- Email: [email protected]
Effect of photosensitizer riboflavin on the fate of 2,4,6-trinitrotoluene in a freshwater environment
Chemosphere 44 (2001) 621±625
www.elsevier.com/locate/chemosphere
Eect of photosensitizer ribo¯avin on the fate of 2,4,6-trinitrotoluene in a freshwater environment Hua Cui
a,b
, Huey-Min Hwang
a,*
, Sean Cook a, Kui Zeng
a
a
b
Department of Biology, Jackson State University, Jackson, MS 39217, USA Department of Chemistry, University of Science and Technology of China, Hefei, Anhui 230026, People's Republic of China Received 10 May 2000; received in revised form 20 July 2000; accepted 5 August 2000
Abstract The eect of ribo¯avin (1 lM) on the fate of TNT (20 mg/l) in a natural water environment was studied. The relative contribution of photolysis, microbial assemblages and freshwater matrix to TNT degradation was examined. The rates, extent and products of TNT and ribo¯avin transformation were compared under dierent experimental conditions. It was found that ribo¯avin signi®cantly enhanced the degradation of TNT in natural water environment. Thus it is a potentially useful photosensitizing agent for the treatment of TNT-contaminated surface water. Furthermore, in the presence of ribo¯avin, two new intermediates with max. absorption wavelength of 230 nm were found, demonstrating that transformation of TNT in the presence of ribo¯avin undergoes dierent pathways. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Ribo¯avin; TNT; Degradation; Freshwater; Fate; Environment
1. Introduction 2,4,6-trinitrotoluene (TNT) is a conventional explosive used by military forces worldwide, causing serious contamination of soil and groundwater (Boopathy et al., 1994). Environmental transformations of TNT and methods of remediation have been extensively studied (Schmelling and Gray, 1995). The photochemical degradation of TNT was considered both as a primary treatment technology and as a pretreatment to biodegradation. Several strategies such as direct solar photolysis, UV-peroxide, UV-ozone, and TiO2 photocatalysis have been developed for the treatment of TNT-contaminated water (Schmelling and Gray, 1995). Exposure of TNT to sunlight or near UV radiation results in rapid conversion of TNT into a variety of aromatic photolysis
*
Corresponding author. Fax: +1-601-979-2778. E-mail address: [email protected] Hwang).
(H.-M.
products such as 2,4,6-trinitrobenzoic acid, 1,3,5-trinitrobenzene, 3,5-dinitroaniline, 2,4,6-trinitrophenol, 4-amino-2,6-dinitrotoluene, 2,4-dinitrobenzoic acid, 2,4,6-trinitrobenzaldehyde, 2,4,6-trinitrobenzyl alcohol, 4,6-dinitroanthranil, 2,4,6-trinitrobenzonitrile, and azoxydimers (Kaplan et al., 1975; Kearney et al., 1983; Schmelling and Gray, 1995; Crosby, 1998; Lang et al., 1998), depending on the photolysis condition. Each method has its own advantages and disadvantages with varying degrees of success. Therefore, there is a need to explore more ecient strategies for the remediation of TNT. Photosensitizers have been used to facilitate the photochemical degradation of some organic compounds. It is proposed that photosensitizers absorb light energy, transform it into chemical energy, and transfer that energy under favorable conditions to photochemically unreactive substances. The mechanism may be involved in redox reactions (Larson et al., 1992). Ribo¯avin as a photosensitizer has been reported to sensitize the photochemical degradation of many
0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 3 3 3 - 7
622
H. Cui et al. / Chemosphere 44 (2001) 621±625
compounds in aqueous solutions (Mopper and Zika, 1987; Hwang et al., 2000). Ribo¯avin is easy to handle and environmentally benign, making it attractive for the treatment of environmental water. In this paper, the eect of ribo¯avin on the fate of TNT in a natural water environment was studied. The relative contribution of photolysis, microbial assemblages and freshwater matrix to TNT degradation was examined in the absence and presence of ribo¯avin. The rates, extent and products of TNT transformation under dierent experimental conditions were determined by high-performance liquid chromatography. The transformation of ribo¯avin under dierent experimental conditions was also compared by use of ¯uorescence analysis. 2. Experimental 2.1. Chemicals and solution 2,4,6-trinitrotoluene, 1,3,5-trinitrobenzene, 2,4,6trinitrobenzoic acid, 4-amino-2,6-dinitrotoluene and 2-amino-4,6-dinitrotoluene were purchased from Chemservice (West Chester, USA). 2,4,6-trinitrophenol, 3,5-dinitroaniline, 2,4-dinitrobenzoic acid and ribo¯avin were purchased from Aldrich (Milwaukee, USA). Pyrogallol and phloroglucinol were purchased from Sigma (ST. Louis, USA). 2 g/l TNT stock solution was prepared in 40% acetonitrile±water solution and kept in dark at 4°C. Nanopure water was prepared by a Barnstead Nanopure In®nity (Dubuque, USA).
