AbstractThe apparent leach kinetics for an ore particle within a heap leaching system depend on the chemical conditions in the fluids around the particle, the mass transport within the particle and the reaction kinetics at the surface of each mineral grain. The apparent rate kinetics thus depend upon the distribution of the mineral grains, in terms of both size and position, within the individual ore particles, as well as the evolution of this distribution. Traditionally this behaviour has been modelled using simplified relationships such as the shrinking core model. In this paper a method for simulating this evolution and the resultant kinetics based directly on 3D XMT images of the internal structure of the particles is presented. The model includes mass transport through the gangue matrix, surface reaction kinetics and the dissolution and subsequent evolution of the individual mineral grains within the ore particle. Different minerals and mineral associations will result in different surface reaction kinetics. One of the key inputs into this model is thus the distribution of the surface rate kinetics. A method for experimentally determining this distribution is presented. The simulation results are compared to the evolution of real particles as they undergo leaching as measured using a time sequence of 3D XMT images of a leaching column. It was found that these simulations are able to accurately predict both the overall leaching trends, as well as the leaching behaviour of mineral grains in classes based on their size and distance to the particle surface. The leaching behaviour did not follow that of a simple shrinking core approximation, with the actual spatial and size distribution of the grains, as well as the distribution of their surface rate kinetics, all impacting the apparent leach kinetics. For the copper ore particles used in this work the best fit to the experiments was achieved at an intermediate value of the dimensionless group that characterises the relative importance of surface kinetics to diffusion indicating that both need to be considered for accurate modelling. This paper thus demonstrates that using 3D XMT to provide both structural and kinetic data and incorporating this information into a particle scale simulator provides an improved basis for predicting particle scale leach performance.