- Email: [email protected]
Effect of photosensitizer diethylamine on the photodegradation of polychlorinated biphenyls
Chemosphere 55 (2004) 879–884 www.elsevier.com/locate/chemosphere
Effect of photosensitizer diethylamine on the photodegradation of polychlorinated biphenyls Yaw-Jian Lin *, Li-Shan Teng, Ampere Lee, Yi-Ling Chen Department of Environmental Science and Engineering, National Pingtung University of Science and Technology, Nei Pu, Pingtung Hsien 91207, Taiwan Received 13 February 2003; received in revised form 11 September 2003; accepted 26 November 2003
Abstract The objective of this study was to investigate the effect of diethylamine and xenon simulated sunlight on the photodegradation of two forms of PCBs including PCBs in transformer oil and PCB congener 138. The result of GC chromatograms illustrated the shifting pattern of higher chlorinated biphenyls in transformer oil degraded to lower chlorinated biphenyls with the extension of exposure time. The effect of diethylamine and xenon simulated sunlight was significant on the photodegradation of both PCBs in transformer oil and congener 138. The initial degradation rates of congener 138 (1.14 · 109 to 4.47 · 109 mol l1 h1 ) were in direct proportion to the initial concentrations of congener 138 which confirmed the pseudo-first-order reaction of PCB photodegradation. The apparent quantum yields (/) of congener 138 using diethylamine in xenon photoreactor were ranged between 2.08 · 102 and 9.8 · 104 . PCB congeners 123, 97, 70, 67, 33, 29, 17, 12, and 9 were detected as the descendants of the photodegradation of congener 138 through dechlorination. The major pathway of congener 138 photodegradation in this study was via para-dechlorination. 2004 Elsevier Ltd. All rights reserved. Keywords: Congener 138; Photolysis, Transformer oil; Xenon simulated sunlight; Apparent quantum yield
1. Introduction Worldwide application of polychlorinated biphenyls (PCB) to a wide variety of industries has been practiced for decades. Serious environmental pollution and detrimental effect of PCB to the ecosystem have raised constant safety concerns due to their extraordinary physical and chemical stability. PCB chronic toxic effects include the damage of heart, liver, kidney, central nerve systems, and reproduction. The safety concerns and environmental contamination have resulted in continuous investigations on the transportation and disposal of PCB. Incineration is the most commonly used technology for the destruction of PCB; however, it often leads
*
Corresponding author. Tel.: +886-8-7740470; fax: +886-87740256. E-mail address: [email protected] (Y.-J. Lin).
to the formation of more toxic oxygenated derivatives such as dioxins and polychlorodibenzofurans (PCDF). Also, incineration is not applicable for the treatments of PCB contaminated water and soil. Photochemical method is an important mechanism for the degradation of environmental pollutants such as pesticides and PCBs (Maletzky and Bauer, 1998; De Laat et al., 1999; Hequet et al., 2001). However, most PCB congeners do not strongly absorb the light energy of wavelength above 300 nm (Hawari et al., 1992). Studies showed that through the use of photosensitizers (e.g. diethylamine, triethylamine, diethyl phenylene diamine, riboflavin, propanol, and dyes) PCB dechlorination could be enhanced (Freeman et al., 1986; Zepp, 1988; Hawari et al., 1992; Lin et al., 1995, 1996; 2000; Cui et al., 2001). Photosensitizers absorb light energy and transform it into chemical energy as the formation of excited state of molecules and then initiate a series of photochemical reactions for the degradation of PCB.
0045-6535/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2003.11.059
880
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
The process is commonly known as indirect photodegradation which often proceeds with higher quantum efficiency than that of direct photodegradation. Xenon lamps used on indirect photodegradation emit the similar patterns of both wavelength and radiant energy compared to those of natural sunlight (Lin et al., 2002). Therefore, xenon photoreactors have been commonly selected as the energy source of simulated sunlight to investigate the photodegradation mechanisms of various environmental pollutants such as pesticides and PCBs (Hawari et al., 1992; Lin et al., 1995, 2002; Nguyenm, 1996; Selli et al., 1999; Parra et al., 2002). PCBs were originally sold in the USA under the trade name of Aroclor by Monsanto company. PCB congener 138 constitutes 9.1% of all Aroclor 1254 congeners by weight (Capel et al., 1985). This study was conducted to determine the effect of diethylamine and xenon simulated sunlight on the photodegradation of two forms of PCBs including PCB in transformer oil and PCB congener 138.
2. Experimental section 2.1. Sample preparation PCB in transformer oil was dissolved in acetone to increase its aqueous solubility. High concentrations (5%, 10%, and 50%) of diethylamine (99.6%, Tedia Co., USA) that acted as a sensitizer were used to ensure the enhancement of PCBs photodegradation (Freeman et al., 1986; Lin et al., 1996). The standards of PCB congener 138 (2,20 ,3,4,40 ,50 -hexachlorobiphenyl, H6 CB) and the mixture of 209 standard congeners were used as stock solution and for the confirmation of the degraded products (descendants) of PCB photodegradation, respectively (AccuStandard Inc., USA). The degraded products of PCB were verified by comparing the relative retention times (RRT) and relative responses (RR) of the pure standard congeners (AccuStandard Inc., USA) (Mullin et al., 1984). Photosensitizer diethylamine that served as an electron donor to enhance the photodegradation was prepared in stock solution prior to the exposure of xenon simulated sunlight. The dechlorination process was described as below (Bunce et al., 1978). ArCl þ hm ! ArCl ArCl þ D ! ArCl þ Dþ ArCl ! Cl þ Ar D : diethylamine; ðC2 H5 Þ2 NH
2.2. Xenon photoreactor A xenon lamp (150 W) was used to simulate natural sunlight in xenon photoreactor (L2274, Hamamatsu
Photonics, Japan). The UV/visible spectrum emitted by the xenon lamp ranged from 280 to 800 nm. The maximum UV/visible absorption of PCB congener 138 was around 240 nm with the range between 210 and 310 nm. The radiant energy of xenon photoreactor was monitored by a LI-250 light meter equipped with a LI-200SA pyranometer sensor (LI-COR Inc., Nebraska, USA). The selected exposure times were 0, 8, 16, 32, and 64 h, and the fluctuations of reactor temperature was monitored. To investigate the degradation pathway and the intermediates of PCB congener 138 photodegradation, shorter exposure times (0.5, 1, 2, 3, 4, 5, 6, 7, and 8 h) were used. The experiments were conducted in three replicates. The results of sample analysis were presented as mean ± standard deviation to show the variation among samples.
