A novel method for NOx and Hg emission control in power plants using existing wet limestone scrubbers

Transcription

1 50 A novel method for NOx and Hg emission control in power plants using existing wet limestone scrubbers R. Krzyzynska Wroclaw University of Technology, Faculty of Environmental Engineering, Wroclaw, Poland Keywords: multipollutant, mercury, nitrogen oxides, limestone scrubbers Abstract - Research was performed in a simulated wet flue gas desulphurization scrubber to enhance NOx (NO and NO 2) and Hg (Hg 2+ and Hg 0 ) removals. Research showed that is a possibility to achieve a multi-pollutant (SO 2, NOx and Hg) removal capacity in a wet flue gas desulphurization scrubber fed by limestone and sodium chlorite solution. Sodium chlorite additive was effective in increasing of NOx and Hg capture. In this paper, investigation is focused mainly on NO 2 capture in the scrubber, which is the most demanding pollutant in respect to control. It was showed, that changing conditions in the wet scrubber could change the process effectiveness. Research showed also how NOx (as NO 2) absorption could be improved in a multi-pollutant scrubber without changing other pollutants (SO 2, Hg) removals capacity. INTRODUCTION Last year, (August 2, 2010) the EPA (The U.S. Environmental Protection Agency) proposed a new transport rule, CATR (Clean Air Transport Rule) to replace the CAIR (Clean Air Interstate Rule), of which, key elements were struck down by the D.C. Circuit Court of Appeals in North Carolina v. EPA, 550 F. 3d 1176 (D.C. Cir 2008) 1. According to the CATR, SO 2 emissions should be reduced by 71% from the 2005 levels by 2014, and NOx emissions should be reduced by 52% in the same period of time 1. Moreover, the United States have also indicated an intention to regulate Hg (mercury), which also comes from coal-fired electric utility boilers, because of its well-known, high toxicity 2,3. Recently (March 16, 2011), U.S. EPA, proposed Mercury and Air Toxics Standards (MATS) for power Plants to limit mercury, acid gases and other toxic pollution from power plants, keeping 91 percent of the mercury in coal from being released to the air 2. Toxic air pollutants from coal- and oil-fired power plants cause serious health impacts. Mercury can harm children's developing brains, including effects on memory, attention, language, and fine motor and visual spatial skills. Mercury and many of the other toxic pollutants also damage the environment and pollute our nation's lakes, streams, and fish. EPA estimated the value of the improvements to health alone total $59 billion to $140 billion in This means that for every dollar spent to reduce pollution from power plants, they get $5 to $13 in health benefits 4. Similar action is being taken in Europe 5,6. NOx and SO 2 emissions from coal-fired power stations are subject to tighten emission limits in the EU countries 5,6. Regulations such as the European Union Directive on Large Combustion Plants (LCP, 2001/80/EC) and the Directive on Industrial Emissions (IED) require significant emission reduction of these pollutants 5,6. Thus, because of these more stringent emission requirements, a significant increase in the use of wet-fgd and NOx control technologies are expected in the new decade. It is estimated that, in 2020 about 60% of the total coal-fired capacity will utilize some sort of wet- FGD technology 7.