creased the degradation of TNT. Thus 1 lM ribo¯avin was chosen for all experiments. 2.3. Sample analysis After the degradation, the samples were ®ltered through an AutovialÒ 0.45-lm nylon membrane ®lter (Whatman, USA). The separation, quantitation and identi®cation of TNT and its degradation products were performed with a Waters 996 HPLC system equipped with a Waters 717 plus Autosampler, a photodiode array detector, a pre-column ®lter (Sigma±Aldrich, USA) and a LichrospherÒ 100 RP-8 column (25 cm 4:0 mm ID, 5 lm) (Hewlett Packard, USA). Mixtures of acetonitrile (A) and water (B) were used as the mobile phase. The following gradient elution was run: 10% to 55% A from 0 to 5 min, 55% to 100% A from 5 to 15 min, held at 100% A from 15 to 22 min, 100% to 10% A from 22 to 24 min. The ¯ow rate was 1.0 ml/min. The detection wavelength was 228 nm. The temperature of the autosampler was controlled at 7°C. The identi®cation of degradation products of TNT was carried out by comparison of the retention time and UV-visible absorption spectra with that of standards. The determination of ribo¯avin was conducted on a Fluomax 2 spectro¯uorometer (Instruments S.A., USA).
2.2. TNT degradation Surface freshwater samples were collected in the spring season from the Ross Barnett Reservoir at Ridgeland, Mississippi. The pH of the water samples ranged from 7.1 to 7.5, and the temperature ranged from 20°C to 25°C. To 150-ml quartz ¯asks, 25 ml of freshwater or nanopure water and freshwater containing certain concentration of ribo¯avin and 250 ll of TNT solution (2 g/l) were added and incubated in duplicates. The ¯asks were suspended in an outdoor tub which contained running water for maintaining the water temperature at '20°C. The water level in the ¯ask was about 3 cm below the surface of the cooling water. The ¯asks of dark exposure group were wrapped with aluminum foil. Killed group was accomplished with the water samples being autoclaved at 121.1°C at 15 psi for 20 min. All bottles were capped with silicone stoppers. When nanopure water was used for the experiments, pH of nanopure water was adjusted to the same pH as the freshwater by adding sodium hydroxide. The eect of dierent concentrations of ribo¯avin (0, 0.1, 1, 10, 100 lM) on the degradation of TNT was tested. It was found that 0.1 and 1 lM ribo¯avin in-
Fig. 1. Change of TNT concentration vs time. Initial TNT concentration 20 mg/l; LL live light group; LK killed light group; LLR live light group with ribo¯avin; LKR killed light group with ribo¯avin; LKw killed light group in nanopure water with ribo¯avin.
H. Cui et al. / Chemosphere 44 (2001) 621±625 Table 1 Initial pseudo-®rst order rate constants and half lives of TNT Groupa
Rate constantsb (min 1 )
Half lives (min)
LL LK LLR LKR LKw
0.0185 0.0004 0.0286 0.0003 0.0289 0.0001 0.0213 0.0006 0.0383 0.0014
37.46 24.24 23.96 32.55 18.09
a
LL live light group; LK killed light group; LLR live light group with ribo¯avin; LKR killed light group with ribo¯avin; LKw killed light group in nanopure water with ribo¯avin. b Mean S.D.
The excitation and emission wavelengths were set at 467 and 524 nm, respectively.
