Mercury removal standards are coming.Where's the technology?

Burning coal in the U.S. is a tough job that continues to get tougher. Every few years, the federal government decides that coal-fired power plants must reduce their emissions of yet another pollutant. First came limits on NOX, SO2, and particulates, which required plants to install cleaner burners, scrubbers, and baghouses or electrostatic precipitators. The latest air-quality nemesis to show up on the radar screen is mercury. But the difference this time is that regulations for removing it are coming not just from Washington but from states as well.

By Dr. Robert Peltier, PE, Editor-in-Chief of POWER

For limiting mercury emissions have to be the most insidious of all the new regulations facing power plant operators. Why? Although the dates on which the rules go into effect are certain, technologies for reliably removing mercury are barely beyond the pilot testing stage. Even worse, there is no single technology you can point to and say, "That will work in my plant." Many plant upgrades for reducing pollutant emissions involve little more than writing a check but at least you can assume that the technology will work. Not so for mercury. It's likely that you'll not only have to write a big check for an unproven technology but also incur a large additional cost to buy, install, and operate it.

Although draft federal mercury regulations won't be out until the end of the year, it already seems that most federal and state legislators will try to put your round plant into a square regulatory hole. Most environmental activists are pushing for a single, across-the-board 90% removal standard by 2007-2009, regardless of fuel type, existing equipment configuration, or plant location.

1. The fate of mercury in power plantsThe mercury life cycle in an average coal-fired power plant begins with the coal pile and ends with fly ash, gypsum, and stack gas. Each year, the U.S. utility industry puts into the air about 50 tons of mercury, representing about 40% of all human-caused mercury emissions in the U.S. Source: EPRI

For some of the 1,140 coal-fired utility boilers in the U.S., mercury removal won't be much of an additional burden, because it is already done to some extent by other gas-cleaning processes and equipment such as selective catalytic reduction (SCR) for NOX control, flue gas desulfurization (FGD) for SC>2 control, and precipitators/baghouses for particulate control. For the rest, however, mercury removal will prove much more difficult, though distinctly possible (see "What are your mercury removal options?" on page 47). Estimates of the industry's cost of meeting the 90% mercury removal standard range from $2 to $5 billion per year.

Power plants are the only industrial source currently exempt from federal clean air standards for mercury. In 1999, coal-fired plants released anywhere from 45 to 57 tons of mercury into the air, depending on what source you believe. But what is certain is that coal plants are the source of about 40% of all human-caused mercury emissions in the U.S. (Figure 1). It is estimated that the U.S. accounts for 3% to 5% of mercury emissions worldwide. However, the Edison Electric Institute says that "recent atmospheric modeling of global mercury transport and deposition suggests that more than 70% of mercury deposited in the U.S. comes from sources beyond our borders."

The U.S. Environmental Protection Agency (EPA) has determined that mercury emissions from power plants must be reduced, and the U.S. Department of Energy (DOE) and the Electric Power Research Institute (EPRI) are evaluating a number of technologies to achieve the desired reductions. The EPA estimates that using technologies intended to remove mercury alone—such as activated carbon injection— could cost as much as $40 million per ton of mercury removed. The reason is that we are talking about microscopic quantities of mercury. One industry pundit likened the difficulty of removing mercury from the stack gas of a power plant to that of finding 30 black ping-pong balls in the Astrodome filled to the brim with white balls.

Legislative debates
Although mercury emissions from coal-fired plants are currently unregulated, the EPA announced in December 2000 that it will propose emission regulations for mercury in December 2003 and issue final regulations in December 2004. While the paperwork shuffle goes on, the EPA has tipped its hand by publicly discussing reduction rates of 50% to 70% by 2005 and 90% by 2010. President Bush's Clear Skies Initiative proposes that mercury emissions be reduced 45% by 2008 and 69% by 2018. An alternative proposal sponsored by Sen. James Jeffords (Ind-Vt.) seeks a 90% reduction by 2007.

