Development of a chemical kinetic model for the homogeneous oxidation of mercury by chlorine species: A tool for mercury emissions control

The U.S. Environmental Protection Agency (EPA) recently announced that mercury emissions from coal burning power plants will be regulated by the year 2004. This impending regulation is a concern to the power industry because inexpensive means of controlling mercury are unavailable. Mercury emissions vary considerably, and the reasons for this variation are not well understood. Mercury exists in the flue gas in elemental and oxidized (HgCl₂) forms. Literature data show that the oxidized form is more likely to be removed from the flue gas by existing pollution control equipment. During combustion, the mercury in the coal is vaporized. In the flame zone, all the mercury is in the elemental form. However, as the flue gas cools, a portion of the mercury becomes oxidized. This oxidation process involves both homogeneous and heterogeneous mechanisms. This research focuses on developing a homogeneous chemical kinetic mechanism that describes mercury oxidation by chlorine species.; Experiments performed using a natural gas tunnel furnace and a heated quartz reactor are compared with similar results from the literature. The possible elementary reactions that may lead to oxidation are reviewed and a chemical kinetic model is proposed. This model yields good qualitative agreement with the data and indicates that mercury oxidation occurs during the thermal quench of the combustion gases and not in the high temperature zone near the flame. In particular, the model suggests superequilibrium chlorine atom as the gas cools. This reactive chlorine then oxidizes the mercury in the quench region, where the reverse decomposition reactions of the oxidized mercury are suppressed. This mechanism appears to resolve the dilemma presented by literature data that show oxidized mercury existing at high temperatures. Instead, all the oxidized mercury is apparently formed as the gases are cooled prior to analysis. Once developed, the model was successful in explaining various features in the literature data. For instance, the model showed why water, a significant flue gas component, interferes with the oxidation process. This model is a first step in understanding the mercury oxidation process and in controlling mercury emissions.