Homologue and Isomer Patterns of Polychlorinated Dibenzo-p-dioxins

May 10, 2005 - JAMES A. MULHOLLAND, ‡. DO HYONG KIM, ‡. AND. MASAO TAKEUCHI †. Combustion Control Group, Energy Technology Research...
0 downloads 0 Views 584KB Size
Download PDF
Environ. Sci. Technol. 2005, 39, 4398-4406

Homologue and Isomer Patterns of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans from Phenol Precursors: Comparison with Municipal Waste Incinerator Data J A E - Y O N G R Y U , * ,† JAMES A. MULHOLLAND,‡ DO HYONG KIM,‡ AND MASAO TAKEUCHI† Combustion Control Group, Energy Technology Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), 16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan, and Environmental Engineering Group, Civil and Environmental Engineering, Georgia Institute of Technology, 311 Ferst Drive, Atlanta, Georgia 30332-0512

The role of phenol precursors in polychlorinated dibenzo-p-dioxin (PCDD) and dibenzofuran (PCDF) formation in municipal waste incinerators is assessed on the basis of homologue and isomer patterns. Homologue and isomer patterns of PCDD and PCDF congeners formed from phenols both in the gas phase and via particle-mediated reactions were studied in an isothermal flow reactor. A mixture of unsubsitituted phenol and 19 chlorinated phenols in relative concentrations found in a municipal waste incinerator (MWI) stack gas was used for this study. PCDD and PCDF homologue and isomer patterns obtained from the phenol experiments were compared with those observed in MWI data. From the phenol experiments, gas-phase formation at 600-700 °C favors PCDF formation whereas particle-mediated formation at 400 °C favors PCDD formation. Unsubstituted phenol, which was present in high concentration, played a significant role in the formation of PCDD/F congeners under both sets of experimental conditions. PCDD/F distributions in MWI flue gas and fly ash samples were different from those observed in the phenol experiments, suggesting that direct phenol condensation was not the primary route of PCDD/F formation at the incinerators. Gas-phase phenol condensation is a source of dibenzofuran, with subsequent particle-mediated chlorination resulting in PCDF formation. In the case of PCDD formation, phenol condensation may be responsible for the formation of certain highly chlorinated congeners. In this paper we demonstrate the use of homologue and isomer patterns for PCDD/F formation mechanism attribution in municipal waste incinerators.

Introduction In combustion processes, polychlorinated dibenzo-p-dioxin (PCDD) and dibenzofuran (PCDF) byproducts are formed * Corresponding author phone: +81-29-861-8229; fax: +81-29861-8228; e-mail: [email protected]. † AIST. ‡ Georgia Institute of Technology. 4398

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

by two general formation pathways: precursor mechanisms and so-called “de novo” synthesis. Precursor pathways involve the formation of PCDD/F congeners from similar structures, especially chlorinated phenols (1-3). Precursor formation can occur under two different conditions: at temperatures between 600 and 800 °C in the gas phase (4) and at temperatures between 200 and 500 °C on ash surfaces (5). Formation by de novo synthesis occurs at temperatures between 200 and 500 °C from carbonaceous material in soot, which releases PCDD/F congeners in the presence of oxygen and catalyst without the gas-phase precursors (6, 7). Chlorinated phenols are known to be important PCDD/F precursors (8-11), and are also among the most abundant aromatic compounds found in municipal waste incinerator (MWI) flue gases (12). PCDD/F formation rates from precursors have been found to be significantly faster than rates from carbonaceous material, or de novo synthesis, under typical incinerator conditions (1, 2, 13). In addition, chlorinated phenols are also implicated as key intermediates in the de novo synthesis (14). Up to now, however, most studies of the phenol condensation pathway have been performed using single chlorinated phenols or mixtures of only a few chlorinated phenols. Many studies have used 2,4,6-trichlorophenol (2,4,6-T3CP), 2,3,4,6-tetrachlorophenol (2,3,4,6-T4CP), and pentachlorophenol (PCP) because these are present in high amounts in municipal waste incinerators and are thermodynamically more stable than other chlorinated phenols with multiple chlorine substituents (14-20). These phenols have been shown to produce the most abundant PCDD congeners found in municipal waste incinerators, which are 1,3,6,8- and 1,3,7,9-tetrachlorodibenzo-p-dioxin (T4CDD), 1,2,4,6,8-, 1,2,4,7,9-, 1,2,3,6,8-, and 1,2,3,7,9-pentachlorodibenzo-p-dioxin (P5CDD), and 1,2,3,4,6,8-hexachlorodibenzo-p-dioxin (H6CDD). Unsubstituted phenol and monochlorinated phenols are typically much more abundant than the 2,4,6-chlorinated phenols in municipal waste incinerators, however (21-25). Unsubstituted phenol is known to be an important precursor in the gas phase for the formation of dibenzofuran (DF) and lower chlorinated PCDFs (26). It has also been shown that unsubstituted phenol can produce PCDD products (27, 28). It is important to understand the role of unsubstituted phenol in PCDD/F congener distributions because it is present in the highest amount in MWI exhaust gas. PCDF isomer patterns at MWIs appear to be consistent with chlorination and/or dechlorination pathways (29-31). However, these pathways are not consistent with PCDD isomer patterns found in MWIs. Little is known about PCDD/F congener distributions from phenol condensation using a complete mixture of 19 chlorinated phenols and unsubstituted phenol in relative amounts typical of incinerator exhaust gases. In this paper, we investigate PCDD/F congener patterns in the gas phase and via particle-mediated reactions using a mixture of 20 phenols with distribution typical of incinerator emissions. Then, we compare PCDD/F congener patterns with those found in municipal waste incinerators. Isomer patterns are not strongly dependent on combustion conditions such as temperature and gas velocity (20, 22, 23, 29-33). That is, PCDD/F isomer patterns are controlled largely by the formation mechanism and distribution of precursors. Complete distributions of PCDD/F congeners provide a rich fingerprint that can be used to identify a dominant formation mechanism. 10.1021/es048224v CCC: $30.25

