Energy & Fuels 2000, 14, 597-602
597
The Effect of Sulfur Dioxide on the Formation of Molecular Chlorine during Co-Combustion of Fuels Ying Xie, Wei Xie, Kunlei Liu, Laura Dicken, Wei-Ping Pan,* and John T. Riley Combustion Laboratory, Department of Chemistry, Western Kentucky University, Bowling Green, Kentucky 42101 Received August 9, 1999
This project was designed to evaluate the combustion performance of and emissions from a fluidized bed combustor during the combustion of mixtures of high sulfur and/or high chlorine coals and municipal solid waste (MSW). The effect of sulfur dioxide on the formation of molecular chlorine during co-combustion of fuels was examined in this study. Sulfur dioxide was shown to be an effective inhibitor for the formation of molecular chlorine through the Deacon Reaction and, subsequently, the formation of chlorinated organics. Theoretically, co-firing high sulfur coals with MSW will decrease the possibility of polychlorodibenzodioxin/furan (PCDD/F) formation during the combustion process. A mixture of coal and PVC pellets was burned in a 0.1 MWth bench-scale fluidized bed system at WKU and no detectable amounts of chlorinated organics were found in the flue gas and bed ash. The results from this study indicated the practical effects of using coal as a combustion support fuel when burning MSW.
Introduction The amount of municipal solid waste produced in the United States each year continues to rise. The total MSW produced in 1993 rose to 187.5 million tons (equivalent to nearly 2.0 kg per day per person) from 179 million tons in 1988, and will reach approximately 200 million tons in the year 2000.1-3 However, the rapidly declining availability of sanitary landfills has forced most municipalities to evaluate alternative waste management technologies in order to reduce the volume of waste sent to landfills. Waste-to-energy technologies are receiving more and more attention as landfill costs and environmental concerns rise. Incineration of MSW, or refuse-derived fuels (RDF) processed from MSW and consisting of combustible material, is one of the alternative waste management strategies to replace landfilling. Such waste-to-energy technology has already displayed a few advantages over conventional methods. In 1993, nearly 30 metric tons (Mt) of MSW generated in the United States was burned in 151 plants, of which 125 were operated as waste-toenergy plants.4 However, extra care needs to be taken in burning RDF and the operating conditions need to be optimized so that combustion can take place in an environmentally acceptable manner. The development of MSW combustors has slowed significantly in recent years, resulting from apprehension over possible emis* Corresponding author. Fax: (270)745-5361. Email:
[email protected]. (1) Ekmann, J. M.; Smouse, S. M.; Winslow, J. C. Co-firing of Coal and Waste, IEA Coal Research, IEACR/90, 1996. (2) Steuteville, R. What is New in the Waste Stream? Biorecycle 1992, 33 (10), 10. (3) Dichristina, M. How We Can Win the War Against Garbage. Popular Sci. 1990, 237 (10), 57. (4) Strong Waste-to-Energy Growth Predicted. Air Waste 1994, 44 (2), p 122-123.
sions of hazardous chlorinated organics, especially the harmful polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs). The emission of PCDD/Fs from incineration processes were first reported in 1977 by Olie and co-workers.5 A number of measurements and evaluations have been carried out in a variety of combustion systems since then. Trace quantities of PCDD/Fs have been detected in many combustion systems involving the burning of organic materials, such as MSW, paper industry wastes, motor vehicle exhausts, etc.6 Fluidized bed combustion (FBC) technology is characterized by intense mixtures between solids and gases during combustion, as well as long residence times for fuel particles in the high-temperature zone, resulting in very little active carbon being produced during the combustion process. Therefore, the possibility of the occurrence of a de novo synthesis is minimum. As indicated in the study by Dickson,7 in the presence of sufficient phenol and chlorine sources, fly ash catalyzed the formation of PCDDs using chlorophenol as precursors. Laboratory evidence demonstrates that transition metal ions of Cu and Fe are capable of catalyzing PCDD/F formation reactions, which take place on the fly ash particles. The small organic molecules can be adsorbed onto the fly ash and subsequently converted to PCDD/Fs.8 The total synthesis involves the Deacon Reaction and is represented by the following steps: (5) Olie, K.; Vermeulen, P. L.; Hutzinger, O. Chemosphere 1997, 6, 455. (6) Buckland, S. J.; Hannah, D. J.; Taucher, J. A.; Allison, R. W. Organohalogen Compd. 1990, 3, 219. (7) Dickson, L. C.; Lenoir, D.; Hutzinger, O. Chemosphere 1989, 19, 77. (8) Born, J. G. P.; Louw, R.; Mulder, P. Organohalogen Compd. 1990, 3, 31.
