Polychlorinated Benzene and Polychlorinated Phenol in

high production of radicals leads to the dramatic increase in ClxBz formation. .... compounds as possible intermediates in the ultimate formation ...
0 downloads 0 Views 203KB Size
Environ. Sci. Technol. 1996, 30, 1009-1013

Polychlorinated Benzene and Polychlorinated Phenol in Heterogeneous Combustion Reactions of Ethylene and Ethane KENNETH L. FROESE† AND OTTO HUTZINGER* Ecological Chemistry and Geochemistry, University of Bayreuth, D-95440 Bayreuth, Germany

Ethylene and ethane were reacted with HCl/air under heterogeneous combustion conditions between 300 and 600 °C. Model catalyst mixtures of SiO2, SiO2/ Al2O3, and SiO2/CuO were compared with municipal waste incinerator (MWI) fly ash. Chlorinated benzenes (ClxBz) and chlorinated phenols (ClxPh) were detected in both gas-phase products and catalystadsorbed products. In C2H4 reactions over fly ash, the amount of ClxBz produced increased to 600 °C (Cl5Bz: 1.6 × 104 ng/g C2H4; HCB: 6 × 103ng/g C2H4), while ClxPh (PCP) production remained constant from 400 to 600 °C (1.5 × 103ng/g of C2H4). For C2H6 reactions at 600 °C, ClxBz production was very similar to the C2H4 reaction (Cl5Bz: 1.6 × 104 ng/g C2H6; HCB: 5 × 103ng/g C2H6), whereas total ClxPh (PCP) production was an order of magnitude greater (1.2 × 104ng/g C2H6). The CuO-catalyzed gas-phase ClxBz patterns most closely matched those produced in the reactions with MWI fly ash. The Al2O3-catalyzed reactions distinguished themselves from the others in unique ClxBz and ClxPh congener patterns that were dominated by the dichloro homologue group and in the observation of non-chorinated aromatic reaction products, in contrast to a predominance of chlorinated aliphatic compounds in the CuO and fly ash-catalyzed reactions.

Introduction Combustion processes are one of the primary sources of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/F) and related persistent organic contaminants in the environment (1-4). Investigations and discussions regarding the mechanisms of their formation have revolved primarily around the theories of de novo synthesis from elemental carbon and formation from aromatic precursors such as pentachlorophenol (5-11). The potential importance of C2 aliphatic units in the high temperature formation of aromatic compounds has † Present address: Aquatic Toxicology, Pesticide Research Center, Michigan State University, East Lansing, MI 48824.

0013-936X/96/0930-1009$12.00/0

 1996 American Chemical Society

been postulated by a number of investigators (1, 4, 12, 13). In gas-phase pyrolysis reactions of ethylene, Tirey et al. (14) observed the production of aromatic compounds. Investigations of heterogeneous fly ash and Cu-catalyzed reactions of ethylene have been reported (15-17). Vinyl chloride and 1,2-dichoroethane were reported as primary reaction products; in one study, chlorinated benzenes were qualitatively identified in 500 °C reactions of C2H4 using Cu as a catalyst (15). Heterogeneous combustion reactions of acetylene (18) established the relevance of this type of reaction for the formation of simple, halogenated aromatics. By extending the investigation to ethylene and ethane, we intend to determine whether similar formation of chlorinated benzenes and chlorinated phenols occurs. Obtaining reaction data for all of the non-chlorinated C2 aliphatics will ultimately provide a valuable portion of the information needed to develop combustion systems in which the formation of PCDD/F and dioxin-type compounds is avoided. Our goal in this paper is to present experimental data that further establishes the importance of short-chain aliphatics as precursors for the formation of aromatic compounds in post-combustion catalytic processes. We do not intend to determine mechanisms for the formation of PCDD/F from short-chain aliphatics. Rather, we will present experimental data for the formation of simple chlorinated aromatics and focus on mechanistic aspects of this process in a future paper.

Experimental Section The combustion apparatus and experimental procedure have been described in detail elsewhere (12, 18). Tables 1 and 2 provide experimental conditions for ethylene and ethane combustion reactions.