623
3. Results and discussion 3.1. Degradation of TNT The degradation of TNT in the freshwater samples was studied under dierent exposure treatments in the absence and presence of ribo¯avin. In order to examine the relative contribution of photolysis, microbial assemblages and freshwater matrix to TNT degradation, the experiments were conducted in ®ve groups: live light group, killed light group, live dark group, killed dark group and killed light group in nanopure water. In the dark, including live dark group and killed dark group, TNT was stable in 3 days tested. Under all light exposure conditions, a decrease in TNT concentration was observed. Fig. 1 shows the change of
Fig. 2. Chromatograms of TNT and its degradation products at dierent degradation time: (a) live light group; (b) killed light group; (c) live light group with ribo¯avin; (d) killed light group with ribo¯avin. Degradation time for (a) and (b) from bottom to top: 0, 15, 45, 90, 150, 200 min; Degradation time for (c) and (d) from bottom to top: 0, 15, 30, 60, 100, 150, 200 min. Retention time: peak 1 1.54 min, peak 2 9.47 min, peak 3 9.84 min, peak 4 (1,3,5-trinitrobenzene) 10.27 min, peak 5 (TNT) 11.13 min, peak 6 (ribo¯avin) 6.27 min.
624
H. Cui et al. / Chemosphere 44 (2001) 621±625
TNT concentration vs time in the absence and presence of ribo¯avin. The rate of TNT disappearance can be described by an apparent ®rst-order kinetic law in the ®rst 100 min. Initial pseudo-®rst-order rate constants and half lives were calculated by linear regression of ln (C0 /C) vs time as shown in Table 1. In the absence of ribo¯avin, the transformation of TNT in killed light group was more rapid than that in live light group. In the presence of ribo¯avin, the degradation rate of TNT was enhanced signi®cantly in live light group compared with the case without ribo¯avin, whereas the rate decreased somewhat in killed light group. TNT in the killed nanopure water was degraded at the highest rate. More than 99% of TNT was lost after 60 min for killed nanopure water, after 100 min for killed light group without ribo¯avin and live light group with ribo¯avin, after 150 min for live light group without ribo¯avin and killed light group with ribo¯avin, respectively. Therefore, sunlight is necessary for the degradation of TNT in the freshwater samples and microbial assemblages slow down the degradation of TNT in the freshwater samples. Ribo¯avin can stimulate the degradation of TNT in natural water environment. However, the freshwater matrix decreases the eciency of ribo¯avin.
Fig. 3. UV-visible absorption spectra of peak 1, 2, 3 and 4.
3.3. Transformation of ribo¯avin The comparative study on transformation of ribo¯avin in nanopure water and freshwater was carried out. Fig. 4 shows the change of ribo¯avin concentration vs time. Initial pseudo-®rst order rate constants and half
3.2. Degradation products of TNT Fig. 2 shows the chromatograms at dierent degradation times under various exposure treatment in the absence and presence of ribo¯avin. In the absence of ribo¯avin, only 1,3,5-trinitrobenzene (peak 4) and one unknown degradation product (peak 1) were observed at 10.27 and 1.54 min, respectively. When comparing Fig. 2(a) with (b), it is easy to see that there is no dierence in the degradation products between live light group and killed light group. In the presence of ribo¯avin, two new intermediates (peaks 2 and 3) were found at 9.47 and 9.84 min, respectively. These two intermediates were more easily degraded than TNT, disappearing completely after 60 min. In nanopure water, these two intermediates were also observed and disappeared after 30 min, indicating that they are less stable in nanopure water than in the fresh water. The results show that TNT transformation undergoes dierent pathways in the presence of ribo¯avin. The eort was made for the identi®cation of peaks 1, 2 and 3 with some standard compounds such as 2,4,6-trinitrobenzoic acid, 2,4-dinitrobenzoic acid, 3,5-dinitroaniline, 2,4,6-trinitrophenol, pyrogallol and phloroglucinol. However, no matching was seen with these unknown peaks in UV-visible absorption spectra and retention time. The corresponding UV-visible spectra of these peaks obtained by photodiode array detector are shown in Fig. 3.
Fig. 4. Change of ribo¯avin concentration vs time. Initial ribo¯avin concentration 1 lM, initial TNT concentration 20 mg/l. PWR ribo¯avin in nanopure water; PWRTNT ribo¯avin + TNT in nanopure water; FWR ribo¯avin in a freshwater; FWRTNT ribo¯avin + TNT in a freshwater. All groups were investigated under light exposure condition.
