- Email: [email protected]
Dissolved organic matter inhibition of sonochemical degradation of aqueous polycyclic aromatic hydrocarbons
Ultrasonics Sonochemistry 6 (1999) 175–183 www.elsevier.nl/locate/ultsonch
Dissolved organic matter inhibition of sonochemical degradation of aqueous polycyclic aromatic hydrocarbons E. Taylor Jr., B.B. Cook, M.A. Tarr * Department of Chemistry, University of New Orleans, New Orleans, LA 70148-2820, USA Received 9 March 1999; received in revised form 20 March 1999
Abstract Sonochemical degradation of aqueous polycyclic aromatic hydrocarbons (PAHs) was found to be rapid in the absence of other dissolved compounds (k=0.006–0.015 s−1). In the presence of 20 mg C l−1 fulvic acid, first-order PAH degradation rate constants decreased from 2.3- to 3.7-fold. Similar results were obtained with added benzoic acid, a crude analog for fulvic acid. In natural waters, PAH degradation was almost completely inhibited. Analysis of the kinetic behavior and reaction products indicates that PAHs are most likely degraded through a radical cation mechanism. Hydroxyl radical appeared to play an insignificant role in the degradation. Inhibited degradation was probably the result of either altered cavitation processes or isolation of the PAH away from cavitation sites. © 1999 Elsevier Science B.V. All rights reserved. Keywords: Degradation; Dissolved organic matter; Fulvic acid; Polycyclic aromatic hydrocarbons; Sonochemistry
1. Introduction Sonochemistry has been demonstrated as a possible tool for degradation of aqueous pollutants. Cavitation during ultrasonic treatment of aqueous solutions results in the concentration of the sonic energy into a very small volume, thereby producing local areas of high energy. Within these localized high-energy regions, chemical transformations become quite probable. The temperature and pressure in the region of cavitation can reach 3000–5000 K and 500–10 000 atm. [1–4]. Three reactive zones exist in the region of cavitation: the gas phase, the gas–liquid interface, and the liquid directly surrounding the cavitation bubble [3]. Numerous reports indicate the formation of hydroxyl radical and hydrogen atoms in the vicinity of the cavitation site [1– 7]. Despite some controversy over the formation of hydrated electrons during sonication of neutral pH water, recent work indicates that hydrated electrons are not formed under these conditions [8]. Some sonochemical transformations that have been observed include degradation of nucleic acid bases [9] and degradation of aqueous carbon tetrachloride [10], 1,1,1-trichloroethane [11], methylene chloride [12], and * Corresponding author. Fax: +1-504-280-6860. E-mail address: [email protected] (M.A. Tarr)
trichloroethylene [12]. In addition, chloroaromatics [13– 15], phenols [14,16 ], substituted benzenes [17], and polycyclic aromatic hydrocarbons (PAHs) [18] have also been degraded by sonochemistry. These studies indicate the possibility of utilizing sonochemistry for degradation of pollutants, including chlorinated hydrocarbons, petroleum products and by-products, and light and dense non-aqueous phase liquids (LNAPLs and DNAPLs). Three mechanisms have been proposed for sonochemical degradation: (1) oxidation by hydroxyl radical, (2) pyrolytic decomposition, and (3) supercritical water oxidation [19]. Despite the extent of sonochemical degradation studies, little work has been conducted on the effects of matrix compounds present in the sample. The presence of matrix components, such as dissolved organic and inorganic matter, may dramatically alter the sonochemical processes. Since both industrial waste streams and contaminated waters (surface and groundwater) generally contain a wide array of both pollutant and nonpollutant chemicals, it is necessary to understand the effect of such matrix compounds before sonochemistry can be successfully utilized as a degradation tool. This study has assessed the effects of natural organic matter (NOM ) on the sonochemical degradation of hydrophobic pollutants. Three polycyclic aromatic hydrocarbons (see structures in Fig. 1) were degraded
1350-4177/99/$ – see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S1 3 5 0 -4 1 7 7 ( 9 9 ) 0 0 01 5 - 2
176
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
Fig. 1. Structures of anthracene (a), phenanthrene (b), pyrene (c), and a proposed average structure [20] of Suwannee River fulvic acid (d).