When the EPA proposed Clear Skies last year, the intent was to achieve the first phase of mercury reductions by providing "co-benefits" to plants that had already installed scrubbers and/or other pollution-control equipment or technology. But the industry now considers that goal overly optimistic. "We have urged the 2010 target to reflect something that can be achieved through co-benefits," an EEI spokesman said. However, the industry has yet to determine what level of reductions it can achieve through them. In addition, the Clear Skies bill's requirement that utilities cut mercury emissions from 48 tons, measured in 2000, down to 26 tons in 2010 might lead to "fuel switching" away from coal or other measures, the spokesman said. To complicate the situation further, the EPA has pointed out that, if Congress passes Clear Skies, the mercury standard it is currently working on (see "Where do mercury regulations come from?" on page 48) would become "irrelevant." The EPA rule-making would require electric utilities to install maximum achievable control technology (MACT) beginning in 2007. But the administration believes that the Clear Skies legislation would allow utilities to meet a comparable goal sooner, without the litigation that EPA rule-making inevitably produces. "One way or another, we have to regulate mercury emissions, either through MACT or Clear Skies," said an EEI spokesman. "Neither one is easy."

States weigh in
While the dance continues in D.C., several states have begun developing their own mercury regulations, and a few specify removal efficiency requirements. Three environmental groups and PSEG Power Connecticut, owner of the 375-MW Bridgeport Harbor coal-fired power plant, recently recommended to the Connecticut General Assembly that it legislate stringent new mercury emission standards for the state's coal-fired power plants.

PSEG Power Connecticut and the environmental groups—Clean Water Action, the Connecticut Coalition for Clean Air, and the Clean Air Task Force—said the proposed legislation would set a national precedent. It would require coal-fired power plants in the state to achieve either an emissions standard of 0.6 pounds of mercury per trillion Btus or 90% mercury removal efficiency. The new requirements would become effective in July 2008.

Martha Keating, an air toxics scientist with the Clean Air Task Force, said the proposed mercury standard "is the first in the nation to set stringent mercury emissions limits within a tight timeframe." Keating, who served for 10 years at the EPA and authored the agency's landmark 1997 report to Congress on mercury, added, "the example set by the Connecticut Legislature, the state's environmental community, and PSEG Power Connecticut sends a clear message to other states, the federal government, and the energy industry that reducing mercury emissions is necessary and achievable."

Just up the road, the financially beleaguered PG&E National Energy Group (NEG) took another blow an environmental one when Massachusetts Governor Mitt Romney announced in February that the state would not grant the company the extra two years it says it needs to meet strict state emissions standards for its 714-MW Salem Harbor plant. In 2001, Massachusetts imposed new emissions rules on Salem Harbor and five other aging power plants in the state with a total capacity of 4,500 MW. They call for the plants to reduce their NOX emissions by 50%, their SOx by as much as 74%, and their CO2 by 10%.

What's more, the rules also require that plants begin stack testing for mercury in preparation for mercury emissions standards that go into force on October 1, 2006. The other plants subject to the requirements are Sithe Energies' Mystic Station, NRG Energy's Montaup Station, NEG's Brayton Point, Northeast Utilities' Mount Tom Station, and Mirant's Canal Electric. Owners cannot use emissions averaging between plants to meet the standards, but they can choose between repowering and installing new technologies. NEG said it is willing to make the necessary $80 million in improvements but needs until 2006, the state's original deadline. However, Romney says he'll give the company only until October 2004.

Complicated chemistry
If you think mercury legislation is complex, it's nothing compared to mercury removal technology. Mercury removal efficiency is highly dependent on a large number of site-specific details, such as a plant's fuel type and constituents, the type of burners it uses, its boiler operating conditions, and the designs of its particulate collection and flue gas treatment system, to name just a few. A key driver is the form of the mercury (Hg) in the flue gas. Physically, mercury may take the form of a gas or a liquid, or it may be associated with solid particles. Chemically, mercury can exist in three oxidation states: elemental mercury, mercurous or monovalent mercury, and divalent mercury. If that isn't complicated enough, mercury also bonds with other chemicals to form inorganic compounds (such as HgCi2--mercuric chloride) and organic compounds (such as methyl mercury).