H. Cui et al. / Chemosphere 44 (2001) 621±625 Table 2 Initial pseudo-®rst order rate constants and half lives of ribo¯avin Groupa
Rate constantsb (min 1 )
Half lives (min)
PWR PWRTNT FWR FWRTNT
0.2368 0.0012 0.2074 0.0007 1.0298 0.0009 0.9320 0.0207
2.93 3.34 0.67 0.74
a PWR ribo¯avin in nanopure water; PWRTNT ribo¯avin + TNT in nanopure killed water; FWR ribo¯avin in a freshwater; FWRTNT ribo¯avin + TNT in a freshwater. All groups were investigated under light exposure condition. b Mean S.D.
lives of ribo¯avin were calculated by linear regression of ln (C0 /C) vs time as shown in Table 2. The degradation rate of ribo¯avin was much faster in freshwater than that in nanopure water, which implies that the components in the freshwater could catalyze the photolysis of ribo¯avin. The degradation rate of ribo¯avin was slightly slowed down by adding TNT. The reason that the degradation rate of TNT is faster in nanopure water in the presence of ribo¯avin than that in the freshwater may be due to the short life time of ribo¯avin in the freshwater. Lumichrome was detected to be the major decomposition product of ribo¯avin by HPLC. Other decomposition products such as lumi¯avin and formylmethyl¯avin reported by Mopper and Zika (1987) were not found in this case. 4. Conclusions Ribo¯avin was found to facilitate TNT degradation in a freshwater environment. Such work may lead to application for the treatment of TNT-contaminated surface water. It is possible to improve the eciency of ribo¯avin if the life time of ribo¯avin in the freshwater could be increased by suitable methods such as changing solution pH, adding metal ions as complex agents and derivatizing the functional group of ribo¯avin. Further work is in progress. Two new intermediates were observed by the addition of ribo¯avin. Thus TNT degradation in the freshwater environment undergoes dierent pathways in the presence of ribo¯avin. The identi®cation of degradation products and the pathways are under investigation with LC-MS.
625
Acknowledgements This research work was supported by: (1) NIHMBRS S06GM08047 (to JSU); (2) Department of Energy #DE-AC03-76SF00098 (to LBL); subcontract #6482515 to JSU; (3) Department of Energy DE-FG0297ER62451 (to UGA); subcontract #RR100-239/ 4891914 to JSU. We thank Dr. Hongtao Yu of Department of Chemistry of Jackson State University, Drs. Herbert Fredrickson and David Ringelberg of Environmental Laboratory of US Army Corps of Engineers, Waterways Experiment Station In Vicksburg, Mississippi for their assistance in this research.
References Boopathy, R., Kulpa, C.F., Manning, J., Montemagno, C.D., 1994. Metabolism of 2,4,6-trinitrotoluene by a Pseudomonas consortium under aerobic conditions. Curr. Microbiol. 28, 131±137. Crosby, D.G. (Ed.), 1998. Environmental Toxicology and Chemistry. Oxford University Press, Oxford, pp. 287±288. Hwang, H.-M., Slaughter, L.F., Cook, S.M., Cui, H., 2000. Photochemical and microbial degradation of 2,4,6-trinitrotoluene in a freshwater environment. Bull. Environ. Contam. Tox. 65, 228±235. Kaplan, L.A., Burlinson, N.E., Sitzmann, M.E., 1975. Photochemistry of TNT: Investigation of the Ôpink waterÕ problem. Part 2, Rep. No. NSWC/WOL/75-152 (AD-A020072), Naval Surface Weapons Center, White Oaks Laboratory, Silver Spring, MD. Kearney, P.C., Zeng, Q., Ruth, J.M., 1983. Oxidative pretreatment accelerates TNT metabolism in soil. Chemosphere 12, 1583±1597. Lang, P.S., Ching, W.-K., Willberge, D.M., Homann, M.R., 1998. Oxidative degradation of 2,4,6-trinitrotoluene by ozone in an electrohydraulic discharge reactor. Environ. Sci. Technol. 32, 3142±3148. Larson, R.A., Stackhouse, P.L., Crowley, T.O., 1992. Ribo¯avin tetraacetate: A potentially useful photosensitizing agent for the treatment of contaminated water. Environ. Sci. Technol. 26, 1792±1798. Mopper, K., Zika, R.G., 1987. Photochemistry of environmental aquatic systems. In: Zika, R.G., Copper, W.J. (Eds.), ACS Symposium Series 327. American Chemical Society, Washington, DC, pp. 174±190. Schmelling, D.C., Gray, K.A., 1995. Photocatalytic transformation and mineralization of 2,4,6-trinitrotoluene (TNT) in TiO2 slurries. Water Res. 29, 2651±2662.
