A predictive numerical model for localized erosion-corrosion of metals in disturbed two-phase flow has been developed. The flow structure is determined by the application of a two-phase flow version of a k − ϵ eddy viscosity, low Reynolds number model of turbulence. The Eulerian approach for the fluid flow is coupled with a Lagrangian approach for particle motion. Local values of fluid velocity and turbulent and molecular transport coefficients are determined along with particle–wall interactions in terms of impact velocity, angle, and frequency. The corrosion component of the model assumes mass-transfer control. The mass-transfer rates are determined by the solution of the mass-transport equation simultaneously with the fluid flow equations. The erosion is determined on the basis of the computed particle–wall interactions and cutting-wear erosion equations.
Simulation results for the erosion of stainless steel and the erosion–corrosion of mild steel at a sudden expansion in a vertical pipe carrying a dilute 2% sand/water slurry are compared with previously measured values. The erosion–corrosion rates at a groove in a steel pipe are also presented to demonstrate the generality of the model.