The species of mercury found at a plant is critical to efforts to remove it, because various control technologies are more effective at removing some species than others. For example, the mercury removal efficiencies of dry processes such as powdered activated carbon (PAC) injection depend strongly on temperature, residence time, and the particle size and chemical characteristics of the carbon. There is also a confounding impact from sulfur, chloride, and other elements that can bind to mercury and block its adsorption by the carbon, dramatically affecting total mercury removal rates. Furthermore, flue gas parameters such as temperature, volume, and composition also vary considerably among plants, and they may eliminate certain technologies from consideration. Fly ash may oxidize elemental mercury to ionic mercury, which a scrubber can readily remove, but the degree of oxidation seems to vary with both the characteristics of the fly ash and contact time.

There seems to be no limit to the number of variables. For example, plants burning the same coal but from different seams could achieve very different mercury removal rates as a result of slight differences in the coal's chlorine and ash content, effects on operation of the combustion system from unburned carbon in the ash, and temperature and residence time in the particulate control device. All affect mercury speciation in the flue gas and the amount of mercury adsorbed on particulate matter in the stack. The effect on both processes has been seen both in FGD and SCR systems.

2. Mercury removal by chemical additionBabcock & Wilcox is developing mercury control technology for coal-fired plants equipped with wet flue gas desulfurization systems with the goal of removing 90% of the mercury in the stack gas for one-quarter to one-half of what today's commercially available powdered activated carbon mercury-removal technologies cost. Courtesy: B&W

Focusing on FGD
Despite the mysteries of mercury removal, equipment vendors are taking a shot at coming up with effective technologies for doing it. Babcock & Wilcox (B&W), Barberton, Ohio, is approaching the problem by narrowing its focus to coal-fired plants equipped with wet FGD (Figure 2). The company is shooting for 90% total mercury removal at a cost of between one-quarter and one-half of today's commercially available activated carbon mercury removal technologies. B&W's preoccupation with wet FGD-equipped plants makes good business sense when you look at the numbers: 85% of all FGD systems are wet, and the 220 plants where they are installed account for about 25% of U.S. generating capacity.

B&W and McDermott Technology Inc. (MTI, formerly the B&W Research and Development Division) are demonstrating a wet-scrubbing mercury removal technology (using very small amounts of a liquid reagent to achieve increased mercury removal) at two plants burning high-sulfur Ohio bituminous coal. B&W/MTI's enhanced mercury removal process adds very small amounts of a proprietary reagent to an existing wet FGD system to increase its mercury removal efficiency.

The big advantages of this approach: The removal equipment is already in place, the approach requires little in the way of additional capital expenditures, and the reagent is significantly less expensive than those used by competing technologies. One pilot plant is Michigan South Central Power Agency's (MSCPA) 55-MW Endicott Station in Litchfield; the other is Cinergy Corp.'s 1,300-MW Zimmer Station near Cincinnati. Already, the Endicott and Zimmer pilots have demonstrated the significant impact of coal size and scrubber chemistry on mercury removal efficiency.

At Endicott, tests showed a consistently high oxidized mercury removal efficiency of 95%, which, considering the elemental mercury that slips through, equates to an overall removal efficiency of 77%. The test results at Zimmer were a bit disappointing. Scientists expected that the reagent would be equally effective in the plant's lime-based FGD system, but that was not the case. Overall, the mercury capture by Zimmer’s FGD system averaged 52%.

Why such a large difference? One reason is that the variability in the mercury content of the coal fired during the four months of operation at Endicott was significant, ranging from 8 to 32 lb/1012 Btu. But it was not completely unexpected, because the plant typically fires up to four different Ohio coals in varying percentages, depending on spot market availability. Despite the wide variation in the coal's mercury content, some of the variation disappears by the time the mercury enters the wet FGD system. Some homogenizing also probably occurs in the turbulent exhaust gas, but it is also possible that mercury is rejected by the pyrite in the pulverizers.

The economics of a wet limestone, forced-oxidization FGD system enhanced for mercury capture indicate that, for a 500-MW unit burning 3% sulfur and 0.23 ppm Hg coal, the costs for 70% removal are $3 million (capital) and around $0.18 million/kWh (operational). The additive costs about $0.21/lb. B&W claims an identically sized powdered activated carbon system approaches $0.85 million/kWh in operating costs.

The addition of the B&W/MTI enhanced mercury capture process to a wet FGD system has the additional benefit of virtually no impact on scrubber operation and gypsum quality. Moreover, it does not adversely affect the acceptability of fly ash for disposal or sale.

