Machine Learning Molecular Dynamics Simulations of Local Structure and Transport Properties of TiO2-FeO-Ti2O3 Melt

There are some problems in the reduction smelting process of ilmenite, such as low reaction rate, poor separation of slag and iron, and inferior quality of titanium slag. The control of the transport properties of titanium slag melt is the key to achieve the efficient preparation of high-quality titanium slag. This work uses neural network to obtain machine learning potential function, and verifies its accuracy according to atomic force and system energy. In this work, the molecular dynamics simulation of the local structure and transport properties of TiO2-FeO-Ti2O3 system is studied by using the obtained machine learning potential function. The results show that TiO68- octahedron and TiO69- octahedron are involved in the construction of network skeleton, and the stability of TiO68- octahedron is higher than that of TiO69- octahedron. Under different FeO contents, the proportions of TiO68- and TiO69- octahedron in the system are dominant. When the content of FeO increases from 5% to 19%, the tricluster oxygen and bridge oxygen in the system are transformed into non-bridge oxygen and free oxygen, the DSC value of the system decreases from 1.37 to 0.62, Q4, Q5 and Q6 are transformed into Q0, Q1, Q2 and Q3, the DOP value decreases from 4.34 to 1.84. When the content of FeO increases from 5% to 19%, the complexity and polymerization degree of the system decrease, the overall strength of the network skeleton decreases, and the viscosity value of the system decreases from 0.043 Pa·s to 0.037 Pa·s. The results will lay a theoretical and technical foundation for the low-carbon and efficient preparation of high-quality titanium slag.