Variational principles and forecasting models of the contents of Ca and Sr during the A356 melting process

Ca and Sr are the most effective modification elements in the casting of the Al-Si alloy, and they are usually added before casting when the alloy is in the form of the middle alloy. However, in the aluminum scrap melting regeneration industry, small amounts of Ca and Sr are often contained in the raw materials; as a result, controlling the mechanisms responsible for their contents in the raw aluminum scrap materials is a prerequisite for the reuse of such aluminum scrap during the melting process. This study aims to present the experimental investigation of the influences of the melting temperature and the holding time on the variational principles of the Ca and Sr percentage in an industrial A356 ingot during the melting process. The results indicate that the variations in the mass fractions of Ca and Sr both show Exp3P2 laws that continued decreasing gradually with the holding time and that the change rates of the mass fractions of Ca and Sr both improve with an increase in the melting temperature. Based on thermodynamic and dynamic analysis, the reactions ofCa andSr with oxygen in the melt occur and CaO and SrO are produced in the first stage of aluminum scrap melting regeneration process; subsequently, Al₂O₃·6CaO and Al₂O₃·SrO are produced by the reactions of CaO and SrO with Al₂O₃, respectively, before the mass fractions of Ca and Sr finally decrease after slagging. In the middle and later stages of the melting regeneration process, the Ca and Sr percentages are reduced byCa andSr diffusing to the melt surface and then deoxidizing Al₂O₃. The apparent activation energies of Ca and Sr oxidation reactions calculated at 660~740℃ during the A356 melting process are 182.6 kJ·mol^-1 and 117.8 kJ·mol^-1, respectively, demonstrating that the both reactions are controlled by the chemical reaction process. The Ca and Sr mass fraction prediction models are established according to their variations and the forecast error is less than 10% by the production verification; thus, both mass fraction prediction models can be used to predict the mass fractions of Ca and Sr in scrap aluminum melting regeneration.