2.3. Sample extraction and analyses Hexane was used for the dilution and extraction of PCB samples. Chrysene-d12 was used as an internal standard to compensate the loss of sample during extraction and analysis (Nelson et al., 1998). After exposure to simulated sunlight, 2 ml of hexane (C6 H14 ) was added to the samples for extraction, and the samples were shaken at 280 rpm for 6 h on an orbital shaker (VWR Scientific, Philadelphia, Pennsylvania). The extraction procedure was repeated once. A gas chromatograph equipped with an electron capture detector (GC-ECD) (HP 6890) and a DB 5, 30 m (i.d. ¼ 0.32 mm) column (0.25 lm film thickness) was used for analyses. The column temperature started at 100 C for 1 min, then ramp 4 C/min to 170 C, then to 230 C at 5 C/min, then to 280 C at 10 C/min. Final holding time was 5 min. The respective temperatures of injector and detector were set at 250 and 280 C. The amount of sample injected was 1 ll. The detection limit of GC analysis for congener 138 was 0.1 lg/l. Concentration of total PCBs in transformer oil was calculated from the sum of the concentration of 209 standard congeners. Substitution method was used to determine the reaction order of congener 138 photodegradation. After the pseudo-first-order reaction of congener 138 photodegradation was confirmed the concentrations were calculated as lnðCt =C0 Þ where Ct and C0 were concentrations at time t and 0 h, respectively. The data of lnðCt =C0 Þ were plotted against exposure times. The half-lives (t1=2 ) of congener 138 were calculated by the equation of (t1=2 ¼ 0:693=k) where the k was the rate constant of congener 138 photodegradation. Apparent quantum yield was calculated as the ratio between the amount of congener 138 degraded and the amount of quantum emitted. The quantum yield can be used to compare with the direct photodegradation rate constants and the half-lives of pollutants present in
k /¼ 2:3 e l I0 I0 : emitted light intensity (mol quantum l1 h1 ); k: reaction constant (h1 ); e: molar extinction constant (l mol1 cm1 ); ediethylamine ¼ 550 l mol1 cm1 (Calvert and Pitts, 1967); l: light path (cm); /: apparent quantum yield. The data were analyzed using a statistical software, Statistix (Win V. 2.0) (Analytical Software Inc., Florida, USA) for statistical analyses including analysis of variance (ANOVA) and least significant difference (LSD) to determine significant differences (P < 0:05) among treatments.
3. Results and discussion 3.1. Xenon exposure The radiant energy of xenon simulated sunlight emitted onto each reaction vessel was 36.6 ± 1.8 mW/cm2 which was equivalent to 4.0 ± 0.2 kJ per hour per sample. Compared with the study of Lin et al. (2000), the energy in this xenon photoreactor at the light path distance of 14 cm was around 75% of the solar energy on average or 30% of the highest solar energy measured in Pingtung, Taiwan (22370 N, 120340 E). With the room temperature at 25 ± 2 C, the temperatures of reaction vessels and samples covered with aluminum foil for quality control were increased and then stabilized around 36 and 34 C, respectively after 8 h of xenon simulated sunlight exposure or longer. 3.2. PCB in transformer oil The total concentration of PCBs in transformer oil used in this study was measured as 20% in comparison with those of the mixture of 209 PCB congeners. The effect of diethylamine and xenon simulated sunlight was significant on the photodegradation of PCB in transformer oil. The percentage of remaining PCBs was decreased with the increase of exposure time for all of the samples with different proportions of diethylamine (Fig. 1). The percent degradation of PCB in transformer oil increased as a function of increasing concentration of the diethylamine. For quality assurance, the roles of xenon lamp and sensitizer diethylamine were investigated. Results showed that there was less than 2% PCB degraded for the samples covered with aluminum foil to avoid light exposure. In the absence of diethylamine, there was 10% PCB degradation contributed by xenon exposure without the sensitizer. Acetone that was used
881
100 80 60 40 20 0
0
10
20
30
40
50
60
70
Exposure Time (h) 0%
5%
10%
50%
Fig. 1. Percent PCB remaining in transformer oil after exposure to xenon simulated sunlight with various percentages of diethylamine.
to increase the solubility of PCBs in transformer oil did not contribute significant effect on the photolysis for the sample without diethylamine (0%) (Fig. 1). The result of GC chromatograms illustrated the shifting pattern of higher chlorinated biphenyls to lower chlorinated biphenyls through photodechlorination. The shifting pattern showed that the concentrations of many higher chlorinated congeners decreased while the concentrations of lower chlorinated congeners increased due to continuous photodechlorination.
3.3. PCB congener 138 Pseudo-first-order reaction was suggested for the photodegradation of PCB congener 138 using diethylamine and xenon simulated sunlight. The photodegradation ½lnðCt =C0 Þ of congener 138 with different concentrations of diethylamine were given in Figs. 2 and 3. The results showed that the concentrations of congener 138 were decreased with the increase of exposure time using three different concentrations of
ln (Ct /Co)
aquatic systems. The calculation of quantum yield is listed below (Dullin and Mill, 1982; Lin et al., 2002).