10.1021/ef990173d CCC: $19.00 © 2000 American Chemical Society Published on Web 03/22/2000
598
Energy & Fuels, Vol. 14, No. 3, 2000
Xie et al.
PVC (in MSW) f HCl (thermal decomposition) (1) 4HCl + O2 f 2H2O + 2Cl2 (Deacon Reaction) C6H5OH (in fuels) + Cl2 f chlorophenols
(2) (3)
condensation of chlorophenols f PCDDs and PCDFs (4) Molecular chlorine produced from the Deacon Reaction is considered as a key intermediate step in the above mechanism. The occurrence of the Deacon Reaction in the MSW incineration process can be supported by numerous studies regarding the effect of oxygen on PCDD/Fs production.9-11 Coal as a co-firing energy source for municipal solid wastes, is able to suppress the formation of chlorinecontaining organic compounds. Scheidle and co-workers demonstrated that adding lignite coal as an auxiliary fuel to paper recycling residues decreased the levels of dioxins in fluidized-bed incinerator emissions.12 Similar results can be inferred from the studies13-16 which showed that co-firing MSW with 60% coal drastically reduced the formation of PCDD/Fs. Ohlsson and coworkers observed that despite the enhancing level of HCl due to the addition of RDF, no PCDD/Fs have been detected when co-firing high sulfur coal and RDF pellets.17 Frankenhaeuser and others also reported the adverse effects of SO2 in the formation of organic chlorides during the co-combustion of plastics with coal.18 On the basis of thermodynamic evaluation and published test data, Griffin proposed that as long as the Cl/S ratio is high, chlorine formation for the elevated production of chlorinated aromatics and PCDD/Fs is prevalent, but in the presence of substantial amounts of sulfur, chlorine production and consequently PCDD/ Fs formation is suppressed.19 Co-combustion seems to have the dual advantage of being a source of energy and having the potential of reducing the formation of chlorinated species in combustor emissions. Several different mechanisms are proposed for the inhibition of PCDD production involving the sulfur species, one of which suggests that in coal combustion the role of sulfur interference with the chlorination step (and hence the formation of PCDDs) is critical. When sulfur is present in excess over chlorine, in any system, the forward reaction predominates: (9) Vogg, H.; Metzger, M.; Stieglitz, L. Waste Manage. 1987, 5, 285. (10) Sakai, S.; Hiraoka, M.; Takeda, M.; Nie, P.; Ito, T. Formation and Degradation of PCDDs/PCDFs in a Laboratory Scale Incineration Plant; Presented at The Kyoto Conference on Dioxins, Problem of MSW Incineration, 1991; Kyoto International Community House, 1991. (11) Gullet, B. K.; Bruce, K. R.; Beach, L. O. Chemosphere 1990, 20 (10-12), 1945. (12) Scheidle, K.; Wurst, F.; Kuna, R. P. Chemosphere 1986, 17, 2089. (13) Gullett, B. K.; Bruce, K. R.; Beach, L. O. Waste Manage. Res. 1992, 8, 203. (14) Grochowalski, A.; Wybraniec, S. Chem. Anal. (Warsaw) 1996, 41 (27), 27. (15) Lindbauer, R. L. Chemosphere 1992, 25 (7-10), 1409. (16) Banaee, J.; Larson, R. A. Waste Manage. 1993, 13, 77. (17) Ohlsson, O. O.; Shepherd, P. In Combustion Modeling, Cofiring and Nox Control; FACT-Am. Soc. Mech. Eng. 1993, 17, 173. (18) Frankenhaeuser, M.; Manninen, H.; Virkki, J.; Kojo, I. The Effect of the Chlorine/Sulfur Ratio on Organic Emissions from the Combustion of Mixed Fuels; NESTE Final Report: Porvoo, Finland, 1992. (19) Griffin, R. D. Chemosphere 1986, 15, 1987.