Results Temperature Dependence. The rate of production of ClxBz increased with temperature in reactions of C2H4 with HCl over fly ash (Figure 1a). There was essentially no ClxBz formed at 300 °C, and a minimum appeared again at 500 °C. At 300 °C, the production of radicals was likely too low for ClxBz formation. An explanation for the apparently anomalous results at 500 °C cannot be provided without further investigation on this phenomenon. At 600 °C, perhaps a high production of radicals leads to the dramatic increase in ClxBz formation. Comparing the 400 and 600 °C patterns, relative ratios of certain isomers were reversed. For example, at 400 °C, 1,2,3-Cl3Bz was the main Cl3Bz peak, whereas 1,2,4-Cl3Bz was not detected; also, the production of HCB was greater than that of PCBz. At 600 °C, 1,2,4-Cl3Bz is now greater than 1,2,3-Cl3Bz and PCBz is now greater than HCB. A number of factors may lead to the dominance of one aromatic formation route over another, resulting in these observations: different radical species formed at the different temperatures, varying degrees of surface-reagent interaction, and perhaps competing electrophilic and radical chlorination mechanisms. However, insufficient information is available for proper evaluation of such mechanisms under the heterogeneous conditions used here.

VOL. 30, NO. 3, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1009

1010

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 3, 1996

1.2 0.9 1.0 1.0 1.1 1.2 1.0 1.1 1.2 3.5 1.2

fly ash 300 fly ash 400 fly ash 500 fly ash 500b fly ash 600a fly ash 600b SiO2 600 SiO2/Al2O3 600 SiO2/CuO 600a SiO2/CuO 600b* SiO2/CuO 600c

2.9 2.4 2.5 2.5 2.8 2.9 2.5 2.6 3.0 8.8 3.0

C2H4 (mMol/h) 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0

(mL/min)

HCl 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5

(mMol/h) 55 55 55 55 55 55 54 54 56 56 56

air (mL/min) 27 27 27 27 27 27 27 27 28 28 28

O2 (air) (mMol/h) 110 110 110 110 110 110 110 110 110 110 110

N2 (air) (mMol/h) 300 400 500 500 600 600 600 600 600 600 600

temp (°C) 30 30 30 30 30 30 30 30 30 30 30

rxn time (min) 1.4 1.2 1.3 1.3 1.4 1.4 1.3 1.3 1.5 4.4 1.5

40 33 35 35 39 40 35 37 42 120 42

total C2H4 (mMol) (mg)

a

0.9 0.9 1.2 1.0 0.9 0.9 0.9 0.9 0.9 1.0

2.3 2.3 3.0 2.5 2.3 2.3 2.3 2.3 2.3 2.5

a-d indicate reaction repetition number.

fly ash 600a fly ash 600b fly ash 600c fly ash 600d SiO2 600 SiO2/Al2O3 600 SiO2/CuO 600a SiO2/CuO 600b SiO2/CuO 600c SiO2/CuO 600d

catalysta

C2H6 (mL/min) (mMol/h) 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0

(mL/min)

HCl 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5

(mMol/h) 55 55 57 57 56 55 55 55 56 56

air (mL/min) 27 27 28 28 28 28 27 28 28 28

O2 (air) (mMol/h)

conditions

110 110 110 110 110 110 110 110 110 110

N2 (air) (mMol/h)

600 600 600 600 600 600 600 600 600 600

temp (°C)

30 30 30 30 30 30 30 30 30 30

rxn time (min)

Experimental Conditions for Heterogeneous Combustion Reactions of Ethane with Various Catalysts

TABLE 2

1.1 1.1 1.5 1.3 1.1 1.1 1.1 1.1 1.1 1.3

34 34 45 38 34 34 34 34 34 38

total C2H6 (mMol) (mg)

2 2 2 2 2.5 2.5 2.5 2.5 2.5 2.5

2.5 2.5 2.5 2.5 2.5 2.5 2.4 2.4 2.4 2.4 2.4

free vol (mL)

rxn res time

2.6 2.6 2.6 2.6 2.7 2.7 2.5 2.5 2.4 2.4 2.4

res time (s)

110 130 150 150 170 170 170 170 180 180 170

flow (mL/min)

1.4 1.2 1.0 1.0 0.9 0.9 0.8 0.9 0.8 0.8 0.8

res time (s)

temp correcteda

2.5 2.5 2.5 2.5 2.4 2.4 2.4 2.4 2.4 2.4

free vol (mL)

rxn res time

2.6 2.6 2.5 2.6 2.4 2.4 2.4 2.4 2.4 2.4

res time (s)

170 170 180 180 170 170 170 170 180 180

flow (mL/min)

0.9 0.9 0.9 0.9 0.8 0.8 0.8 0.8 0.8 0.8

res time (s)

temp corrected

a-c indicate reaction repetition number. An asterisk

catalyst (cm)

b

2 2 2 2 2 2 2.5 2.5 2.5 2.5 2.5

catalyst (cm)

a Calculations of the temperature corrected total gas flow and the resulting residence time in the reaction tube were based on the free volume in the reaction tube. (*) indicates that the concentration of C2H4 was significantly higher than in run SiO2/CuO 600a; therefore, the reaction was repeated (SiO2/CuO 600c).

