EP A/600/A-98/005 DIOXIN FORMATION FROM SOOT-BASED DEPOSITS K. Raghunathan Acurex Environmental Corporation Research Triangle Park, NC 27709 Chun Wai Lee and Jim Kilgroe Environmental Protection Agency National Risk Management Research Laboratory Research Triangle Park, NC 27711 ABSTRACT In many commercial incineration facilities, history or "memory effects" play an important role in the formation of polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzo furans (PCDFs). This study aims to explain wide variations in PCDD and PCDF formation from tests performed under similar conditions but in facilities subjected to different surface deposits. Specifically, the effect of soot and copper deposits on subsequent PCDD and PCDF formation is evaluated. In a bench-scale setup, fuel oil doped with and without copper was combusted under sooting conditions. The generated soot species were carried by the hot flue gas and deposited on quartz tubes. In a separate setup, a pilot-scale combustor was fired with natural gas, and 1,6-dichlorohexane was injected into the flame to supply chlorine. Each of the above quartz tube was heated to 320 °C, and a slip stream of the flue gas from the pilot-scale combustor was passed through the tube and sampled for PCDDs and PCDFs. Results show that the combined soot-copper deposit caused substantial PCDD and PCDF formation. On the other hand, a deposit of soot or copper alone did not promote such high formation. Soot-free combustion of No. 2 fuel oil containing varying levels (less than 100 ppmw) of copper and chlorine also generated very little PCDDs and PCDFs. The PCDD and PCDF formation potential of the soot-copper deposit appears to decay with time, and the presence of sulfur dioxide inhibits formation. The role of chlorine concentration is also evaluated. INTRODUCTION Emissions of PCDDs and PCDFs, often referred to as simply dioxins and furans, pose serious health concerns. These compounds are formed when organic precursors/carbon and a chlorine (CI) source are present, under the influence of a metal catalyst, especially copper. Even when these species are present in very small quantities, PCDD/PCDF formation is a problem since they are toxic even at low concentrations. A typical waste incineration/combustion process can leave deposits or contaminants in the facility which can alter the subsequent PCDD/PCDF emission trends for the facility. Many field units exhibit hysteresis effects toward PCDD/PCDF emissions; i.e., recent past facility history has an effect on subsequent PCDD/PCDF emissions. In a pilot-scale combustor, wall effects on PCDD/PCDF yields were observed (1). In a study of chlorofluorocarbon (CFC) incineration, high PCDD/PCDF emissions were reported for a facility previously used for incineration of metal-containing wastes (2). These high emissions were attributed to residual copper that was retained in the facility. Subsequent studies (3) confirmed that PCDD/PCDF levels were higher after injection of copper into the facility. Catalytic effects of copper in PCDD/PCDF formation are well established (4,5,6). 1 ------- Carbon can also promote PCDD/PCDF formation due to the de novo mechanism (5,7). Carbon may be deposited in the form of soot or retained with the ash in the system as unburnt solid-phase carbon from the fuel. The extent to which soot can promote PCDD/PCDF formation is unclear. During shake-down tests prior to this work, it was observed that soot deposition in a pilot-scale combustor led to increased PCDD and PCDF formation for CI concentrations of 500 ppm. Prior to the sooting, the combustor had been used for a variety of tests, and the inner walls had been regularly exposed to coal and municipal solid waste fly ash containing various metals in trace amounts. The combination of catalytic trace metals, soot, and the high CI concentration in the gas phase possibly resulted in the higher PCDD and PCDF levels. In this study, effects of combined