A new viscosity model for the high-iron coal slag based on the bond distribution

Abstract

The coal slag viscosity, intrinsically determined by the slag structure, is the key to the smooth slag tapping process of an entrained flow coal gasifier. However, the slag structure parameters obtained by the experiments (Raman, NMR, and XPS) or by the theoretical calculation (Molecular dynamics simulation and First-principles calculation) are very time-consuming. This study employed thermodynamic modeling as a convenient method to determine the short-range-ordered structure of high-iron coal slag. The viscosity prediction performance of the structure parameters obtained by the thermodynamic modeling was compared with that obtained by the experiments (Raman and XPS). Our results demonstrated that the short-range-ordered structure of coal slag was described as the distribution of the seven bonds determined by thermodynamic modeling. A new structure parameter (RNF/NM) based on the bond distribution was proposed to quantify the network depolymerization degree of slag. The activation energy (Eη) for the viscous flow, which was predicted by RNF/NM, was much more accurate than the structure parameters (Qn distribution, Al coordination, and NBO/(BO + NBO)) that were determined by the experiments, even the effect of Fe3+-tetrahedron was considered. Finally, a novel slag viscosity model as a function of RNF/NM was proposed. This study provides new insights into the structure analysis and viscosity modeling of coal slag by thermodynamic modeling. Additionally, this method can be applied to the viscosity prediction of other slag systems, such as metallurgical slag, glass, etc.