(mL/min)

catalystb

conditions

Experimental Conditions for Heterogeneous Combustion Reactions of Ethylene with Various Catalysts

TABLE 1

FIGURE 1. (a) Temperature dependence of total ClxBz production in the reaction of ethylene/HCl with fly ash (dhx¯(500°C) ) 21%, n ) 2; dhx¯(600°C) ) 8%, n ) 2. (b) Temperature dependence of total ClxPh formation in the reaction of ethylene/HCl over fly ash (dhx¯(500°C) ) 70%, n ) 2; dhx¯(600°C) ) 24%, n ) 2).

A definitive temperature effect was not apparent for the production of ClxPh in the reaction of ethylene/HCl over fly ash (Figure 1b). With the exception of PCP, chlorophenols were not formed in the 300 °C sample. At 500 °C, the 2,4,6-Cl3Ph isomer dominated the reaction products. An inverse temperature relationship was observed for 2,4-Cl2Ph. Catalyst Effects. To determine specific effects of each catalyst, the ClxBz results of the reactions of ethylene/HCl over each catalyst at 600 °C are presented in Figure 2. ClxBz congeners were not detected in the reactions of ethylene/ HCl over pure SiO2, thus it is not included in the figure. The Al2O3 reaction showed high intensity for the dichloro isomers, decreasing for the higher substituted homologues. Comparison of the production amounts revealed that the Cl2Bz congeners were nearly equal in the Al2O3 and fly ash reactions and a factor of 4-5 greater than in the CuO reaction. For Cl3-6Bz, fly ash was 5-10 times greater, with Cl5Bz in particular dominating the congener pattern. The relative abundance of the dichlorinated congeners decreased from their dominance in the Al2O3 sample to less than the other homologues in the fly ash-catalyzed reactions. Of the catalysts tested in these experiments, CuO produced a congener pattern more comparable to the fly ash results than did Al2O3. In 600 °C reactions of ethane/HCl (Figure 3), the congener pattern produced by SiO2/Al2O3 was again dominated by the Cl2Bz congeners and the more highly chlorinated congeners decreased in intensity, as observed in the Al2O3-catalyzed C2H4 reaction. Quantities of the Cl2Bz are within a factor of 1-2 of those produced in the fly ash sample, but a factor of 20-40 greater than those in the

FIGURE 2. ClxBz isomer pattern from 600 °C reactions of ethylene/ HCl with various catalysts: dhx¯(CuO) ) 9%, n ) 2; dhx¯(fly ash) ) 2%, n ) 2).

CuO sample. The pattern shift trend toward higher substituted homologues from Al2O3 to CuO and fly ash was also observed. Once again, the CuO reaction was the closest match with the fly ash sample in terms of isomer distribution; however, the congener quantities in the CuO-catalyzed reactions were at least an order of magnitude less than in the fly ash reactions. Chlorinated Phenols. Similar ClxPh congeners were observed in products from reactions of ethylene and ethane; 2,4-, 2,4,6-, and 2,3,4,6-substituted congeners together with PCP produce a unique pattern in most of the reactions. These congeners were observed in acetylene reactions (18) and for TCE reactions (12). While CuO again appeared to provide the best comparability to fly ash, Al2O3 produced a unique, Cl2-dominated congener pattern for both aliphatics, also observed for the reaction of acetylene with Al2O3 (18), suggesting that it has a unique catalytic effect in the high-temperature reactions of C2 aliphatics. Little

VOL. 30, NO. 3, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1011

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 3, 1996

++ ++

+ +

+

++ ++ ++ + + + + + SiO2 6 SiO2/Al2O3 6a SiO2/CuO 6a SiO2/CuO 6b fly ash 6a

+

+ ++ ++ ++ ++

+ +

+ + + + + + + SiO2 6 SiO2/Al2O3 6 SiO2/CuO 6a fly ash 4 fly ash 6a fly ash 6Ae fly ash 6Be

+

+ +

+

+ + + +

+

+ + +

+

+ + + + +

++

Ethane

Ethylene

+ + + +

+ + +

+

+

++

+ + + +

cis-1,2-C2H2Cl2d (cis-1,2-dichloroethylene) trans-1,2-C2H2Cl2d (trans-1,2-dichloroethylene) C2H3Clc (chloroethylene) C2Cl2c (dichloroacetylene) C2HClc (chloroacetylene) CCl4c (carbon tetrachloride) CHCl3b (chloroform) CH2Cl2b (dichloromethane)

a (+) identification; (++) major peak aside from argon and CO . b Identification by comparison with authentic standards. c Tentative identification from MS data and spectral library searches. d Combination of footnote 2 c and consideration of molecule characteristics such as relative retention time (i.e., cis- and trans-1,2-dichloroethylene). e EthFa 6A and 6B: molecular sieve cartridges were attached in series; breakthrough obviously occurred.

