Print
Email
Save
By William J. Gretta, Song Wu and Hirofumi Kikkawa, Hitachi Power Systems America
In March 2005, the U.S. Environmental Protection Agency (EPA) announced two rules for air pollution that apply to coal-fired power plants: the Clean Air Interstate Rule (CAIR) and the Clean Air Mercury Rule (CAMR). CAIR applies to states in which EPA atmospheric chemistry and transport models have demonstrated that nitrogen oxides (NOX) and sulfur oxides (SOX) emissions contribute to high levels of ambient O3 and PM2.5. CAMR separately addressed the reduction of mercury (Hg) emissions from U.S. power plants. Although the CAMR was eventually vacated, mercury reductions are now being initiated on a state by state basis. Mercury emission reductions up to 90 percent are being required.
Although a number of technologies have been demonstrated to reduce mercury emissions such as activated carbon injection (ACI), it has been determined that significant Hg removal can be achieved in flue gas desulfurization FGD systems. This removal is, however, significantly affected by the amount of oxidized mercury (Hg2+) entering the FGD. It is well known that increasing the proportion of Hg2+, which is present in the form of water-soluble mercuric chloride (HgCl2), allows for high Hg emission reduction because HgCl2 can be removed in downstream equipment such as electrostatic precipitator (ESP) and FGD systems.
Increased mercury oxidation upstream of the FGD system will facilitate higher overall mercury removal for the plant. Selective catalytic reduction (SCR) catalyst has been shown to significantly increase mercury oxidation by converting elemental mercury (Hg0) to Hg2+ in coal combustion flue gases. Many U.S. plants either have SCR and FGD systems already in place, or will need to install these systems for future CAIR NOX and SO2 control requirements, respectively. Therefore, utilizing the co-benefits of this equipment for mercury removal can result in significant capital and operating cost savings as compared to other mercury control technologies such as ACI.
Ideally, the SCR catalyst should produce high mercury oxidation, without increasing SO2 oxidation, which forms SO3 that can cause air heater fouling, flue corrosion and visible stack plumes. Several downstream SO3 mitigation technologies have become commercially available in recent years, but these systems can have both high initial and operating costs and maintenance concerns. Therefore an advanced SCR catalyst, which can achieve high Hg0 oxidation with low SO2 to SO3 conversion, will be the most economical solution for bituminous coal-fired power plants.
The effect of SCR catalyst on Hg0 oxidation appears to depend on coal type. Power plants burning eastern bituminous coals, which generally contain high amounts of chlorine (Cl) and sulfur (S), tend to show relatively high Hg0 to Hg2+ conversion across the SCR catalyst compared with those burning Powder River Basin (PRB) coals because HCl can promote Hg0 oxidation. Other flue gas components such as NO, NH3, H2O and SO2, however, decrease Hg0 oxidation within the typical SCR temperature range. In particular, SO2 strongly reduces Hg0 oxidation at high temperatures (>662 F). It will therefore be difficult to achieve high Hg0 oxidation efficiency through conventional SCR catalysts under high temperature and high SO2 concentration conditions when firing high-sulfur bituminous coals.
Since 2003, Hitachi has been conducting studies utilizing their in-house Air Quality Control Systems (AQCS) pilot test facility for evaluating technologies that can be applied to meet the CAIR and mercury reduction requirements. This facility in Akitsu, Japan, includes air pollution control devices such as SCR, ESP, fabric filter (FF) and wet flue-gas desulphurization (WFGD) systems.
Hitachi has also studied the effect of SO2 concentration and Hg0 oxidation across SCR catalysts in a laboratory-scale apparatus. Hitachi has successfully developed a new type of SCR catalyst which satisfies the high Hg0 oxidation and low SO2 oxidation requirements under high temperatures (716 to 770 F). The Hg0 oxidation performance of the new SCR catalyst has been tested in a laboratory-scale apparatus and also in the pilot-scale test facility.