in pure water, in solutions of Suwannee River fulvic acid, in the presence of benzoic acid (a crude analog for fulvic acid), and in two natural surface waters. Fulvic acids are a class of naturally occurring polyphenolic compounds that are typically derived from plant lignins. Fulvic acids, as well as other classes of natural organic matter, are ubiquitous in surface water, ground water, soil, and sediment. An average structure for fulvic acid is given in Fig. 1.
2. Experimental Purified water (NP) was obtained by further purification of distilled water with a Barnstead (Dubuque, IA) NanopureUV water treatment system. Natural water samples were collected from sites in Southeast Louisiana at Crawford Landing on the West Pearl River (PR) and from a small water body connecting Lake Pontchartrain and Lake Maurepas (LM ). All natural water samples were filtered using pre-combusted 0.5-mm glass fiber filters (Rundfilter MN, Machery-Nagel, Alberta, Canada) and were stored in the dark at 4°C. Suwannee River fulvic acid standard material was obtained from the International Humic Substances Society (http://www.ihss.gatech.edu/index.html ). Pyrene (99%), anthracene (99+%), phenanthrene (99.5+%), benzoic acid (99.5%), and p-hydroxybenzoic acid (99+%) were purchased from Aldrich. All chemicals were used as received. Aqueous PAH solutions were prepared in the following manner. First, a more concen-
trated (mM ) stock solution was prepared in hexane. A small aliquot of this solution was transferred to a clean, dry volumetric flask. The solvent was evaporated under a stream of nitrogen. NP water was added to the flask, and the solute was dissolved by low-energy (60 W ) sonication in a bath sonicator for ~30 min. Final PAH concentrations of the resulting aqueous solutions were in the 0.1–0.5 mM range and were well below the solubility limit for the PAH. Fulvic acid solutions were prepared by mixing aqueous fulvic acid solutions with the aqueous PAH solution. These solutions were prepared so that the PAH concentration remained constant for all sonicated samples, but the fulvic acid concentration was varied. Natural water solutions were prepared by mixing the aqueous PAH solution with an equal volume of the natural water. To avoid possible alteration of the natural organic matter, natural waters and fulvic acid solutions were not placed into the bath sonicator. All solutions were stored in the dark, and natural waters and fulvic acid solutions were kept refrigerated (4°C ). All samples were sonicated using an Ace Glass ( Vineland, NJ ) 600 W sonochemical apparatus operating at 20 kHz. A 0.5-inch-diameter titanium probe was used with the power supply set to maximum amplitude. Sonication was carried out with a 10 or 20% duty cycle (sonication on for 1 s, off for 9 s or on for 1 s, off for 4 s). The low duty cycle was used to avoid excess heating. Samples were contained in an all-glass, waterjacketed reaction vessel with the cooling water maintained at 20°C. The vessel was sealed by O-rings and glass stoppers during sonication. Solutions were air-
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
equilibrated, and were sonicated under a headspace of air. To avoid any photochemical reactions, all samples were kept in the dark during and after sonication. Samples were typically 50–100 ml in volume, and 1– 5-ml aliquots were removed for analysis at regular time intervals. The tip of the titanium probe was polished on a regular basis to maintain a smooth surface. PAH fluorescence was monitored with a PTI (Monmouth Junction, NJ ) Quantamaster spectrofluorometer. Anthracene and phenanthrene were excited at 251 nm, and pyrene was excited at 319 nm. The fluorescence emission intensity was measured at 377 nm for anthracene, 345 nm for phenanthrene, and 370 nm for pyrene. Two- or 5-nm excitation and emission slits were used. For all samples, background fluorescence spectra were acquired from solutions without PAH and were subtracted from the spectra of the PAHs. Due to the relatively short measurement times, no significant photobleaching of the PAHs was observed during fluorescence measurements. Some of the sonicated samples were analyzed by HPLC with fluorescence detection. A Varian 5000 chromatograph was used with a 25 cm×4.6 mm, 5-mm Spherisorb ODS-2 column (Alltech Associates, Deerfield, IL). The mobile phase was 90% methanol and 10% water at a flow rate of 1 ml min−1. Analytes were detected using a Dionex (Sunnyvale, CA) FD300 fluorescence detector. Samples were injected using a 100-ml loop. The hydroxyl radical was quantitated using benzoic acid as a chemical probe [21]. The formation of phydroxybenzoic acid was monitored by HPLC with absorbance detection (254 nm) [21,22]. The observed first-order rate constant for p-hydroxybenzoic acid formation was divided by the second-order rate constant for the hydroxyl radical–benzoic acid reaction to calculate the hydroxyl radical concentration.
