nd in Fe-Mo-Y-B-Nd and Fe-Nb-Y-B-Nd alloy systems, Y content has an optimum valuefor improving GFA. In order to find that value in the range2-6%a B/Nd ration is assumed to benear21.5%/6.5%in the Fe-Mo-Y-B-Nd system and near22.2%/5.8%in the Fe-Nb-Y-B-Nd alloysystem for the fixed Fe+Mo (or Fe+Nb) content of68%. It is also realized that contrarily to Fe-Mo-Y-B-Nd system; there are more chances to have composition with good GFA at high Ndcontent in the Fe-Nb-Y-B-Nd.The theoretical analysis in this work was partly verified by the experiments. Various Fe-Mo-Y-B-Nd alloys predicted by the theory were prepared by rapid solidification techniques,including melt spinning and ejection casting. The results showed that the theoretical model cansuccessfully predict the alloys with good GFA. All the four alloys predicted to have a GFAallowing at least2mm amorphous rods, a2.3mm diameter amorphous rod could be obtained inexperiment. Also it was approved that reasonable conditions can be set to ensure magneticproperties. However, in order to get excellent magnetic properties the heat treatment conditions need to be optimized. Nevertheless, this work provides useful guidelines for the compositionoptimization and preparation of nanocomposite permanent magnets by direct casting. |
PDF Full Text Request