2 51 One concept recently introduced is the multi-pollutant control where SO 2, NO x, and other pollutants are simultaneously controlled in a single air pollutant control device. Apparently, the multipollutant control across a wet- FGD is an option 8. This paper presents research on multi-pollutant control in wet limestone enhanced by an oxidizing agent. Research showed that there is possibility to achieve a multi-pollutant (SO 2, NOx and Hg) removal capacity in a wet flue gas desulphurization scrubber fed by limestone and sodium chlorite solution, used as an oxidizer agent. Sodium chlorite additive was effective in increasing of NOx and Hg capture. In this paper, attention is focused mainly on NO 2 capture in the scrubber, which is the most demanding pollutant in respect to control. It was shown that changing conditions in the wet scrubber could change the process effectiveness. Research showed also how NOx (as NO 2) absorption could be improved in a multi-pollutant scrubber without changing other pollutants (SO 2, Hg) removals capacity. EXPERIMENTAL METHODOLOGY AND PROCEDURES Bench-Lab Stage Experimental Set Up The performance of NaClO 2 in promoting multi-pollutant control was first investigated with bench-scale tests. The bench-scale experimental apparatus (Fig. 1), made use of a flow-through gas-liquid impinger to simulate a wet FGD scrubber. The system included a gas blending system for the makeup of synthetic flue gas, a flow-through gas-liquid impinger, and an online gas analysis system for measurement of the flue gas stream constituents. Slurry was pumped through the impinger, being maintained a constant slurry level through overflow mechanics. The scrubber was operated at 55 o C by immersing the scrubber into a constant temperature water bath. Sodium chlorite (NaClO 2) and calcium carbonate (CaCO 3) were mixed prior to being introduced to the scrubber through separate ports. The constant addition of NaClO 2 was manipulated via weight losses in the beaker. The alkali solution addition rate was monitored via the weight gain from the overflow of the scrubber. Simulated flue gas was introduced to the experimental system from gas cylinders. The gas mix was comprised of 11 % CO 2, 8% O 2, 1500 ppmv SO 2, 200 ppmv NO, 206 μg/m 3 Hg 0, and balanced with N 2. The flow of the simulated flue gas was maintained at 2 L/min (STP). Elemental mercury vapor was supplied from a VICI Metronics Dynacalibrator permeation oven, maintained at C and was transported by N 2. The high level of mercury was necessary to meet the required sensitivity of the mercury analyzer. Prior to analysis, the sampling gas passes through an ice cooler and a Nafion dryer where moisture is removed. Hg 0 vapor was measured using a continuous cold vapor atomic absorption (CVAA) analyzer (BUCK model 400A). The interference of SO 2 to mercury analyzer was determined and corrected together with the downstream SO 2 analyzer. The SO 2 species were measured using a continuous SO 2 fluorescence analyzer (model 100AH, Advanced Pollution Instrumentation, Inc.). The NO X species were measured using a continuous chemiluminescence NO X analyzer (model 200AH, Advanced Pollution Instrumentation, Inc.). Slurry was feeding simultaneously with oxidizer to the impinger until it started to overflow. A magnetic stirrer was used to promote the mixing of slurry. The agitation was maintained at a constant speed throughout all runs. Flue gas was then introduced to the impinger and the test was continued for 60 minutes. Fresh NaClO 2/CaCO 3 was added at a constant rate during the scrubbing period. Pilot-Lab Stage Experimental Set Up The pilot-scale apparatus is given in Fig. 2. The spray tower consists of three absorbers intended for scrubbing NO, Hg, and SO 2. Each absorber is 10 cm in diameter, 92 cm in length, and contains a 20 cm deep bed of plastic hollow balls (2 cm in diameter), which are supported by a grid at the bottom of each absorber and are fluidized by the upward flue gas. The system was operated in the forced oxidation mode during experiments. Simulated flue gas was generated

3 52 from a down-fired cylindrical furnace, known as an innovative furnace reactor (IFR). The fuel was introduced at the top of the furnace and combusted with air from axial and tangential directions. Since the natural gas-derived flue gas contained no SO 2 and only small amounts of NO, these components were added from gas cylinders to achieve the desired flue gas concentrations. When needed, a mercuric chloride (HgCl 2) solution (i.e., a dilute aqueous solution of HgCl 2) was used as the source of Hg 2+ and was delivered by a peristaltic pump to the middle stage of the spray tower at a constant rate (5 ml/min). Gaseous Hg 0 was produced in a permeation oven (VICI Metronics Dynacalibrator) and was carried by air into the duct before the scrubber. Unless otherwise indicated, the simulated flue gas contains approximately 7 percent O 2, 7 percent CO 2, 550 to 2000 ppm SO 2, 220 ppm NO, and 17 μg/m 3 Hg. The total flow of flue gas is controlled at 800±50 L/min. Measurements of Hg 0 were performed using ultraviolet (UV) spectrometers (Seefelder-Hg 3000) at the scrubber inlet and outlet and at the outlet of the forced oxidation tower. The NOx species were measured using a continuous chemiluminescence NOx analyzer (model 200EH, Teledyne Technologies Company). The SO 2 species were measured using a continuous Fluorescence SO 2 analyzer (model 100 AH, Advanced Pollution Instrumentation, Inc.). Figure 1. Schematic of the bench-scale wet scrubber system. 1) Hg Permeation Oven; 2) Gas Inlet; 3) Gas Outlet; 4) Filter; 5) Nafion Dryer; 6) Data Acquisition System; 7) Oxidant w. Balance; 8) Limestone Slurry w. Stir Plate; 9) Scrubber; 10) Heat and Stir Plate; 11) Hold Tank w. Balance; 12) Base Tank w. Balance; 13) Acid Tank w. Balance; 14) KiBC Bottle; 15) Silica Bottle; 16) Rotameter ; 17) Slurry Inlet