+

+

C6H6b (benzene) C2H3Cl3f (1,1,2-trichloroethane) C2H4Cl2d (1,2-dichloroethane) C2H5Clc (chloroethane) C2Cl4b (tetrachloroethylene) 1012

CH3Cl2c (chloromethane)

Many of the sample eluates were qualitatively analyzed for additional chlorinated and non-chlorinated aliphatic and aromatic compounds. Molecular sieve adsorption tubes were analyzed for volatile reaction products (Table 3) using headspace GC/MS as described elsewhere (18). Only two volatile compounds, chloromethane and benzene, were observed in the molecular sieve-absorbed products of the Al2O3-catalyzed reactions. Reactions over CuO or fly ash, however, produced most of the chlorinated methyl compounds as well as many chlorinated C2 aliphatics. Dichloroacetylene (C2Cl2) was observed in many of the samples. Acetylenic compounds and radicals have been linked by many researchers to direct polymeric formation of chlorinated aromatic compounds (4, 14, 19-23). Benzene was not detected in CuO or fly ash reactions. Semivolatile reaction products were detected by full scan GC/MSD

TABLE 3

Additional Products

reaction conditions

(,10%) or no products were detected in eluates from the catalysts in the ethylene reactions. Catalyst-adsorbed products were more significant, however, for the ethane reactions, comprising from one-third to approximately three-fourth of the total congener amounts detected contrasting the lack of ClxBz in the same elutions. The isomer distribution patterns and production amounts were very similar between the CuO and fly ash samples.

Volatile Compounds Qualitatively Identified in Headspace GC/MS Analyses of Molecular Sieve Adsorbenta

FIGURE 3. ClxBz isomer patterns for 600 °C reactions of ethane/HCl with various catalysts: (s¯x¯(CuO) ) 52%, n ) 4; s¯x¯(fly ash) ) 41%, n ) 4).

analysis. In this case as well, the Al2O3-catalyzed reactions produced different compounds than those observed in the CuO and fly ash reactions. Non-chlorinated naphthalene and alkylbiphenyl compounds were predominant in the Al2O3-catalyzed reactions. In contrast, chlorinated and perchlorinated C3-C5 alkyl compounds dominated the CuO and fly ash reaction products. These included hexachloropropene, hexachlorobutadiene, and hexachlorocyclopentadiene, all of which have been considered in this form or as non-chlorinated analogues as potential key intermediates in the formation of first aromatic rings in combustion systems (14, 20, 23-25). The absence of the acetylenic and olefinic compounds in the Al2O3 reactions suggests that Al2O3 catalyses the 600 °C C2 reactions in a unique way, causing the selective production of nonchlorinated compounds as possible intermediates in the ultimate formation of chlorinated aromatic compounds such as ClxBz and ClxPh. The composition of the fly ash used in these experiments (12) reflects the apparent importance of Cu in the catalytic action of the fly ash: despite the fact that Al contributes 4% (wt) to the fly ash, the reaction product patterns observed using SiO2 model mixtures of these metals do not match those produced with fly ash. Cu contributes only 0.1% (wt) to the mass of the fly ash. It is known to catalyze the Deacon process, which is accepted as a major route of formation of Cl2(g) from HCl. Cl2(g) has been shown to be the primary chlorinating agent in the subsequent formation of chlorinated aliphatic and aromatic hydrocarbons (15, 26-28). SiO2 and Al2O3 may also promote the production of Cl2 via the Deacon process (29); however, the results of these experiments indicate that the final chlorinating agent or the ultimate mechanism of chlorination must be different from that which occurs in the CuO and fly ash reactions. Cupric chloride (CuCl2), which may be formed from CuO, has also been proposed as the active chlorinating agent in fly ash and Cu-catalyzed reactions of ethylene (30). One supposition is that even very small amounts of CuCl2 achieve the characteristic catalytic action and that this characteristic action is dominant over or excludes the chlorination reactions promoted by Al or Fe.

