Fundamentals of Mercury Speciation Kinetics: A Theoretical and Experimental StudyEPA Grant Number: R828168
Title: Fundamentals of Mercury Speciation Kinetics: A Theoretical and Experimental Study
Investigators: Wendt, Jost O.L. , Blowers, Paul
Institution: University of Arizona
EPA Project Officer: Shapiro, Paul
Project Period: September 1, 2000 through August 31, 2002 (Extended to August 31, 2004)
Project Amount: $225,000
RFA: Exploratory Research - Engineering, Chemistry, and Physics) (1999) RFA Text | Recipients Lists
Research Category: Engineering and Environmental Chemistry , Water , Land and Waste Management , Air
This is a fundamental research project and it is concerned with developing validated detailed kinetic models to describe the inter-conversion of mercury species in combustion flue gases. To this end the project contains novel experimental and theoretical components. Within the experimental component, a sub-objective is to measure gaseous combined mercury species directly, using a mass spectrometer. This has not been accomplished heretofore because the probable concentrations that can be measured using mass spectrometry (MS) are very much higher than those of practical interest for coal combustion, although they may have application in incineration. We believe mass spectrometers will measure species in the 10-100 ppm range, rather than the ppb range, which is of interest for pulverized coal combustion. We circumvent this difficulty by using the multi-mercury species measurements at high concentrations to validate a realistic kinetic model in which concentrations of several mercury species may be required for validation. This model will then be used to extrapolate to low concentrations to yield ppb level predictions. Consistency with these ppb level predictions will then be verified experimentally using mercury metal only measurements by UV absorption, supplemented by wet chemistry. Note that UV absorption can measure Hg metal only, and wet chemical methods, at best, can separate metallic mercury from all other mercury species, and that this separation is insufficient to glean fundamental knowledge about detailed chemical mechanisms. Hence, data of multiple mercury species is required, and that can currently only be obtained at the higher Hg levels.
Within the theoretical component of this research, a sub-objective is to predict gas phase elementary rate coefficients between mercury and chlorine molecules and chlorine atoms. This is no easy task, due to relativistic effects that are important in describing the intermolecular interaction function for heavy metals. Methods to accomplish these predictions are discussed below. With the rate coefficients so determined, we shall develop a CHEMKIN-based kinetic model including them and others from the C, H, N, O, system, and calibrate this against experimental measurements of pertinent mercury species in a plug flow reactor.
This research has two novel attributes: first, it attempts to measure the temporal profile of various mercury species directly, using a mass spectrometer, rather than inferring them by difference; second, it involves theoretical estimation of gas phase mercury reaction rate constants and subsequent modeling of detailed kinetics of the Hg, Cl, S, C, H, O, N system. The direct measurements using mass spectrometry will be for Hg, HgCl2, HgO, HgS, and other (possibly organic) mercury species, as appropriate.
Experimental work employs a downflow quartz reactor with combustion gases doped with mercury at both ppm levels for the mass spectrometry measurements, and at ppb levels, for relevance to practical situations. It is known that mass spectrometry requires the concentrations to be in the ppm range, which is greater than the ppb range found in coal combustion units. We use the high concentration experiments to validate a realistic kinetic model, and use the model to extrapolate to ppb conditions. Selected ppb conditions are then experimentally investigated using existing UV absorption spectroscopy for Hg metal and traditional "difference" techniques. The result should be a validated kinetic model describing the inter-conversion and formation of mercury chloride, mercury oxide, and mercury sulfide (and other sulfur compounds) as functions of reactor conditions (temperature, presence of flame, stoichiometric ratio, mercury concentration, chlorine and sulfur concentrations). Experiments with and without flame present are important since these, together with detailed kinetic modeling, can elucidate the role of free radicals, including the role of atomic chlorine, which is thought to be very important. This research focuses on homogeneous gas phase mercury speciation kinetics. Capabilities exist for the addition of real and simulated fly ash into the reactor, but for the present, heterogeneous reactions are outside the scope of this research, but will be the subject of future research.
Premixed flames will be supported at the top of the heated quartz reactor on a sintered quartz flame holder. Flames will be in a non-sooting mode, in order to avoid deposition on the walls. Some experiments will also be performed in the absence of flames. Temperatures will be controlled by independent flow control of oxygen, nitrogen and methane. Mercury will be added from permeation tubes (less than 1 ppm Hg @1 slpm flue gas) or from direct Hg vaporization (greater than 1ppm). Chlorine and sulfur will be admixed to the incoming mixture. An existing Extrel C-50 Mass Spectrometer with a 300 l/s turbo pump will be used for the mass spectrometry measurements. A suitable sampling system for this will be constructed and employed, as described in the proposal.
Theoretical estimation of rate constants will be based on previous theoretical work of one of the PIs. This work has heretofore focused on developing more accurate correlations for activation energies by, for example, addition of repulsive interactions between the reactants. To assess the importance of repulsive interactions for heavy metal species like mercury, quantum mechanical calculations will be performed for reactions involving mercury with chlorine, hydrogen, and oxygen. These will use effective core potentials (ECP), which replace the core electrons on mercury with a simpler description using an effective potential. Only the valence electrons are still treated explicitly during the calculations. Also, with elements as heavy as mercury, relativistic effects on the electrons within the effective core approximation must be taken into account. This suggests that the ECP of Moysagin or Dolg, for example, will be more accurate than the ECPs which do not include relativistic effects. The rate coefficients so determined will be used in conjunction with CHEMKIN based software to predict mercury species profiles in the tubular reactor configuration, and these predictions will be compared to measurements.
The first output is a method for measuring ppm concentration levels of mercury species in combustion flue gases, and the actual measurements. The second output is a realistic theoretical kinetic model describing mercury speciation inter-conversion kinetics, where the model is based on detailed chemical kinetics and fundamentally grounded predictions of pertinent rate coefficients.