 2005 American Chemical Society Published on Web 05/10/2005

FIGURE 1. Experimental apparatus.

Methods Experimental Details. The experimental apparatus is shown in Figure 1. Experiments were conducted in an electrically heated, quartz tube flow reactor, 400 mm in length and 17 mm in diameter. A mixture of 20 phenols dissolved in benzene was synthesized to perform this study. Liquid reactants containing the mixture of all phenols were continuously injected into a heated glass vessel (vaporizer) by means of a motorized microsyringe pump (Sage Instruments model 341B, Boston, MA) via a gastight syringe, where they were immediately vaporized in a gas stream of nitrogen (92 vol %) and oxygen (8 vol %) before entering the reactor. The feed rate of phenol and 19 chlorinated phenols was 3.18 × 10-5 mol/min in the gas-phase and particle-mediated experiments. The concentration of benzene was 0.4% in the feed gas. The temperature of the reactor was controlled using an electric heater. Particle-mediated experiments were conducted at 400 °C and gas-phase experiments were performed at 600-700 °C because previous work shows that PCDD/F yields are greatest at these temperatures (20, 34). Sampling was carried out for 4 h. The gas velocity was 2.7 cm/s (0.3 s of organics and particle contact time) for the duration of the particle-mediated experiments. The retention time and velocity were 10 s and 2.7 cm/s, respectively, for the gasphase experiments. For particle-mediated experiments, a 1 g particle bed of 1 cm height, consisting of silicon dioxide (SiO2; 99.6% purity, 325 mesh, Aldrich) and 0.5% (Cu mass) copper chloride (CuCl2; anhydrous, 99.999+% purity, Aldrich) was prepared by mechanical mixing and was located at the center of the reactor. Gas-phase experiments with benzene only produced phenol, but no PCDD/Fs. Particle-mediated experiments with benzene only produced chlorinated benzenes and perchloroethylene, but no PCDD/Fs. All phenols were purchased (Aldrich Chemicals), and a mixture of 20 phenols in benzene was synthesized for this

study. The molar distribution of unsubstituted phenol and chlorinated phenols was obtained from published data (22) in which all phenol distributions were measured at four locations (furnace-out, electrostatic precipitator inlet, electrostatic precipitator outlet, and fly ash samples) from a municipal waste incinerator. The concentrations of chlorinated phenols were very different at each sampling point, but most abundant chlorinated phenols were largely constant, showing high concentrations of 2-chlorophenol (2CP), 4-chlorophenol (4-CP), 2,4-dichlorophenol (2,4-DCP), 2,4,6-T3CP, 2,3,4,6-T4CP, and PCP. In this study, molar concentrations of chlorinated phenols measured at furnaceout were used. Similar phenol distributions are also found in studies of other researchers from MSW combustion (9, 10, 25). We used the same molar ratio for this study except for unsubstituted phenol. The molar concentration of unsubstituted phenol was 100 times higher than the sum of other chlorinated phenols from a municipal waste incinerator (22). In this study, we used 10 times higher concentration of phenol than chlorinated phenols. The molar distribution of 20 phenols used in this study is shown in Figure 2. Total concentrations of phenols used in this study were 10 000 times higher than that found in a municipal waste incinerator (22). While the use of a complete set of phenols in these experiments provides a complete set of dioxin products for comparison with field data, interpretation of the results is limited by the discrepancies between experimental and field conditions that can affect dioxin distributions. The high phenol concentrations, lack of combustion byproducts and fly ash in the gas stream, and simple particle composition in the experiments were necessary to produce dioxin products unambiguously from phenol condensation. The entire product gas stream was rapidly quenched, and samples were collected in an ice-cooled dichloromethane VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4399