www.elsevier.com/locate/chemosphere
Eect of photosensitizer ribo¯avin on the fate of 2,4,6-trinitrotoluene in a freshwater environment Hua Cui
a,b
, Huey-Min Hwang
a,*
, Sean Cook a, Kui Zeng
a
a
b
Department of Biology, Jackson State University, Jackson, MS 39217, USA Department of Chemistry, University of Science and Technology of China, Hefei, Anhui 230026, People's Republic of China Received 10 May 2000; received in revised form 20 July 2000; accepted 5 August 2000
Abstract The eect of ribo¯avin (1 lM) on the fate of TNT (20 mg/l) in a natural water environment was studied. The relative contribution of photolysis, microbial assemblages and freshwater matrix to TNT degradation was examined. The rates, extent and products of TNT and ribo¯avin transformation were compared under dierent experimental conditions. It was found that ribo¯avin signi®cantly enhanced the degradation of TNT in natural water environment. Thus it is a potentially useful photosensitizing agent for the treatment of TNT-contaminated surface water. Furthermore, in the presence of ribo¯avin, two new intermediates with max. absorption wavelength of 230 nm were found, demonstrating that transformation of TNT in the presence of ribo¯avin undergoes dierent pathways. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Ribo¯avin; TNT; Degradation; Freshwater; Fate; Environment
1. Introduction 2,4,6-trinitrotoluene (TNT) is a conventional explosive used by military forces worldwide, causing serious contamination of soil and groundwater (Boopathy et al., 1994). Environmental transformations of TNT and methods of remediation have been extensively studied (Schmelling and Gray, 1995). The photochemical degradation of TNT was considered both as a primary treatment technology and as a pretreatment to biodegradation. Several strategies such as direct solar photolysis, UV-peroxide, UV-ozone, and TiO2 photocatalysis have been developed for the treatment of TNT-contaminated water (Schmelling and Gray, 1995). Exposure of TNT to sunlight or near UV radiation results in rapid conversion of TNT into a variety of aromatic photolysis
*
Corresponding author. Fax: +1-601-979-2778. E-mail address: [email protected] Hwang).
(H.-M.
products such as 2,4,6-trinitrobenzoic acid, 1,3,5-trinitrobenzene, 3,5-dinitroaniline, 2,4,6-trinitrophenol, 4-amino-2,6-dinitrotoluene, 2,4-dinitrobenzoic acid, 2,4,6-trinitrobenzaldehyde, 2,4,6-trinitrobenzyl alcohol, 4,6-dinitroanthranil, 2,4,6-trinitrobenzonitrile, and azoxydimers (Kaplan et al., 1975; Kearney et al., 1983; Schmelling and Gray, 1995; Crosby, 1998; Lang et al., 1998), depending on the photolysis condition. Each method has its own advantages and disadvantages with varying degrees of success. Therefore, there is a need to explore more ecient strategies for the remediation of TNT. Photosensitizers have been used to facilitate the photochemical degradation of some organic compounds. It is proposed that photosensitizers absorb light energy, transform it into chemical energy, and transfer that energy under favorable conditions to photochemically unreactive substances. The mechanism may be involved in redox reactions (Larson et al., 1992). Ribo¯avin as a photosensitizer has been reported to sensitize the photochemical degradation of many
0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 3 3 3 - 7
622
H. Cui et al. / Chemosphere 44 (2001) 621±625
compounds in aqueous solutions (Mopper and Zika, 1987; Hwang et al., 2000). Ribo¯avin is easy to handle and environmentally benign, making it attractive for the treatment of environmental water. In this paper, the eect of ribo¯avin on the fate of TNT in a natural water environment was studied. The relative contribution of photolysis, microbial assemblages and freshwater matrix to TNT degradation was examined in the absence and presence of ribo¯avin. The rates, extent and products of TNT transformation under dierent experimental conditions were determined by high-performance liquid chromatography. The transformation of ribo¯avin under dierent experimental conditions was also compared by use of ¯uorescence analysis. 2. Experimental 2.1. Chemicals and solution 2,4,6-trinitrotoluene, 1,3,5-trinitrobenzene, 2,4,6trinitrobenzoic acid, 4-amino-2,6-dinitrotoluene and 2-amino-4,6-dinitrotoluene were purchased from Chemservice (West Chester, USA). 2,4,6-trinitrophenol, 3,5-dinitroaniline, 2,4-dinitrobenzoic acid and ribo¯avin were purchased from Aldrich (Milwaukee, USA). Pyrogallol and phloroglucinol were purchased from Sigma (ST. Louis, USA). 2 g/l TNT stock solution was prepared in 40% acetonitrile±water solution and kept in dark at 4°C. Nanopure water was prepared by a Barnstead Nanopure In®nity (Dubuque, USA).
