Development of a thermodynamic database for copper smelting and converting

Introduction of thermodynamic modeling into the practice of the pyrometallurgical copper production may significantly increase the economic efficiency and environmental friendliness of the industry. A reliable thermodynamic database in combination with the Gibbs energy minimizing software is able to predict the energy balance, distribution of elements, phase transformations during the chemical process. Implemented in the control/simulation package, such thermodynamic database and software will be able to raise the process automation to a new level.;In particular, liquid metal and matte phases were described as one solution using the Modified Quasichemical Formalism in the pair approximation. The solution was built using one sublattice and uncharged species: (Cu I, CuII, FeII, FeIII, O, S). In this way, the model is able to describe wide deviations from stoichiometry in the sulfide phase towards excess metal or excess sulfur and oxygen. The oxygen solubility in the sulfide phase is correctly described. The model takes into account the existence of compositions of maximum first nearest neighbor short-range ordering near Cu2O, CuO, FeO, Fe2O 3, Cu2S and FeS and predicts the drastic changes in P(O 2) and P(S2) at these compositions. It accurately estimates the solubilities of sulfur and oxygen in metal region of the solution. It can be used to predict solid-liquid equilibria in chemical systems containing sulfide and oxide phases at low and high oxygen partial pressures.;The slag solution was modeled within the Quasichemical Formalism in quadruplet approximation. Cations are charged and placed in a separate sublattice from anions: (Al3+, Ca2+, Cu1+, Fe 2+, Fe3+, Mg2+, Si+4)(O 2-, S2-). The model of this type takes into account the second nearest neighbor short-range ordering between basic (Ca 2+, Fe2+, etc.) and acidic (Si4+) cations, while the first nearest neighbor short range ordering between metals and non-metals is assumed to be complete. Previously optimized, re-optimized and newly obtained databases for oxide subsystems were combined together and complemented with sulfide components. The resulting slag database is able to describe phase relations in oxide systems, as well as sulfide capacities. The effect of sulfur on the solubility of copper in slag is demonstrated and modeled. This is of primary importance for the industry which is trying to avoid copper losses into slag. The effect of calcium on the solubility of copper and sulfur in fayalite slag is modeled quantitatively for the first time..;Copper and calcium were added into the spinel database in a thermodynamically consistent way. The spinel database may be used to estimate the wearing of refractory materials and to monitor the conditions to prevent the spinel precipitation, which might cause occlusions of reactors. It is able to predict cation distribution between sublattices, which is important for electronic applications.;The goal of this thesis is the development of the database able to predict the thermodynamic properties and phase equilibria in the basic chemical system for the pyrometallurgical copper production, which is the Al--Ca-- Cu--Fe--Mg--O--S--Si system. This involves the critical evaluation and thermodynamic optimization of available experimental data for all phases. The database is self-consistent, i.e. solid, liquid and gas phases are assessed simultaneously. 4 binary, 5 ternary, 7 quaternary, 2 quinary and 1 six-component system were evaluated or re-evaluated. The project is performed in close collaboration with the Pyrosearch group (University of Queensland, Australia), who performed some crucial experiments. Some of the above-mentioned systems were optimized together with T. Hidayat. In addition, 4 ternary and 5 quaternary systems were optimized by T. Hidayat in co-operation with the present author and reported in another Ph.D. thesis [1].;The predictive ability of the database was tested during the simulation of a smelting furnace. The calculation results are compared with the plant data. Very good agreement is demonstrated in the estimation of energy and mass balance and in the prediction of element distributions between outputs of the furnace.;Thus, the thermodynamic database for the applications in the pyrometallurgical production of copper, developed during the course of this project, is the most accurate and complete in the world. It is intended to be used with the FactSage(TM) software. Using the methods, refined in this study, more elements can be easily added to the database. The work currently continues on the addition of Co, Pb, Ni and Zn to the database.