Scale Tests
Laboratory scale tests were conducted with simulated flue gas for evaluating Hg0 and SO2 oxidation characteristics of SCR catalysts. A schematic diagram of the test apparatus is shown in Figure 1. This apparatus consists of the SCR reactor heated to the typical SCR temperature range (662 to 752 F), the mercury generation unit, the gas preheating (752 F) and remixing sections and the online NOX and SO2 analyzers. Flue gas components such as O2, CO2, SO2, NO and N2 were supplied to the SCR reactor through a preheating furnace. Moisture and HCl, were also supplied to the preheating furnace as HCl solution using a tube pump. NH3 was injected directly upstream of the SCR catalysts. Mercury concentration was adjusted by adding a variable volume of mercury-saturated gas to the carrier gas.
 |
New SCR catalysts have been developed for bituminous and PRB coal-fired boiler applications resulting in improved mercury (Hg0) oxidation activity. This new catalyst, TRAC (triple action catalyst) can significantly enhance mercury oxidation and help to reduce or eliminate the need for additional mercury control measures such as activated carbon injection in some applications.
For PRB applications, the coal’s low halogens will significantly reduce the mercury oxidation potential of conventional PRB SCR catalyst. New catalyst has been developed, tested and applied in a commercial PRB-fired boiler with high mercury oxidation results, without the need for halogens to be added to the fuel feed or flue gas.
New SCR catalyst has also been developed for bituminous fuel applications, resulting in improved mercury (Hg0) oxidation activity while maintaining low SO2 to SO3 conversion. The performance of this new catalyst has been confirmed using laboratory-scale test apparatus and a 1 MW thermal pilot plant. Recent laboratory studies revealed that mercury oxidation rates across the SCR catalyst decreased with increasing concentrations of H2O and SO2, and the decline of mercury oxidation was especially prominent at high flue gas temperatures.
The decline of mercury oxidation may be explained by the reduction of Hg2+ to Hg0, which is accelerated by H2O and SO2, especially at high gas temperatures. Therefore it is necessary to restrain the mercury reduction reaction in order to improve the overall mercury oxidation rate of SCR catalyst. The new SCR catalyst has been designed to suppress the mercury reduction reaction with H2O and SO2 by controlling the kinetics of competing mercury reactions.
Laboratory scale tests showed that the Hg0 oxidation activity of the developed catalyst was 1.5~2.0 times higher and the SO2 to SO3 conversion activity about half of conventional catalyst. Pilot-scale test results were in good agreement with the laboratory-scale test results and confirmed that the Hg0 oxidation activity of the newly developed catalyst was 1.4~1.7 times higher than that of conventional SCR catalyst. Large test modules of this new catalyst have been installed as part of a full size SCR reactor in a power plant to determine long term catalyst performance.
Bituminous Mercury Oxidation Catalyst
The new SCR catalyst and conventional catalysts were evaluated in the pilot-scale test facility, equipped with pollution control devices including SCR, DESP (dry-ESP), wet-FGD and WESP (wet-ESP) systems. Two types of eastern bituminous coals, which contained high amounts of chlorine and sulfur, were used in the tests. A schematic diagram of the pilot test facility is in Figure 1 (page 50).
The combustor is a vertical-type furnace with a burner installed at the top. The coal combustion rate was about 120-150 kg/h (220 lb/hr) which was equivalent to roughly 1 MWt. The flue gas temperature at the inlet of the ESP’s was controlled to 320 F by utilizing a gas-gas heater and a gas cooler. Sampling was conducted simultaneously at seven points through flue gas stream: at the inlet and outlet of SCR reactor, the inlet and outlet of ESP, the outlet of the wet-FGD, and the stack (that is, the outlet of WESP). Figure 2 (page 52) indicates the sampling points.
 |
Ever since SCR was first applied in U.S. power plants firing bituminous fuels, Hitachi has made continuous efforts to improve the catalyst properties and to lower SO2 to SO3 conversion. Through these ongoing efforts, improved SCR catalyst has been applied for commercial use, and resulting SO2 conversion rates have decreased to about 20 percent of the original values when SCR was first applied. All of these reductions have taken place with increasing NOX removal requirements (≥90 percent). This low SO2 conversion, however, has also resulted in low Hg0 oxidation because there is a close correlation between Hg0 oxidation activity and SO2 conversion activity.
