Research On The Application Of The Molecular Interaction Volume Model In The Fe-based Solid Alloy Systems

In this thesis, the thermodynamic properties of 37 binary solid alloys have been evaluated and fitted by the molecular interaction volume model (MIVM) and the regular solution model. The results show that the evaluation effect of MIVM is less than that of regular solution model, but the fitting effect of MIVM is better than those of both models. Both models are suitable for the positive deviation systems, weak negative and coexisted deviation ones, but not for the strong negative and coexisted deviation ones. With the increasing of deviation between experimental values and ideal values, the evaluation and fitting effects of both models become less. Generally speaking, both models are more acceptable in the proportional solid solutions than the limited solid solutions, but not in the systems which include main group element specially. At the same time, the roots of B_ijB_ji > 1 are safety that have been gotten andproved after analyzing the selecting methods of roots, in which systems there are three group roots.Then based on the results of binary solid alloys, thermodynamic properties of 4 ternary and 1 quaternary iron-based solid alloys are predicted and compared by MIVM and the regular solution model. The results show that the predicted activities of MIVM are better agreement with the experimental results in the multi-component systems, in which the corresponding binary systems are strong coexisted deviation systems. On the contrary, MIVM is less than the regular solution model in the multi-component systems, in which the corresponding binary systems are strong coexisted deviation systems. As to MIVM, the prediction effect is not good if the evaluation effects of the corresponding binary systems are not good in multi-component systems, vice versa. Both MIVM and the regular solution model have a significant feature which can predict the thermodynamic properties of multi-component solid alloys from the corresponding binaries. MIVM needs only the corresponding binary infinite dilute activity coefficients, whereas regular solution model has to use some continuous analytical expressions of experimental data of the binaries, so MIVM has a simpler method to evaluate. And MIVM is closer to real solution and clearer physical meaning than regular models.Finally, the Miedema model and MIVM are applied to predict thermodynamics properties of iron-based multi-component solid alloys. The results show that MIVM is not suitable for multi-component solid alloys in this way. We think that the failed reason lies in that the mixing enthalpy and the infinite dilution activity coefficient of Miedema model can't reflex the real solution completely, and MIVM can't predict thermodynamic properties of multi-component solid alloys with the parameters that are fitted by Miedema model.