当前位置：文档库 › Development of Mercury Emissions Control Technologies for the Power Industry

Development of Mercury Emissions Control Technologies for the Power Industry

Paul S. Nolan Babcock & Wilcox Barberton, Ohio, U.S.A.

BR-1685

Presented to:

EPRI-DOE-EPA Combined Utility Air Pollutant Control Symposium

August 16-20, 1999

Atlanta, Georgia, U.S.A.

Abstract

Concern over the difficulty of limiting the emission of trace amounts of hazardous air pollutants in large flue gas streams to meet potential environmental regulations has been a driving force for the development and characterization of control tech-nologies for utility boilers. In cooperation with the U.S. De-partment of Energy and the Ohio Coal Development Office,Babcock & Wilcox/McDermott Technology, Inc. has been con-ducting the Advanced Emissions Control Development Program to investigate the formation and control of these emissions from coal combustion in the 10 MW e -equivalent Clean Environment Development Facility. This paper discusses results recently obtained on mercury speciation and removal, and their poten-tial implications on the applicability of such conventional tech-nologies as electrostatic precipitators, baghouses, sorbent in-jection, and wet flue gas desulfurization for cost-effective con-trol of mercury emissions.

Introduction

The combustion of coal in U.S. power generation facilities has historically represented the source potentially responsible for the largest quantities of pollutant emissions in the country.Because of this, the power industry has been, and continues to be, the focal point for the development and implementation of emission control technologies. The control of mercury emis-sions presents technical challenges not faced in earlier devel-opmental efforts to capture fly ash, sulfur dioxide (SO 2), and nitrogen oxides (NO x ) that are all normally generated at rela-tively high concentrations in flue gas.

Visible particulate emissions, resulting from coal’s ash con-tent in the range of 3–20 weight percent, first received atten-tion decades ago and are now routinely controlled at efficien-cies of 99.7 percent and more by electrostatic precipitators (ESPs) and baghouses. Concern over the acid rain precursors SO 2 (generated at concentrations of several hundred to several thousand parts per million (ppm) in proportion to the 0.2–6weight percent of sulfur in the fuel) and NO x (compounds also typically found in the hundreds of ppm concentration range,usually as a function of combustion conditions) fostered the development of control methods for each.

Wet flue gas desulfurization (FGD) technology has evolved from a typical 90 percent removal capability in the 1970s to more reliable and lower cost systems that can achieve 95–98percent capture in systems being installed today. Because high NO x levels are typically more a combustion byproduct than something resulting only from fuel nitrogen content, efforts focused first on identifying low-NO x burner designs. This,sometimes combined with modifications of combustion tech-niques such as “reburn” technology, results in roughly 50 to 75percent emissions reduction over those characteristic of ear-lier combustion techniques. Beyond these, overall NO x reduc-tions in the 90 to 95 percent range are possible today with the addition of Selective Non-catalytic Reduction (SNCR, typically designed at 30 to 50 percent reduction) and/or Selective Cata-lytic Reduction systems (SCR, typically designed at 70 to 90percent reduction).

Passage of the Clean Air Act Amendments of 1990 signifi-cantly raised the degree of sophistication required for identifi-cation and possible control of extremely low concentrations of

George A. Farthing

Deborah Madden Yurchison

Michael J. Holmes

Babcock & Wilcox /McDermott Technology, Inc.

Alliance, Ohio, U.S.A.

the elements and compounds collectively known as Hazardous Air Pollutants (HAPs, sometimes also referred to as “air toxics”). The power industry, governmental agencies, and equip-ment and technology providers have been developing the higher level of technical expertise to assure a reasonable and quanti-tative representation of the species to be targeted, the environ-mental risk they pose, and the means, effectiveness, and cost of control. On the national level, the United States Depart-ment of Energy (DOE), the United States Environmental Pro-tection Agency (EPA) and the Electric Power Research Insti-tute (EPRI) have been engaged in extensive programs aimed at characterizing HAPs’ emissions and their environmental and health impacts.