SO2 + Cl2 + H2O T SO3 + 2HCl
(5)
Thus the chlorinating agent, chlorine, is converted into HCl, which is very unlikely to undergo aromatic substitution reactions to form PCDD and PCDF precursors. In the study reported in this paper, this reaction was examined by conducting experiments in a tube furnace. The results indicate the apparently inhibiting effect of sulfur on the Deacon Reaction. The combustion and thermal decomposition processes of other materials that comprise MSW and their blends were examined in our study using TG/FTIR/MS and GC/ MS trapping techniques. The combination of TG/FTIR and TG/MS offers complementary techniques for detection and identification of evolved gases. The major advantages of these techniques lie in their abilities to obtain information concerning combustion products on line at the elevated temperatures. On the basis of the static results from TG/FTIR/MS, two coals with different chlorine contents were co-fired with PVC in a bench scale 0.1 MWth FBC system. The effects of temperature, with and without limestone, and S/Cl mole ratios on PCDD/Fs formation in ash were investigated. Experimental Section TGA/FTIR/MS System. To study the combustion performance of MSW co-fired with coal, a small amount (about 1020 mg) of sample (coal or PVC or the blend of coal and PVC) was placed in the TGA and heated to 1000 °C at different heating rates in an air atmosphere. The gaseous products were analyzed by the TGA/FTIR/MS system, and the FTIR spectra and MS profiles were recorded.20 The three major components of this system are a Model 951 Thermogravimetric Analyzer, a Model 1650 Fourier Transform Infrared Spectrophotometer, and a VG Thermolab Gas Analysis System. The TGA is interfaced to the FTIR using an insulated Teflon tube which is heated to a temperature of 150 °C by a Powerstat variable autotransformer in order to prevent possible condensation of the gaseous products. The mass spectrometer is coupled with the TGA by means of a fused silica capillary sampling inlet that is heated to approximately 170 °C. Studies with a Tube Furnace. This series of tests were performed in a quartz reaction tube inserted into a horizontally mounted electric Lindberg furnace. Prior to the introduction of samples into the reactor, the furnace was preheated to the desired temperature. To simulate the conditions used in the AFBC system, a mixture of flue gas atmosphere including CO2 (15%), CO (0.2%), O2 (5%), and H2O vapor (5%) in N2 was adopted. The compositions of the process gases were adjusted by calibrated Teflon flow meters. The reaction products were swept into a cooled trap containing a chosen absorbent. After the reaction was complete the trapping solution was concentrated and then subjected to GC/MS analysis. A Shimadzu QP 5000 system with a NIST/EPA/NIH 62,000 compound database was used for GC/MS analysis. Aliquots of 2 µL of sample were injected in the splitless mode onto a RTX-5 fused silica capillary column (60 m × 0.32 mm and a stationary phase thickness of 1.0 µm). Helium was used as the carrier gas. The surrounding temperature for injector, interface, and detector was 230 °C. The mass spectrometer was operated in the selected ion monitoring (SIM) mode. The identification of compounds was accomplished by using a computerized library search and by comparison with literature mass spectra. Moreover, comparison to the GC retention time for pure (20) Lu, R.; Purushothama, S.; Yang, X.; Hyatt, J.; Pan, W.-P.; Riley, J.; Lloyd, W. G. Fuel Process. Technol. 1999, 59, 35.
SO2 Effect on Cl2 Formation during Fuel Combustion
Energy & Fuels, Vol. 14, No. 3, 2000 599
Figure 1. Profiles of HCl and Cl2 evolved during the combustion of PVC. compounds was also used to confirm the identification of unknowns. Standard materials were tested to establish the detection limits for the experimental setup, for calibration, and determination of the quantitative working range for the compounds. The detection limit is 0.1 ppb when the selected ion monitoring (SIM) mode is chosen. Bench Fluidized Bed Combustion. The inner diameter of combustor21 is 0.3 m and the height is 4.4 m. The freeboard zone of the combustor is 2.5 m high, providing adequate residence time for the combustion of fine particles which may be entrained in the gases. The fuel (coal and PVC) is injected into the combustor through a pressurized underbed feed mechanism. To determine if any chlorinated organic compounds were formed during the combustion reactions, three samples each of flue gas, fly ash, and bed ash were obtained for each run condition. The combustion gases were collected in a Tenax trap and glass fiber filters for 24 h with a 300 mL/ min flow rate. The line between combustor and adsorption tubes was heated to 450 °C to minimize any condensation. The collected samples were Soxhlet extracted with CH2Cl2 for 6 h. Extracts were then concentrated in micro Kuderna-Denish apparatus to 0.5 mL before GC/MS analysis through the selected ion monitoring (SIM) mode.
