WEATHERING OF PYRITIC TAILINGS IN UNSATURATED COLUMNS: EXPERIMENTAL AND REACTIVE TRANSPORT MODELING Patricia Acero, Carlos Ayora, Jordi Cama and Jesús Carrera Earth Sciences Institute Jaume Almera, CSIC, c/ Lluis Solé Sabarís s/n,, E-08028 Barcelona Abstract Several columns filled up with pyritic tailings were forced to desiccate by heating, and were dismantled at four successive stages of dryness. Porewaters and solid phases collected at different depths were characterized, and the results were interpreted by thermo-hydro-geochemical reactive transport modeling. Oxygen diffusion in the unsaturated zone leads to pyrite weathering and subsequent silicate dissolution. Evaporation at the top of the columns drove capillary upflow and mass transport to the surface, producing a crust of sulfates. This crust, which has also been identified in the field, caused a decrease in the evaporation rate of the columns, and promoted downwards water vapor flux, which condensed at depth and diluted porewater solutes. Introduction To prevent weathering and Acid Mine Drainage, sulphide tailings are usually kept under a layer of water during mining operations. However, once the exploitations are abandoned, this layer may disappear due to evaporation, which enhances the access of oxygen to sulfide minerals. This is especially relevant in the case of abandoned mines placed in sub-arid areas, where scarce rainfall favor evaporation and allow the access of oxygen to greater depths into the tailings. Earlier works focused on the geochemistry of porewaters and mineral phases in the water-saturated and vadose zones of tailings (Blowes et al., 1990; Gussinger et al., 2006, among many others). Also, an increasing number of works include reactive transport modeling (for revision, refer to Mayer et al., 2002). These studies provide snapshots of the spatial variability of the mine tailings geochemistry at different depths or areas, but no data on the evolution of porewaters and solid phases. On the other hand, column experiments have focused on the evolution of effluent waters and mineral association once the experiment has been dismantled (Malmström et al., 2006 and references therein). The present study is an attempt to study unsaturated pyrite-bearing wastes during weathering taking into account not only vertical variations of porewaters and mineral phases in the tailings profile, but also their time evolution. This is expected to unravel coupled heat, water and gas flow processes and mineral reactions which have not been described to now. Methods Ten columns with inner diameter of 14.2 cm, and length of 30 cm were filled with fresh and water-saturated mine tailings from a tailings impoundment. Each column was placed with the top surface at 60 cm below an infrared bulb, used as a heat source. Bulbs were kept switched on throughout the whole experiment, which lasted for 125 days. Water loss by evaporation was controlled by measuring the weight variations of the columns with time. Variations in the moisture content, mineralogy and porewater chemical composition with depth were investigated at four dryness stages. Each dismantling column was divided in depth into four sub-samples. Solids were separated for SEM and XRD and also for sequential extractions. The extraction of porewater was carried out by squeezing in the three first stages of dryness. Solute concentrations were measured by ICP-AES and ICP- MS, and Fe(II)/(III) by colorimetry. Moreover, the O isotope ratio was determined by equilibrium with CO 2 and subsequent analysis by IRMS. Reactive transport calculations were performed with CODEBRIGHT-RETRASO (Saaltink et al., 2004). The columns are represented by an initially homogeneous 1D porous medium. The main thermohydraulic processes controlling the evolution of our system are represented in Figure 1. The evolution of water contents was calculated from the calculated liquid and gas pressures through a Van Genuchten retention curve. Vapor pressure is calculated in equilibrium with the liquid taking into account the temperature and solute concentrations. Gas dissolution was always considered in equilibrium. Hydraulic conductivity, diffusivity and the thermal conductivity are considered to change with porosity and water saturation. From the geochemical point of view, a homogeneous mixture of minerals has been assumed as initial condition. The reactive area of each mineral phase has been obtained by distributing the measured BET specific surface between the different phases according to their volumetric fraction. The initial composition of the column porewaters has been assumed to correspond to one of the squeezing samples from the first dismantled column. IMWA Symposium 2007: Water in Mining Environments, R. Cidu & F. Frau (Eds), 27th - 31st May 2007, Cagliari, Italy