Fortunately, most of the trace elements generated by coal combustion are in the solid phase at the operating tempera-tures of electrostatic precipitators (ESPs) and baghouses. As a result, existing control equipment can provide high efficiency removal for most HAPs. The exceptions to this as a general rule are those elements and compounds whose volatility per-mits them to pass through the particulate collection device in the gaseous state or with the fine, uncollected fly ash. Several mercury, arsenic, and selenium species and dioxins exhibit rela-tively high volatility, making them somewhat more difficult to collect than many of the other HAPs. After several years of study, the Mercury Study Report to Congress and the Study of Hazardous Air Pollutant Emissions from Electric Utility Steam Generating Units – Final Report to Congress identified mer-cury and its compounds as the HAPs of primary concern due to the potential for multi-pathway exposure risks and possible eventual bioaccumulation in the food chain.[1,2] As a result, the EPA has deferred regulatory determination for mercury until additional information can be gathered, including a major ini-tiative entitled the Electric Utility Steam Generating Unit Mercury Emissions Information Collection Effort.[3] The adequacy of the sampling and analytical techniques used for determinations, often at or near the detection limits, was also one of the more formidable issues faced as the various programs unfolded. While it is not the purpose of this paper to describe or even summarize the work done at several sites and laboratories to validate these procedures, it is important to keep in mind that this required at least as significant an effort just to develop improvements in the techniques in order to achieve greater accuracy and precision in the measurements. This was particularly true for mercury, whose chemistry is more com-plex than most, due to the ease with which it transforms from one oxidation state to another. EPA Method 29 had long been recognized as a validated method for measuring total mercury, but it did not permit one to identify the different mercury spe-cies making up this total. The significance of this was particu-larly important as anomalous variations in the relative amounts of elemental and divalent, “oxidized” forms of mercury were routinely being found in measurements from field sites. Only recently has a variation of EPA Method 29 known as the Ontario Hydro Method come to be recognized as the preferred tech-nique to distinguish between elemental mercury and its oxi-dized forms.[3]

Both the EPA and EPRI estimate that the annual uncontrolled mercury emissions from coal-fired utility boilers in the United States total about 50 to 55 tons. Based on field sampling at utility sites, the concentrations in the flue gas are generally in the range of 5 to 30 mg/dscm, equivalent to annual emissions of roughly one-third to one pound of mercury per megawatt of generating capacity. To provide some perspective on the chal-

lenge the industry has been facing, municipal waste combus-tors (MWC), another significant point source with uncontrolled mercury concentrations in the 200 to 1000 mg/dscm range, most of which typically exists in the more easily removed oxidized form, are currently required to control their emissions below 80 mg/dscm.

Advanced Emission Control Development Program Overview

As part of the research and development conducted for the environmental equipment product line, Babcock & Wilcox (B&W) and McDermott Technology, Inc. (MTI) entered into a cooperative effort with the DOE and the Ohio Coal Develop-ment Office (OCDO) within the Ohio Department of Develop-ment. Known as the Advanced Emissions Control Development Program (AECDP), the multi-year project has been directed toward demonstration of practical, cost-effective strategies for reducing HAPs emissions from coal-fired boilers using con-

ventional particulate and SO

control equipment. In evaluating mercury emissions control, the work sought to understand and maximize the performance of conventional flue gas emissions control systems, in addition to that obtained by advanced sys-tems under development by B&W.

A substantial portion of the AECDP experimental work took place in B&W/MTI’s state-of-the-art Clean Environment De-velopment Facility (CEDF). Fig. 1 provides an isometric view of the CEDF. The 100 million Btu/hr CEDF (approximately 10

equivalent) integrates combustion and post-combustion testing capabilities to facilitate the development of the next generation of power generation equipment. The furnace has been designed to yield combustion zone temperatures, flow patterns, and residence times representative of commercial boilers. Boiler convection pass and air heater simulators main-tain conditions representative of the entire boiler system. Back-end systems include both a baghouse and an electrostatic pre-cipitator for particulate control, sorbent injection systems, and

wet and dry scrubbers for SO

control.