(3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29)

Acknowledgments We would like to thank the Natural Science and Engineering Research Council of Canada for financial support.

Literature Cited (1) Altwicker, E. R. Sci. Total Environ. 1991, 104, 47-72. (2) Choudhry, G. G.; Hutzinger, O. Mechanistic aspects of the thermal formation of halogenated organic compounds including poly-

(30)

chlorinated dibenzo-p-dioxins; Current Topics in Environmental and Toxicological Chemistry; Gordon and Breach Science Publishers: New York, 1983. Hagenmaier, H.; Kraft, M.; Brunner, H.; Haag, R. Environ. Sci. Technol. 1987, 21, 1080-1084. Tsang, W. Combust. Sci. Technol. 1990, 74, 99-116. Vogg, H.; Stieglitz, L. Chemosphere 1986, 15, 1373-1378. Stieglitz, L.; Zwick, G.; Beck, J.; Roth, W.; Vogg, H. Chemosphere 1989, 18, 1216-1226. Stieglitz, L.; Vogg, H.; Zwick, G.; Beck, J.; Bautz, H. Chemosphere 1991, 23, 1255-1264. Milligan, M. S.; Altwicker, E. Environ. Sci. Technol. 1993, 27, 1595-1601. Altwicker, E. R.; Konduri, R. K. N. V.; Lin, C.; Milligan, M. S. Combust. Sci. Technol. 1993, 88, 349-368. Karasek, F. W.; Dickson, L. C. Science 1987, 237, 754-756. Dickson, L. C.; Lenoir, D.; Hutzinger, O. Environ. Sci. Technol. 1992, 26, 1822-1828. Froese, K. L.; Hutzinger, O. Chemosphere 1994, 28, 1977-1987. Ballschmiter, K.; Swerev, M. Fresenius Z. Anal. Chem. 1987, 328, 125-127. Tirey, D. A.; Taylor, P. H.; Kasner, J.; Dellinger, B. Combust. Sci. Technol. 1990, 74, 137-157. Born, J. G. P.; de Lijser, H. J. P.; Ahonkhal, S. I.; Louw, R.; Mulder, P. Chemosphere 1991, 23, 1213-1220. Gel’perin, E. I.; Bakshi, Yu M.; Avetisov, A. K.; Gel’bshtein, A. I. Kinet. Catal. 1978, 19, 1241-1246. Gel’perin, E. I.; Bakshi, Yu M.; Avetisov, A. K.; Gel’bshtein, A. I. Kinet. Catal. 1979, 20, 102-108. Froese, K. L.; Hutzinger, O. Environ. Sci. Technol. 1996, 30, 9981008. Tirey, D. A.; Taylor, P. H.; Dellinger, B. In Emissions from Combustion Processes: Origin, Measurement, Control; Clement, R., Kagel, R., Eds.; Lewis Publishers, Inc.: Chelsea, MI, 1990. Miller, J. A.; Melius, C. F. Combust. Flame 1992, 91, 21-39. Mochida, I.; Aoyagi, Y.; Yatsunami, S.; Fujjitsu, H. J. Anal. Appl. Pyrolysis 1991, 21, 95-102. Chambon, M.; Marquaire, P.-M.; Come, G.-M. C1 Mol. Chem. 1987, 2, 47-59. Aubrey, N. E.; Van Wazer, J. R. J. Am. Chem. Soc. 1964, 86, 43804383. Cole, J. A.; Bittner, J. D.; Longwell, J. P.; Howard, J. B. Combust. Flame 1984, 56, 51-70. Mulder, P.; Jarmohamed, W. Organohalogen Compd. 1993, 11, 273-276. Gullet, B. K.; Bruce, K. R.; Beach, L. O. Waste Manage. Res. 1990, 8, 203-214. Bruce, K. R.; Beach, L. O.; Gullet, B. K. Waste Manage. Res. 1991, 11, 97-102. Bruce, K. R.; Gullet, B. K.; Beach, L. O. In Procedings of the 1991 Incineration Conference, 13-17 May 1991; EPA/RREL: Knoxville, TN, 1991. Shaub, W. M.; Tsang, W. Environ. Sci. Technol. 1983, 17, 721730. Magistro, A. J.; Cowfer, J. A. J. Chem. Educ. 1986, 63, 1056-1058.

Received for review July 5, 1995. Revised manuscript received November 2, 1995. Accepted November 3, 1995.X ES9504810 X

Abstract published in Advance ACS Abstracts, January 15, 1996.

VOL. 30, NO. 3, 1996 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1013