FIGURE 2. Molar distribution of unsubstituted phenol and 19 chlorinated phenols used in this study. (DCM) trap. For particle-mediated formation study, the particles were thoroughly vigorously rinsed after each experiment. Gaseous products and rinse solutions were combined together, and then filtered for analysis. For gasphase formation experiments, gaseous products were directly obtained from DCM. Sample solutions were then filtered and analyzed. In all experiments, the quartz tube reactor was replaced with a new one. Sample analysis was performed using GC/MS (HP 6890 series gas chromatograph with a model 5973 mass-selective detector, EI type) equipped with an HP-5MS capillary column with length 30 m, i.d. 0.25 mm, and phase 0.25 µm film of cross-linked 5% phenylmethylsiloxane (J&W Scientific, California). With helium carrier gas, the following GC temperature program was used: 38 °C for 2 min, 3 °C/min to 180 °C and held for 2 min, 5 °C/min to 250 °C and held for 5 min, 6 °C/min to 280 °C and held for 3 min, and 2 °C/min to 300 °C. MS conditions were as follows: resolution, low; ionization energy, 70 eV; ionization current, 300 µA; ion source temperature, 230 °C. For identification the mass spectrometer was operated in the scan mode and for quantification in selected ion mode. Preliminary identification of PCDD/F products was based on published relative retention times for similar columns (35, 36). Final identification was based on available standards and gas-phase synthesis experiments from single precursors and precursor pairs. A total of 74 of 75 PCDD congeners and 121 of 135 PCDF congeners have been synthesized from phenol pairs. All 4 monochlorodibenzofuran (MCDF) isomers were separated, as were 13 of 16 dichlorodibenzofurans (DCDFs), 25 of 28 trichlorodibenzofurans (T3CDFs), 34 of 38 tetrachlorodibenzofurans (T4CDFs), 22 of 28 pentachlorodibenzofurans (P5CDFs), 12 of 16 hexachlorodibenzofurans (H6CDFs), 2 of 4 heptachlorodibenzofurans (H7CDFs), and octachlorodibenzofuran (OCDF). CDD isomer peak separation was as follows: 2 peaks/2 monochlorodibenzo-p-dioxin (MCDD) isomers, 7 peaks/10 dichlorodibenzo-p-dioxin (DCDD) isomers, 10 peaks/14 trichlorodibenzo-p-dioxin (T3CDD) isomers, 15 peaks/22 T4CDD isomers, 12 peaks/14 P5CDD isomers, 7 peaks/10 H6CDD isomers, 2 peaks/2 heptachlorodibenzo-p-dioxin (H7CDD) isomers. For quan4400

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

tification of DD (dibenzo-p-dioxin), DF, and PCDD/Fs, DD and DF were used as external standards, respectively. Municipal Solid Waste Incineration Data. To assess the potential role of phenol condensation pathways in PCDD/F formation, PCDD/F isomer fractions were obtained from municipal solid waste incinerators, regrouped due to different analytical methods (different columns), and compared with PCDD/F isomer distributions observed from gas-phase and particle-mediated phenol condensation experiments. PCDD/F Data in MWI Fly Ash Samples. MWI data were obtained for fly ash samples from eight incinerators in Japan, as described elsewhere (30, 31). Four of these were fluidized bed incinerators, and four were stoker-type incinerators. While PCDD/F homologue distributions varied very widely, the distributions of congeners were very similar, as shown elsewhere (30, 31). PCDD/F homologue and isomer distributions were compared with those obtained by phenol experiments. PCDD/F Data in Flue Gas Samples. MWI data were obtained from the flue gas of a stoker-type MWI in Norfolk, VA. The flue gas samples were collected prior to the gas cleaning system. The 13 test conditions are described in detail elsewhere (37). While total PCDD/F emissions and homologue profiles varied widely, the distributions of PCDD/F isomers were very similar (29). PCDD/F homologue and isomer distributions were compared with those obtained by phenol experiments. In the MWI tests, the effect of cofiring coal with municipal waste on PCDD/F formation was studied. High and low sulfur coals were used. As previously reported (37), total PCDD/F emissions and the PCDD-to-PCDF ratio were affected by the operating parameters. Here, we examine the distributions of homologues using data from all 13 tests, and the distributions of isomers using data from 9 of the 13 tests. Samples from the other four tests were not available for isomer-specific analysis. The 13 test conditions are described in detail elsewhere (37).

Results and Discussion PCDD/F Yields. Yields of PCDD/Fs from gas-phase and particle-mediated experiments are presented in Figure 3,