177
3. Results and discussion Observation of the PAH fluorescence spectrum as a function of sonication time indicated a decrease in fluorescence intensity but no change in the spectral shape. The consistency of the fluorescence spectral shape indicated that there were no fluorescent interferences with the excitation and emission wavelengths used. We therefore used the fluorescence intensity at the emission wavelength maximum as a measure of PAH concentration. Fluorescence data for degradation of phenanthrene in pure water are presented in Fig. 2. Anthracene and pyrene gave similar results. All three PAHs studied exhibited pseudo-first-order loss during sonolysis in pure water. Fig. 3 illustrates the first-order behavior observed for phenanthrene degradation in pure water. All the PAHs degraded fairly rapidly in pure water, with firstorder rate constants ranging from 0.006 to 0.015 s−1 (see Table 1). These rate constants are expressed in terms of actual sonication time and do not include the ‘off ’ period of the duty cycle. In order to prevent volatilization losses in our experiments, a sealed reaction vessel was used. Any volatilization of PAH would have resulted in deposition of the PAH back to the solution or onto the glassware. Postsonication extraction of the reaction vessel with acetone indicated no loss of PAH onto the reaction vessel. Therefore, it was assumed that volatilization losses were negligible, and all observed losses of PAH were due to chemical transformation. This result is reasonable, based on the low volatility of the PAHs used and the sealed design of the reaction vessel. For all three PAHs studied, the addition of fulvic acid resulted in a decrease in the PAH degradation rate constant. Fig. 4 illustrates the observed first-order rate constants as a function of fulvic acid concentration. Phenanthrene showed decreasing rate constants as the
Fig. 2. Fluorescence spectra of phenanthrene during sonication. From top to bottom, traces are after 0, 1, 2, 3, and 4 min of sonication.
178
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
Fig. 3. Observed first-order degradation of phenanthrene during sonication. A log plot with regression data is shown in the inset.
fulvic acid concentration increased. Pyrene and anthracene showed an initial drop in rate constant upon addition of a small amount of fulvic acid, but showed little additional change in rate constant with increasing fulvic acid concentration. The observations for pyrene and anthracene cannot be explained by a simple reactive transient scavenging model. During sonication, reactive transients, such as hydroxyl radicals, are formed. These transients may then act as oxidants for the degradation of pollutants present. However, scavenging of reactive transients would result in a continuously decreasing rate constant with added scavenger (fulvic acid in this case), with the rate constant eventually approaching zero. We have modeled a simple scavenging system with a single reactive transient using numerical methods. A fixed reactive transient production rate and a fixed background scavenging rate were used. The relative rate constants of the added scavenger and the PAH were varied. The model was then adjusted for each PAH to approximate the change in observed k at the lowest level of added FA. Although this model is only qualitative, it yields valuable information about the expected shape of the k vs. [FA] curves for simple scavenging behavior. The predicted behavior for reactive transient scavenging
is illustrated in Fig. 4 along with the experimental observations. None of the three compounds studied showed experimental curves that matched those predicted for scavenging behavior. Using this simple model, no conditions could be found that reproduced the experimental results. Therefore, the changes in k observed upon addition of FA cannot be simply due to scavenging of a reactive transient by the FA, and some other mechanism(s) must be responsible. One important reactive transient present in sonochemical systems is the hydroxyl radical. The concentration of hydroxyl radical in these systems was measured using benzoic acid as a radical trap [21,22]. Results indicated that the hydroxyl radical concentration present is not sufficient to explain the observed PAH degradation kinetics. For pyrene, after 3 min of sonication, ~60% of the pyrene had been degraded. Based on the measured hydroxyl radical concentration in this system (~4×10−16 M ) and a pyrene-hydroxyl radical rate constant of ~1×1010 [23], reaction with hydroxyl radical accounts for <1% loss of pyrene. Similar results were obtained for the other PAHs, clearly indicating that the hydroxyl radical is not playing a major role in the sonochemical degradation of these compounds.