Percent Remaining (%)
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
0.5 0 -0.5 -1 -1.5 -2 -2.5 -3
0
10
20
30
40
50
60
70
Exposure Time (h) 500 µg/l
250 µg/l
50 µg/l
Fig. 2. Photodegradation of PCB congener 138 (C0 : 50 lg=l) with different concentrations of diethylamine using xenon photoreactor.
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
ln (Ct /Co)
882 0.5 0 -0.5 -1
in the range of 0.45–1.46 (Lin et al., 2002). Dullin and Mill (1982) reported that sunlight actinometer such as nitrobenzaldehyde typically have the high value of quantum yield (/ > 0:5).
-1.5 -2 -2.5 -3
3.4. Photodegradation pathway of PCB congener 138 0
10
20
30
40
50
60
70
Exposure Time (h) 500 µg/l
250 µg/l
50 µg/l
Fig. 3. Photodegradation of PCB congener 138 (C0 : 25 lg=l) with different concentrations of diethylamine using xenon photoreactor.
diethylamine. The effect of diethylamine on congener 138 photodegradation was reduced after 32 h of xenon lamp exposure for the samples with higher concentration of PCB congener 138 (Figs. 2, 3). The initial degradation rates of congener 138 ranged from 1.14 · 109 to 4.47 · 109 mol l1 h1 were in direct proportion to the concentrations of congener 138 (Table 1). The shorter half-lives of congener 138 were found in the samples with diethylamine concentration 10 times higher than that of congener 138 (Table 1). The finding corroborated that the concentration of diethylamine 10 times higher than those of xenobiotics was sufficient to provide electrons and to facilitate photodechlorination (Freeman et al., 1986; Lin et al., 1996). The apparent quantum yields (/) of congener 138 photodegradation using diethylamine in xenon photoreactor were ranged between 2.08 · 102 and 9.8 · 104 which was similar to the findings of Dullin and Mill (1982), Choudhry et al. (1990), and Lin et al. (1995) on the related chlorinated xenobiotics. However, using the same xenon photoreactor, the apparent quantum yields of chlorinated xenobiotics such as PCB and PCDF were much lower than that of nitrobenzaldehyde which was
The photodegradation pathway of congener 138 was proposed through a series of dechlorination reactions (Fig. 4). The mechanism of congener 138 dechlorination involved nucleophilic aromatic substitution. The chain reaction enhanced by xenon simulated sunlight was initiated by the transfer of electrons from the electron donor (diethylamine) to congener 138. The higher chlorinated congeners were found at the beginning of congener 138 dechlorination. With the extension of exposure time, lower chlorinated congeners were detected. Two pentachlorobiphenyls (P5 CB) (congeners 123 and 97) were identified after 30 min of exposure of congener 138 (hexachlorobiphenyls, H6 CB) to xenon simulated sunlight (Fig. 4). Congeners 123 and 97 were formed through the ortho- and para-dechlorination of congener 138, respectively. After 1 h of exposure, in addition to congeners 123 and 97, congeners 70 and 67 (both tetrachlorobiphenyls, T4 CB) were detected via ortho- and para-dechlorination. Meta-dechlorination was only discovered after 2 h of exposure with the formation of the descendants of congeners 33, 29, and 17 (all trichlorobiphenyls, T3 CB). After 3 h of exposure, congeners 12 and 9 (both dichlorobiphenyls, D2 CB) were found as the degraded products through orthoand para-dechlorination, respectively. Overall, the major pathway of congener 138 photodegradation in this study was via para-dechlorination. The descendants of congener 138 (congeners 123, 97, 70, 67, 33, 29, 17, 12, and 9) were found in the samples both with and without diethylamine. The finding indicated that photosensitizer diethylamine only enhanced both the quantum efficiency and degradation rate of PCB congener 138, but it did
Table 1 Rate constant (k), initial degradation rate, half-life, and apparent quantum yield of congener 138 with different concentrations of diethylamine PCB138 (lg/l)
Diethylamine (lg/l)
k (·102 ) (h1 )
Rate (·109 ) (mol l1 h1 )
t1=2 (h)
/ (·103 )
50 50 50 50 25 25 25 25
500 250 50 0 500 250 50 0
3.23 ± 0.36 2.84 ± 0.23 2.54 ± 0.25 0.17 ± 0.01 2.39 ± 0.19 2.65 ± 0.19 1.65 ± 0.15 0.11 ± 0.01
4.47 ± 0.49 3.93 ± 0.31 3.52 ± 0.35 0.24 ± 0.01 1.66 ± 0.13 1.84 ± 0.13 1.14 ± 0.10 0.08 ± 0.00
(21.5 ± 2.4)a (24.4 ± 1.9)a (27.3 ± 2.7)b (407.6 ± 20.4)d (29.0 ± 2.3)b (26.2 ± 1.8)b (42.0 ± 3.8)c (630.0 ± 31.5)e
(2.65 ± 0.3)c (4.67 ± 0.4)d (20.8 ± 2.1)f (3.2 · 104 )a (0.98 ± 0.1)b (2.18 ± 0.2)c (6.80 ± 0.6)e (2.1 · 104 )a
Means with different letters are significantly different from each other at P < 0:05.
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
tected. para-dechlorination was the major pathway of congener 138 dechlorination in this study.
Cl Cl
Cl Cl
Cl 138
Cl
Cl
Cl
Acknowledgements
Cl - para
-ortho Cl
This research was funded by the National Science Council (NSC) of Taiwan, ROC under grant NSC892211-E020-013. The authors thank its financial support.