FIGURE 3. Yields of DD/DF and PCDD/Fs. Error bars represent (1 standard deviation. expressed in units of total percent phenol conversion to PCDD/Fs. The total PCDD/F concentration measured in MWI flue gas samples is also shown. The average and (1 standard deviation are shown from replicated experiments and measurements of 13 test conditions. DD and DF were not measured in the MWI flue gas samples. From gas-phase experiments, average yields of DF and PCDFs were 1.9% and 0.13% and average yields of DD and PCDD were 0.000038% and 0.00071%. From particle-mediated experiments, average yields of DF and PCDFs were 0.74% and 0.080% and average yields of DD and PCDDs were 0.029% and 0.031%. Comparison with MWI data shows that the PCDF/PCDD ratio in MWI flue gas is more like that of the particle-mediated experiments. The average PCDF concentration in MWI flue gas was 5 or 6 times higher than the PCDD concentration. The PCDF/PCDD molar ratio was about 200 and 3 in the gas-phase and particle-mediated experiments, respectively. DF yields in the two experimental conditions were greater than total PCDF yields, and DD yields in the two experiments were less than total PCDD yields, although only slightly less in particle-mediated formation. DF was most abundant from both experimental conditions. Several researchers have found the DF concentration to be 100-200 times higher than the sum of total PCDFs in MWI samples (22, 23, 25). In this study, the DF concentration was about 10 times higher than the sum of PCDFs. The concentration of unchlorinated phenol was 10 times higher than the sum of other chlorinated phenols in this study. The concentration of unchlorinated phenol in the MWI flue gases is typically 30-100 times higher than that of chlorinated phenols (22, 23, 25). Recently, Wikstrom et al. (38) performed experiments with soot deposit as a carbon source for PCDD/F formation (de novo synthesis), and reported that PCDD/Fs are produced but DF was below the detection limit. These results support previous evidence (26) that unchlorinated phenol condensation plays an important role in the formation of DF. DD and PCDD formation was greater in the particlemediated experiments at 400 °C, whereas formation of DF and PCDFs was greater in the gas-phase experiments at 600700 °C. DF formation in the gas-phase experiments was about 3 times higher than in the particle-mediated experiments at the same reactant concentration. PCDF formation was about 2 times higher in the gas-phase experiments than in the particle-mediated experiments. In the gas phase, phenol was formed from benzene, used as a solvent, contributing to DF formation, consistent with previous results (34). In the particle-mediated experiments, benzene reacted to produce chlorinated benzenes and perchloroethylene, but not phenol, consistent with previous results (27). For the formation of DF and PCDF, one ortho site without chlorine is needed on each phenoxy radical to have carbon-carbon coupling,

followed by enolization and water elimination. In the gas phase, chlorinated phenols lose chlorine at temperatures between 500 and 700 °C, preferentially at ortho sites, resulting in formation of more phenol precursors without o-chlorine. Thus, formation of dibenzofuran and less chlorinated PCDD/F congeners is favored by gas-phase phenol condensation (25, 39, 40). DD and PCDD formation in the particle-mediated experiments was 40-700 times higher than in the gas-phase experiments. Here, phenols are chlorinated on particle surfaces, preferentially at ortho sites. PCDD formation by phenol condensation is important in particlemediated processes (41), but not gas-phase processes. PCDD/F Homologue Patterns. Figure 4 shows the homologue fraction of PCDFs (top) and PCDDs (bottom) formed from phenols in the gas phase and via particle-mediated reaction. Homologue fractions found in MWI flue gas and fly ash samples are also shown. Complete homologue data were not measured at the MWIs. That is, DD and DF were not measured in the MWI flue gas samples, and DD/DF through T3CDD/F congeners were not measured in the MWI fly ash samples. PCDF homologue distributions are similar for both the gas-phase and particle-mediated phenol condensation experiments. The DF fraction was 0.95 and 0.90 for gas-phase and particle-mediated formation, respectively. Wikstrom and Marklund (23) have shown the DF fraction was more than 99% in municipal waste combustion. The PCDF homologue fraction decreased with increasing number of chlorine substituents in both gas-phase and particle-mediated experiments. This is not the case for the MWI samples. Homologue fractions found in flue gas samples show that T3CDF is maximum, and then decreased with increasing number of chlorine substituents. The DD fraction was 0.49 in the particle-mediated experiments, and the PCDD homologue fraction also decreased with increasing number of chlorine substituents. However, the PCDD homologue fraction increases with increasing number of chlorine substituents in the MWI samples. The particle-mediated PCDD/F products contained more chlorine than the gas-phase products in the phenol condensation experiments. In both phenol condensation experiments as well as both sets of MWI samples, PCDD products contain more chlorine than PCDF products. Unsubstituted phenol, which was present in high concentration, played a significant role in the distributions of PCDD/F homologues in the phenol condensation experiments. Homologue patterns are sensitive to combustion conditions, such as a chlorine/hydrogen ratio and temperature (42). PCDD/F Isomer Patterns. To further assess the role of phenol condensation pathways in PCDD/F formation, PCDD/F isomer distributions are compared to those obtained from MWI flue gas and fly ash samples. Isomer distributions are much less variable than homologue distributions with varying combustion conditions. Figures5 and 6 show PCDF and PCDD isomer distributions from MWI flue gas, MWI fly ash, and gas-phase and particlemediated phenol condensation experiments. Data were not available for mono- through trichlorinated isomers in MWI fly ash samples. Isomer groupings were developed on the basis of combinations of coeluting PCDD/F isomers from researchers using different analytical methods. Means and standard deviations are given for nine measurements of MWI flue gas samples, eight measurements of MWI fly ash samples, three measurements of formation via particle-mediated phenol condensation, and three measurements of formation via gas-phase phenol condensation. PCDF Isomer Patterns. In four isomer sets (tetra through hepta), the MWI flue gas and MWI fly ash sample patterns are very similar. MCDF through T3CDF isomer data were not VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4401