ECO's three-step process
Another company seeking to solve the mysteries of mercury removal is Powerspan Corp., New Nashua, N.H. Powerspan is building a 50-MW plant to demonstrate its multi-pollutant ECO (Electro-Catalytic Oxidation) technology (Figure 3) at FirstEnergy Corp.'s R.E. Burger Plant near Shadyside, Ohio. The unit, which is scheduled to be operational by the end of the year, is being installed as a slipstream on a 156-MW boiler and is designed to process approximately one-third of the boiler's flue gas. The project is estimated to cost $18 million, to which the Ohio Coal Development Office is contributing $4.5 million. FirstEnergy and Powerspan are paying for the remainder.

Meanwhile, Powerspan continues to operate a 1-MW equivalent pilot test unit at the same site (Figure 4). The pilot was first constructed in 1998 and modified in February 2002 to incorporate an ammonia scrubber and associated liquid handling equipment. It treats 1,500 to 3,000 scfm of flue gas drawn from the Burger Plant's Unit 4 or 5 boilers (upstream of the plant's electrostatic precipitator) and has demonstrated 80% to 90% mercury removal when operating on eastern bituminous coals.

Powerspan anticipates that, in commercial operation, ECO would be installed downstream of a plant's existing electrostatic precipitator (ESP) or fabric filter. The technology treats flue gas in a three-step process:

Liquid effluent produced by the ammonia scrubber is sent to a co-product recovery sys-tem, which includes filtration to remove ash and activated carbon adsorption for mercury removal. The treated co-product stream, free of mercury and ash, can be processed to form ammonium sulfate nitrate fertilizer.

In June 2001 the DOE awarded Powerspan $2.8 million to study the mercury removal capabilities of the ECO process, one of six processes selected as promising by the DOE's National Energy Technology Laboratory (NETL). The objective of the project is to optimize the mercury removal capability of the ECO process while maintaining its ability to remove other pollutants such as NOX , SO2, and PM2.5. As part of the effort, the mercury removed from the flue gas is isolated for disposal, to avoid any impact on the usability of the co-product.

Although there have been significant challenges during operation of the flue gas mercury monitoring equipment, tests at the pilot have consistently given ECO high marks for mercury removal. Levels have been verified using two methods: near real-time measurement using mercury continuous emission monitoring systems operated in semi continuous mode and single-day testing by an independent stack testing company using the "Ontario Hydro" method. The latter pegged average mercury removal efficiency at 88%.

The mercury concentration in the liquid effluent from the pilot has been successfully reduced to below minimum detectable limits in the fertilizer co-product: from 200 ppb to less than 20 ppb. Fertilizer crystals processed from the liquid effluent have had no detectable mercury.
The cost of removing mercury from the liquid effluent has been estimated at $733 per pound. Additional testing is now being conducted at the pilot on higher concentrations of elemental mercury to understand its impact on total mercury removal.

3. Electro-Catalytic OxidationThe Electro-Catalytic Oxidation (ECO) technique removes mercury from flue gas via a three-step process: a dielectric barrier discharge reactor oxidizes gaseous pollutants; ammonia scrubbing removes unconverted SO2 and NOX produced in the reactor; and a wet electrostatic precipitator captures acid aerosols, fine particulate matter, and oxidized mercury. Courtesy: Powerspan

4. ECO pilot testPowerspan constructed a 1 -MW-equivalent pilot at FirstEnergy's R.E. Burger Plant. The ECO technology has demonstrated 80% to 90% mercury removal during pilot testing when burning eastern bituminous coals. Courtesy: Powerspan

Sorbents yield good results
At about the same time as B&W was running tests at the Endicott and Zimmer stations, ADA-Environmental Solutions (ADA-ES), Littleton, Colo., began running full-scale (150-MW equivalent) PAC injection tests at four plant sites: Alabama Power Co.'s E.G. Gaston Station, Wisconsin Electric Co.'s (now We Energies') Pleasant Prairie Power Plant, and PG&E NEC's Brayton Point and Salem Harbor plants. Under a DOE NETL cooperative agreement, ADA-ES is working in partnership with a number of power generators and equipment vendors on a field evaluation program of injecting various sorbents upstream of existing particulate control devices. The objective: determine the type and quantity of sorbents required for a given level of mercury removal and estimate the system's capital and operating costs.