Cl Cl
Cl 123
Cl 97
Cl - para
- para Cl
Cl
References
-ortho
Cl Cl
Cl
-meta -meta
Cl Cl 70
- meta
Cl
883
67
Cl
Cl
-meta
Cl
Cl
Cl Cl
Cl
Cl
33
-ortho Cl
29
Cl 17
- para
Cl
Cl
Cl 12 9
Cl
Fig. 4. Proposed degradation pathway of PCB congener 138.
not affect either the pathway of dechlorination or the types of degraded product.
4. Conclusions Effect of diethylamine and xenon simulated sunlight was significant on the photodegradation of both PCBs in transformer oil and PCB congener 138. Concentration of total PCBs in transformer oil was calculated as 20% from the sum of concentration of 209 standard congeners. The percent degradation of PCB in transformer oil increased as a function of increasing exposure time. The initial degradation rates of congener 138 samples were in direct proportion to their initial concentrations. Pseudo-first-order reaction was confirmed for the photodegradation of congener 138. The apparent quantum yields (/) of congener 138 using diethylamine in xenon photoreactor were similar to those of other related studies. PCB congeners 123, 97, 70, 67, 33, 29, 17, 12, and 9 were identified as the descendants of congener 138 photodegradation through dechlorination. The higher chlorinated congeners were found in the initial photodegradation of congener 138. With the extension of exposure time, lower chlorinated congeners were de-
Bunce, N.J., Kumar, Y., Ravanal, L., Safe, S., 1978. Photochemistry of chlorinated biphenyls in isooctane solution. J. Chem. Soc. 2, 880–884. Calvert, J.G., Pitts, J.N., 1967. Photochemistry. John Wiley & Sons Inc., New York. pp. 450, 479. Capel, P.D., Rapaport, R.A., Eisenreich, S.J., Looney, B.B., 1985. Computerized quantification of total PCB and congeners in environmental samples. Chemosphere 14, 439–450. Choudhry, G.G., Foga, M., Webster, G., 1990. Quantum yields of the direct phototransformation of 1,2,3,4,7,8-hexachlorodibenzofuran in aqueous acetonitrile and their sunlight half-lives. Toxicol. Environ. Chem. 26, 181–195. Cui, H., Hwang, H.M., Cook, S., Zeng, K., 2001. Effect of photosensitizer riboflavin on the fate of 2,4,6-trinitrotoluene in a freshwater environment. Chemosphere 44, 621–625. De Laat, J., Gallard, H., Ancelin, S., Legube, B., 1999. Comparative study of the oxidation of atrazine and acetone by H2 O2 /UV, Fe(III)/H2 O2 /UV and Fe(II) or Fe(III)/H2 O2 . Chemosphere 39, 2693–2706. Dullin, D., Mill, T., 1982. Development and evaluation of sunlight actinometers. Environ. Sci. Technol. 16, 815–820. Freeman, P.K., Srinivasa, R., Campbel, J.A., Denzer, M.L., 1986. The photochemistry of polyhaloarenes. 5. Fragmentation pathways in polychlorobenzene radical anions. J. Am. Chem. Soc. 108, 5531–5536. Hawari, J., Demeter, A., Samson, R., 1992. Sensitized photolysis of polychlorobiphenyls in alkaline 2-propanol: dechlorination of Aroclor 1254 in soil samples by solar radiation. Environ. Sci. Technol. 26, 2022–2027. Hequet, V., Gonzalez, C., Le Cloirec, P., 2001. Photochemical processes for atrazine degradation: methodological approach. Water Res. 35, 4253–4260. Lin, Y.J., Gupta, G., Baker, J., 1995. Photodegradation of polychlorinated biphenyl congeners using simulated sunlight and diethylamine. Chemosphere 31 (5), 3323–3324. Lin, Y.J., Gupta, G., Baker, J., 1996. Photodegradation of Aroclor 1254 using diethylamine and simulated sunlight. J. Hazard. Mater. 45, 259–264. Lin, Y.J., Lin, C., Yeh, K.J., Lee, A., 2000. Photodegradation of the herbicides butachlor and ronstar using natural sunlight and diethylamine. Bull. Environ. Contam. Toxicol. 64 (6), 780–785. Lin, Y.J., Lee, A., Teng, L.S., Lin, H.T., 2002. Effect of experimental factors on nitrobenzaldehyde photoisomerization. Chemosphere 48, 1–8. Maletzky, P., Bauer, R., 1998. The photo-fenton method–– degradation of nitrogen containing organic compounds. Chemosphere 37, 899–909.
884
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
Mullin, M.D., Pochini, C.M., McCrindle, S., Romkes, M., Safe, S., Safe, L.M., 1984. High-resolution PCB analysis: synthesis and chromatographic properties of all 209 PCB congeners. Environ. Sci. Technol. 18, 468–472. Nelson, E.D., McConnell, L.L., Baker, J.E., 1998. Diffusive exchange of gaseous polycyclic aromatic hydrocarbons and polychlorinated biphenyls across the air–water interface of the Chesapeake Bay. Environ. Sci. Technol. 32, 912–919. Nguyenm, A.L., 1996. Detection of polychlorinated biphenyls by photoactivated fluorescence and by sensitized laser photolysis. Anal. Chim. Acta 319, 221–229.
Parra, S., Olivero, J., Pulgarin, C., 2002. Relationships between physicochemical properties and photoreactivity of four biorecalcitrant phenylurea herbicides in aqueous TiO2 suspension. Appl. Catal. B Environ. 36, 75–85. Selli, E., Baglio, D., Montanarella, L., Bidoglio, G., 1999. Role of humic acids in the TiO2 -photocatalyzed degradation of tetrachloroethene in water. Water Res. 33 (8), 1827– 1836. Zepp, R.Z., 1988. Factors affecting the photochemical treatment of hazardous waste. Environ. Sci. Technol. 22, 256– 257.