4 53 Figure 2. Schematic of the pilot-scale wet scrubber system. RESULTS AND DISCUSSION Oxidant Additives Initial oxidant screening tests were conducted with a variety of water-soluble oxidants in order to identify the most promising scrubber additive(s). These oxidants and the percent removal of each of the pollutants of interest are given in Fig. 3. Initial results showed that only potassium permanganate (KMnO 4), sodium chlorite (NaClO 2), sodium hypochlorite (NaOCl) and calcium hypochlorite (Ca(OCl)) 2 showed significant Hg removal - 100%, 95%, 100% and 56%, respectively. NaClO 2, KMnO 4 and NaOCl showed encouraging NO oxidation (62%, 34% and 14%, respectively) and NOx (as NO 2) absorption (36%, 33% and 10%, respectively). However, sodium chlorite (NaClO 2) showed the most promising results. The sodium chlorite concentration was the lowest of all oxidants tested, but its pollutant removal efficiencies were the highest. Therefore, sodium chlorite was chosen for further tests because it was the most promising scrubber additive. NaClO 2-Enhanced Wet Limestone Scrubber Because the initial scoping tests using sodium chlorite as the oxidizing additive were the most promising, a more detailed series of tests were performed. Solution of sodium chlorite was injected simultaneously with calcium carbonate slurry (CaCO 3). A range of sodium chlorite concentrations was tested. The influence of chlorite additive on pollutant removals in benchscale tests is given in Fig. 4. As it has been shown in Fig. 4 the concentration of the sodium chlorite additive in the scrubber strongly affected pollutant removal efficiencies. Bench-scale experiments showed nearly complete removals of Hg and NO and approximately 50% NOx (as NO 2) with low levels of sodium chlorite additive into the scrubbing liquor (~0.005M). An excess supply of additive did not improve NO and Hg removal but slightly improved NOx absorption (to approximately 60% at 0.025M ClO 2- ). In all tests, the SO 2 removal was at or near 100%. Initial pilot-scale results showed lower, but still promising pollutants removal (approximately 70%

5 54 Hg, 30% NO and 15% NOx) (Fig. 5). Further improvement of NOx absorption is subject of following research. Figure 3. Removal of pollutants (% at 1 h) in bench-scale tests using oxidant additives. Figure 4. Influence of chlorite additive on pollutant removals in bench-scale tests. Parameters Affecting Pollutant s Removal in the Oxidant-Enhanced Scrubber Experiments showed, that presence of SO 2 is one of the most critical parameters in respect to Hg and NO removals, what is presented in Figs 6 and 7. It was discovered that a lack of SO 2 in a flue gas negatively affected Hg and NO removals, causing a dramatic decrease in Hg and NO removal in a bench- and pilot-scale experiments. A small amount of SO 2 in the flue gas is essential to achieve complete Hg and NO removals, however, too high SO 2 concentration (>2000ppm) in flue gas caused decreasing it.

6 55 Figure 5. Influence of chlorite additive on pollutant removals in pilot-scale tests. Figure 6. Influence of SO 2 concentration in flue gas on NO/NOx removals in bench-and pilot-scale tests. The SO 2 is an abundant pollutant, and it is quite clear that lack of SO 2 in flue gas causes increasing of slurry ph. Decreasing of pollutant removals, in case when SO 2 was not present in flue gas, can be linked with increasing of slurry ph. Therefore series of experiments determining the optimum ph for satisfactory Hg 0 and NO oxidation with maximum NO X adsorption were performed. Experiments showed that the ph strongly influences the chemical mechanism in the scrubber and affects the pollutant removal (Fig. 8). Sodium chlorite is a suitable additive for the simultaneous removals of SO 2 and NO and Hg from the flue gas in a wide ph range (4-7.0) and that a tight ph control is not required.

