Experimental and Numerical Investigations on Solidification Thermodynamics of H13 Steel with Multi components

Thermodynamic data is of great significance to investigate the formation and control mechanisms of solidification defects during the casting process of H13 steel which is high in Si, Cr, Mo, and V elements. It has been proven that the conventional Ueshima model based on the equilibrium phase diagrams of Fe-X (X = C, Si, Mn, P, S, Cr, Mo, and V) binary alloys cannot accurately predict the phase transition in the solidification of H13 steel with multi components. So, the pseudo-binary phase diagrams of Fe-X alloys at different initial concentrations were calculated via Thermo-Calc software. And, the datasets of liquidus and δ/γ phase transition temperatures were obtained. Then, a backpropagation (BP) neural network model was developed to predict the δ/γ phase transition temperature. While, the slopes of liquidus lines were fitted. These updates were implanted into the Ueshima model. And, the BP-Ueshima model was validated with the phase transition temperatures measured via the differential scanning calorimetry (DSC) test. Subsequently, the phase transition and solute micro-segregation behaviors in the solidification of H13 steel were analyzed as well as the influences of solute elements. The results show that the predicted liquidus temperature (T_L) and solidus temperature (T_S) of H13 steel via BP-Ueshima model agree with the experimental results. As the cooling rate increases from 10 to 20 °C/min, the phase transition temperatures change slightly. Both the solidus and liquidus temperatures decrease with increase of the initial contents of solute elements. Increasing the initial contents of C and Mn can enhance T_P and T_δ (the vanishing temperature of δ phase), whereas the trend is reversed for the other solute elements. Changes of the phase transition temperatures depends on the segregation behaviors of solute elements. The micro-segregation ratios of solute elements in the liquid phase at the end of solidification decreases in the order of S, P, Si, Mo, C, V, Mn, and Cr, respectively. It is determined by the redistributive capacity at the solid/liquid interface and the back diffusion in the solid phase of solute elements.