FIGURE 4. PCDD/F homologue patterns formed from phenols. Error bars represent (1 standard deviation. available from MWI fly ash samples. However, PCDF isomer patterns formed from phenol condensation in the gas-phase and particle-mediated formation were different from those observed from MWIs. From phenol condensation experiments, the largest MCDF and DCDF isomers were 2-MCDF and 2,4-DCDF. 1,2,4-T3CDF and 1,2,3,4-T4CDF were the largest isomers of the tri- and tetrachlorinated congeners. The dominance of 2,4-DCDF, 1,2,4-T3CDF, and 1,2,3,4-T4CDF demonstrates the importance of unchlorinated phenol in the formation of PCDFs by the phenol condensation pathway. Mono- and dichlorinated phenols, particularly the 2-, 4-, and 2,4-congeners, which are present in high amounts, play an important role in the distributions of P5CDF and H6CDF congeners. On the other hand, the MWI distributions do not show a dominance of particular PCDF isomers. In the case of the H6CDF and H7CDF distributions, the 1,9substituted PCDF isomers are formed in low amounts relative to other isomers. The distributions of PCDF isomers formed by phenol condensation depend on the distribution of phenols. Using a typical distribution of phenols found in municipal waste incinerators, the PCDF isomer distributions from the phenol condensation pathway were largely different from those observed in MWI flue gas and fly ash samples. We conclude that the phenol condensation pathway is not a dominant route of PCDF formation in MWIs. Gas-phase phenol condensation is a source of DF, with subsequent particlemediated chlorination possibly resulting in PCDF formation (23, 29). Recently, Tuppurainen et al. (11) reviewed PCDD/F formation pathways from MWIs and other combustion processes and speculated that the chlorination pathway is 4402

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

the main pathway of PCDF formation. De novo pathways may also be important in PCDF formation (14, 43). PCDD Isomer Patterns. In four isomer sets (tetra through hepta), isomer patterns of the MWI flue gas and MWI fly ash samples are similar. PCDD formation in the gas-phase experiments was very small, and a limited number of PCDD isomers were observed (not shown here). While gas-phase PCDD formation from phenols is not an important process, particle-mediated PCDD formation may be (44). We compare the PCDD isomer distributions produced from particlemediated experiments with those obtained in MWIs. In addition, an experiment was performed with a mixture of 10 phenols in the particle-mediated formation at 400 °C. This mixture includes the 10 most abundant phenols with two or more chlorine atoms such as 2,6-dichlorophenol, 2,4,6trichlorophenol, 2,3,4,6-tetrachlorophenol, and pentachlorophenol. The distributions of 10 phenols were obtained from published data (8). The 1,3-DCDD isomer was most abundant in the phenol condensation experiments, whereas the 2,7/2,3/2,8-DCDD peak (coeluting isomers) was the largest in the MWI samples. The two largest T3CDD isomer peaks were 1,3,7/1,3,8- and 1,2,4/2,3,7/1,3,9/1,2,6-T3CDD isomers from the MWI samples, whereas the 1,2,3- and 1,2,4/2,3,7/1,3,9/1,2,6-T3CDD isomers were most abundant from the phenol condensation pathway. In the MWI samples, about 81-100% of this isomer group was 2,3,7-T3CDD; in the phenol condensation experiments, about 97% was 1,2,4-T3CDD. The major mono- through trichloro congeners found in MWI samples are 2-MCDD, 2,7/2,3/2,8-DCDD, and 2,3,7-T3CDD (45). The 1,3,6,8- and 1,3,7,9-T4CDD isomers are the most abundant in MWI

FIGURE 5. Comparison of PCDF isomers. Error bars represent (1 standard deviation. samples. The most abundant P5CDD isomers produced in MWIs are as follows: 1,2,4,6,8-/1,2,4,7,9-, 1,2,3,6,8-, and 1,2,3,7,9-P5CDD. In the phenol condensation experiments, on the other hand, the 1,2,6,9/1,2,3,4/1,2,3,6-T4CDD peak (97% of this group) and the 1,2,3,4,7-P5CDD isomer (98% of this group) were most abundant. As with PCDFs, the MWI samples have a broader distribution of PCDD products. The isomer patterns clearly demonstrate the importance of unchlorinated phenol in the formation of PCDDs by the phenol condensation pathway. Previous studies (27, 46) have shown that only one o-chlorine

is needed to produce PCDD products. Thus, unchlorinated phenol can contribute to PCDD formation, and this appears to be a significant path in the phenol condensation mechanism. PCDD isomer distributions obtained from particle-mediated phenol condensation, using the 10 most abundant chlorinated phenols with two or more chlorine substituents show many similarities to the MWI distributions. The most abundant T4CDD isomers produced from a mixture of 10 chlorinated phenols were the 1,3,6,8- and 1,3,7,9-T4CDD isomers. This isomer pair is produced by condensation of VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4403

