skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Confounding effects of aqueous-phase impinger chemistry on apparent oxidation of mercury in flue gases

Abstract

Gas-phase reactions between elemental mercury and chlorine are a possible pathway to producing oxidized mercury species such as mercuric chloride in combustion systems. This study examines the effect of the chemistry of a commonly used sample conditioning system on apparent and actual levels of mercury oxidation in a methane-fired, 0.3 kW, quartz-lined reactor in which gas composition (HCl, Cl{sub 2}, NOx, SO{sub 2}) and quench rate were varied. The sample conditioning system included two impingers in parallel: one containing an aqueous solution of KCl to trap HgCl{sub 2}, and one containing an aqueous solution of SnCl{sub 2} to reduce HgCl{sub 2} to elemental mercury (Hg{sup 0}). Gas-phase concentrations of Cl{sub 2} as low as 1.5 ppmv were sufficient to oxidize a significant fraction of the elemental mercury in the KCl impinger via the hypochlorite ion. Furthermore, these low, but interfering levels of Cl{sub 2} appeared to persist in flue gases from several doped rapidly mixed flames with varied post flame temperature quench rates. The addition of 0.5 wt% sodium thiosulfate to the KCl solution completely prevented the oxidation from occurring in the impinger. The addition of thiosulfate did not inhibit the KCl impinger's ability to capture HgCl{sub 2}. The effectivenessmore » of the thiosulfate was unchanged by NO or SO{sub 2}. These results bring into question laboratory scale experimental data on mercury oxidation where wet chemistry was used to partition metallic and oxidized mercury without the presence of sufficient levels of SO{sub 2}. 23 refs., 5 figs., 1 tab.« less

Authors:
; ; ; ; ;  [1]
  1. University of Utah, Salt Lake City, UT (United States). Department of Chemical Engineering
Publication Date:
OSTI Identifier:
21049831
Resource Type:
Journal Article
Resource Relation:
Journal Name: Environmental Science and Technology; Journal Volume: 42; Journal Issue: 7; Other Information: brydgercauch@yahoo.com
Country of Publication:
United States
Language:
English
Subject:
01 COAL, LIGNITE, AND PEAT; MERCURY; FLUE GAS; OXIDATION; BENCH-SCALE EXPERIMENTS; CHLORINE; SIMULATION; NITRIC OXIDE; CONCENTRATION RATIO; SULFUR DIOXIDE; THIOSULFATES; CHEMICAL STATE; AQUEOUS SOLUTIONS; MERCURY CHLORIDES; POTASSIUM CHLORIDES; TIME DEPENDENCE; IMPINGEMENT; CHEMICAL REACTIONS