soot-copper deposits on PCDD/PCDF formation are studied. Effects of gas-phase CI concentration and the presence of sulfur dioxide (SO2) are discussed. Tests were also conducted to assess the PCDD/PCDF formation potential of No. 2 fuel oil combustion under soot-free conditions. EXPERIMENTAL Soot-Based Formation Experiments involved two steps: (1) a contamination step in which a contaminated quartz tube is prepared, and (2) a PCDD/PCDF formation step in which PCDD/PCDF produced by the contaminated tube is measured. Contamination Step A schematic of the contaminant generation system is shown in Figure 1. It consists of a quartz burner with two concentric tubes, heated by a Lindberg three-zone electrical furnace to 1000*C. The contaminant solution is metered in by a peristaltic pump and injected into the inner tube of the quartz burner by a two-fluid atomizing nozzle at a rate of 0.8 mL/min. Air entering the reactor is preheated as it passes through the annular space before contacting the injected mist. Connected to the exit of the quartz burner is another quartz tube which provides the surface onto which the generated contaminants will deposit. An electric heat tape maintains this quartz tube at above 120 °C to prevent moisture condensation. For most of the tests, the contaminant solution was prepared by adding copper (II) naphthenate solution (8% copper by weight) to No. 2 fuel oil to make up a solution containing 0.5% copper by weight. Combustion of the copper-fuel oil contaminant solution was under fuel-rich conditions, producing soot. The quartz tube is coated with the generated copper-containing soot along its length. The contamination step was carried out for about 45 minutes. PCDD/PCDF Formation Step In this step, a chlorine-containing slip stream from a pilot-scale reactor is passed through a heated contaminated tube into a Method 23 train to measure the PCDD and PCDF yields. The pilot-scale facility used is EPA's Innovative Furnace Reactor (IFR). A schematic of the pilot-scale IFR setup is shown in Figure 2. The IFR is a down-fired, refractory-lined cylindrical unit with a length of about 3 m and an internal diameter (ID) of 20.3 cm, nominally rated at 48.8 kW. Ports along the length of the IFR facilitate addition of reactants and 2 ------- sampling. In this study, the IFR was fired at about 29 kW, with natural gas. The chlorine source was 1,6-dichlorohexane, injected just below the burner. The resulting chloride (CI) concentration in the flue gas was 500 ppm. For studies on SO2 effects, gaseous SO2 was doped into the IFR, yielding a concentration of 600 ppm in the flue gas. For tests with varying hydrogen chloride (HC1) concentration in the furnace, a mixture of hexane and 1,6-dichlorohexane at different proportions was used. The contaminated quartz tube (150 cm long, 25 mm ID), prepared as described above, is inserted into one of the IFR ports as shown in Figure 2. The temperature at this port was about 600 °C, and the temperature inside the tube was maintained at 320 °C by a temperature controller. The tube exit was connected to a Method 23 train for PCDD/PCDF sampling. . Initially, the IFR was fired with natural gas, and the contaminated tube was flushed with the chlorine-free IFR slip stream for 1 hour. Then, the flow through the tube was shut off, and 1,6-dichlorohexane was introduced into the IFR. After allowing 45 minutes for equilibration, the slip stream was sampled through the tube into the Method 23 train for about 1 hour. The sampling flow rate through the tube was such that the gas residence time in the tube was about 2 seconds. Soot-Free Formation Another set of experiments was carried out to determine if soot-free combustion of No. 2 fuel oil under varying chlorine and copper contents in oil would form significant PCDD and PCDF. The experimental setup shown in Figure 1 was used for these tests; the conditions in the burner were oxygen-rich with no soot formation, and