Testing in Phase I (Facility Modification and Benchmarking) of the AECDP project sought to characterize the CEDF system in order to verify that it would produce representative combus-tion conditions and HAPs data. The concentrations of uncon-trolled trace metal emissions compared well with levels pre-dicted by correlations developed by the EPA on the basis of utility boiler data gathered in the DOE- and EPRI-supported studies. The agreement obtained indicated that CEDF results can be reliably used to predict the air toxics emissions perfor-mance of full-scale systems.[4]

Phase II (Optimization of Conventional Systems) consisted of three series of tests focused on the measurement of the quan-tity and distribution of mercury species downstream of the boiler and emissions control equipment. Reports of significant varia-tions in mercury removal for some commercial systems, coupled with the continuing development of mercury speciation mea-surement methods, drove the test program toward testing that might explain the performance variations being observed in the field. The emphasis in Series 1 included evaluation of the ana-lytical techniques for mercury speciation, the performance of a near-continuous mercury monitor and of continuous Fourier Transform Infrared (FTIR) spectrophotometry for gaseous hy-drochloric acid, and on characterization of ESP and baghouse

HAPs emission control. Series 2 examined the control of mer-cury by wet limestone scrubber systems over a range of typical operating conditions.[5] Series 3 provided data on the impact of a number of raw and cleaned Ohio coals on mercury emissions.As Phase II ended, the analytical technique for mercury specia-tion commonly known as the Ontario Hydro Method came to be recognized as reliable by a number of laboratories, leaving much of the removal efficiency variations to be explained by some-thing other than analytical uncertainty.

Phase III (Coal Comparisons and Advanced Concepts) had as its objectives the extension of the database on the expected HAPs emissions from Ohio coals and investigation of methods for improving the control of mercury emissions. Using coals selected for their lower chlorine-to-mercury (Cl/Hg) ratios per-mitted the project to test the observation that a low Cl/Hg ratio favors the formation of a greater proportion of insoluble, el-emental mercury, which in turn would result in diminished mer-cury capture.[6] The other major experimental activity in this phase explored advanced control concepts primarily for the purpose of reducing gas-phase mercury emissions. This effort focused on two major segments of the coal-fired utility market,differentiated primarily according to whether or not they em-ploy FGD, specifically:

·Unscrubbed systems that represent about 75 percent of the coal-fired utility installations, most of which consist of par-ticulate collection by ESPs, though baghouses are gaining in popularity in some areas.

·Scrubbed systems that represent the remaining 25 per-cent. The majority of these scrubbed systems consist of an ESP followed by a wet scrubber.

Mercury Speciation and Coal Characteristics

Accurate measurement of the mercury species is essential to the development of mercury control options because of the im-pact on the removal efficiency. Mercury is generally present either as elemental mercury, Hg o , or as oxidized compounds such as mercuric chloride, HgCl 2, and mercuric oxide, HgO. Indus-try experience to date suggests that Hg o and HgCl 2 are the domi-nant species in the flue gas from coal-fired boilers. The oxi-dized form of mercury is much more soluble in the aqueous solution present in FGD systems than elemental mercury and is, therefore, more likely to be removed from the flue gas. El-emental mercury tends to remain in the vapor state at the oper-ating temperature of conventional emissions control equipment.

Figure 1

Isometric view of the Clean Environment Development Facility.

A relatively higher proportion of oxidized mercury present as HgCl 2 would be expected to favor higher removal efficiency in a FGD system. Mercury measurements during most of Phase II and all of Phase III in the AECDP project were made using the EPA Method 29 for total mercury, while the Ontario Hydro Method was used for the individual elemental and oxidized forms. A comparison of the two for total mercury emissions appears in Fig. 2.

That the oxidized form of mercury can be removed in FGD

systems to a greater extent than elemental mercury is recog-nized throughout the industry and attributed to the higher solu-bility of the ionic species in the aqueous slurries of wet and spray dryer FGD systems. The lower removal efficiency ob-tained on at least some western fuels appeared to correlate with a low Cl/Hg ratio in the coal.[6] For this reason, the Ohio 5/6/7blend and Clarion 4A coals selected for Phase III were chosen for their lower Cl/Hg ratio in anticipation of generating a higher elemental mercury concentration. As can be seen from compari-son of Figs. 3 and 4, the relative amounts of oxidized and el-emental mercury, ranging from 83 to 95 percent of the ionic species, are approximately the same as had been seen in the Phase II tests with the Ohio 5/6 blend, Ohio 6A, and Meigs Creek coals. Mercury removal in the wet scrubber was like-wise unchanged from values that have been reported earlier.[5]It appears that even though the Cl/Hg ratio for these coals was about half of what it had been in the earlier tests, the fact is that the actual weight ratio is still in the range of 2,000 to 4,000. These high concentrations of chlorine may mask any ef-fect that this element has on the speciation of mercury.

