Model for Ferric Sulfate Leaching of Copper Ores Containing a Variety of Sulfide Minerals" Part I. Modeling Uniform Size Ore Fragments BRADLEY C. PAUL, H.Y. SOHN, and M.K. McCARTER A computer model was constructed for bacterial ferric sulfate leaching of the major sulfides found in porphyry copper deposits. Leaching occurs by reactions with ferric ion diffusing into the rock fragment. The model incorporates the reaction kinetics of the individual minerals and keeps material and energy balances. The model is needed to aid in the design of modified in situ leaching operations and combines desirable features found in previous models with extensions needed for the described study of in situ leaching. In modeling the leaching of single ore frag- ments, it is shown that the rate of ferric ion generation by bacteria can limit the rate of copper recovery. The transition from kinetic to diffusion rate limitation is different for each mineral and ore fragment size. The width of the leaching reaction zone is different for each mineral, and many reaction zones cannot be considered narrow. Minerals do not leach in proportion to their concentration in ore fragments. I. INTRODUCTION A study was undertaken to evaluate the technological and economic feasibility of recovering copper from cop- per sulfide porphyry type orebodies by a method referred to as flood drain leach cell (FDLC) mining. The process is briefly illustrated in Figure 1. Cells 60 by 75 m in plan view and 70 m in height are created throughout the ore body using modified vertical retreat mining with 25 pct removal of swell material. The swell material is leached in finger dumps on the surface, while the cells underground are bulkheaded, flooded, drained, and ar- tificially aerated to promote bacterial ferric sulfate leach- ing. Solutions are cycled through several cells to reach a pregnant grade of 1 g of copper per liter of solution. These solutions are then pumped to an underground sol- vent extraction plant and then to a surface electro- winning tankhouse. A key to both the technical design and economic fea- sibility of this method is the prediction of leaching rates and the resulting heat and oxygen consumption loads on the ventilation system. A search was made for a model that could provide a simulation of the leaching process and output the needed information for design. A model prepared by Cathles and Schlitt t~ provides output on the temperature profile and oxygen concentration in waste dumps when air moves by convection. The air flow as- sumptions of the Cathles-Schlitt model do not fit forced ventilation of confined stopes, but more importantly, the model assumes that copper is extracted from a single narrow reaction zone (a shrinking core) in each of the fragments of the ore mass. As demonstrated later in Sec- tion V, many reaction zones may not be narrow and the BRADLEY C. PAUL, Assistant Professor of Mining Engineering, is with the Department of Mining Engineering, Southern Illinois University at Carbondale, Carbondale, IL 62901. H.Y. SOHN, Professor of Metallurgical Engineering, Department of Metallurgical Engineering, and M.K. McCARTER, Professor of Mining Engineering, Department of Mining Engineering, are with the University of Utah, Salt Lake City, UT 84112. Manuscript submitted July 1, 1991. amount of oxygen used per unit of copper produced may not be a constant, as assumed in the Cathles-Schlitt model. Madsen and Wadsworth 121 developed a model that pro- vides separate reaction kinetics for pyrite, chalcopyrite, chalcocite, and covellite, thus overcoming the single narrow reaction zone problem. Bornite, enargite, cuban- ite, native copper, and pyrrohtite are sometimes signif- icant minerals in porphyry-like systems, but no model identified at the time deals with the bacterial leaching of these minerals. The implicit solution scheme used in the Madsen-Wadsworth model is subject to numerical prob- lems associated with ill-conditioning and requires that kinetic equations be approximated as linear functions of ferric ion concentration. Bartlett pl has developed a model for pressure leaching that overcomes the numerical ap- proximations of the Madsen-Wadsworth model using a rigorous explicit scheme. Lin and Sohn t41 simplified the numerical scheme while still retaining most of the vir- tues of the explicit solution. The Lin-Sohn model of 1987, however, was for oxygen pressure leaching of chalco- pyrite only. None of the models reviewed considers that the rate at which bacteria can produce ferric ion might become rate-limiting. It was thus determined that the authors would create a new model that would incorporate oxygen consump- tion and heat generation, like the Cathles-Schlitt model, multiple mineral kinetics, like the Madsen-Wadsworth model, and the simplified explicit numeric solution scheme of the Lin-Sohn model of 1987. To this model would be added separate kinetics for many previously neglected but useful minerals and a mass balance on ferric iron in the bulk solution that considers the rate at which the iron can be oxidized by bacteria. The leaching of sulfides in a typical porphyry ore is an exothermic process. The bacterially catalyzed oxidation of ferrous to ferric ion consumes oxygen. If an environment conducive to typ- ical active ferric ion generating bacteria is to be main- tained, oxygen must be supplied and heat removed. In FDLC mining, forced ventilation is used to cool the rub- ble and provide oxygen to all parts of the cells. This METALLURGICAL TRANSACTIONS B VOLUME 23B, OCTOBER 1992--537