Results and Discussion Characterization of Raw Materials and Their Blends. To understand the fundamental processes and mechanisms of thermal decomposition, the thermal behavior of coal, MSW, and their blends were investigated using TGA/FTIR/MS at a fast rate of 100 °C/min. Part of the results have been presented by Lu and coworkers.20 These results are important to the analysis and control of the performance of a FBC system. Upon interpretation of the MS spectrum obtained from the combustion of PVC, one finds a very notable result, that is the production of molecular chlorine accompanies the release of a large amount of HCl during the combustion of PVC. As illustrated from Figure 1, masses 36 and 38 (21) Xie, W.; Pan, W.-P.; Riley, J. Energy Fuel 1999, 13, 585.
are formed at the same time, and the integrated ratio of their ion intensity strongly suggests the presence of isotopes of H35Cl and H37Cl. Furthermore, three additional m/z peaks appear at exactly the same point, with apparent masses of 70, 72, and 74 corresponding to 35Cl2, 35Cl37Cl, and 37Cl2. This is strong evidence suggesting that some fraction of the abundant HCl may be undergoing a thermal Deacon Reaction to produce molecular chlorine. Following the in-situ generation of Cl2, the aromatic compounds can be readily attacked to form the chlorinated organics such as chlorobenzene, which corresponds to masses of 112 and 114. It is a plausible starting point for the formation of chlorinated organics from the combustion of chlorine-rich fuel mixtures. When changing the atmosphere from air to nitrogen, chlorine is not identified in the products from the thermal decomposition of PVC. This can be ascribed to the absence of oxygen, a necessary reactant in the Deacon Reaction. However, instead of Cl2, HCl is still the major product from the combustion of PVC, even in air. As determined from the FTIR data, as shown in Figure 2, the chlorine and hydrocarbon species formed during the combustion of blends are released at the same time. In fact, the heating rate in an AFBC system is much higher than 100 °C/min. Thus, one can expect there are greater possibilities for the production of chlorinated organic compounds during co-firing coals with RDF in an AFBC system. Mechanism for the Formation of PCDD/Fs during the Combustion of MSW. Four grams of different raw materials (newspaper, cellulose, and RDF) were burned in air in a tube furnace. The furnace was preheated to a temperature of 850 °C before the sample was introduced. The gaseous products were trapped in chilled CH2Cl2 and analyzed by GC/MS. In summary, the results of replicate analyses show phenol as one of the major organic products released during the combus-
600
Energy & Fuels, Vol. 14, No. 3, 2000
Xie et al.
Figure 2. 3D FTIR spectra of combustion products of PVC heated at a fast heating rate. Table 1. Chlorination and Condensation Reactions of Phenol run
temperature (°C)
time (min)
tentative product identification
1 2 3 4 5 6 7
250 425 600 600 700 800 800
30 30 15 30 30 15 30
dibenzofurans, mono-, di-, and trichlorophenols chlorinated phenols and dibenzofuran dibenzofurans, no chlorophenols dibenzofurans, mono-, di-, and trichlorophenols dibenzofurans, mono-, di-, and trichlorophenols dibenzofurans, no chlorophenols dibenzofurans, mono-, di-, and trichlorophenols
Table 2. Products from the Study of the Condensation Reactions of Chlorophenols sample
temperature (°C)
tentative product identification
2,4-dichlorophenol
700
2,4-dichlorophenol
400
4-chlorophenol 2-chlorophenol
700 700
tetrachlorodibenzodioxin, tetrachlorofuran, dibenzodioxin, 2-chlorophenol, 2,6-dichlorophenol, 2,4,6-trichlorophenol, dichlorobenzene trichlorobenzene, dichlorodibenzofuran trichlorodibenzodioxin, tetradichlorodibenzofuran, dichlorodibenzodioxin dichlorodibenzofuran, benzene, chlorobenzene dichlorodibenzofuran, phenol
tion of newspaper, cellulose, and RDF. From the GC/ MS data, chlorophenol was determined to be a major product when blends are burned.22 To examine the gas-phase chlorination of phenol, as illustrated by eq 3, 100 mg portions of phenol were placed in a heated tube and evaporated in the presence of a constant flow of 0.5% Cl2 in nitrogen. The reaction products were cooled by liquid nitrogen and condensed upon exiting from the reaction tube, carefully washed by methylene chloride, and analyzed by GC/MS. The test results is listed in Table 1. The chlorination of phenol began at a temperature around 250 °C and produced 2-chlorophenol, 4-chlorophenol, and 2,4-chlo(22) Pan, W. P.; Riley, J. T. Co-Firing High Sulfur Coal with Refuse Derived Fuels; Final Report, Project No. DE-FG-94PC94211, Nov, 1997.
