|About the Book|
This dissertation describes the design, evaluation and utilization of a reactor used for the investigation of the gas-phase reactions of mercury with halogens. Reactor conditions are tailored to be representative of industrial coal-fired utilityMoreThis dissertation describes the design, evaluation and utilization of a reactor used for the investigation of the gas-phase reactions of mercury with halogens. Reactor conditions are tailored to be representative of industrial coal-fired utility boilers. Experiments were performed to verify the ability of the reactor to produce data suitable for validation of detailed kinetic models and mechanisms.-The reactor used in this investigation employs a 300-W quartz glass burner to generate 6 slpm of combustion gases. Gases pass through a quartz tube 47 mm in diameter and 132 cm in length and experience temperatures from 1100 to 350°C. All reactants are introduced through the flame with mercury concentrations of 25 mug/m3 and chlorine concentrations ranging from 0 to 600 ppmv as Cl. A residence time of approximately 6 seconds is achieved and is dependent on temperature profile. Mercury analysis is performed using a Tekran 2537A mercury vapor analyzer and a wet chemical conditioning system.-Mercury oxidation has been investigated using two temperature profiles with quench rates of 210 and 440 K/s. With a chlorine concentration of 200 ppmv, 72 percent oxidation is achieved with the high-quench temperature profile and 20 percent oxidation with low-quench. Quartz surfaces have been shown to attract chlorine species, probably the Cl radical, and subsequently adsorb mercury. This phenomenon was demonstrated to inhibit gas-phase mercury oxidation, but this effect is small and may be neglected under the conditions examined here.-NO and SO2 in the flue gas were identified as causing bias in speciated mercury measurements. Injection of these species into the KCl impinger of the conditioning system caused reduction of Hg2+, resulting in an apparent reduction in the amount of oxidized mercury. Injection of SO2 into the KCl impinger results in complete reduction (that is all mercury reported as elemental). The effects of NO are identical whether injection occurs at the impinger or through the burner. The potential impact of this measurement bias on currently accepted mercury measurement techniques should be investigated further.-A detailed kinetic model was successful in predicting the effects of chlorine concentration, quench rate and oxygen concentration on mercury oxidation demonstrated in experiment. The model was also used to suggest that effects of NO and SO2 on Hg oxidation were probably caused by liquid-phase reactions in the impinger.