Mercury removal from contaminated groundwater: Performance and limitations of amalgamation through brass shavings

Brass shavings have been proposed as a cost-effective filter material to remove Hg from contaminated groundwater. This method, which is based on the reduction of reactive Hg(II) and subsequent formation of amalgams, has been shown to be fast and effective in the short term. However, the effectiveness of brass filters and their stability over the long term, especially if used in passive filter systems such as permeable reactive barriers (PRB) under high flow conditions, is unknown. To evaluate the performance and limitations of brass shavings for Hg removal from contaminated groundwater, we performed longterm pilot scale filtration tests (6 and 28 months) at two former wood impregnation sites with severe groundwater contamination (up to 870 mu g L-1 Hg). The results showed that even under high flow conditions (>60 m d(-1)), 60-80% of the Hg was removed in the first 8 mm of the brass shavings filter bed. The kinetics of filtration, Hg total removal performance (>99.95%), and loading capacity (164 g L-1) surpassed those of a Hg-specific synthetic resin (LEWATI (R) MonoPlus TP-214). However, under natural pH conditions (pH 6.4 and 6.7), Zn was leached from the brass and exceeded the threshold value (0.5 mg L-1) in the filter outflow by up to a factor of 40. Increasing pH (>8.5) decreased the Zn concentration (<0.05 mg L-1) but affected Hg removal due to the formation of Zn-hydroxide/carbonate coatings on the brass (up to 15% performance reduction). Thus, the use of brass shavings as an exclusive filter material in PRBs is restricted to aquifers with high pH. However, brass is ideal as a low-cost, thin-bed prefilter in onsite systems to remove the main Hg load from groundwater when Zn release is managed. (C) 2016 Elsevier Ltd. All rights reserved.