Geochemical and hydrological aspects of interactions between water and mine waste

Abstract: This thesis presents laboratory studies of sulphidic mine tailings remediated by flooding, and a pilot-scale study of dry cover for remediation of unoxidised sulphide-rich tailings. At the Kristineberg Zn-Cu mine in northern Sweden, sulphide-rich, poorly buffered tailings have been deposited in five impoundments located along a valley. By increasing pH by liming, thereby reducing metal mobility, a water-covered downstream impoundment is made to function as a trap for metals released from the till-covered impoundments upstream. As a result of the liming, a calcite-gypsum sludge has been formed on the tailings. The potential metal release from the sludge during resuspension events and in a long-term perspective was investigated by performing a shake flask test and sequential extraction of the sludge. A laboratory mixing experiment was performed to simulate the flocculation processes that occur in limed tailings ponds if stream water is diverted through a pond as a part of a remediation programme. The laboratory studies showed that the sequentially extracted carbonate and oxide fractions together contained ≥ 97% of the total amount of Cd, Co, Cu, Ni, Pb and Zn in the sludge. The association of these metals with carbonates and oxides appears to result from sorption and/or co-precipitation reactions at the surfaces of calcite and Fe, Al and Mn oxyhydroxides forming in the impoundment. In the shake flask test, remobilisation of Zn, Cu, Cd and Co (at a pH of 7-9) from the sludge resulted in dissolved concentrations of these metals that were significantly lower than those predicted to result from dissolution of the carbonate fraction of the sludge. This may suggest that cationic Zn, Cu, Cd and Co remobilised from dissolving calcite, gypsum and Al oxyhydroxides were readsorbed onto Fe oxyhydroxides remaining stable under oxic conditions. The laboratory mixing experiment showed that the flocculants that settled were rich in C (18 wt%) and acid leachable Fe (14-19wt%). Thus, organic matter and Fe oxyhydroxides appear to form substantial fractions of the flocculants. Trace metal uptake in the flocculants that settled in the mixing bottles resulted in removal of Cd, Co, Cu and Zn from the dissolved phase. Another part this thesis work focuses on the hydrological aspects of dry covers. The water balance in well-defined, pilot-scale systems with various types of dry cover applied on mine tailings at the Kristineberg site has been studied. Experimental studies of water infiltration through dry covers on sulphidic tailings can be used for predicting cover performance and cover design parameters relating to the same climate conditions and the same type of tailings. For this purpose, different multilayer covers have been applied in pilot-scale test cells. Clayey till, sewage sludge, Trisoplast (a mixture of a polymer, bentonite and tailings sand) and fine-grained apatite are used as sealing layers, on top of which a protective cover of unspecified till has been applied. To be able to evaluate the behaviour of the multilayer covers in tailings-impoundment scale, and to predict how the different barrier constructions influence infiltration rate and water balance, the experimental soil covers have been monitored for water percolation, climate conditions and frost penetration. The different infiltration rates in different cells were discussed. The initial results for the first two years of monitoring suggest that the apatite concentrate as a sealing layer is more promising in reducing net infiltration than a 0.3 m thick clayey till. The high fractions of water percolated through the sealing layers (~30% of precipitation) may be explained by the absence of natural run-off from the experimental cells. The results show that snowmelt, freezing of the soil and differences in soil hydraulic properties appear to have effects on the differences in water balance in the cells.