the quartz tube was maintained at 320 °C. The effluent from the quartz tube was passed through a Method 23 train. The No. 2 fuel oil injected into the burner was doped with 1,6-dichlorohexane and copper naphthenate to yield the following fuel compositions (by weight): 0 ppm Cu and 0 ppm CI 15 ppm Cu and 20 ppm CI 15 ppm Cu and 100 ppm CI 30 ppm Cu and 100 ppm CI The PCDD and PCDF sampling duration was about 2 hours. Again, the gas-phase residence time in the quartz tube was about 2 seconds at 320 °C. The Method 23 samples were analyzed in EPA's in-house Organics Support Laboratory (OSL), using procedures described elsewhere (8). RESULTS AND DISCUSSION PCDD and PCDF levels from tubes contaminated with either copper alone or soot alone were negligible. However, the combined copper-soot deposit resulted in significant PCDD and PCDF levels. It should be pointed out, however, that the conditions under which such yields were measured do not simulate conditions typically found in fuel-oil-fired boilers. The CI levels in the gas phase were high (500 ppm) and so were the average amounts of copper injected (as copper naphthenate) for contaminating the quartz tube (a surface deposit of about 1.5 g/m2). Various combinations of fuels (No. 2 fuel oil, No. 5 fuel oil, or natural gas), all under sooting conditions, and copper compounds (cupric oxide, cuprous oxide, or copper naphthenate) were tried, and the high PCDD and PCDF formation potential was observed for 3 ------- all these cases. Results of repeated tests with the same tube are shown in Figures 3 and 4. Figure 3 shows the initial high yields from the contaminated tube at 500 ppm CI. When the CI in the IFR gas stream has been shut off (CI = 0), the yields are much lower but still there is residual activity toward PCDD and PCDF formation from the earlier experiment. It is likely that the prior exposure to CI has allowed formation of copper chlorides which served as the CI source even when there is no CI in the gas phase. Both figures indicate that initially, the PCDD and PCDF yields are higher, but the yields decay with time. Sintering of the catalyst, loss of copper through vaporization of its chlorides, or slow oxidation of the soot are possible mechanisms for the activity to decrease with time. At the end of each series of tests shown in Figures 3 and 4, gaseous SO2 was injected into the IFR in addition to the 1,6-dichlorohexane, and the corresponding PCDD and PCDF formation from the contaminated tube was measured. The resulting SO2 concentration was about 600 ppm, providing a SO2/HCI ratio of 1.2. Figures 5 and 6 show that the presence of SO2 decreases formation. Although it is true that a decrease in yields is also caused by repeated testing with the same contaminated tube, it appears that the decrease due to SO2 (Figure 6) is much sharper than the decaying trend (Figure 4). SO2 has been shown to inhibit PCDD and PCDF formation in municipal waste combustion systems (8). It was not possible to ensure that different contaminated tubes, prepared as shown in Figure 1, had the same contaminant characteristics and hence the same PCDD and PCDF formation potential. Therefore, since the yields also decay with repeated testing, it was not possible to conduct a systematic study to establish the dependence of CI or Cu concentration on the formation potential. However, back-to-back tests with a contaminated tube were conducted in which the gas-phase CI concentrations were 10 and 500 ppm, respectively, and the results are shown in Figure 7. The figure shows that, even at a CI concentration of 10 ppm, the PCDD and PCDF formation takes place, but at low levels. The figure also shows that there is a substantial increase in formation at the higher CI, in spite of possible decay in activity after the 10 ppm test. For the soot-free No. 2 fuel oil combustion tests, the burner tube was thoroughly rinsed with solvents (toluene, methanol, and methylene chloride) and