A laboratory investigation of PCB removal from stormwater using biochar-amended engineered soils in columns

The purpose of this study was to investigate the benefit of using biochar as an amendment to a common bioretention soil media (BSM) to improve the removal of polychlorinated biphenyls (PCBs) from urban stormwater. Five biochar products were admixed with BSM at a one-to-three ratio by volume and placed in 7.5-inch-diameter glass columns to a depth of 18 inches, typical of bioretention field installations. One additional column was prepared as a control and filled with 18 inches of BSM meeting the specifications from the Bay Area Stormwater Management Agencies Association (BASMAA). Stormwater was collected in the San Francisco Bay Area and applied to the columns in four experimental runs. Three runs used undiluted stormwater on all columns and the fourth run used stormwater diluted at a one-to-nine ratio in a single column to test removal effectiveness at lower influent pollutant concentrations. Influent and effluent samples were analyzed for total PCBs, suspended solids concentration (SSC), and turbidity. Column study results were compared to field results obtained from a biochar-amended bioretention pilot site installed and monitored as part of the Clean Watersheds for a Clean Bay Project (CW4CB). The results of this column study are intended to be used to choose a biochar for further study in field and laboratory tests. Compared to the standard BSM column, all five biochar-amended BSM columns provided greater total PCB and SSC percent removal. Standard BSM percent removal ranged from 75 to 90 percent, with a mean of 83 percent. All five biochar-amended BSM percent removal ranged from 54 to 96 percent, with a mean of 88 percent. Due to data variability and relatively few values, the aggregated biochar-amended BSM dataset was not statistically different from the standard BSM results. Two products, Sunriver and Rogue, provided the best treatment, based on total PCB effluent concentrations normalized by influent concentrations. However, the performances of both products are based on only two data points each because some data were deemed unreliable due to unusually high infiltration rates that suggested short-circuiting in some column tests. When their data are combined, the aggregate dataset is statistically distinguishable from standard BSM. In addition, Rogue provided the best SSC removal. Based on this study, Rogue and Sunriver biochars are strong candidates for further field and laboratory testing for the removal of PCBs.