Changing the active composition can typically control the catalyst activity of Hg0 oxidation and NOX removal, but the SO2 to SO3 conversion activity is also dependent on the active composition of the catalyst. The Hg oxidation and DeNOX reactions take place exclusively on the surface of the catalyst, whereas SO2 conversion rate is very slow and thus may increase with total active materials. With conventional catalysts, by adding active components to increase Hg0 oxidation activity, the SO2 to SO3 conversion activity will also increase as shown in Figure 3 because Hg0 oxidation and SO2 oxidation are promoted by the same active sites in the catalyst.
 |
The fundamental reaction mechanism of Hg0 oxidation and SO2 to SO3 conversion as well as the impact of SO2 concentration and Hg0 oxidation across SCR catalysts was investigated in a laboratory-scale apparatus. This testing was done to ascertain the most appropriate catalyst composition, conformation, and manufacturing methods for the new catalyst based on the characteristics of each reaction mechanism.
The reaction of Hg0 oxidation with hydrogen chloride (HCl) and O2 can be considered a diffusion-controlled reaction of Hg0 in the catalyst’s pores because the reaction rate of Hg0 oxidation is faster than the diffusion velocity of Hg0 through the catalyst’s pores. On the other hand, the reaction rate of SO2 to SO3 conversion is slower than the diffusion velocity of SO2 through the catalyst. In other words, the reaction of SO2 conversion is controlled by the oxidation rate, which has a very close relationship with amounts of active components, that is, the number of active sites in the catalyst.
The effect of SO2 on Hg0 oxidation through SCR catalyst was subsequently studied using a conventional catalyst in the laboratory-scale test. The rate of this reverse reaction on SCR catalyst depends on the concentrations of SO2 and H2O in the flue gas, and also depends on the flue gas temperature. The overall Hg0 oxidation activity across SCR catalyst is determined by the competing reactions of Hg0 oxidation with HCl and H2O and of HgCl2 reduction with SO2 and H2O at a given gas temperature.
Pilot-Scale Test Results of Developed Catalyst
The developed catalyst and conventional catalyst were evaluated in the pilot-scale test facility using two types of high-sulfur eastern bituminous coals.
All the tests were conducted under the constant condition of 90 percent NOX removal efficiency by controlling the NH3/NOX molar ratio to 0.9. Hg0 oxidation rate of the developed catalyst was higher than that of the conventional catalyst when firing coal type A. Results from the pilot-scale test are in good agreement with the laboratory-scale test results and indicate that the Hg0 oxidation activity of the developed catalyst was 1.4~1.7 times higher than that of the conventional catalyst for these tests with high sulfur coals.
In order to ascertain the performance of the developed catalyst in an actual operating unit, testing has been carried out at two U.S. plants burning high-sulfur coal. For Plant A the reactor includes two modules (blocks) of the developed catalyst in both layers of the existing reactor. Above and below each catalyst layer is a partition plate that is used to separate the gas entering and leaving the test catalyst. This allows for a proper determination of the inlet and outlet flue composition which include measurements for NOX, SO2, SO3, NH3, Hg and HCl. The testing will identify the mercury oxidation across the catalyst along with the resulting SO2 oxidation. For Plant B, catalyst is installed in an operating SCR and catalyst plates are periodically sampled and tested at Hitachi’s laboratory to confirm long term durability.
Hitachi’s new TRAC catalyst can achieve high Hg0 oxidation with low SO2 to SO3 conversion at high temperatures for power plants burning high sulfur bituminous coals. The Hg0 oxidation activity of the developed catalyst in laboratory scale tests was 1.5~2.0 times higher than that of the conventional catalyst and the SO2 to SO3 conversion activity was about half of conventional catalyst.
Pilot-scale test results were in good agreement with the laboratory-scale test results and showed that the Hg0 oxidation activity of the developed catalyst was 1.4~1.7 times higher than that of the conventional catalyst. Test data confirm that the developed catalyst has sufficiently robust catalytic performance to be applied to high sulfur bituminous coal-fired power plant applications.
Full-scale testing has been conducted and the performance has been determined for the TRAC catalyst in an actual operating unit burning eastern bituminous coal. TRAC catalyst for PRB applications has also been developed, tested and proven successfully in a commercial plant, with excellent results.
Authors: William J. Gretta is the Director of SCR Systems Engineering, Hitachi Power Systems, Basking Ridge, NJ. Dr. Song Wu is Director of Advanced Technologies, Commercial Operations, Hitachi Power Systems, Basking Ridge, NJ. Dr. Hirofumi Kikkawa is General Manager of Environmental Research, Kure Research Laboratory, Babcock-Hitachi K.K., Japan.