acids (concentrated nitric and sulfuric acids) before each test to remove residual copper and soot deposits. As discussed earlier, the setup shown in Figure 1 was used for these tests, and 1,6-dichlorohexane and copper naphthenate were added in ppm quantities to the No. 2 fuel oil injected into the burner. For the range of Cu and CI concentrations in the fuel oil, there was no evidence of PCDD and PCDF formation. Therefore, it appears that clean, soot-free combustion of No. 2 fuel oil, even with Cu and CI concentrations up to 30 ppmw and 100 ppmw, respectively, does not form PCDDs/PCDFs. CONCLUSIONS A quartz tube with high levels of combined soot and copper deposits resulted in significant PCDD and PCDF formation in a flue gas stream containing CI. The PCDD and PCDF yields decline with time. Presence of SO2 in the gas phase decreases PCDD and PCDF formation. Soot-free combustion of No. 2 fuel oil containing low levels of copper and chlorine does not produce PCDD and PCDF. 4 ------- ACKNOWLEDGMENTS The authors appreciate the logistical support of Richard E. Valentine (U.S. EPA/APPCD). Charles B. Courtney, Dennis Tabor, Ann Preston, and Paul W. Groff (Acurex Environmental Corporation), and Jeff V. Ryan (U.S. EPA/APPCD) provided extensive sampling, technical, and analytical assistance. REFERENCES 1. Gullett, B.K. and Raghunathan, K., "Observations on the Effect of Process Parameters on Dioxin/Furan Yield in Municipal Waste and Coal Systems," Chemosphere, 34, 1027, 1997. 2. Hassel, G.R., "Experimental Investigation of PIC Formation in CFC Incineration," EPA- 600/7/91-010 (NTIS PB92-126952), Research Triangle Park, NC, December 1991. 3. Lee, C.W., Ryan, J.V., Hall, R.E., Kryder, G.D. and Springsteen, B.R., "Effects of Copper Contamination on Dioxin Emissions from CFC Incineration," Combust. Sci. Tech. Vols. 116-117 (455) 1996. 4. Hagenmaier, H., Kraft, M., Brunner, H. and Haag, R., "Catalytic Effects of Fly Ash from Waste Incineration Facilities on the Formation and Decomposition of Poly chlorinated Dibenzo-p-dioxins and Polychlorinated Dibenzofurans," Environ. Sci. Technol., 21,1080, 1987. 5. Stieglitz, L., Zwick, G., Beck, J., Roth, W. and Vogg, H. "On the de-novo Synthesis of PCDD/PCDF on Fly Ash of Municipal Waste Incinerators," Chemosphere, 18,1219, 1989. 6. Gullett, B.K., Bruce, K.R. and Beach, L.O., "The Effect of Metal Catalysts on the Formation of Polychlorinated Dibenzo-p-dioxin and Polychlorinated Dibenzofuran Precursors," Chemosphere, 20, 1945, 1990. 7. Stieglitz, L., Zwick, G., Beck, J. and Roth, W., "Carbonaceous Particles in Fly Ash - A Source for the de-novo Synthesis of Organochlorocompounds," Chemosphere, 19,283, 1989. 8. Raghunathan, K. and Gullett, B.K., "Role of Sulfur in Reducing PCDD and PCDF Formation," Environ. Sci. TechnoL, 30(6): 1827 (1996). 5 ------- LIST OF FIGURES Figure 1. Experimental setup for contamination step Figure 2. Experimental set up for PCDD/PCDF formation step Figure 3. Results of repeated tests with the same tube (Tube A) Figure 4. Results of repeated tests with the same tube (Tube B) Figure 5. Effect of S02 on PCDD and PCDF yields from a contaminated tube (Tube A) Figure 6. Effect of S02 on PCDD and PCDF yields from a contaminated tube (Tube B) Figure 7. PCDD and PCDF yields from a contaminated tube at different CI concentrations 6 ------- atomizing nitrogen furnace quartz burner fuel oil/copper naphthenate stirrer to continuous emission monitor tube a thermo- couples heat tape to dioxin trap /exhaust Figure 1. Experimental setup for contamination step ------- natural gas dichlorohexane PILOT-SCALE REACTOR to continuous emission monitor f combustion air SOo contaminated quartz tube temperature controller ~ ~ thermocouples to Method 23 train Figure 2. Experimental setup for PCDD/PCDF formation step ------- tetra-octa yield (ng/dry cubic meter) 5E+4 4E+4 3E+4 2E+4 1E+4 0E+0 34,91 11,425 / PCDD ~ PCDF z 285 837 9,536 3,552 500 0 500 gas-phase CI concentration (ppm) Figure 3. Results of repeated tests with the same tube (Tube A) ------- tetra-octa yield (ng/dry cubic meter) 1.5E+5 1 E+5 5E+4 0E+0 z 34,087 H PCDD ~ PCDF 98,697 71 82,406 23,255 / 500 500 gas-phase CI concentration (ppm) Figure 4. Results of repeated tests with the same tube (Tube B) ------- tetra-octa yield (ng/dry cubic meter) 1.5E+4 1E+4 5E+3 0E+0 0 600 gas-phase S02 concentration (ppm) Figure 5. Effect of S02 on PCDD and PCDF yields from a contaminated tube (Tube A) ------- tetra-octa yield (ng/dry cubic meter) 1.5E+5 1E+5 5E+4 0E+0 23,255 PCDD ~ PCDF 82,406 4,784 // (/ 10,816 0 600 gas-phase S02 concentration (ppm) Figure 6. Effect of S02 on PCDD and PCDF yields from a contaminated tube (Tube B) ------- tetra-octa yield (ng/dry cubic meter) 1E+5 8E+4 6E+4 4E+4 2E+4 0E+0 10 500 gas-phase CI concentration (ppm) Figure 7. PCDD and PCDF yields from a contaminated tube at different CI concentrations ------- _ _... TECHNICAL REPORT DATA III llll II III! NR MR L- RT P- P- 2 41 (Please read Instructions on the reverse before complei III llll II llll III llll III 1. REPORT Ni _ . EPA/600/A-98/005 3. PB98 hi in i ¦¦ hi -137474 4. TITLE AND SUBTITLE Dioxin Formation from Soot-based Deposits 5. REPORT DATE 6. PERFORMING ORGANIZATION CODE 7. AUTHOR(S) K. Raghunathan (Acurex) and C.W.Lee and J.Kilgroe (EPA) 8. PERFORMING ORGANIZATION REPORT NO. 9. PERFORMING ORGANIZATION NAME AND ADDRESS Acurex Environmental Corporation P. 0. Box 13109 Research Triangle Park, North Carolina 27709 10. PROGRAM ELEMENT NO. 11. CONTRACT/GRANT NO. 12. SPONSORING AGENCY NAME AND ADDRESS EPA, Office of Research and Development Air Pollution Prevention and Control Division Research Triangle Park, NC 27711 13. TYPE OF REPORT AND PERIOD COVERED Published paper; 14. SPONSORING AGENCY CODE EPA/600/13 is.supplementary notes^ppcD project officer is Chun Wai Lee, Mail Drop 65, 919/541- 7663. Presented at Int. Conf. on Incineration and Thermal Treatment Technologies, Oakland. CA. 5/12-16/97. i6. abstractpap6r discusses a study aimed at explaining wide variations in poly- chlorinated dibenzo-p-dioxing (PCDD) and polychlorinated dibenzofuran (PCDF) for- mation from tests under similar conditions but in facilities exhibiting different sur- face deposits. Specifically, the effect of soot and copper deposits on subsequent PCDD and PCDF formation is evaluated. (NOTE: In many commercial incineration facilities, history or "memory effects" play an important role in the formation of PCDDs and PCDFs.) In a bench-scale setup, fuel oil doped with and without copper was combusted under sooting conditions. The generated soot species were carried by the hot flue gas and deposited on quartz tubes. In a separate setup, a pilot-scale com- bustor was fired with natural gas, and 1, 6-dichlorohexane was injected into the flame to supply chlorine. Each of the above quartz tubes was heated to 320 C, and a slip- stream of the flue gas from the pilot-scale combustor was passed through the tube and sampled for PCDDs and PCDFs. Results show that the combined soot-copper de- posits caused substantial PCDD and PCDF formation. On the other hand, a deposit of soot or copper alone did not promote such formation. The PCDD and PCDF forma- tion potential of the soot-copper deposit appears to decay with time, and the presence of sulfur dioxide inhibits formation. 17. KEY WORDS AND DOCUMENT ANALYSIS a. DESCRIPTORS b.IDENTIFIERS/OPEN ENDED TERMS c. COSATI Field/Group Pollution Emission Fuel Oil Halohydrocarbons Natural Gas Furans Combustion Soot Copper Incinerators Pollution Control Stationary Sources Dioxin Polychlorinated Diben- zo-p-dioxins (PCDDs) Polychlorinated Diben- zofurans (PCDFs) 13 B 14 G 2 ID 07C 2 IB 07B 18. DISTRIBUTION STATEMENT _ , , Reproduced from 00^ Release to Public best available copy.fpp 19. SECURITY CLASS (This Report) Unclassified 21. NO. OF PAGES 20. SECURITY CLASS (This page) Unclassified 22. PRICE EPA Form 2220-1 (9-73) ------- |