rophenol. At higher temperatures, dibenzofuran was produced. The combustion of chlorinated phenols, which may lead to the reaction illustrated by eq 4, was examined by heating 100 mg portions of 2,4-chlorophenol in the presence of air in the tube furnace. The results of the analysis of the reaction products are presented in Table 2. The GC/MS results show the products from the combustion of 2,4-dichlorophenol include 2,4,6-trichlorophenol, tetrachlorodibenzofuran, and dichlorodibenzodioxin. Tetrachlorodibenzofuran and dichlorodibenzodioxin were also formed below 400 °C. The results from this series of experiments strongly suggest the possibility that PCDD/Fs form from the condensation of combustion products of chlorophenol.
SO2 Effect on Cl2 Formation during Fuel Combustion
Energy & Fuels, Vol. 14, No. 3, 2000 601 Table 3. Proximate and Ultimate Analysis Data for Raw Materials
Figure 3. Relative yield of chlorinated phenols as a function of S/Cl ratio.
The Effect of Sulfur Species on the Deacon Reaction. In a study conducted by Gullet and coworkers, it was reported that the reaction of Cl2 with SO2 to form HCl (reaction 5) is not measurable below 800 °C.23 These results are not apparent from thermodynamic calculations of the free energy change. Although equilibrium calculations suggest that the reaction is favored over the full range of temperatures tested, the kinetics of the reaction may prevent observation of measurable product until the higher temperatures are reached. The possible effect of SO2 upon the formation of Cl2 through the Deacon Reaction was examined at 800 °C. The flow rate of HCl (1% in nitrogen), SO2 (4.86% in nitrogen), and air were adjusted using calibrated Teflon flow meters. The evolved gas was trapped by an absorbent, which was prepared by dissolving 50 mg of phenol in 25 mL of methylene chloride. The amount of phenol in the trapping solution was accurately controlled to within (0.0001 g. After each experiment the trapped solutions were concentrated to 1 mL and injected into the GC/MS system for analysis. In this quantitative study, the concentrations of HCl (250 ppm) and O2 (5%) in the gaseous mixtures were fixed and only the concentration of SO2 was changed in the range from 0 to 1230 ppm. The results of tests under varying conditions are summarized in Figure 3, and all data points presented are an average of at least three runs. As clearly shown in this figure, without SO2 in the reaction system the relative yield of chlorinated phenol (represented by Chlorinated Phenol/Phenol) is significant. With the addition of SO2 the production of chlorinated phenol, which results from the production of molecular chlorine, decreases. When the S/Cl ratio approached 2.5/1, the chlorinated phenol production was reduced to less than 8% of that produced in the absence of SO2. Before the S/Cl reaches 2.5/1, the yield of chlorinated phenols versus S/Cl ratio shows a linear relationship. After that point the extra addition of sulfur, even doubled, has no apparent influence on the chlorine species formation, as shown by the plateau in Figure 3. The results shown in this figure indicate that the molecular chlorine produced through the Deacon Reaction in this reaction system was depleted by SO2, as indicated by reaction 5. (23) Gullet, B. K.; Bruce, K. R.; Beach L. O. Environ. Sci. Technol. 1992, 26, 1938.
parameters
95010
95031
PVC
Wood
coal seam rank of coals moisturea ash vol. matter fixed carbon carbon hydrogen nitrogen sulfur oxygen chlorine, ppm cal. value (Btu/lb)
blend A 2.32 7.22 39.97 52.82 79.38 5.31 1.63 0.67 5.69 1039 14077
IL No. 6 B 8.32 10.78 37.21 52.02 72.61 4.82 1.54 2.38 7.57 3065 12842
0.00 0.36 99.64 0.00 38.71 4.2 0.07 0.22 0 54.65b 8556
4.71 1.14 80.6 14.3 49.9 6.13 0.08 0.12 42.55 748 8343
a Moisture is as-determined. All other analyses are reported on a dry basis. The rank of each coal is high volatile A, B, or C bituminous. b The unit for chlorine in PVC is percent.