FIGURE 6. Comparison of PCDD isomers. Error bars represent (1 standard deviation. two 2,4,6-T3CPs. The most abundant P5CDD isomers produced were 1,2,4,6,8-/1,2,4,7,9-, 1,2,3,6,8-, and 1,2,3,7,9-P5CDD. These four isomers are produced by condensation of 2,4,6-T3CP and 2,3,4,6-T4CP. In the case of the H6CDD isomer distribution, the two largest peaks are for the 1,2,4,6,7,9-/ 1,2,4,6,8,9-/1,2,3,4,6,8-H6CDD isomer group and the 1,2,3,6, 7,9-/1,2,3,6,8,9-H6CDD isomer. Four of these isomers are produced by condensation of two 2,3,4,6-T4CPs, and the 1,2,3,4,6,8-isomer is produced by condensation of 2,4,6-T3CP and PCP. Without involvement of unsubstituted phenol and lower chlorinated phenols, the PCDD isomer distributions for tetra- through hexachloro homologues from phenol 4404

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

condensation are consistent with those observed in MWI samples. Results presented in this paper show that phenol and less chlorinated phenols, which are abundant in MWI exhaust gas, such as 2-CP, 4-CP, and 2,4-DCP, play a significant role in the distributions of PCDD/F congeners formed by phenol condensation. The PCDD/F distributions produced by phenol condensation are not consistent with the PCDD/F distributions observed in MWI samples. The phenol condensation pathway may contribute, however, to PCDD/F formation. In the case of PCDF formation, gas-phase formation of DF by phenol condensation may lead to particle-mediated forma-

tion of PCDFs by chlorination. In the case of PCDD formation, MWI samples show relatively high concentrations of isomers chlorinated at alternating sites (e.g., 1,3,6,8- and 1,3,7,9-T4CDDs and 1,2,4,6,8- and 1,2,4,7,9-P5CDDs). These congeners are produced from 2,4,6-trichlorophenol and 2,3,4,6-tetrachlorophenol, which are among the most abundant multichlorinated phenols in MWI exhaust. PCDD/F congener patterns obtained from phenol precursors in this study provide a first attempt at using full distributions as a diagnostic of the dioxin formation mechanism in municipal waste incinerators. To be used more effectively, more information is needed on the PCDD/F congener patterns from other mechanisms. Results presented here on PCDD/F congener patterns from phenol condensation are limited to complex waste feed streams that result in a complete distribution of phenols. This approach is limited to the degree that PCDD/F congener patterns are affected by combustion conditions and to the degree to which multiple PCDD/F formation mechanisms contribute.

Acknowledgments J.-Y.R.’s work was supported by the United States Environmental Protection Agency under Grant QT-OH-99-000537 and National Science Foundation under Grant CTS-0210089 and in part by the New Energy and Industrial Technology Development Organization (NEDO) in Japan. We are thankful for the time and comments provided by anonymous reviewers for improving this paper.

Literature Cited (1) Altwicker, E. R.; Milligan, M. S. Formation of dioxins: competing rates between chemically similar precursors and de novo reactions. Chemosphere 1993, 27, 301-307. (2) Luijk, R.; Akkerman, D.; Slot, P.; Olie, K.; Kepteijn, F. Mechanism of formation of polychlorinated dibenzo-p-dioxins and dibenzofurans in the catalyzed combustion of carbon. Environ. Sci. Technol. 1994, 28, 312-321. (3) Altwicker, E. R. Relative rates of formation of polychlorinated dioxins and furans from precursors and de novo reactions. Chemosphere 1996, 33, 1897-1904. (4) Ballschmitter, K.; Zoller, W.; Buchert, H.; Class, Th. Correlation between substitution pattern and reaction pathway in the formation of polychlorodibenzofurans. Fresenius’ Z. Anal. Chem. 1985, 322, 587-594. (5) Dickson, L. C.; Lenoir, D.; Hutzinger, O. Surface-catalyzed formation of chlorinated dibenzodioxins and dibenzofurans during incineration. Chemosphere 1989, 19, 277-282. (6) Stieglitz, L.; Vogg, H. On formation conditions of PCDD/PCDF in fly ash from municipal waste incinerators. Chemosphere 1987, 16, 1917-1922. (7) Huang, H.; Buekens, A. On the mechanisms of dioxin formation in combustion processes. Chemosphere 1995, 31, 4099-4117. (8) Ballschmitter, K.; Braunmiller, I.; Niemczyk, R.; Swerev, M. Reaction pathways for the formation of polychloro-dibenzodioxins (PCDD) and -dibenzofurans (PCDF) in combustion processes: II. chlorobenzenes and chlorophenols as precursors in the formation of polychloro-dibenzodioxins and -dibenzofurans in flame chemistry. Chemosphere 1988, 17, 995-1005. (9) Tuppurainen, K.; Marjaleena Aatamila.; Ruokojarvi, P.; Halonen, I.; Ruuskanen, J. Effect of liquid inhibitors on PCDD/F formation. Prediction of particle-phase PCDD/F concentrations using PLS modeling with gas-phase chlorophenol concentrations as independent variables. Chemosphere 1999, 38, 2205-2217. (10) Tuppurainen, K.; Ruokojarvi, P.; Asikainen, A.; Marjaleena Aatamila.; Ruuskanen, J. Chlorophenols as precursors of PCDD/ Fs in incineration processes: correlations, PLS modeling, and reaction mechanisms. Environ. Sci. Technol. 2000, 34, 49584962. (11) Tuppurainen, K.; Asikainen, A.; Ruokojarvi, P.; Ruuskanen, J. Perspectives on the formation of polychlorinated dibenzo-pdioxins and dibenzofurans during municipal solid waste (MSW) incineration and other combustion processes. Acc. Chem. Res. 2003, 36, 652-658. (12) Dempsey C. R.; Oppelt, E. T. Incineration of Hazardous Waste: A critical review update. Air Waste 1993, 43, 25-73.

