The Electric Power Research Institute (EPRI) is conducting research to investigate mercury removal in utility flue gas using sorbents. Bench-scale and pilot-scale tests have been conducted to determine the abilities of different sor-bents to remove mercury in simulated and actual flue gas streams. Bench-scale tests have investigated the effects of various sorbent and flue gas parameters on sorbent performance. These data are being used to develop a theoretical model for predicting mercury removal by sorbents at different conditions. This paper describes the results of parametric bench-scale tests investigating the removal of mercuric chloride and elemental mercury by activated carbon. Results obtained to date indicate that the adsorption capacity of a given sorbent is dependent on many factors, including the type of mercury being adsorbed, flue gas composition, and adsorption temperature. These data provide insight into potential mercury adsorption mechanisms and suggest that the removal of mercury involves both physical and chemical mechanisms. Understanding these effects is important since the performance of a given sorbent could vary significantly from site to site depending on the coal- or gas-matrix composition.

Sorbents for removing mercury from flue gases of coal-fired power plants are presently being evaluated due to potential regulation of mercury emissions under Title III of the 1990 Clean Air Act Amendments. Laboratory tests have been conducted to evaluate the adsorption characteristics of potential sorbents and the effects of flue gas constituents on these characteristics. This paper presents a theoretical model that combines the adsorption characteristics measured in the lab with mass transfer considerations to predict mercury removal by the duct injection process in actual flue gas streams. The model was used to determine the effect of various sorbent properties on mercury removal when injecting a powdered sorbent upstream of either an electrostatic precipitator (ESP) or fabric filter. Mercury removal is expected to differ between these configurations since the mass transfer conditions are different in an ESP and fabric filter. The model was used to determine when mercury removal is limited by mass transfer and when it is limited by sorbent capacity. This information defines conditions when removal can be improved by reducing particle size or increasing sorbent capacity. In both cases, removal can be increased by injecting more sorbent.

Coal-derived activated carbons (CDACs) were tested for their suitability in removing trace amounts of vapor-phase mercury from simulated flue gases generated by coal combustion. CDACs were prepared in bench-scale and pilot-scale fluidized-bed reactors with a three-step process, including coal preoxidation, carbonization, and then steam activation. CDACs from high-organicsulfur Illinois coals had a greater equilibrium Hg 0 adsorption capacity than activated carbons prepared from a low-organic-sulfur Illinois coal. When a low-organic-sulfur CDAC was impregnated with elemental sulfur at 600 °C, its equilibrium Hg 0 adsorption capacity was comparable to the adsorption capacity of the activated carbon prepared from the high-organicsulfur coal. X-ray diffraction and sulfur K-edge X-ray absorption near-edge structure examinations showed that the sulfur in the CDACs was mainly in organic forms. These results suggested that a portion of the inherent organic sulfur in the starting coal, which remained in the CDACs, played an important role in adsorption of Hg 0 . Besides organic sulfur, the BET surface area and micropore area of the CDACs also influenced Hg 0 adsorption capacity. The HgCl 2 adsorption capacity was not as dependent on the surface area and concentration of sulfur in the CDACs as was adsorption of Hg 0 . The properties and mercury adsorption capacities of the CDACs were compared with those obtained for commercial Darco FGD carbon.