Mercury Emission Control for Unscrubbed Systems

Approximately seventy-five percent of today’s U.S. coal-fired power plants are not equipped with FGD systems. Be-cause these units control only particulate emissions, the im-portance of developing cost-effective mercury capture technol-

ogy for these installations is one of the major drivers of HAPs research efforts across the nation. A substantial portion of this work is devoted to the optimization of carbon adsorption in one form or another. The ability of activated carbon to adsorb trace amounts of many chemical species is well established.The more effective materials tend to come at a high price for the methods in which they might most easily be applied for the treatment of large volumes of flue gas containing such low con-centrations of mercury. Analysis of mass transfer limitations points to the dependence of the theoretical minimum required carbon-to-mercury(C/Hg) ratio on the sorbent particle size,mercury concentration in the flue gas, and residence time.[7]

Much of what has been done mimics the use of activated carbon for municipal waste combustors (MWC). In that appli-cation however, the technology is geared toward achieving a regulated 80 μg/dscm (at 7 percent oxygen in the flue gas) or 85 percent reduction of mercury emissions. For these MWC applications where uncontrolled concentrations are on the or-der of 500 - 1,000 μg/dscm, the EPA analysis reports ranges of control costs for activated carbon injection ($211 - 870/lb Hg removed), carbon filter beds ($513 - 1,083/ lb Hg removed),and “polishing” wet scrubbers ($1,600 - 3,320/lb Hg removed).For utility boiler applications, the EPA and DOE developed corresponding cost projections based on data available at the time showing significantly higher ranges: activated carbon in-jection ($14,200 - 70,000/lb Hg removed), carbon filter beds ($32,700 - 37,800/ lb Hg removed), and spray coolers with ac-tivated carbon injection and a baghouse ($17,400 - 38,600/lb Hg removed).

[1]

Figure 2Total uncontrolled mercury emissions: EPA method

29 vs. the Ontario Hydro Method.Figure 3

Range of coal chrlorine-to-mercury mass ratios.

Figure 4Mercury speciation for coals of varying chlorine-to-mercury ratios.

Not surprisingly, research efforts have continued to seek ways of overcoming these projected high costs of dry mercury cap-ture. The studies span a wide range of potential methods. At one end of the spectrum are the possibilities of taking advan-tage of adsorption on unburned carbon in the fly ash, or of us-ing a relatively large amount of a low-cost, activated carbon derived from coal.[7] At the other end, some are exploring the use of relatively small amounts of highly active, promoted sor-bents such as iodine-impregnated carbon.[8] Still others have gone to sorbents other than carbon, recovery processes, or pro-cesses that seek to enhance the adsorptive qualities of an acti-vated carbon.[9,10,11]

B&W/MTI has explored various aspects of HAPs emission control and mercury capture outside of the AECDP project. With the perspective gained as a major supplier of competitively priced boiler and environmental equipment and processes for utilities, project personnel selected some of the more promis-ing (i.e ., more cost-effective) alternatives for mercury https://www.wendangku.net/doc/761012786.html,ing concepts for low-cost mercury capture for which patent protection is currently being sought, the Phase III tests for mercury emissions control in unscrubbed systems were designed to afford comparison of these alternatives with “standard” ac-tivated carbon injection technology. Because potential propri-etary positions are still being established, the following describes the results without full disclosure of the means and mechanisms involved.

The activated carbon selected for the base case injection test was a lignite-based sorbent commercially known as Darco FGD carbon obtained from Norit Americas. The carbon was injected at a C/Hg mass ratio of 9,000:1 into 400F flue gas upstream of the ESP that was operating at an inlet temperature of 345F.Mercury capture under these conditions was 53.7 percent over-all from an inlet concentration of 21.6 mg/dscm. This included 4.4 mg/dscm of particulate mercury, virtually all of which was captured. Under the assumption that little interconversion of species took place, oxidized mercury was reduced 46.6 per-cent from 14.6 mg/dscm to 6.8 mg/dscm, and elemental mer-cury remained effectively unabsorbed at the relatively high tem-perature condition of the test.

The results of tests using alternative dry methods of reduc-ing mercury emissions are presented in Fig. 5 where the costs of materials, normalized to unit cost for “standard” activated carbon injection, are also presented. No operational problems were encountered in the course of the tests.

Figure 5Mercury emission reduction for lower cost alterna-tives to standard activated carbon injection.Activated Carbon Alternative

A Alternative

Figure 6Effect of liquid-to-gas ratio on Mercury removal for

tray and open spray towers.

