GEOCHEMICAL AND ISOTOPIC CHARACTERIZATION OF THE ENVIRONMENTAL IMPACTS OF COAL COMBUSTION PRODUCTS: LESSONS FROM THE TENNESSEE VALLEY AUTHORITY COAL ASH SPILL

Our long-term investigation of the largest coal ash spill in US history at the Tennessee Valley Authority (TVA) coal-fired power plant in Kingston, TN has revealed several conclusions about the potential impacts of coal combustion products (CCPs) on the environment. Our initial investigation of the TVA coal ash spill showed that CCPs could leach contaminants into the environment, and mobilization of leachable coal ash contaminants (LCACs) such as boron, arsenic, selenium, strontium, and barium depend on ash composition combined with environmental factors such as pH and redox conditions. Release of CCPs to surface water in areas of restricted water exchange near the spill revealed high levels of LCACs (e.g., As: 9- 95 µg/L), but remediation and removal of ash from this area combined with diversion of surface water led to a reduction of the LCAC concentrations with time. Throughout the clean up, systematic sampling of the water and sediments reveal that the Clinch and Emory rivers downstream of the spill had low LCAC concentrations below the EPA maximum contaminant level (As=10 µg/L), but with levels (e.g., As ~4 µg/L) above the baseline of the upstream rivers. Dredging activities to remove the ash from the river had little impact on the concentration of LCACs in the surface water. The data show that dilution effect plays a significant role in the concentration of LCACs in affected surface water. In contrast, pore water extracted from bottom sediments of the downstream rivers had significantly higher LCAC levels (e.g., As 9-2010 µg/L), which were associated with some degree of anoxic conditions that can affect the mobilization of some LCACs.

Field data and independent leaching experiments indicate that boron is a sensitive indicator for CCP leaching, with boron content up to 1518 µg/L in pore water relative to upstream river water (6 to 9 µg/L) with a distinctive isotope fingerprint that is significantly different from that of meteoric boron in the upstream rivers. Pore water and surface water (sampled downstream of the TVA spill) had a δ¹¹B value of -11‰ and -12‰ respectively (relative to the NIST951), while most natural waters have a more positive δ¹¹B value (+10 to +20‰). The results from this investigation will allow for better environmental management practices and better monitoring of the potential impacts CCPs on the environment.