Table 1 First-order rate constants (s−1) for PAH sonochemical degradation (all values ×10−3)
Anthracene Phenanthrene Pyrene
NP
PR
LM
FA (10 mg C l−1)
FA (20 mg C l−1)
15±6 (7)a 5.9±0.3 (3) 6±3 (7)
b b 2.3 (1)
b b 1 (1)
5.72±0.03 (2) 3.4±0.3 (2) NDc
4.1 (1) 1.8 (1) 2.58±0.06 (2)
a The error is ±1SD. Parenthetical values indicate the number of measurements. b Little or no degradation observed; did not follow first order kinetics. c ND, not determined.
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
179
(a)
(b)
(c)
Fig. 4. Observed first-order rate constants for anthracene (a), phenanthrene (b), and pyrene (c) sonochemical degradation as a function of added Suwannee river fulvic acid concentration. The dashed lines indicate the predicted behavior for a single reactive transient which is scavenged by the added fulvic acid. Error bars indicate one standard deviation. Error was not determined for the points at 21 mg C l−1. For anthracene (a), the error bars for the point at 10 mg C l−1 are smaller than the marker.
180
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
Previously reported studies on halocarbons have also revealed that hydroxyl radical concentrations were not sufficient to explain degradation rates and therefore concluded that pyrolysis was the likely mechanism [24]. Using HPLC with fluorescence detection, a single fluorescent product was observed for the sonochemical degradation of pyrene in pure water. This product was also degraded under continued sonication to yield no additional fluorescent products. No fluorescent products were observed for anthracene or phenanthrene. However, it is likely that all three PAHs have oxygenated products that likely further degrade under continued sonication. These intermediate degradation products are probably very important since oxygenated PAHs are often more toxic and carcinogenic than the parent PAH [18,25,26 ]. The pyrene fluorescent product has not yet been identified; however, it does have the same retention time as the photochemical degradation product of aqueous pyrene. Given the selectivity of fluorescence detection and the coincident retention times, there is little doubt that the sonochemical and photochemical degradation products are the same compound. The shorter retention time of this product indicates that it is likely more polar than pyrene. Based on these observations, it is postulated that the product is a dihydroxypyrene, a dihydrodihydroxypyrene, a pyrenedione, or another similar oxygenated pyrene. A phenanthrene-diol has been previously observed upon sonication of phenanthrene [18], and similar oxygenated compounds have been previously reported from photochemical degradation of PAHs [27– 29]. Furthermore, since both sonochemistry and photochemistry result in the same pyrene degradation product, it is likely that the product is formed by a similar mechanism. A plausible mechanism involves the formation of the pyrene radical cation, followed by reaction with water or molecular oxygen (R. Dabestani, personal communication). Degradation of pyrene by hydroxyl radicals (generated by Fenton’s reagent [22,30]) did not produce any observable fluorescent products. The above evidence indicates that the hydroxyl radical is not an important species and suggests a sonochemical mechanism involving formation of PAH radical cation species via high-temperature pyrolysis. Such reactions would require the PAH to be in very close proximity to the cavitation site or even within the cavitation bubble. The presence of FA results in a significant inhibition of the degradation of PAHs in aqueous solution. In the presence of ~20 mg C l−1 fulvic acid (~50 mM FA based on the average MW for Suwannee River fulvic acid of ~800 g mol−1 [31]), the first-order rate constant for anthracene dropped by a factor of 3.7, the rate constant for phenanthrene decreased by a factor of 3.3, and the rate constant for pyrene decreased by a factor of 2.3. Significant decreases in PAH rate constants were observed at fulvic acid concentrations as low as ~1 mM.
