Mercury exists in trace amounts in fossil fuels, vegetation, crustal material, and waste products [77]. Mercury vapor can be released to the atmosphere through combustion or natural processes where it can drift for a year or more, spreading over the globe. It has been estimated that 5500 short tons of mercury were emitted globally in 1995 from both natural and anthropogenic sources, with coal-fired power plants in the United States contributing about 48 short tons, or <1% of the total [77]. The complexity of the mercury control issue is illustrated in a simple example from the DOE: If the Houston Astrodome were filled with ping-pong balls representing the quantity of flue gas emitted from coal-fired power plants in the United States each year, 30,000,000,000 ping-pong balls would be required. Mercury emissions would be represented by 30 colored ping-pong balls, and the challenge by industry is to remove 21 of the 30 colored balls (for 70% compliance) from among the 30,000,000,000 balls. Technologies under development or being demonstrated that involve the removal of mercury from flue gas include sorbent injection, particulate collection systems, catalysts, or chemical additives to promote the oxidation of elemental mercury and facilitate its capture in par-ticulate and sulfur dioxide control systems, as well as fixed structures in flue gas ducts that adsorb mercury.

Mercury in U.S. Coal

Over 40,000 fuel samples were analyzed as part of the ICR, and a summary of the ICR coal data, by point of origin for six regions and corresponding coal rank, is provided in Table 6-11 [78]. Appalachian bituminous coal and Western subbituminous coal accounted for ~75% of U.S. coal production in 1999 and over 80% of the mercury entering coal-fired power plants. The composition of these coals is quite different, which can affect their mercury emissions. Appalachian coals typically have high mercury, chlorine, and sulfur contents and low calcium content, resulting in a high percentage of oxidized mercury (i.e., Hg2+); in contrast, Western subbituminous coals typically have low concentrations of mercury, chlorine, and sulfur contents and high calcium content, resulting in a high percentage of elemental mercury (i.e., Hg°).

Emissions from Existing Control Technologies from Coal-Fired Power Plants

Estimates for mercury emissions from coal-fired power plants with various control technologies, based on the 1999 ICR data, are given in Table 6-12 [78]. These data show that mercury emissions were estimated to be ~49 short tons in 1999. This estimate is based on 84 units tested in the third phase of the ICR (out of more than 1100 units in the United States); a question of bias has been raised based on the number of samples from Eastern versus Western coal-fired boilers, so the various estimates of mercury emissions range from 40 to 52 short tons/year [78]. The ICR data indicate that the speciation of mercury exiting the stack of the boilers is primarily gas-phase oxidized (43%) or elemental (54%) mercury, with some particulate-bound (3%) mercury present [77]. Table 6-12 provides information on the influence of various existing air pollution control devices (APCDs) on mercury removal; however, mercury capture across the APCDs can vary significantly based on coal properties, fly ash properties (including unburned carbon), specific APCD configurations, and other factors [77]. Mercury removals across cold-side ESPs averaged 27%, compared to 4% for hot-side ESPs [78]. Removals for fabric filters were higher, averaging 58% due to additional gas-solid contact time for oxidation. Both wet and dry FGD systems removed 80 to 90% of the gaseous oxidized mercury, but elemental mercury was not affected. High mercury removals (i.e., 86%) in fluidized-bed combustors with fabric filters were attributed to mercury capture on high carbon content fly ash.

Pavlish et al. [78] provide an in-depth review of mercury emissions from existing control technologies, but the differences by coal rank are among the most significant findings of the ICR; specifically, units burning subbitumi-nous coal and lignite frequently demonstrate worse mercury capture. For example, removal across a cold-side ESP averaged 35% for bituminous coal compared to 10% for low-rank coal. This is further illustrated in Figure 6-24, which shows the range of removal efficiencies across cold-side ESPs for

0 0

Post a comment