Effect of photosensitizer diethylamine on the photodegradation of polychlorinated biphenyls Yaw-Jian Lin *, Li-Shan Teng, Ampere Lee, Yi-Ling Chen Department of Environmental Science and Engineering, National Pingtung University of Science and Technology, Nei Pu, Pingtung Hsien 91207, Taiwan Received 13 February 2003; received in revised form 11 September 2003; accepted 26 November 2003
Abstract The objective of this study was to investigate the effect of diethylamine and xenon simulated sunlight on the photodegradation of two forms of PCBs including PCBs in transformer oil and PCB congener 138. The result of GC chromatograms illustrated the shifting pattern of higher chlorinated biphenyls in transformer oil degraded to lower chlorinated biphenyls with the extension of exposure time. The effect of diethylamine and xenon simulated sunlight was significant on the photodegradation of both PCBs in transformer oil and congener 138. The initial degradation rates of congener 138 (1.14 · 109 to 4.47 · 109 mol l1 h1 ) were in direct proportion to the initial concentrations of congener 138 which confirmed the pseudo-first-order reaction of PCB photodegradation. The apparent quantum yields (/) of congener 138 using diethylamine in xenon photoreactor were ranged between 2.08 · 102 and 9.8 · 104 . PCB congeners 123, 97, 70, 67, 33, 29, 17, 12, and 9 were detected as the descendants of the photodegradation of congener 138 through dechlorination. The major pathway of congener 138 photodegradation in this study was via para-dechlorination. 2004 Elsevier Ltd. All rights reserved. Keywords: Congener 138; Photolysis, Transformer oil; Xenon simulated sunlight; Apparent quantum yield
1. Introduction Worldwide application of polychlorinated biphenyls (PCB) to a wide variety of industries has been practiced for decades. Serious environmental pollution and detrimental effect of PCB to the ecosystem have raised constant safety concerns due to their extraordinary physical and chemical stability. PCB chronic toxic effects include the damage of heart, liver, kidney, central nerve systems, and reproduction. The safety concerns and environmental contamination have resulted in continuous investigations on the transportation and disposal of PCB. Incineration is the most commonly used technology for the destruction of PCB; however, it often leads
*
Corresponding author. Tel.: +886-8-7740470; fax: +886-87740256. E-mail address: [email protected] (Y.-J. Lin).
to the formation of more toxic oxygenated derivatives such as dioxins and polychlorodibenzofurans (PCDF). Also, incineration is not applicable for the treatments of PCB contaminated water and soil. Photochemical method is an important mechanism for the degradation of environmental pollutants such as pesticides and PCBs (Maletzky and Bauer, 1998; De Laat et al., 1999; Hequet et al., 2001). However, most PCB congeners do not strongly absorb the light energy of wavelength above 300 nm (Hawari et al., 1992). Studies showed that through the use of photosensitizers (e.g. diethylamine, triethylamine, diethyl phenylene diamine, riboflavin, propanol, and dyes) PCB dechlorination could be enhanced (Freeman et al., 1986; Zepp, 1988; Hawari et al., 1992; Lin et al., 1995, 1996; 2000; Cui et al., 2001). Photosensitizers absorb light energy and transform it into chemical energy as the formation of excited state of molecules and then initiate a series of photochemical reactions for the degradation of PCB.
0045-6535/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2003.11.059
880
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
The process is commonly known as indirect photodegradation which often proceeds with higher quantum efficiency than that of direct photodegradation. Xenon lamps used on indirect photodegradation emit the similar patterns of both wavelength and radiant energy compared to those of natural sunlight (Lin et al., 2002). Therefore, xenon photoreactors have been commonly selected as the energy source of simulated sunlight to investigate the photodegradation mechanisms of various environmental pollutants such as pesticides and PCBs (Hawari et al., 1992; Lin et al., 1995, 2002; Nguyenm, 1996; Selli et al., 1999; Parra et al., 2002). PCBs were originally sold in the USA under the trade name of Aroclor by Monsanto company. PCB congener 138 constitutes 9.1% of all Aroclor 1254 congeners by weight (Capel et al., 1985). This study was conducted to determine the effect of diethylamine and xenon simulated sunlight on the photodegradation of two forms of PCBs including PCB in transformer oil and PCB congener 138.
2. Experimental section 2.1. Sample preparation PCB in transformer oil was dissolved in acetone to increase its aqueous solubility. High concentrations (5%, 10%, and 50%) of diethylamine (99.6%, Tedia Co., USA) that acted as a sensitizer were used to ensure the enhancement of PCBs photodegradation (Freeman et al., 1986; Lin et al., 1996). The standards of PCB congener 138 (2,20 ,3,4,40 ,50 -hexachlorobiphenyl, H6 CB) and the mixture of 209 standard congeners were used as stock solution and for the confirmation of the degraded products (descendants) of PCB photodegradation, respectively (AccuStandard Inc., USA). The degraded products of PCB were verified by comparing the relative retention times (RRT) and relative responses (RR) of the pure standard congeners (AccuStandard Inc., USA) (Mullin et al., 1984). Photosensitizer diethylamine that served as an electron donor to enhance the photodegradation was prepared in stock solution prior to the exposure of xenon simulated sunlight. The dechlorination process was described as below (Bunce et al., 1978). ArCl þ hm ! ArCl ArCl þ D ! ArCl þ Dþ ArCl ! Cl þ Ar D : diethylamine; ðC2 H5 Þ2 NH
2.2. Xenon photoreactor A xenon lamp (150 W) was used to simulate natural sunlight in xenon photoreactor (L2274, Hamamatsu
Photonics, Japan). The UV/visible spectrum emitted by the xenon lamp ranged from 280 to 800 nm. The maximum UV/visible absorption of PCB congener 138 was around 240 nm with the range between 210 and 310 nm. The radiant energy of xenon photoreactor was monitored by a LI-250 light meter equipped with a LI-200SA pyranometer sensor (LI-COR Inc., Nebraska, USA). The selected exposure times were 0, 8, 16, 32, and 64 h, and the fluctuations of reactor temperature was monitored. To investigate the degradation pathway and the intermediates of PCB congener 138 photodegradation, shorter exposure times (0.5, 1, 2, 3, 4, 5, 6, 7, and 8 h) were used. The experiments were conducted in three replicates. The results of sample analysis were presented as mean ± standard deviation to show the variation among samples.
