Search

Efforts to discern the role of an activated carbon's surface functional groups on the adsorption of elemental mercury (Hg0) and mercuric chloride demonstrated that chlorine (Cl) impregnation of a virgin activated carbon using dilute solutions of hydrogen chloride leads to increases (by a factor of 2-3) in fixed-bed capture of these mercury species. A commercially available activated carbon (DARCO FGD, NORITAmericas Inc. [FGD])was Cl-impregnated (Cl-FGD) [5 lb (2.3 kg) per batch] and tested for entrained-flow, short-time-scale capture of Hg0. In an entrained flow reactor, the Cl-FGD was introduced in Hg0-laden flue gases (86 ppb of Hg0) of varied compositions with gas/solid contact times of about 3-4 s, resulting in significant Hg0 removal (80-90%), compared to virgin FGD (10-15%). These levels of Hg0 removal were observed across a wide range of very low carbon-to-mercury weight ratios (1000-5000). Variation of the natural gas combustion flue gas composition, by doping with nitrogen oxides and sulfur dioxide, and the flow reactor temperature (100-200 degrees C) had minimal effects on Hg0 removal bythe Cl-FGD in these carbon-to-mercury weight ratios. These results demonstrate significant enhancement of activated carbon reactivity with minimal treatment and are applicable to combustion facilities equipped with downstream particulate matter removal such as an electrostatic precipitator.

Efforts to develop multipollutant control strategies have demonstrated that adding certain oxidants to different classes of Ca-based sorbents leads to a significant improvement in elemental Hg vapor (Hg0), SO2, and NOx removal from simulated flue gases. In the study presented here, two classes of Ca-based sorbents (hydrated limes and silicate compounds) were investigated. A number of oxidizing additives at different concentrations were used in the Ca-based sorbent production process. The Hg0, SO2, and NOx capture capacities of these oxidant-enriched sorbents were evaluated and compared to those of a commercially available activated carbon in bench-scale, fixed-bed, and fluid-bed systems. Calcium-based sorbents prepared with two oxidants, designated C and M, exhibited Hg0 sorption capacities (approximately 100 microg/g) comparable to that of the activated carbon; they showed far superior SO2 and NOx sorption capacities. Preliminary cost estimates for the process utilizing these novel sorbents indicate potential for substantial lowering of control costs, as compared with other processes currently used or considered for control of Hg0, SO2, and NOx emissions from coal-fired boilers. The implications of these findings toward development of multipollutant control technologies and planned pilot and field evaluations of more promising multipollutant sorbents are summarily discussed.