creased the degradation of TNT. Thus 1 lM ribo¯avin was chosen for all experiments. 2.3. Sample analysis After the degradation, the samples were ®ltered through an AutovialÒ 0.45-lm nylon membrane ®lter (Whatman, USA). The separation, quantitation and identi®cation of TNT and its degradation products were performed with a Waters 996 HPLC system equipped with a Waters 717 plus Autosampler, a photodiode array detector, a pre-column ®lter (Sigma±Aldrich, USA) and a LichrospherÒ 100 RP-8 column (25 cm 4:0 mm ID, 5 lm) (Hewlett Packard, USA). Mixtures of acetonitrile (A) and water (B) were used as the mobile phase. The following gradient elution was run: 10% to 55% A from 0 to 5 min, 55% to 100% A from 5 to 15 min, held at 100% A from 15 to 22 min, 100% to 10% A from 22 to 24 min. The ¯ow rate was 1.0 ml/min. The detection wavelength was 228 nm. The temperature of the autosampler was controlled at 7°C. The identi®cation of degradation products of TNT was carried out by comparison of the retention time and UV-visible absorption spectra with that of standards. The determination of ribo¯avin was conducted on a Fluomax 2 spectro¯uorometer (Instruments S.A., USA).
2.2. TNT degradation Surface freshwater samples were collected in the spring season from the Ross Barnett Reservoir at Ridgeland, Mississippi. The pH of the water samples ranged from 7.1 to 7.5, and the temperature ranged from 20°C to 25°C. To 150-ml quartz ¯asks, 25 ml of freshwater or nanopure water and freshwater containing certain concentration of ribo¯avin and 250 ll of TNT solution (2 g/l) were added and incubated in duplicates. The ¯asks were suspended in an outdoor tub which contained running water for maintaining the water temperature at '20°C. The water level in the ¯ask was about 3 cm below the surface of the cooling water. The ¯asks of dark exposure group were wrapped with aluminum foil. Killed group was accomplished with the water samples being autoclaved at 121.1°C at 15 psi for 20 min. All bottles were capped with silicone stoppers. When nanopure water was used for the experiments, pH of nanopure water was adjusted to the same pH as the freshwater by adding sodium hydroxide. The eect of dierent concentrations of ribo¯avin (0, 0.1, 1, 10, 100 lM) on the degradation of TNT was tested. It was found that 0.1 and 1 lM ribo¯avin in-
Fig. 1. Change of TNT concentration vs time. Initial TNT concentration 20 mg/l; LL live light group; LK killed light group; LLR live light group with ribo¯avin; LKR killed light group with ribo¯avin; LKw killed light group in nanopure water with ribo¯avin.
H. Cui et al. / Chemosphere 44 (2001) 621±625 Table 1 Initial pseudo-®rst order rate constants and half lives of TNT Groupa
Rate constantsb (min 1 )
Half lives (min)
LL LK LLR LKR LKw
0.0185 0.0004 0.0286 0.0003 0.0289 0.0001 0.0213 0.0006 0.0383 0.0014
37.46 24.24 23.96 32.55 18.09
a
LL live light group; LK killed light group; LLR live light group with ribo¯avin; LKR killed light group with ribo¯avin; LKw killed light group in nanopure water with ribo¯avin. b Mean S.D.
The excitation and emission wavelengths were set at 467 and 524 nm, respectively.
