Environ. Sci. Technol. 2002, 36, 3924-3927
Role of Inorganic Chlorides in Formation of PCDDs, PCDFs, and Coplanar PCBs from Combustion of Plastics, Newspaper, and Pulp in an Incinerator AKIO YASUHARA Research Center for Material Cycles and Waste Management, National Institute for Environmental Studies, 16-2, Onogawa, Tukuba, Ibaraki 305-0061, Japan TAKEO KATAMI Gifu Prefectural Institute of Health and Environmental Sciences, 1-1 Fudogaoka, Naka, Kagamigahara, Gifu 504-0838, Japan TOSHIKAZU OKUDA Fuji Seiku Kogyosho, Co., Ltd., 6-18, Honmachi, Kano, Gifu 500-8474, Japan TAKAYUKI SHIBAMOTO* Department of Environmental Toxicology, University of California, Davis, California 95616
The total amounts of dioxins found in exhaust gases from combustion of polyethylene (PE), polystyrene (PS), and poly(ethylene terephthalate) (PET) with approximately 3% (w/ w) NaCl were 6.07, 17.7, and 28.9 ng/g, respectively. Plastics containing benzene rings produced more dioxins than plastic containing no benzene ring. The amounts of dioxin formed in the exhaust gases from the combustion of newspapers impregnated with CaCl2, KCl, and NaCl were 18.6, 28.6, and 49.0 ng/g, respectively. Dioxin formation was associated with the bond energy between metal atom and chlorine atom. When newspapers impregnated with four different amounts of NaCl were combusted, the highest NaCl content newspaper (chlorine content 4.08%, w/w; lignin content 19.8%, w/w) produced the greatest amount of dioxins (174 ng/g). Pulp with NaCl (chlorine content 4.25%, w/w; lignin content 0.69%, w/w) produced more dioxins (6.71 ng/g) than pulp alone (0.799 ng/g) did upon combustion. The lignin content in a combustion sample correlated with the amount of dioxin formation. The results exhibited that combustion conditions with low CO concentration (2.0 83
X pulp
XI pulp+NaCl
>0.01 3.20 647 702 525 204 15.5 4.4 77 1.2
4.25 3.07 594 684 504 209 16.2 3.8 159 46
Newspaper.
TABLE 3. Analytical Results of PCDDs, PCDFs, and Coplaner PCBs in Exhaust Gases from Combustion of Samples (ng/g) I II III IV V VI VII VIII IX PE+NaCl PS+NaCl PET+NaCl NPa+CaCl2 NP+KCl NP+NaCl NP+NaCl NP+NaCl NP+NaCl T4CDDs P5CDDs H6CDDs H7CDDs O8CDDs total PCDD T4CDFs P5CDFs H6CDFs H7CDFs O8CDFs total PCDF coplaner PCBs grand total a
0.427 0.26 0.141 0.036 0.021 0.885 2.73 1.35 0.549 0.141 0.014 4.78 0.408 6.07
2.22 0.928 0.803 0.162 0.057 4.17 9.92 2.64 0.375 0.052 0.014 13.00 0.526 17.7
0.340 0.632 1.28 2.24 2.75 7.242 4.36 5.42 5.55 4.37 1.49 21.19 0.498 28.9
0.509 0.578 0.783 0.964 0.861 3.695 4.74 3.06 2.27 2.18 1.69 13.94 0.972 18.6
0.584 0.602 0.595 0.785 0.761 3.327 9.59 5.42 3.12 2.93 3.23 24.29 1.02 28.6
2.17 2.09 1.89 1.52 1.12 8.79 16.6 11.9 5.33 2.19 0.85 36.87 3.32 49.0
0.285 0.423 2.12 6.14 5.84 14.81 5.00 8.45 19.0 36.2 36.6 105.25 1.06 121
1.19 2.72 9.82 16.3 11.3 41.33 21.4 28.4 37.1 32.3 12.2 131.4 1.21 174
0.039 0.035 0.045 0.076 0.101 0.296 0.210 0.207 0.348 0.640 0.804 2.209 0.174 2.68
X pulp
XI pulp+NaCl
0.044 0.044 0.034 0.017 0.008 0.147 0.209 0.187 0.108 0.035 0.006 0.545 0.107 0.799
0.113 0.188 0.297 0.343 0.228 1.169 1.73 1.46 1.12 0.709 0.265 5.284 0.258 6.71
Newspaper.
