Laboratory and field evaluation of a flushable oxic limestone drain for treatment of net-acidic drainage from a flooded anthracite mine, Pennsylvania, USA

Abstract

This paper demonstrates the use of dissolution-rate data obtained in the laboratory to indicate the potential quality of effluent from a field-scale oxic limestone drain (OLD) treatment system for neutralization of dilute acidic mine drainage (AMD). Effluent from the Reevesdale Mine South Dip Tunnel, a large source of AMD and base flow to the Wabash Creek and Little Schuylkill River in the Southern Anthracite Coalfield of east-central Pennsylvania, is representative of AMD with low concentrations but high loadings of dissolved Fe, Al and other metals because of a high flow rate. In January 2003, rapid neutralization of the AMD from the Reevesdale Mine was achieved in laboratory tests of its reaction rate with crushed limestone in closed, collapsible containers (Cubitainers). The tests showed that net-alkaline effluent could be achieved with retention times greater than 3 h and that effluent alkalinities and associated dissolution rates were equivalent for Fe(OH)3-coated and uncoated limestone. On the basis of the laboratory results, a flushable OLD containing 1450 metric tons of high-purity calcitic limestone followed by two 0.7-m deep wetlands were constructed at the Reevesdale Mine. During the first year of operation, monthly data at the inflow, outflow and intermediate points within the treatment system were collected (April 2006-2007). The inflow to the treatment system ranged from 6.8 to 27.4 L/s, with median pH of 4.7, net acidity of 9.1 mg/L CaCO3, and concentrations of dissolved Al, Fe and Mn of 1.0, 1.9 and 0.89 mg/L, respectively. The corresponding effluent from the OLD had computed void-volume retention times of 4.5-18 h, with median pH of 6.6, net acidity of -93.2 mg/L CaCO3, and concentrations of dissolved Al, Fe and Mn of <0.1, 0.08 and 0.52 mg/L, respectively. The wetlands below the OLD were effective for retaining metal-rich solids flushed at monthly or more frequent intervals from the OLD, but otherwise had little effect on the effluent quality. During the first year of operation, approximately 43 metric tons of limestone were dissolved and 2 metric tons of Al, Fe and Mn were precipitated within the OLD. However, because of the accumulation of these metals within the OLD and possibly other debris from the mine, the effectiveness of the treatment system declined. Despite the installation of a flush-pipe network at the base of the OLD to remove precipitated solids, the limestone bed clogged near the inflow. Consequently, a large fraction of the AMD bypassed the treatment system. To promote flow through the OLD, the flush pipes were open continuously during the last 4 months of the study; however, this effluent was only partially treated because short-circuiting through the pipes decreased contact between the effluent and limestone. A reconfiguration of the flow path through the limestone bed from horizontal to vertical upward could increase the limestone surface area exposed to the metal-laden influent, increase the cross-sectional area perpendicular to flow, decrease the flow path for solids removal, and, consequently, decrease potential for clogging.