2.3. Sample extraction and analyses Hexane was used for the dilution and extraction of PCB samples. Chrysene-d12 was used as an internal standard to compensate the loss of sample during extraction and analysis (Nelson et al., 1998). After exposure to simulated sunlight, 2 ml of hexane (C6 H14 ) was added to the samples for extraction, and the samples were shaken at 280 rpm for 6 h on an orbital shaker (VWR Scientific, Philadelphia, Pennsylvania). The extraction procedure was repeated once. A gas chromatograph equipped with an electron capture detector (GC-ECD) (HP 6890) and a DB 5, 30 m (i.d. ¼ 0.32 mm) column (0.25 lm film thickness) was used for analyses. The column temperature started at 100 C for 1 min, then ramp 4 C/min to 170 C, then to 230 C at 5 C/min, then to 280 C at 10 C/min. Final holding time was 5 min. The respective temperatures of injector and detector were set at 250 and 280 C. The amount of sample injected was 1 ll. The detection limit of GC analysis for congener 138 was 0.1 lg/l. Concentration of total PCBs in transformer oil was calculated from the sum of the concentration of 209 standard congeners. Substitution method was used to determine the reaction order of congener 138 photodegradation. After the pseudo-first-order reaction of congener 138 photodegradation was confirmed the concentrations were calculated as lnðCt =C0 Þ where Ct and C0 were concentrations at time t and 0 h, respectively. The data of lnðCt =C0 Þ were plotted against exposure times. The half-lives (t1=2 ) of congener 138 were calculated by the equation of (t1=2 ¼ 0:693=k) where the k was the rate constant of congener 138 photodegradation. Apparent quantum yield was calculated as the ratio between the amount of congener 138 degraded and the amount of quantum emitted. The quantum yield can be used to compare with the direct photodegradation rate constants and the half-lives of pollutants present in
k /¼ 2:3 e l I0 I0 : emitted light intensity (mol quantum l1 h1 ); k: reaction constant (h1 ); e: molar extinction constant (l mol1 cm1 ); ediethylamine ¼ 550 l mol1 cm1 (Calvert and Pitts, 1967); l: light path (cm); /: apparent quantum yield. The data were analyzed using a statistical software, Statistix (Win V. 2.0) (Analytical Software Inc., Florida, USA) for statistical analyses including analysis of variance (ANOVA) and least significant difference (LSD) to determine significant differences (P < 0:05) among treatments.
3. Results and discussion 3.1. Xenon exposure The radiant energy of xenon simulated sunlight emitted onto each reaction vessel was 36.6 ± 1.8 mW/cm2 which was equivalent to 4.0 ± 0.2 kJ per hour per sample. Compared with the study of Lin et al. (2000), the energy in this xenon photoreactor at the light path distance of 14 cm was around 75% of the solar energy on average or 30% of the highest solar energy measured in Pingtung, Taiwan (22370 N, 120340 E). With the room temperature at 25 ± 2 C, the temperatures of reaction vessels and samples covered with aluminum foil for quality control were increased and then stabilized around 36 and 34 C, respectively after 8 h of xenon simulated sunlight exposure or longer. 3.2. PCB in transformer oil The total concentration of PCBs in transformer oil used in this study was measured as 20% in comparison with those of the mixture of 209 PCB congeners. The effect of diethylamine and xenon simulated sunlight was significant on the photodegradation of PCB in transformer oil. The percentage of remaining PCBs was decreased with the increase of exposure time for all of the samples with different proportions of diethylamine (Fig. 1). The percent degradation of PCB in transformer oil increased as a function of increasing concentration of the diethylamine. For quality assurance, the roles of xenon lamp and sensitizer diethylamine were investigated. Results showed that there was less than 2% PCB degraded for the samples covered with aluminum foil to avoid light exposure. In the absence of diethylamine, there was 10% PCB degradation contributed by xenon exposure without the sensitizer. Acetone that was used
881
100 80 60 40 20 0
0
10
20
30
40
50
60
70
Exposure Time (h) 0%
5%
10%
50%
Fig. 1. Percent PCB remaining in transformer oil after exposure to xenon simulated sunlight with various percentages of diethylamine.
to increase the solubility of PCBs in transformer oil did not contribute significant effect on the photolysis for the sample without diethylamine (0%) (Fig. 1). The result of GC chromatograms illustrated the shifting pattern of higher chlorinated biphenyls to lower chlorinated biphenyls through photodechlorination. The shifting pattern showed that the concentrations of many higher chlorinated congeners decreased while the concentrations of lower chlorinated congeners increased due to continuous photodechlorination.
3.3. PCB congener 138 Pseudo-first-order reaction was suggested for the photodegradation of PCB congener 138 using diethylamine and xenon simulated sunlight. The photodegradation ½lnðCt =C0 Þ of congener 138 with different concentrations of diethylamine were given in Figs. 2 and 3. The results showed that the concentrations of congener 138 were decreased with the increase of exposure time using three different concentrations of
ln (Ct /Co)
aquatic systems. The calculation of quantum yield is listed below (Dullin and Mill, 1982; Lin et al., 2002).
