Mechanisms of dissolved arsenic removal by biochemical reactors: A bench- and field-scale study

A series of two biochemical reactors (BCRs) in Trail, British Columbia, removes dissolved As(V), Zn2+ and SO 4 2 - from leachate water collected from historic mine waste landfills. Each reactor cell is over 2000 m3 and contains a biosolid matrix which acts as a substrate for microbial activity. The current study used representative samples collected from the Trail BCRs as well as bench scale models of the first Trail BCR to demonstrate that As sulfide precipitation is a mechanism of As removal on both the bench- and field-scale and that the presence of Zn is important in As removal. Six reactors containing a matrix of biosolids were treated with an inlet solution containing 0.113 mM As(V) (8.5 mg L−1) and 0.812 mM SO 4 2 - (78 mg L−1) at a rate of 85 mL day−1; another six reactors containing biosolids and two reactors containing sand were treated with a similar inlet solution containing 0.352 mM Zn2+ (23 mg L−1) in addition to As(V) and SO 4 2 - . X-ray absorption near edge structure (XANES) results indicated the potential presence of a mixture of As(III)–O and disordered As2S3 compounds in the solid substrate from both sets of reactors containing biosolids following completion of the experiment as well as in the solid representative samples from the Trail BCRs, indicating As sulfide precipitation as a mechanism of As removal on both the bench- and field-scale. Based on other recent literature, soluble As(III)–S species may also be present in the samples, particularly in the samples collected from the field-scale BCRs, since the XANES peak for those samples is broad and slightly higher than that of the orpiment standard. Among the bench scale reactors containing biosolids, those treated with the inlet solution containing Zn2+ removed more As(V) than those not treated with Zn2+ (p < 0.001). The mechanism by which Zn enhances As removal is unclear, but several possibilities have been identified. Zinc may sequester excess free sulfide by forming Zn sulfides, thus preventing the dissolution of amorphous orpiment; Zn–As complexation may lower the solubility of As; and kottigite may form in the aerobic inlet area of the reactors. Reactors containing only sand also removed As from the inlet water, but XANES results showed that the main form of As in the sand matrix is As(V), indicating that the sand reactors likely work by As(V) adsorption to Fe oxyhydroxides and/or by the formation of Zn arsenate. Samples from all bench scale reactors and from the Trail BCRs were tested using the Toxicity Characterization Leaching Procedure (TCLP) and results indicated that the reactor contents are not considered hazardous waste.