7 56 Figure 7. Influence of SO 2 concentration in flue gas on Hg removals in bench-and pilot-scale tests. Another important parameter affecting pollutant removal is the location at which the oxidant is applied in the flue gas cleaning control profile. This parameter appears to be very important, because the chemical conditions and the flue gas composition are different in cases in which the oxidant is applied before, in or after the wet scrubber. For instance, presence or lack of some flue gas components (like SO 2) can change chemical conditions, which are very important for pollutant removal process as presented above. How the place of sodium chlorite application affects pollutants removal is described in details elsewhere 9. Figure 8. Effect of ph solution on pollutant removals in the bench- and pilot- tests

8 57 CONCLUSIONS This paper presents experiments on simultaneous removal of SO 2, NOx and Hg (both type Hg 0 and Hg 2 ) from simulated coal-fired flue gas in a wet flue gas desulphurization system (WFGD) in a bench- and pilot -scale. The multipollutant capacity of the scrubber was enhanced with the addition of an oxidizing agent. Sodium chlorite (NaClO 2) was identified as the most effective additive. It was demonstrated, that injection of NaClO 2 solution into the limestone wet scrubber can strongly enhance the removal of NO, NOx (as NO 2) and Hg (both type Hg 0 and Hg 2+ ). At the same time, injection of additive did not change conditions for effective SO 2 removal in the wet scrubber. Series of experiments were performed to improve NOx/Hg capture, understand chemical mechanism and optimize the process. Results showed that removal efficiencies of NOx, Hg and SO 2 strongly depend on ph of the slurry, flue gas composition and place of oxidant injection. It was showed that minimum amount of SO 2 in the flue gas is necessary to obtain high Hg and NO/NOx removals. However, too high SO 2 concentration (>2000ppm) in flue gas caused decreases all pollutants removal efficiencies. It was found that the acidic solution is favorable for NO and Hg removal. Increasing the ph above 7.0 decreased the NO, NOx and mercury removal. It has been discovered that the position at which the sodium chlorite is applied (before, in or after the wet scrubber) greatly influences pollutant control. The effect of sodium chlorite is related to the different chemical conditions (i.e. ph, absence/presence of some particular gases) that will occur in different locations of the flue gas cleaning system profile. ACKNOWLEDGMENT This work was supported by funding from the U.S. EPA (Office of Air and Radiation (OAR) and Office of Research and Development (ORD)) and by Polish Ministry of Science and Higher Education. The authors acknowledge the contributions of Dr. Nick D. Hutson from U.S. Environmental Protection Agency, Office of Air Quality Planning & Standards, Research Triangle Park, USA, and Dr. Yongxin Zhao of ARCADIS, an on-site contractor to EPA/ORD in Research Triangle Park, NC. Mention of trade names of commercial products and companies does not constitute endorsement or recommendation for use. REFERENCES 1. CATR (The Clean Air Transport Rule) Federal Register, Vol. 75, No. 147, Monday, August 2, 2010, Proposed Rules. 2. U.S. EPA proposed rule on mercury and air toxics standards for power plants 3. Sloss, L. L., Economics of Mercury Control, report of Clean Coal Center and International Energy Agency, accessed Directive 2001/80/WE (LCP) of the European Parliament and of the Council of 23 October 2001 on the limitation of emission of certain pollutants into air from large combustion plants. 6. The EU Industrial Emissions Directive (IED). European Parliament (press release): Stricter rules on industrial emissions (7 July 2010). 7. Srivastava, S. K.; Hutson, N. D.; Martin, G. B.; Princiotta, F.; Staudt, J. Control of Mercury Emissions from Coal-Fired Electric Utility Boilers. Environ. Sci. Technol. 2006, 41, Hutson, N.D.; Krzyzynska, R.; Srivastava, S.K.: Simultaneous Removal of SO 2, NO X, and Hg from a Simulated Coal Flue Gas using a NaClO 2-enhanced Wet Scrubber, Ind. Eng. Chem. Res., 47 (16), 5825, Krzyzynska, R.; Hutson, N.D.: The Importance of The Location of Sodium Chlorite Application in a Multi-Pollutant Flue Gas Cleaning System, J. Air Waste Manag. Assoc., 2012, 62, 707.