This added concentration of FA is comparable to the pyrene and phenanthrene concentrations used (0.25 mM ). Even trace amounts (mM ) of matrix species are sufficient to dramatically reduce the degradation efficiency of trace pollutants. The observed changes in rate constant upon addition of fulvic acid could be due to (1) changes in the cavitation process (caused for example by altered surface tension) or (2) sequestering of the PAH away from cavitation sites. Addition of salt has reportedly enhanced the sonochemical degradation of aromatic compounds by increasing their residence in or at the surface of cavitation bubbles [14]. In the current study, the matrix components may have removed the PAHs from the site of cavitation. Binding of the PAHs to FA may have this effect. However, even at the highest concentration of fulvic acid used, the PAHs were predicted to be less than 30% bound to FA based on reported and calculated K values [32–34]. This degree of binding to FA is not oc sufficient to explain the roughly threefold decrease in rate constants observed. Furthermore, the shape of the k vs. [FA] curves (Fig. 4) does not follow the shape expected if binding to FA were the primary mechanism for decreased reactivity. Partitioning of the fulvic acid across the air–water interface of the cavitation bubble could produce a barrier for entry of the PAH into the interior of the bubble, thereby minimizing the PAH reactivity. Alternatively, the presence of non-pollutant organic compounds in the gas phase within cavitation bubbles may lower the temperature upon bubble collapse [35], resulting in decreased PAH oxidation. Benzoic acid was also added to solutions of anthracene prior to sonication. Benzoic acid was used because it is a known hydroxyl radical scavenger and since it is an aromatic carboxylic acid, somewhat representative of the groups present in fulvic acid. The rate constant for anthracene degradation as a function of benzoic acid concentration is depicted in Fig. 5. This plot has the same shape as the analogous plots for fulvic acid. As stated above, the shape of these curves cannot be explained by a scavenging model. Addition of a free radical scavenger would result in decreasing rate constants that would approach zero as the scavenger concentration increased. With the addition of either benzoic acid or fulvic acid, both anthracene and pyrene showed initial decreases in rate constant followed by fairly constant non-zero rate constants with further increases in benzoic acid or fulvic acid concentration. Furthermore, the molarity of BA was significantly higher than FA because BA has a much lower molecular weight (at 21 mg C l−1, [BA]=172 mM and [FA]#26 mM ). Despite this large difference in molarity, the addition of BA or FA gave similar results. These observations indicate that scavenging processes are not the primary cause of decreased rate constants. The most likely expla-
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
181
Fig. 5. Observed first-order rate constants for anthracene as a function of added benzoic acid concentration. Error was determined only for the 0 mg C l−1 point and is reported as one standard deviation.