Dry injection of a sorbent such as PAC into the flue gas upstream of the particle control equipment is probably the most mature approach for controlling mercury emissions from coal-fired plants. The gas-phase mercury in the flue gas contacts the sorbent and attaches to its surface. Existing particle control equipment, either an ESP or a fabric filter, collects the sorbent, to which mercury as well as fly ash is attached.

The type of particulate control equipment is a key parameter. It both defines the amount of sorbent that is required and sets an upper limit on the amount of mercury that can be removed. In an ESP, the carbon is collected on plates parallel to the flow of gas. Although the residence time in the ESP can be several seconds, there is a limited degree of contact between the gas and the collected particles, because the gas can be as far as four inches from the plates. By comparison, fabric filters provide an optimal opportunity for gas-sorbent interaction, because the sorbent is collected on the filter. Some sites with fabric filters may achieve higher levels of mercury removal at lower sorbent usage rates.

Following are synopses of results to date bituminous coal and uses a Compact Hybrid Particulate Collector (COHPAC) baghouse to collect resulting carbon and fly ash. The evaluation was conducted on one-half of the gas stream to the hot-side ESP followed by a COHPAC fabric filter installed into the casing of an abandoned cold-side ESP. Depending on the operating condition of the hot-side ESP, nominally 97% to more than 99% of the fly ash is collected in the ESP; the remainder is collected by the baghouse.

Soon after starting the injection of the PAC, outlet mercury levels began to drop and continued to drop over the next six hours as the PAC formed a buildup on the bags. In fact, the outlet mercury remained low even after the PAC injection was stopped, indicating that the PAC on the bags was continuing to work. It took about six to eight hours for the outlet mercury to return to baseline levels even after cleaning the bags several times. Mercury removal measured in short-term tests increased nearly linearly with an injection rate of up to 2 Ibs/million acf and then leveled off at about 90% removal. Long-term testing is scheduled to continue through 2004.

5. Schematic of sorbent injectionSorbent injection testing began in the spring of 2001 at Alabama Power's E.G. Gaston Station. Sorbents were injected downstream of the hot-side ESP and before the COH¬PAC fabric filter. Mercury removal rates of 90% were measured. Courtesy: ADA-Environmental Solutions

We Energies' Pleasant Prairie Power Plant (PPPP), located near Kenosha, Wise., was the site of the second round of tests during the fall of 2001 (Figures 6 and 7). Tests were conducted on one-quarter of 600-MW Unit 2, which fires a variety of Powder River Basin (PRB) coals and uses an ESP to collect the carbon and flyash. The primary particulate control equipment consists of Research-Cottrell weighted wire, cold-side ESPs with sulfur trioxide (SOs) flue gas conditioning. Sorbent for mercury control was injected into the ductwork downstream of the SOs injection grid and had about three-fourths of a second residence time in the duct before entering the ESP. A spray cooling system upstream of sorbent injection adjusts flue gas temperature.

6. Silo that housed sorbentDuring testing at the Pleasant Prairie Power Plant in 2001, mercury removal increased nearly linearly with sorbent injection but leveled off at about 73%. Courtesy: ADA-Environmental Solutions

7. Reagent routeFlexible hoses carried the reagent from the feeders to distribution manifolds located on the ESP inlet duct, feeding the injection probes at Pleasant Prairie. Each manifold supplied up to six injectors. Long-term testing is scheduled to continue through 2006. Courtesy: ADA-Environmental Solutions

During baseline and parametric tests, boiler load was held steady at full-load conditions. Once again, the outlet mercury concentration began to drop almost immediately after the start of injection and continued to fall slightly over the next several hours. In contrast to Gaston's baghouse test, in which mercury continued to be captured after injection was halted, mercury capture in the ESP disappeared almost immediately after PAC injection was stopped. This indicated that, for this test’ condition, most of the mercury is captured "in flight" with little additional capture by the carbon collecting on the plates.