TABLE 4. Toxicity Equivalency Quantity in Exhaust Gases from Incinerated Substances amount of compound (ng of TEQ/g of sample) I PE+NaCl
II PS+NaCl
III PET+NaCl
IV NPa+CaCl2
V NP+KCl
VI NP+NaCl
VII NP+NaCl
VIII NP+NaCl
IX NP+NaCl
X pulp
XI pulp+NaCl
0.157
0.258
0.707
0.371
0.452
1.08
1.40
3.12
0.026
0.018
0.158
a
Newspaper.
and PCDF were produced among dioxins formed in the exhaust gases from the combustion of PE and PS with NaCl (Samples I and II). PCDFs were approximately 75% of the total dioxin formed in the case of Samples I and II. When PE and PS were combusted alone, 3.08 and 1.36 ng/g of dioxins were formed (10), respectively. Therefore, the addition of NaCl increased dioxin formation by 2-fold for PE and by 13-fold for PS. In the case of PET, the higher the amount of chlorine in the matrix, the higher the amount of dioxin formed. However, the Cl-6 isomer was produced the most among the PCDFs formed. Also, dioxin formation from PET increased by 19-fold with the addition of NaCl. When NaCl is added, dioxin formation increased slightly in the case of PE, whereas it increased by over 10 times in the case of PS and PET. This suggests that plastics such as PS and PET containing benzene rings produce more dioxins, in addition to NaCl, than plastics such as PE containing no benzene rings. Results of Dioxin Analysis in CaCl2-, KCl-, and NaClImpregnated Newspapers. The amounts of dioxin formed in the exhaust gases from the combustion of CaCl2-, KCl-, and NaCl-impregnated newspapers (Samples IV, V, and VI) were 18.6, 28.6, and 49.0 ng/g, respectively (Table 3). The 3926
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 18, 2002
total PCDFs formed ranged from 75% to 85% of the total dioxin formed. The higher the amount of chlorine, the less PCDD and PCDF were produced among the dioxins formed. Table 4 shows toxicity equivalency in the exhaust gases from incinerated substances. The TEQ value of NaCl-impregnated newspaper was the greatest among these three samples. The amount of dioxin formed from the burning of the three samples was: CaCl2-impregnated newspaper < KCl-impregnated newspaper < NaCl-impregnated newspaper. This may be due to the bond energy differences between the metal and the Cl atom. The bond energies of CaCl2, KCl, and NaCl at ground state are 455 ( 3, 422 ( 1, and 407 ( 3 kJ/mol, respectively. Therefore, NaCl dissociates Cl most readily among the three inorganic compounds. Chlorine Content in Combustion Samples and Dioxin Formation. The relation between chlorine content and dioxin formation may be established when a combustion sample contains over 1% chlorine (5). On the other hand, there is a report that even when the chlorine content was less than 1%, the more chlorine content, the more dioxin formation (8). In the present study, the relationship between the chlorine content of a combustion sample and the subsequent formation of dioxin was investigated using newspapers impreg-
nated with various amounts of NaCl. When newspapers containing 0.99% (Sample VI), 1.88% (Sample VII), and 4.08% (Sample VIII) of chlorine were combusted, 49.0, 121, and 174 ng/g of dioxins were formed, respectively (Table 3). When newspaper (chlorine content 0.028%, w/w) was combusted alone under the same conditions, 0.93 ng/g (0.013 ng of TEQ/ g) of dioxins was formed. Among the samples combusted with NaCl, PE (Sample I; chlorine content 3.14%, w/w) and pulp (Sample XI; chlorine content 4.25%, w/w) yielded about the same value of dioxin TEQ (0.158 ng of TEQ/g) upon combustion. On the other hand, the TEQ values of PS (Sample II; chlorine content 2.85%, w/w) and PET (Sample III; chlorine content 2.80%, w/w) upon combustion were 0.258 and 0.707 ng of TEQ/g, respectively. Also, the sample with less chlorine content, NaCl-impregnated newspaper (Sample VI; chlorine content 0.99%, w/w), yielded higher values of TEQ (1.08 ng of TEQ/g) than the above samples upon combustion. Using these values, a regression equation, Y ) 42.7X + 11.7, R2 ) 0.97, for chlorine content (X) and dioxin formation (Y) was obtained. A regression equation for TEQ (Y) and chlorine content (X) is Y ) 0.738X + 0.115, R2 ) 0.99. The results indicate that there is an obvious relationship between chlorine content and dioxin formation and TEQ value as long as samples consist of the same ingredients (e.g., newspaper and NaCl). On the other hand, the chlorine content in a combustion sample does not correlate with the TEQ value when combusting samples consist of different ingredients. For example, among the samples combusted with NaCl, PE (Sample I; chlorine content 3.14%, w/w) and pulp (Sample XI; chlorine content 4.25%, w/w) yielded about same the value of dioxin TEQ (0.158 ng of TEQ/g) upon combustion. On the other hand, the TEQ values of PS (Sample II; chlorine content 2.85%, w/w) and PET (Sample III; chlorine content 2.80%, w/w) upon combustion were 0.258 and 0.707 ng of TEQ/g, respectively. Also, the sample with less chlorine content, NaCl-impregnated newspaper (Sample VI; chlorine content 0.99%, w/w), yielded higher values of TEQ (1.08 ng of TEQ/g) than the above samples upon combustion. When NaCl-impregnated newspaper (chlorine content 4.08%) was combusted with two burners (Sample IX), dioxin formation (2.68 ng/g, 0.026 ng of TEQ/g) was much less than when it was combusted with one burner (Sample VIII, 174 ng/g, 3.12 ng of TEQ/g). In this case, CO concentration was less than 2 ppm (Table 3). The results suggest that dioxin formation could be minimized by appropriate combustion conditions, even when the samples have a high chlorine content. Lignin Content and Dioxin Formation. Because it has been suggested that substances containing a benzene ring
play a significant role in dioxin formation, the relationship between lignin, which contains benzene rings, and dioxin formation was investigated. The results of dioxin formation from low lignin content samples (Samples X and XI, lignin content 0.69%) and a high lignin content sample (Sample VIII, lignin content 19.8%, w/w) were examined. Pulp containing less than 0.01% chlorine formed 0.80 ng/g of dioxins upon combustion (Sample X); that amount is comparable to the amount formed from newspaper alone (0.93 ng/g) (11). Their TEQ values were conspicuously low: 0.018 ng of TEQ/g from pulp and 0.009 ng of TEQ/g from newspaper. On the other hand, when NaCl-impregnated pulp (Sample XI, chlorine content 4.25%, w/w) was combusted, 6.71 ng/g (0.158 ng of TEQ/g) of dioxins was formed, that is, approximately 8 times that formed from pulp alone (Sample X). However, 174 ng/g of dioxins was formed from NaClimpregnated newspapers (Sample VIII; chlorine content 4.08%, w/w; lignin content 19.8% w/w); that is, approximately 26 times that from NaCl-impregnated pulp (Sample XI; chlorine content 4.25%, w/w; lignin content 0.69%, w/w). The results suggest that the lignin content in a combustion sample correlates with the amount of dioxin formation.
Acknowledgments We thank Y. Tanaka, E. Noguchi, Y. Hashimoto, and M. Kawamura for their assistance with sample preparations, and N. Seki for his lignin analyses.
Literature Cited (1) Zeng, M.; Liu, P.; Piao, M.; Bing, Z.; Xu, X. Organohalogen Compd. 2001, 50, 415-417. (2) Kuzuhara, S.; Kasai, E.; Nakamura, T.; Shibata, E. Organohalogen Compd. 2001, 50, 422-425. (3) Muto, H.; Sugawara, T. Chemosphere 2001, 45, 145-150. (4) Wikstrom, E.; Marklund, S. Chemosphere 2001, 43, 227-234. (5) Wikstrom, E.; Lofvenius, G.; Rappe, C.; Marklund, S. Environ. Sci. Technol. 1996, 30, 1637-1644. (6) Mattila, H.; Virtanen, T.; Vartiainen, T.; Ruuskanen, J. Chemosphere 1992, 25, 1599-1609. (7) Carroll, W. F., Jr. Chemosphere 2001, 45, 1173-1180. (8) Hatanaka, T.; Imagawa, T.; Takeuchi, M. Environ. Sci. Technol. 2000, 34, 3920-3924. (9) Japanese Standards Association. Testing method for lignin in wood for pulp, 1976, JIS P 8008. (10) Katami, T.; Yasuhara, A.; Okuda, T.; Ohno, N.; Shibamoto, T. Environ. Sci. Technol. 2002, in press. (11) Yasuhara, A.; Katami, T.; Okuda, T.; Shibamoto, T. Environ. Sci. Toxicol. 2001, 35, 1373-1378.
Received for review February 15, 2002. Revised manuscript received June 19, 2002. Accepted June 27, 2002. ES020602D
VOL. 36, NO. 18, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3927