623
3. Results and discussion 3.1. Degradation of TNT The degradation of TNT in the freshwater samples was studied under dierent exposure treatments in the absence and presence of ribo¯avin. In order to examine the relative contribution of photolysis, microbial assemblages and freshwater matrix to TNT degradation, the experiments were conducted in ®ve groups: live light group, killed light group, live dark group, killed dark group and killed light group in nanopure water. In the dark, including live dark group and killed dark group, TNT was stable in 3 days tested. Under all light exposure conditions, a decrease in TNT concentration was observed. Fig. 1 shows the change of
Fig. 2. Chromatograms of TNT and its degradation products at dierent degradation time: (a) live light group; (b) killed light group; (c) live light group with ribo¯avin; (d) killed light group with ribo¯avin. Degradation time for (a) and (b) from bottom to top: 0, 15, 45, 90, 150, 200 min; Degradation time for (c) and (d) from bottom to top: 0, 15, 30, 60, 100, 150, 200 min. Retention time: peak 1 1.54 min, peak 2 9.47 min, peak 3 9.84 min, peak 4 (1,3,5-trinitrobenzene) 10.27 min, peak 5 (TNT) 11.13 min, peak 6 (ribo¯avin) 6.27 min.
624
H. Cui et al. / Chemosphere 44 (2001) 621±625
TNT concentration vs time in the absence and presence of ribo¯avin. The rate of TNT disappearance can be described by an apparent ®rst-order kinetic law in the ®rst 100 min. Initial pseudo-®rst-order rate constants and half lives were calculated by linear regression of ln (C0 /C) vs time as shown in Table 1. In the absence of ribo¯avin, the transformation of TNT in killed light group was more rapid than that in live light group. In the presence of ribo¯avin, the degradation rate of TNT was enhanced signi®cantly in live light group compared with the case without ribo¯avin, whereas the rate decreased somewhat in killed light group. TNT in the killed nanopure water was degraded at the highest rate. More than 99% of TNT was lost after 60 min for killed nanopure water, after 100 min for killed light group without ribo¯avin and live light group with ribo¯avin, after 150 min for live light group without ribo¯avin and killed light group with ribo¯avin, respectively. Therefore, sunlight is necessary for the degradation of TNT in the freshwater samples and microbial assemblages slow down the degradation of TNT in the freshwater samples. Ribo¯avin can stimulate the degradation of TNT in natural water environment. However, the freshwater matrix decreases the eciency of ribo¯avin.
Fig. 3. UV-visible absorption spectra of peak 1, 2, 3 and 4.
3.3. Transformation of ribo¯avin The comparative study on transformation of ribo¯avin in nanopure water and freshwater was carried out. Fig. 4 shows the change of ribo¯avin concentration vs time. Initial pseudo-®rst order rate constants and half
3.2. Degradation products of TNT Fig. 2 shows the chromatograms at dierent degradation times under various exposure treatment in the absence and presence of ribo¯avin. In the absence of ribo¯avin, only 1,3,5-trinitrobenzene (peak 4) and one unknown degradation product (peak 1) were observed at 10.27 and 1.54 min, respectively. When comparing Fig. 2(a) with (b), it is easy to see that there is no dierence in the degradation products between live light group and killed light group. In the presence of ribo¯avin, two new intermediates (peaks 2 and 3) were found at 9.47 and 9.84 min, respectively. These two intermediates were more easily degraded than TNT, disappearing completely after 60 min. In nanopure water, these two intermediates were also observed and disappeared after 30 min, indicating that they are less stable in nanopure water than in the fresh water. The results show that TNT transformation undergoes dierent pathways in the presence of ribo¯avin. The eort was made for the identi®cation of peaks 1, 2 and 3 with some standard compounds such as 2,4,6-trinitrobenzoic acid, 2,4-dinitrobenzoic acid, 3,5-dinitroaniline, 2,4,6-trinitrophenol, pyrogallol and phloroglucinol. However, no matching was seen with these unknown peaks in UV-visible absorption spectra and retention time. The corresponding UV-visible spectra of these peaks obtained by photodiode array detector are shown in Fig. 3.
Fig. 4. Change of ribo¯avin concentration vs time. Initial ribo¯avin concentration 1 lM, initial TNT concentration 20 mg/l. PWR ribo¯avin in nanopure water; PWRTNT ribo¯avin + TNT in nanopure water; FWR ribo¯avin in a freshwater; FWRTNT ribo¯avin + TNT in a freshwater. All groups were investigated under light exposure condition.