There was a rapid increase in mercury removal with PAC injection up to an injection concentration of about 5 Ibs/million acf. Increasing the sorbent injection rate from 5 to 10 Ibs/million acf produced a 10% increase in mercury removal. No significant additional removal was observed when the rate of sorbent injection was raised above 10 Ibs/million acf and the mercury removal rate was frozen at about 73%. There was also no impact of either SOs injection or spray cooling on mercury removal.

The estimated uninstalled cost for a sorbent injection system and storage silo for 612-MW Unit 2 is $720,000 + 30%. Sorbent costs for 60% to 70% mercury removal were estimated based on a long-term PAC injection concentration of 10 Ibs/million acf. For PPPP Unit 2, this would require a nominal injection rate of 1,400 Ibs/hr. Assuming a unit capacity factor of 80% and a delivered cost for PAC of $0.50/lb, the annual sorbent cost for injecting PAC into the existing ESP would be about $5,000,000. PAC costs for 50% control at an injection concentration of 1 Ib/million acf would be about $600,000.

The cream-colored ash from PPPP is sold as a cement powder substitute in concrete and is considered a valuable by-product. Unfortunately, the PAC injection darkened the color of the fly ash, making it less marketable, and the fly ash failed a foam index test (related to freeze thaw specs) meaning the plant would not only lose revenue from ash sales but would also have to pay landfill fees for ash disposal. For any plant currently selling its fly ash for use in concrete, this mercury removal technique would incur both lost ash revenues and landfill fees. Although these costs vary from site to site, for PPPP Unit 2 they are estimated at around $5 million/yr.

PG&E NEC's Brayton Point Station was the site of the third test completed during the summer of 2002. It was conducted on Unit 1, a 245-MW tangential boiler firing West Virginia low-sulfur bituminous coal. This unit has an unusual configuration of two ESPs in series and an Epricon SO^ system.
During the test, PAC was injected between the first and the second ESP. The average gas temperature at this location was 320F, with variations from 280 to 340F across the duct. Baseline testing showed most of the native mercury capture occurred in the first ESP and in the fly ash. PAC injection was then added at the entrance to the second ESP.

As with the other plants, mercury removal increased as the injection rate increased. However, unlike at PPPP, there was no cap on the amount of mercury that could be removed. When the injection rate was increased from 10 Ib/million acf to 20 Ib/million acf, mercury removal increased from 70% to 90%. It is believed that the higher removal levels are due to the presence of as much as 150 ppm of HC1. The PRB coal burned at PPPP has HC1 on the order of only 1 ppm.

PG&E's NEC Salem Harbor was the site of the fourth test, conducted in the fall of 2002. Salem Harbor fires low-sulfur bituminous coals and uses an ESP and a selective non-catalytic reduction (SNCR) system for NOX control. Baseline tests showed mercury removal was already at 90% without adding any PAC. Work was directed at trying to understand the mechanisms for the high removal rates. Initial results indicate that very high (30%) carbon concentrations in the ash and relatively low (<300F) operating temperatures seem to be the cause.

MerCAPtures attention
Another potential option for mercury removal could be an EPRI-patented method in the first stages of development called Mercury Capture by Adsorption Process (MerCAP). Apogee Scientific Inc., Englewood, Colo., has been funded by EPRI to design, fabricate, and operate MerCAP probes at several power plants. Early results of pilot-scale testing conducted over the past year indicate that the technique may be an effective alternative to sorbent injection. Results at one site showed that the process initially removed 90% of the mercury from flue gas and after six months was still removing 40%. Laboratory tests and field tests at several power plants are continuing.

MerCAP relies on the use of fixed structures formed of or coated with materials capable of adsorbing mercury. Gold, silver, and zinc are possible materials, because each forms an amalgam with mercury. EPRI envisions the structures being plates or banks of tubes placed in the ductwork downstream of a plant's air heater, where temperatures are normally less than 400F. Mercury is adsorbed as flue gas passes over the metal-coated plates. When their surfaces become saturated with mercury, the mercury can be removed by heating, and the plates can be reused.