Percent Remaining (%)
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
0.5 0 -0.5 -1 -1.5 -2 -2.5 -3
0
10
20
30
40
50
60
70
Exposure Time (h) 500 µg/l
250 µg/l
50 µg/l
Fig. 2. Photodegradation of PCB congener 138 (C0 : 50 lg=l) with different concentrations of diethylamine using xenon photoreactor.
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
ln (Ct /Co)
882 0.5 0 -0.5 -1
in the range of 0.45–1.46 (Lin et al., 2002). Dullin and Mill (1982) reported that sunlight actinometer such as nitrobenzaldehyde typically have the high value of quantum yield (/ > 0:5).
-1.5 -2 -2.5 -3
3.4. Photodegradation pathway of PCB congener 138 0
10
20
30
40
50
60
70
Exposure Time (h) 500 µg/l
250 µg/l
50 µg/l
Fig. 3. Photodegradation of PCB congener 138 (C0 : 25 lg=l) with different concentrations of diethylamine using xenon photoreactor.
diethylamine. The effect of diethylamine on congener 138 photodegradation was reduced after 32 h of xenon lamp exposure for the samples with higher concentration of PCB congener 138 (Figs. 2, 3). The initial degradation rates of congener 138 ranged from 1.14 · 109 to 4.47 · 109 mol l1 h1 were in direct proportion to the concentrations of congener 138 (Table 1). The shorter half-lives of congener 138 were found in the samples with diethylamine concentration 10 times higher than that of congener 138 (Table 1). The finding corroborated that the concentration of diethylamine 10 times higher than those of xenobiotics was sufficient to provide electrons and to facilitate photodechlorination (Freeman et al., 1986; Lin et al., 1996). The apparent quantum yields (/) of congener 138 photodegradation using diethylamine in xenon photoreactor were ranged between 2.08 · 102 and 9.8 · 104 which was similar to the findings of Dullin and Mill (1982), Choudhry et al. (1990), and Lin et al. (1995) on the related chlorinated xenobiotics. However, using the same xenon photoreactor, the apparent quantum yields of chlorinated xenobiotics such as PCB and PCDF were much lower than that of nitrobenzaldehyde which was
The photodegradation pathway of congener 138 was proposed through a series of dechlorination reactions (Fig. 4). The mechanism of congener 138 dechlorination involved nucleophilic aromatic substitution. The chain reaction enhanced by xenon simulated sunlight was initiated by the transfer of electrons from the electron donor (diethylamine) to congener 138. The higher chlorinated congeners were found at the beginning of congener 138 dechlorination. With the extension of exposure time, lower chlorinated congeners were detected. Two pentachlorobiphenyls (P5 CB) (congeners 123 and 97) were identified after 30 min of exposure of congener 138 (hexachlorobiphenyls, H6 CB) to xenon simulated sunlight (Fig. 4). Congeners 123 and 97 were formed through the ortho- and para-dechlorination of congener 138, respectively. After 1 h of exposure, in addition to congeners 123 and 97, congeners 70 and 67 (both tetrachlorobiphenyls, T4 CB) were detected via ortho- and para-dechlorination. Meta-dechlorination was only discovered after 2 h of exposure with the formation of the descendants of congeners 33, 29, and 17 (all trichlorobiphenyls, T3 CB). After 3 h of exposure, congeners 12 and 9 (both dichlorobiphenyls, D2 CB) were found as the degraded products through orthoand para-dechlorination, respectively. Overall, the major pathway of congener 138 photodegradation in this study was via para-dechlorination. The descendants of congener 138 (congeners 123, 97, 70, 67, 33, 29, 17, 12, and 9) were found in the samples both with and without diethylamine. The finding indicated that photosensitizer diethylamine only enhanced both the quantum efficiency and degradation rate of PCB congener 138, but it did
Table 1 Rate constant (k), initial degradation rate, half-life, and apparent quantum yield of congener 138 with different concentrations of diethylamine PCB138 (lg/l)
Diethylamine (lg/l)
k (·102 ) (h1 )
Rate (·109 ) (mol l1 h1 )
t1=2 (h)
/ (·103 )
50 50 50 50 25 25 25 25
500 250 50 0 500 250 50 0
3.23 ± 0.36 2.84 ± 0.23 2.54 ± 0.25 0.17 ± 0.01 2.39 ± 0.19 2.65 ± 0.19 1.65 ± 0.15 0.11 ± 0.01
4.47 ± 0.49 3.93 ± 0.31 3.52 ± 0.35 0.24 ± 0.01 1.66 ± 0.13 1.84 ± 0.13 1.14 ± 0.10 0.08 ± 0.00
(21.5 ± 2.4)a (24.4 ± 1.9)a (27.3 ± 2.7)b (407.6 ± 20.4)d (29.0 ± 2.3)b (26.2 ± 1.8)b (42.0 ± 3.8)c (630.0 ± 31.5)e
(2.65 ± 0.3)c (4.67 ± 0.4)d (20.8 ± 2.1)f (3.2 · 104 )a (0.98 ± 0.1)b (2.18 ± 0.2)c (6.80 ± 0.6)e (2.1 · 104 )a
Means with different letters are significantly different from each other at P < 0:05.
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
tected. para-dechlorination was the major pathway of congener 138 dechlorination in this study.
Cl Cl
Cl Cl
Cl 138
Cl
Cl
Cl
Acknowledgements
Cl - para
-ortho Cl
This research was funded by the National Science Council (NSC) of Taiwan, ROC under grant NSC892211-E020-013. The authors thank its financial support.