H. Cui et al. / Chemosphere 44 (2001) 621±625 Table 2 Initial pseudo-®rst order rate constants and half lives of ribo¯avin Groupa
Rate constantsb (min 1 )
Half lives (min)
PWR PWRTNT FWR FWRTNT
0.2368 0.0012 0.2074 0.0007 1.0298 0.0009 0.9320 0.0207
2.93 3.34 0.67 0.74
a PWR ribo¯avin in nanopure water; PWRTNT ribo¯avin + TNT in nanopure killed water; FWR ribo¯avin in a freshwater; FWRTNT ribo¯avin + TNT in a freshwater. All groups were investigated under light exposure condition. b Mean S.D.
lives of ribo¯avin were calculated by linear regression of ln (C0 /C) vs time as shown in Table 2. The degradation rate of ribo¯avin was much faster in freshwater than that in nanopure water, which implies that the components in the freshwater could catalyze the photolysis of ribo¯avin. The degradation rate of ribo¯avin was slightly slowed down by adding TNT. The reason that the degradation rate of TNT is faster in nanopure water in the presence of ribo¯avin than that in the freshwater may be due to the short life time of ribo¯avin in the freshwater. Lumichrome was detected to be the major decomposition product of ribo¯avin by HPLC. Other decomposition products such as lumi¯avin and formylmethyl¯avin reported by Mopper and Zika (1987) were not found in this case. 4. Conclusions Ribo¯avin was found to facilitate TNT degradation in a freshwater environment. Such work may lead to application for the treatment of TNT-contaminated surface water. It is possible to improve the eciency of ribo¯avin if the life time of ribo¯avin in the freshwater could be increased by suitable methods such as changing solution pH, adding metal ions as complex agents and derivatizing the functional group of ribo¯avin. Further work is in progress. Two new intermediates were observed by the addition of ribo¯avin. Thus TNT degradation in the freshwater environment undergoes dierent pathways in the presence of ribo¯avin. The identi®cation of degradation products and the pathways are under investigation with LC-MS.
625
Acknowledgements This research work was supported by: (1) NIHMBRS S06GM08047 (to JSU); (2) Department of Energy #DE-AC03-76SF00098 (to LBL); subcontract #6482515 to JSU; (3) Department of Energy DE-FG0297ER62451 (to UGA); subcontract #RR100-239/ 4891914 to JSU. We thank Dr. Hongtao Yu of Department of Chemistry of Jackson State University, Drs. Herbert Fredrickson and David Ringelberg of Environmental Laboratory of US Army Corps of Engineers, Waterways Experiment Station In Vicksburg, Mississippi for their assistance in this research.
References Boopathy, R., Kulpa, C.F., Manning, J., Montemagno, C.D., 1994. Metabolism of 2,4,6-trinitrotoluene by a Pseudomonas consortium under aerobic conditions. Curr. Microbiol. 28, 131±137. Crosby, D.G. (Ed.), 1998. Environmental Toxicology and Chemistry. Oxford University Press, Oxford, pp. 287±288. Hwang, H.-M., Slaughter, L.F., Cook, S.M., Cui, H., 2000. Photochemical and microbial degradation of 2,4,6-trinitrotoluene in a freshwater environment. Bull. Environ. Contam. Tox. 65, 228±235. Kaplan, L.A., Burlinson, N.E., Sitzmann, M.E., 1975. Photochemistry of TNT: Investigation of the Ôpink waterÕ problem. Part 2, Rep. No. NSWC/WOL/75-152 (AD-A020072), Naval Surface Weapons Center, White Oaks Laboratory, Silver Spring, MD. Kearney, P.C., Zeng, Q., Ruth, J.M., 1983. Oxidative pretreatment accelerates TNT metabolism in soil. Chemosphere 12, 1583±1597. Lang, P.S., Ching, W.-K., Willberge, D.M., Homann, M.R., 1998. Oxidative degradation of 2,4,6-trinitrotoluene by ozone in an electrohydraulic discharge reactor. Environ. Sci. Technol. 32, 3142±3148. Larson, R.A., Stackhouse, P.L., Crowley, T.O., 1992. Ribo¯avin tetraacetate: A potentially useful photosensitizing agent for the treatment of contaminated water. Environ. Sci. Technol. 26, 1792±1798. Mopper, K., Zika, R.G., 1987. Photochemistry of environmental aquatic systems. In: Zika, R.G., Copper, W.J. (Eds.), ACS Symposium Series 327. American Chemical Society, Washington, DC, pp. 174±190. Schmelling, D.C., Gray, K.A., 1995. Photocatalytic transformation and mineralization of 2,4,6-trinitrotoluene (TNT) in TiO2 slurries. Water Res. 29, 2651±2662.