Citation Formats

Brydger Cauch, Geoffrey D. Silcox, Joann S. Lighty, Jost O.L. Wendt, Andrew Fry, and Constance L. Senior. Confounding effects of aqueous-phase impinger chemistry on apparent oxidation of mercury in flue gases. United States: N. p., 2008. Web. doi:10.1021/es702490y.
Brydger Cauch, Geoffrey D. Silcox, Joann S. Lighty, Jost O.L. Wendt, Andrew Fry, & Constance L. Senior. Confounding effects of aqueous-phase impinger chemistry on apparent oxidation of mercury in flue gases. United States. doi:10.1021/es702490y.
Brydger Cauch, Geoffrey D. Silcox, Joann S. Lighty, Jost O.L. Wendt, Andrew Fry, and Constance L. Senior. 2008. "Confounding effects of aqueous-phase impinger chemistry on apparent oxidation of mercury in flue gases". United States. doi:10.1021/es702490y.
@article{osti_21049831,
title = {Confounding effects of aqueous-phase impinger chemistry on apparent oxidation of mercury in flue gases},
author = {Brydger Cauch and Geoffrey D. Silcox and Joann S. Lighty and Jost O.L. Wendt and Andrew Fry and Constance L. Senior},
abstractNote = {Gas-phase reactions between elemental mercury and chlorine are a possible pathway to producing oxidized mercury species such as mercuric chloride in combustion systems. This study examines the effect of the chemistry of a commonly used sample conditioning system on apparent and actual levels of mercury oxidation in a methane-fired, 0.3 kW, quartz-lined reactor in which gas composition (HCl, Cl{sub 2}, NOx, SO{sub 2}) and quench rate were varied. The sample conditioning system included two impingers in parallel: one containing an aqueous solution of KCl to trap HgCl{sub 2}, and one containing an aqueous solution of SnCl{sub 2} to reduce HgCl{sub 2} to elemental mercury (Hg{sup 0}). Gas-phase concentrations of Cl{sub 2} as low as 1.5 ppmv were sufficient to oxidize a significant fraction of the elemental mercury in the KCl impinger via the hypochlorite ion. Furthermore, these low, but interfering levels of Cl{sub 2} appeared to persist in flue gases from several doped rapidly mixed flames with varied post flame temperature quench rates. The addition of 0.5 wt% sodium thiosulfate to the KCl solution completely prevented the oxidation from occurring in the impinger. The addition of thiosulfate did not inhibit the KCl impinger's ability to capture HgCl{sub 2}. The effectiveness of the thiosulfate was unchanged by NO or SO{sub 2}. These results bring into question laboratory scale experimental data on mercury oxidation where wet chemistry was used to partition metallic and oxidized mercury without the presence of sufficient levels of SO{sub 2}. 23 refs., 5 figs., 1 tab.},
doi = {10.1021/es702490y},
journal = {Environmental Science and Technology},
number = 7,
volume = 42,
place = {United States},
year = 2008,
month = 4
}
  • A flue gas generator has been built in order to study mercury reactions. The generator consists of a propane-fueled furnace and a 12 m long temperature-controlled steel duct with a fabric filter. When vaporized (elemental) mercury was added to the propane flame in a concentration of 150 {mu}g/m{sup 3} and with 8% excess of oxygen, 20-30% of the mercury was oxidized after 0.8 s residence time in the furnace and the duct (T > 500{degree}C). In the presence of HCl(g) in the flue gas, most of the elemental mercury was oxidized after 0.8 s. The reaction product is assumed tomore » be mercuric chloride. A rereduction of the oxidized mercury occurs when the temperature has decreased below 200{degree}C. This reduction is probably a heterogeneous reaction at the surface of the flue gas duct. Experiments with mercury and activated carbon powder (0.5-1.0 g/m{sup 3}, at NTP, specific surface 792 m{sup 2}/g (BET)) in the flue gas duct have also been performed. The results indicate that activated carbon may act as a catalyst in the oxidation of mercury.« less
  • The extent of Hg oxidation determines the portion of Hg in the flue gas from a coal-fired power station that can be removed in SO{sub 2} scrubbers. This article evaluates predicted extents of Hg oxidation from a detailed chemical reaction mechanism, emphasizing the data from 1 and 29 MW pilotscale furnaces for diverse coal types. The proposed mercury (Hg) oxidation mechanism consists of a 168-step gas phase mechanism that accounts for interaction among all important flue gas species and a heterogeneous oxidation mechanism on unburned carbon (UBC) particles, similar to established chemistry for dioxin production under comparable conditions. The mechanismmore » was incorporated into a gas cleaning system simulator to predict the proportions of elemental and oxidized Hg species in the flue gases, given relevant coal properties (C/H/O/N/S/Cl/Hg), flue gas composition (O{sub 2}, H{sub 2}O, HCl), emissions (NOx, SOx, CO), the recovery of fly ash, fly ash loss-on-ignition (LOI), and a thermal history. Predictions are validated without parameter adjustments against datasets from lab-scale and from pilot-scale coal furnaces at 1 and 29 MWt. Collectively, the evaluations cover 16 coals representing ranks from sub-bituminous through high-volatile bituminous, including cases with Cl{sub 2} and CaCl{sub 2} injection. The predictions are, therefore, validated over virtually the entire domain of Cl-species concentrations and UBC levels of commercial interest. Additional predictions identify the most important operating conditions in the furnace and gas cleaning system, including stoichiometric ratio, NOX, LOI, and residence time, as well as the most important coal properties, including coal-Cl. 33 refs., 4 figs., 3 tabs.« less
  • Trace element loss between collection and analysis from liquid environmental samples may be large. This paper describes losses to polyethylene container walls of As, Hg, Sb, Se, Cd and Zn, elements likely to be present in the vapor phase in flue gas from pulverized coal utility boilers. Losses were measured in four types of impinger solutions commonly used to absorb volatile elements. Adsorption losses were measured by adding radiochemical tracers to the selected impinger solutions (double distilled H/sup 2/O, and 10% solutions of H/sup 2/O/sup 2/, Na/sup 2/CO/sup 3/ and ICI) and measuring the amounts adsorbed onto their container wallsmore » over a period of up to 100 days. Arsenic, Sb, and Se showed only small losses. Zinc, Cd, and Hg were stable in some solutions and unstable in others. Freeze dried samples had no measurable losses.« less