DYNAMIC RESPONSE OF A PLANT-SCALE GIBBSITE PRECIPITATION CIRCUIT

Predicting gibbsite precipitation circuit response to changes in operating parameters is very challenging because of the nonlinear nature and long circuit response times in precipitation that can extend to several months. A dynamic population balance based precipitation model developed in this work provides a powerful tool for studying the response of such plant precipitation circuits. The mathematical model, built for a plant precipitation circuit with fine and coarse seed recycles delivering an alumina production of 0.8 Mt per annum, is used in this work to investigate operating parameters affecting the operating stability of gibbsite precipitation circuits. The precipitation process in each continuous precipitation tank is modelled using a dynamic precipitation population balance equation incorporating crystal growth, nucleation, and agglomeration kinetics. The implemented precipitation kinetics are based on more than a decade of research. The precipitation circuit model predicts a dynamic response of full particle size distribution, which is crucial for evaluating the overall precipitation yield and product properties. It is shown here that the dynamic precipitation circuit model leads to a unique steady-state solution regardless of the initial conditions applied. It is also demonstrated that the performance of the classification unit can significantly affect both the precipitation yield and the crystalline product properties.