The big advantage of this technique is that "MerCAP requires no reagent and produces virtually no waste streams," says EPRI's Dr. Ramsey Chang. "If successful, MerCAP would be most applicable to units firing coal that produce flue gas dominated by elemental mercury, since this species is difficult to remove by wet or dry SC>2 scrubbers." Tests to date of MerCAP have been conducted in the laboratory and on pilot-scale slipstreams at seven power plants burning lignite, PRB coal, and bituminous coals.

Mercury removal effectiveness appears to track theoretical predictions fairly well. In a recent test with a 10-ft-long MerCAP probe at the outlet of a spray dryer fabric filter, 92% mercury removal was achieved at a velocity of 50 ft/sec. After 4,580 in-service hours, the probe was still removing 40% of incoming mercury. A MerCAP probe installed at another plant site in the scrubber bypass flue gas did even better; its mercury removal rate was measured at 90%. MerCAP with gold-coated plates tested in non-scrubbed Western coal flue gas had low mercury removals (<20%); the causes remain under investigation.

Dry scrub fails
The purpose of the EPRI tests, conducted at Great River Energy's Stanton Station in North Dakota, was to evaluate mercury removal at sites that burn low-rank fuels. The tests confirmed that one of the most difficult applications for mercury control will be plants that burn PRB coal and use a spray dryer absorber (SDA) to capture sulfur dioxide, in spite of the lower temperature of the fabric filter associated with the spray dryer units. For example, at a carbon federate of 3 to 4 Ib/million acf, capture of mercury was reduced from 90% to 50% by the presence of a spray dryer.

The tests also confirmed that the injection of iodated carbon could produce excel-lent mercury control under spray dryer conditions. Mercury removal levels exceeding 90% were obtained at injection rates of 1 Ib/million acf and above. Unfortunately, the price of iodated carbon (approximately S5/lb) is 10 times that of standard powdered activated carbon, making it cost-prohibitive. However, the tests also confirmed that it is possible to modify PAC to perform in flue gas without HC1.

Once the tests confirmed the impact of HC1 on mercury removal, EPRI funded a series of further tests to evaluate the effectiveness of artificially increasing the HC1 content. Different chloride compounds were injected into the boiler, raising the level of HC1 in the gas stream and the levels of oxidized mercury. However, these tests had to be halted after a few hours because of corrosion in the boiler and plugging of the air preheater.

R&D continues
Promising mercury removal techniques are now enjoying the support of DOE's $2-bil-lion Clean Coal Power initiative. This January, the department announced that it would contribute to the funding of three projects aimed at reducing emissions of multiple pollutants from coal-fired power plants:

At the 150-MW Ray D. Nixon Power Plant, the City of Colorado Springs is teaming with Foster Wheeler Power Group Inc. (Clinton, NJ.) to install a circulating fluidized bed combustor with a fully integrated emissions-control technology. The DOE will fund $30 million of the $301-million project, and its share will be used to demonstrate whether the technology is capable of reducing mercury levels by 90% and SOx emissions by up to 98%. The DOE contributed to the funding of a power plant with a similar design that began operation for the Jacksonville Electric Authority in October 2002 and that was the 2002 POWER Magazine Plant of the Year. I At a 524-MW unit of its Ghent Generating Station in Carrollton, Ky., LG&E Energy Corp. of Louisville will be installing an advanced air pollution control system based on a new technology called the "airborne process." It is believed to be capable of removing 99.5% of the plant's SO2 emissions, 90% of its NOX discharges, and 90% of the mercury in its coal. The DOE plans to pay $31 million of the project's $120 million cost. Any waste generated by the plant will be turned into fertilizer, according to the DOE. At its Presque Isle Power Plant near Marquette, Mich., We Energies will demonstrate integrated mercury and particulate matter emissions control system on three units. The demonstration project, named Toxecon, will also investigate the ability of the proposed system to control SO2 and NOX emissions. The project's goal is to confirm EPRI claims that its patented Toxecon process can remove 90% of the mercury from coal-fired plants, including those firing PRB coals, while recovering at least 90% of the mercury captured in the ash. Toxecon will make use of only one baghouse for three small boilers, increasing the cost-effectiveness of the integrated system. The DOE will subsidize half of the project's $50-million cost. It said the project "represents the best low-cost option for control of greater than 80% of mercury from coal-fired plants."

above article was published in POWER Magazine in its May2003 edition