(13) Dickson, L. C.; Lenoir, D.; Hutzinger, O. Quantitative comparison of de novo and precursors formation of polychlorinated dibenzop-dioxins under simulated municipal solid waste incinerator postcombustion conditions. Environ. Sci. Technol. 1992, 26, 1822-1828. (14) Hell, K.; Stieglitz, L.; Altwicker, E. R.; Addink, R.; Will R. Reactions of 2,4,6-trichlorophenol on model fly ash: oxidation to CO and CO2, condensation to PCDD/F and conversion into related compounds. Chemosphere 2001, 42, 697-702. (15) Born, J. G. P.; Mulder, P.; Louw, R. Fly ash mediated reactions of phenol and mono-chlorophenols: oxychlorination, deep oxidation, and condensation. Environ. Sci. Technol. 1993, 27, 1849-1863. (16) Addink, R.; Cnubben, P. A. J. P.; Olie, K. Formation of polychlorinated dibenzo-p-dioxins/dibenzofurnas on fly ash from precursors and carbon model compounds. Carbon 1995, 33, 1463-1471. (17) Sidhu, S. S.; Maqsud, L.; Dellinger, B.; Mascolo, G. The homogeneous, gas-phase formation of chlorinated and brominated dibenzo-p-dioxin from 2,4,6-trichloro- and 2,4,6tribromophenols. Combust. Flame 1995, 100, 11-20. (18) Cains, P. W.; Mccausland, L.; Fernandes, A. R.; Dyke, P. Polychlorinated dibenzo-p-dioxins and dibenzofurans formation in incineration: effects of fly ash and carbon source. Environ. Sci. Technol. 1997, 31, 776-785. (19) Hell, K.; Altwicker, E. R.; Stieglitz, L.; Addink, R. Comparison of 2,4,6-trichlorophenol conversion to PCDD/PCDF on a MSWIfly ash and a model fly ash. Chemosphere 2000, 40, 995-1001. (20) Mulholland, J. A.; Ryu, J.-Y. Formation of polychlorinated dibenzo-p-dioxins by CuCl2-catalyzed condensation of 2,6 chlorinated phenols. Combust. Sci. Technol. 2001, 169, 107126. (21) Jay, K.; Stieglitz, L. Identification and quantification of volatile organic components in emissions of waste incineration plants. Chemosphere 1995, 30, 1249-1260. (22) Weber, R.; Hagenmaier, H. PCDD/PCDF formation in fluidized bed incineration. Chemosphere 1999, 38, 2643-2654. (23) Wikstrom, E.; Marklund, S. Secondary formation of chlorinated dibenzo-p-dioxins, dibenzofurans, biphenyls, benzenes and phenols during MSW combustion. Environ. Sci. Technol. 2000, 34, 604-609. (24) Xie, Y.; Xie, W.; Liu, K.; Dickson, L.; Pan, W.-P.; Riley, J. T. The effect of sulfur dioxide on the formation of molecular chlorine during co-combustion of fuels. Energy Fuels 2000, 14, 597-602. (25) Zimmermann, R.; Blumenstock, M.; Heger, H. J.; Schramm, K.W.; Kettrup, A. Emission of nonchlorinated and chlorinated aromatics in the flue gas of incineration plants during and after transient disturbances of combustion conditions: delayed emission effects. Environ. Sci. Technol. 2001, 35, 1019-1030. (26) Ryu, J.-Y.; Mulholland, J. A.; Oh, J.-E.; Nakahata, D.-T.; Kim, D.-H. Prediction of polychlorinated dibenzofuran congener distribution from gas-phase phenol condensation pathways. Chemosphere 2004, 55, 1447-1455. (27) Ryu, J.-Y.; Mulholland, J. A. Metal-mediated chlorinated dibenzopdioxin (CDD) and dibenzofuran (CDF) formation from phenols. Chemosphere 2005, 58, 977-988. (28) Wiater, I.; Louw, R. Dibenzofuran and dibenzodioxin patterns from (chloro)phenols in combustion reactions. Organohalogen Compd. 1999, 41, 77-81. (29) Ryu, J.-Y.; Mulholland, J. A.; Dunn, J. E.; Iino, F.; Gullett, B. K. Potential role of chlorination pathways in PCDD/F formation in a municipal waste incinerator. Environ. Sci. Technol. 2004, 38, 5112-5119. (30) Iino, F.; Imagawa, T.; Gullett, B. Dechlorination-controlled polychlorinated dibenzo-furan isomer patterns from municipal waste incinerators. Environ. Sci. Technol. 2000, 34, 3143-3147. (31) Iino, F.; Tsuchiya, K.; Imagawa, T.; Gullett, B. An isomer prediction model based on dechlorination kinetics for polychlorinated naphthalenes, polychlorinated dibenzo-p-dioxins, and polychlorinated biphenyls from municipal waste incinerators. Environ. Sci. Technol. 2001, 35, 3175-3181. (32) Schramm, K. W.; Wehrmeier, A.; Lenoir, D.; Henkelmann, B.; Hahn, K.; Zimmermann, R.; Kettrup, A. Deduction of a typical pattern for the isomers of tetra- to heptachloro-dibenzo-pdioxins and -furans formed in combustion processes. Organohalogen Compd. 1996, 27, 196-200. (33) Wehrmeier, A.; Lenoir, D.; Schramm, K.-W.; Zimmermann, R.; Hahn, K.; Henkelmann, B.; Kettrup, A. Patterns of isomers of chlorinated dibenzo-p-dioxins as tool for elucidation of thermal formation mechanism. Chemosphere 1998, 36, 2775-2801. VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4405