Figure 7Effect of liquid-to-gas ratio on mercury emissions

at common operating pH values.

Mercury Emission Control for Scrubbed Systems

Studies on mercury emission control by wet FGD systems have received far less attention than those directed toward dry methods of reducing mercury emissions, due in part to the thought that success in achieving the desired removal in an up-stream particulate device would minimize the need for capture in a scrubber. The work that has been done has focused on con-version of elemental mercury to the water-soluble, oxidized form. Laboratory and pilot work has shown that a carbon-based catalyst can promote oxidation of the elemental mercury, though testing was limited by low inlet elemental mercury concentra-tions and operational difficulties associated with the pressure drop created by the amount of catalyst that was being used.[12]Additional laboratory studies have concentrated on the effec-tiveness of various oxidizing agents either upstream of or in the scrubber.[13,14]

Again because of B&W/MTI’s position as a major supplier of FGD systems, test plans developed for the AECDP project tended to concentrate on those aspects that are likely to influ-ence performance most directly. During Phase II of the project,the primary independent variables for the tests were those typi-cally used for operation of commercial wet scrubbers. They in-cluded such critical variables as slurry pH, liquor-to-gas ratio (L/G), and basic tower design (tray tower vs . spray tower). Phase III results first confirmed the results of the earlier work,[5] indi-cating that both wet scrubber configuration and operational fac-tors impact mercury emissions. Figs. 6 and 7 depict typical re-sults obtained during this portion of the testing. The data show

that higher total mercury control efficiency can be achieved with wet FGD scrubbers than the 17 percent average removal ini-tially attributed to them by the EPA. These characterization tests suggest perhaps a minimum of about 50 percent as an average baseline wet FGD system mercury removal level as representa-tive of existing scrubbers, with a realization that significant dif-ferences in mercury removal efficiency have been observed.As testing progressed into Phase III, certain variations in mercury removal that had been thought of as fitting within a broad range of experimental error appeared to be more repeat-able than one would otherwise have imagined. This led to test-ing several hypothetical mechanisms considered to be respon-sible, and continued preparation of invention disclosures to es-tablish patent protection. As was the case for the unscrubbed units, Fig. 8 describes the results of three of the approaches used without full disclosure of the details.

These results, obtained with some degree of control of spe-ciation effects, indicate that it will be possible to provide im-proved mercury emission reduction over and above that obtained with conventional wet scrubbing technology. The means used are expected to be low in cost, are considered to be well within the bounds of normal power plant practices, use materials that will not require any more unusual storage and safety precau-tions than plants already practice, and will not impose any par-ticularly unusual requirements in material handling or opera-tional controls.

Summary and Conclusions

Faced with the directives presented in the Clean Air Act Amendments of 1990, the power industry, with the cooperative support of government, academia, and industrial concerns, con-tinues to pursue development of cost-effective HAPs and mer-cury emission controls. B&W/MTI has participated in this ef-fort through the AECDP project that has provided first a test facility that permits representative simulation of full-scale sys-tems in which well-controlled conditions can be maintained.Though not a primary objective of the project itself, the project also afforded an opportunity to participate in some of the de-veloping analytical aspects related to understanding the role mercury speciation plays in process chemistry.

Following overall HAPs characterization and validation stud-ies, test campaigns focused on mercury emissions control for coal-fired boilers. This effort identified alternatives to “stan-dard” carbon injection technology for the roughly 75 percent of

U.S. units that employ only particulate collection devices for flue gas emissions control. Work directed at mercury emission control for the remaining 25 percent of units with wet FGD sys-tems produced several approaches for significantly improving and controlling the mercury capture these systems already pro-vide. Extrapolating the observed effectiveness of both the wet and dry methods across the boiler population, the enhancements would be expected to result in roughly a 50 percent overall re-duction of mercury emissions – beyond that already being real-ized with existing emission control technologies.

As has been pointed out since the first discussions of HAPs emission control, capturing the relatively small amounts of mercury emissions at the very low concentrations is expected to cost far more than that for fly ash, SO 2, and NO x . Fig. 9 pro-vides a very preliminary estimate of the annual levelized cost for mercury capture as it compares with that of other controlled emissions. Although the quantities involved are several orders of magnitude lower than those of fly ash and the acid gases, the difficulty and cost of control are correspondingly higher when all are viewed on the cost per unit weight removed basis. The AECDP program offers encouraging results that there may be some lower cost alternatives to standard activated carbon injec-tion for mercury emission control for coal-fired power plants.