Cl Cl
Cl 123
Cl 97
Cl - para
- para Cl
Cl
References
-ortho
Cl Cl
Cl
-meta -meta
Cl Cl 70
- meta
Cl
883
67
Cl
Cl
-meta
Cl
Cl
Cl Cl
Cl
Cl
33
-ortho Cl
29
Cl 17
- para
Cl
Cl
Cl 12 9
Cl
Fig. 4. Proposed degradation pathway of PCB congener 138.
not affect either the pathway of dechlorination or the types of degraded product.
4. Conclusions Effect of diethylamine and xenon simulated sunlight was significant on the photodegradation of both PCBs in transformer oil and PCB congener 138. Concentration of total PCBs in transformer oil was calculated as 20% from the sum of concentration of 209 standard congeners. The percent degradation of PCB in transformer oil increased as a function of increasing exposure time. The initial degradation rates of congener 138 samples were in direct proportion to their initial concentrations. Pseudo-first-order reaction was confirmed for the photodegradation of congener 138. The apparent quantum yields (/) of congener 138 using diethylamine in xenon photoreactor were similar to those of other related studies. PCB congeners 123, 97, 70, 67, 33, 29, 17, 12, and 9 were identified as the descendants of congener 138 photodegradation through dechlorination. The higher chlorinated congeners were found in the initial photodegradation of congener 138. With the extension of exposure time, lower chlorinated congeners were de-
Bunce, N.J., Kumar, Y., Ravanal, L., Safe, S., 1978. Photochemistry of chlorinated biphenyls in isooctane solution. J. Chem. Soc. 2, 880–884. Calvert, J.G., Pitts, J.N., 1967. Photochemistry. John Wiley & Sons Inc., New York. pp. 450, 479. Capel, P.D., Rapaport, R.A., Eisenreich, S.J., Looney, B.B., 1985. Computerized quantification of total PCB and congeners in environmental samples. Chemosphere 14, 439–450. Choudhry, G.G., Foga, M., Webster, G., 1990. Quantum yields of the direct phototransformation of 1,2,3,4,7,8-hexachlorodibenzofuran in aqueous acetonitrile and their sunlight half-lives. Toxicol. Environ. Chem. 26, 181–195. Cui, H., Hwang, H.M., Cook, S., Zeng, K., 2001. Effect of photosensitizer riboflavin on the fate of 2,4,6-trinitrotoluene in a freshwater environment. Chemosphere 44, 621–625. De Laat, J., Gallard, H., Ancelin, S., Legube, B., 1999. Comparative study of the oxidation of atrazine and acetone by H2 O2 /UV, Fe(III)/H2 O2 /UV and Fe(II) or Fe(III)/H2 O2 . Chemosphere 39, 2693–2706. Dullin, D., Mill, T., 1982. Development and evaluation of sunlight actinometers. Environ. Sci. Technol. 16, 815–820. Freeman, P.K., Srinivasa, R., Campbel, J.A., Denzer, M.L., 1986. The photochemistry of polyhaloarenes. 5. Fragmentation pathways in polychlorobenzene radical anions. J. Am. Chem. Soc. 108, 5531–5536. Hawari, J., Demeter, A., Samson, R., 1992. Sensitized photolysis of polychlorobiphenyls in alkaline 2-propanol: dechlorination of Aroclor 1254 in soil samples by solar radiation. Environ. Sci. Technol. 26, 2022–2027. Hequet, V., Gonzalez, C., Le Cloirec, P., 2001. Photochemical processes for atrazine degradation: methodological approach. Water Res. 35, 4253–4260. Lin, Y.J., Gupta, G., Baker, J., 1995. Photodegradation of polychlorinated biphenyl congeners using simulated sunlight and diethylamine. Chemosphere 31 (5), 3323–3324. Lin, Y.J., Gupta, G., Baker, J., 1996. Photodegradation of Aroclor 1254 using diethylamine and simulated sunlight. J. Hazard. Mater. 45, 259–264. Lin, Y.J., Lin, C., Yeh, K.J., Lee, A., 2000. Photodegradation of the herbicides butachlor and ronstar using natural sunlight and diethylamine. Bull. Environ. Contam. Toxicol. 64 (6), 780–785. Lin, Y.J., Lee, A., Teng, L.S., Lin, H.T., 2002. Effect of experimental factors on nitrobenzaldehyde photoisomerization. Chemosphere 48, 1–8. Maletzky, P., Bauer, R., 1998. The photo-fenton method–– degradation of nitrogen containing organic compounds. Chemosphere 37, 899–909.
884
Y.-J. Lin et al. / Chemosphere 55 (2004) 879–884
Mullin, M.D., Pochini, C.M., McCrindle, S., Romkes, M., Safe, S., Safe, L.M., 1984. High-resolution PCB analysis: synthesis and chromatographic properties of all 209 PCB congeners. Environ. Sci. Technol. 18, 468–472. Nelson, E.D., McConnell, L.L., Baker, J.E., 1998. Diffusive exchange of gaseous polycyclic aromatic hydrocarbons and polychlorinated biphenyls across the air–water interface of the Chesapeake Bay. Environ. Sci. Technol. 32, 912–919. Nguyenm, A.L., 1996. Detection of polychlorinated biphenyls by photoactivated fluorescence and by sensitized laser photolysis. Anal. Chim. Acta 319, 221–229.
Parra, S., Olivero, J., Pulgarin, C., 2002. Relationships between physicochemical properties and photoreactivity of four biorecalcitrant phenylurea herbicides in aqueous TiO2 suspension. Appl. Catal. B Environ. 36, 75–85. Selli, E., Baglio, D., Montanarella, L., Bidoglio, G., 1999. Role of humic acids in the TiO2 -photocatalyzed degradation of tetrachloroethene in water. Water Res. 33 (8), 1827– 1836. Zepp, R.Z., 1988. Factors affecting the photochemical treatment of hazardous waste. Environ. Sci. Technol. 22, 256– 257.