(34) Nakahata, D.-T.; Mulholland, J. A. Effect of dichlorophenol substitution pattern on furan and dioxin formation. Proc. Combust. Inst. 2000, 28, 2701-2707. (35) Hale, M. D.; Hileman, F. D.; Mazer, T.; Shell, T. L.; Noble, R. W.; Brooks, J. J. Mathematical modeling of temperature programmed capillary gas chromatography retention indexes for PCDFs. Anal. Chem. 1985, 57, 640-648. (36) Ryan, J. J.; Conacher, H. B. S.; Panopio, L. G.; Lau, B. P.-Y.; Hardy, J. A. Gas chromatographic separations of all 136 tetrato octa-polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans on nine different stationary phases. J. Chromatogr. 1991, 541, 131-183. (37) Gullett, B. K.; Dunn, J. E.; Raghunathan, K. Effect of co-firing coal on formation of poly-chlorinated dibenzo-p-dioxins and dibenzofurans during waste combustion. Environ. Sci. Technol. 2000, 34, 282-290. (38) Wikstrom, E.; Ryan, S.; Touati, A.; Gullett, B. K. In situ formed soot deposit as a carbon source for polychlorinated dibenzop-dioxins and dibenzofurans. Environ. Sci. Technol. 2004, 38, 2097-2101. (39) Weber, R.; Sakurai, T. Formation characteristics of PCDD and PCDF during pyrolysis processes. Chemosphere 2001, 45, 11111117. (40) Yasuhara, A.; Katami, T.; Shibamoto, T. Formation of PCDDs, PCDFs, and coplanar PCBs from incineration of various woods

4406

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

(41) (42)

(43)

(44) (45)

(46)

in the presence of chlorides. Environ. Sci. Technol. 2003, 37, 1563-1567. Stieglitz, L. Selected topics on the de novo synthesis of PCDD/ PCDF on fly ash. Environ. Eng. Sci. 1998, 15, 5-18. Iino, F. Formation mechanisms of dioxins from municipal waste incinerators. Ph.D. Dissertation, University of Tokyo, Japan, 1999. Hell, K.; Stieglitz, L.; Zwick, G.; Will R. Mechanistic aspects of the de-novo synthesis of PCDD/PCDF on model fly ash. Organohalogen Compd. 1997, 31, 492-496. Shaub, W. M.; Tsang, W. Dioxin formation in incinerators. Environ. Sci. Technol. 1983, 17, 721-730. Ryu, J.-Y.; Mulholland, J. A.; Chu, B. Chlorination of dibenzofuran and dibenzo-p-dioxin vapor by copper (II) chloride. Chemosphere 2003, 51, 1031-1039. Ryu, J.-Y.; Mulholland, J. A. Dioxin and furan formation on CuCl2 from chlorinated phenols with one ortho chlorine. Proc. Combust. Inst. 2002, 29, 2455-2461.

Received for review November 12, 2004. Revised manuscript received March 29, 2005. Accepted March 30, 2005. ES048224V