Griffin, from a study of co-incineration of coal and municipal solid wastes, suggested that dibenzodioxins would not form when the S/Cl ratio was greater than 10, and proposed increasing the sulfur of the wastes in co-combustion with coal in order to decrease dioxin formation.19 However, in the study reported in this paper in which the S/Cl ratio was less than 2.5/1, dramatic decreases in the major chlorine-containing products of combustion were observed. Co-Combustion of PVC with Coal in a FBC System. Based on the laboratory studies concerning the PCDD/Fs formation, mixtures of coal and major components of MSW were burned in a 0.1 MWth AFBC facility in order to investigate co-combustion performance. The effects of the S/Cl molar ratio and the kinds of coal on PCDD/Fs formation were studied under the same combustion temperature, gas velocity and excess air rate. PVC and wood pellets were selected as cocombustion fuels for this study, since they are the major source of chlorine during MSW incineration. The experimental run conditions were: Fuel compositions: (1) 100% coal; (2) 89% coal, 1% PVC, and 10% wood pellets; (3) 86.7% coal, 3.3% PVC, and 10% wood pellets. Fuel feed rates: 9.94 kg/h; Ca/S ratio: 3.0; bed temperature: 850 °C; gas velocity:1.25 m/s; excess air ratio: 1.25. The analytical values for the raw materials used in this study are shown in Table 3. The results from the combustion run showed that no chlorinated organics were detected in the flue gases and the bed ashes under all co-firing experimental conditions in this study. The effects of the PVC/fuel weight ratio on the PCDD/Fs concentration in the fly ashes are presented in Figures 4 and 5. It can be seen from these figures that the amount of PCDD/Fs in fly ash increases with an elevated PVC/fuel ratio, and the yield of PCDD is always higher than that of PCDF. This fact provides additional evidence for the importance of chlorine content in fuel on the PCDD/Fs formation. Figure 4 is a comparison of tetrachlorodibenzodioxin (TCDD) concentration in fly ash for two different kinds of coal. As shown in the figures, with an increase of the PVC/fuel ratio the TCDD concentration in the fly ash is enhanced for both coals. TCDD emission from the combustion of coal 95031 increased gradually, whereas the TCDD emission from the combustion of coal 95011 remained relatively constant as the PVC/fuel ratio is increased. However, for tetrachlorodibenzofuran (TCDFs), the
602
Energy & Fuels, Vol. 14, No. 3, 2000
Xie et al. Table 4. The Effect of Limestone on the Distribution of 2,3,7,8-TCDD and 2,3,7,8-TCDF
2,3,7,8-TCDD gas phase fly ash (ng/kg) bed ash 2,3,7,8-TCDF gas phase fly ash (ng/kg) bed ash
Figure 4. The effect of PVC/fuel ratio on 2,3,7,8-TCDF content in fly ash.
Coal 95010 with 3.3 wt % PVC
Coal 95031 with 3.3 wt % PVC
with limestone
no limestonea
with limestone
no limestonea
b 2009 b
b 5171 b
b 1200 b
b 3733 b
b 1533 b
b 2491 b
b 1241 b
b 3755 b
a No new limestone was fed with fuel, the original bed material is limestone. b Under detection limit.
chloride in flue gas are also revealed. The effects of limestone on PCDD/Fs content in fly ash were listed in Table 4. With more limestone being injected into combustor, more hydrogen chloride was captured by limestone. HCl concentration in flue gas decreases significantly. Thus, the production of PCDD/Fs were reduced remarkably as shown in Table 4. Conclusions
Figure 5. The effect of PVC/fuel ratio on 2,3,7,8-TCDD content in fly ash.
trend for the two coals is the same, as is shown in Figure 5. Since coal 95010 has a higher sulfur content then coal 95031 has, these results indicate the promising future for co-firing MSW with high sulfur coal. On the basis of different trends obtained for TCDD and TCDF emissions for two coals, the different mechanisms for PCDD and PCDF formation in the presence of excess
In a tube furnace, chlorophonel was identified as a major organic product when blends are burned and SO2 is an effective inhibitor of the formation of molecular chlorine through the Deacon Reaction. There is no evidence to identify gas-phase PCDDs and/or PCDFs in this study under the bench fluidized bed combustion system. The other parameters (such as the amount of limestone, the concentration of SO2 in the flue gas, bed temperature, and fluidized velocity) may also influence the formation of chlorinated organic compounds in the fly ash. Acknowledgment. The authors thank the U.S. Department of Energy for the financial support through grant number DE-FG-94PC 94211. EF990173D