References

1.Mercury Study Report to Congress , U.S. Environmental Protection Agency, EPA-452/R-97-003, December 1997.

2.Study of Hazardous Air Pollutant Emissions from Elec-tric Utility Steam Generating Units – Final Report to Congress ,U.S. Environmental Protection Agency, EPA-453/R-98-004a,February 1998.

3.Electric Utility Steam Generating Unit Mercury Emis-sions Information Collection Effort , U.S. Environmental Pro-tection Agency/RTI Web Site at: https://www.wendangku.net/doc/761012786.html,/

backgroundinfo.cfm accessed 06/15/99.

Figure 8Observed mercury removal enhancement for Wet FGD systems.

Figure 9Relative annual levelized costs of various pollutant emission control technologies.

No part of this work may be published, translated or reproduced in any form or by any means, or incorporated into any information retrieval system,without the written permission of the copyright holder. Permission requests should be addressed to: Market Communications, The Babcock &Wilcox Company, P.O. Box 351, Barberton, Ohio, U.S.A. 44203-0351.Disclaimer

Although the information presented in this work is believed to be reliable, this work is published with the understanding that The Babcock & Wilcox Company and the authors are supplying general information and are not attempting to render or provide engineering or professional services.Neither The Babcock & Wilcox Company nor any of its employees make any warranty, guarantee, or representation, whether expressed or implied,with respect to the accuracy, completeness or usefulness of any information, product, process or apparatus discussed in this work; and neither The Babcock & Wilcox Company nor any of its employees shall be liable for any losses or damages with respect to or resulting from the use of, or the inability to use, any information, product, process or apparatus discussed in this work.

4. A.P. Evans, K.E. Redinger, and G.A. Farthing, “Air Toxics Benchmarking Tests on a 10 MW e Coal-fired Utility Boiler Simulator,” Proceedings of the 21st International Tech-nical Conference on Coal Utilization and Fuel Systems,Clearwater, FL (March 1996).

5. A.P. Evans, et al ., “Parametric Testing of FGD Mercury Control”, Proceedings of the POWER-GEN International ’97Conference, Dallas, TX (December 1997).

6.K. Felsvang, et al ., “Air Toxics Control by Spray Dryer Absorption Systems,” Proceedings of the Second International Conference on Managing Hazardous Air Pollutants, Washing-ton, DC (July 1993).

7.M. Rostam-Abadi, et al ., “Novel Vapor Phase Mercury Sorbents,” Proceedings of the EPRI-DOE-EPA Combined Util-ity Air Pollutant Control Symposium, Washington, DC (August 1997).

8.S. Haythornthwaite, et al ., “Demonstration of Dry Car-bon-based Sorbent Injection for Mercury Control in Utility ESPs and Baghouses,” Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollutant Control Symposium, Washington, DC (Au-gust 1997).

9.C-Y. Wu, et al ., “Novel In-situ Generated Sorbent Meth-odology and UV Irradiation for Capture of Mercury in Com-bustion Environments,” Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollutant Control Symposium, Washing-ton, DC (August 1997).

10. D.L. Roberts, R.M. Stewart, and T.E. Broderick, “Cap-turing and Recycling Part per Billion Levels of Mercury Found in Flue Gases,” Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollutant Control Symposium, Washington, DC (Au-gust 1997).

11. D.J. Helfritch, P.L. Feldman, and S.J. Pass, “A Circulat-ing Fluid Bed Fine Particulate and Mercury Control Concept,”Proceedings of the EPRI-DOE-EPA Combined Utility Air Pol-lutant Control Symposium, Washington, DC (August 1997).12.O.W. Hargrove, Jr., et al ., “Factors Affecting Control of Mercury by Wet FGD,” Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollutant Control Symposium, Washing-ton, DC (August 1997).

13. C.D. Livengood, “Improved Mercury Control in Wet Scrubbing through Modified Speciation,” Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollutant Control Sym-posium, Washington, DC (August 1997).

14.L.L. Zhao and G.T. Rochelle, “Mercury Absorption in Aqueous Hypochlorite,” Proceedings of the EPRI-DOE-EPA Combined Utility Air Pollutant Control Symposium, Washing-ton, DC (August 1997).