nations for the observed behavior are either (1) a change in cavitation processes due to altered surface tension or other solvent properties or (2) sequestering of the PAH away from the cavitation site. This latter explanation could involve partitioning of the benzoic acid or fulvic acid across the air–water interface of the initially formed cavitation bubble, thereby excluding other molecules from access to high-temperature regions occurring upon collapse of the cavitation bubble. A previous study has indicated competitive effects when an aqueous mixture of chlorobenzene and 4-chlorophenol was sonicated [13]. This report indicated that chlorophenol degradation was inhibited until chlorobenzene was degraded to a low concentration. Chlorobenzene was believed to be degraded by pyrolysis within the cavitation bubble, whereas chlorophenol was believed to be degraded by hydroxyl radicals outside the bubble. Although the effects observed in the previous report are likely different than those observed here, the previous work does illustrate that the presence of matrix compounds can dramatically affect pollutant degradation rates. Although the above data clearly demonstrate that natural organic matter significantly reduces the sonochemical degradation efficiency of PAHs, two natural water samples were also studied in order to extend these results to real-world samples. In addition to NOM content, these natural waters contained dissolved inor-
ganic matter that may also interfere with degradation by altering the cavitation process or scavenging reactive transients. Table 2 presents some compositional data for the natural waters used. In contrast to the rapid loss of PAH in pure water sonications, degradation rates were significantly retarded in the natural water samples. Fig. 6 illustrates the PAH concentrations as a function of sonication time in pure water and the natural waters. Due to background fluorescence from the natural waters, increased noise in the PAH concentration measurements was observed despite background correction. This result was caused by slight changes in the NOM fluorescence upon sonication. In pure water, phenanthrene was almost completely degraded after 5 min of sonication, whereas in the natural waters, little or no loss of phenanthrene was observed. Similar results were observed for anthracene and pyrene, as depicted in Fig. 6. First-order rate constants for each PAH in pure and natural water are presented in Table 1. The natural water matrices resulted in significant to nearly complete inhibition of the PAH degradation. The natural waters used showed much greater inhibition of PAH degradation than was observed for the fulvic acid studies. It is clearly illustrated that the ability to degrade PAHs by sonochemistry in natural waters is severely limited by matrix compounds.
Table 2 DOC and Na content of natural waters used
4. Conclusions
Pearl River Lake Maurepas
DOC (mg C l−1)
Na (ppm)
15 61
6.5 293
The results of this study indicate that sonochemical degradation of hydrophobic pollutants can be significantly inhibited by other dissolved species. Three possible explanations have been hypothesized: (1) matrix
182
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
(a)
(b)
(c) Fig. 6. PAH concentration as a function of sonication time for NP water (&), PR water ($), and LM water (+).
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
compounds scavenge reactive transients preventing PAH degradation, (2) the presence of additional dissolved species alters the cavitation process, and (3) dissolved matrix species may inhibit PAH access to cavitation sites. The third explanation is currently believed to be the most likely. Hydroxyl radicals, commonly present during sonication of aqueous solutions, were not present in sufficient quantities to account for the degree of PAH degradation. This result, along with the observation of a polar product that appears to be the same as the photochemical product, suggest that pyrolysis and/or combustiontype reactions proceeding via a PAH radical cation may be predominant in the sonolysis of PAHs in aqueous solution. The utility of sonochemical treatment for the degradation of aqueous hydrophobic pollutants is likely severely limited by the presence of dissolved matrix compounds. Although the current study focused on natural organic matter, similar results are expected for matrix compounds in industrial waste streams.
Acknowledgements This work was supported by the Louisiana Board of Regents under grant LEQSF(1998-01)-RD-A-34 and by the Academy of Applied Sciences Research and Engineering Apprenticeship Program (subgrant 705).
References [1] K.S. Suslick, D.A. Hammerton, R.E. Cline Jr., J. Am. Chem. Soc. 108 (1986) 5641. [2] K.L. Suslick, Science 247 (1990) 1439. [3] P. Riesz, D. Berdahl, C.L. Christman, Environ. Hlth Persp. 64 (1985) 233. [4] V. Misik, N. Miyoshi, P. Riesz, J. Phys. Chem. 99 (1995) 3605. [5] K. Makino, M.M. Mossoba, P. Riesz, J. Phys. Chem. 87 (1983) 1369. [6 ] K. Makino, M.M. Mossoba, P. Riesz, J. Am. Chem. Soc. 104 (1982) 3537. [7] P. Riesz, T. Kondo, Free Rad. Biol. Med. 13 (1992) 247. [8] V. Misik, P. Riesz, J. Phys. Chem. A 101 (1997) 1441.
183
[9] J.R. McKee, C.L. Christman, W.D. O’Brien, S.Y. Wang, Biochemistry 16 (1977) 4651. [10] P.K. Chendke, H.S. Fogler, J. Phys. Chem. 87 (1983) 1362. [11] M.S. Toy, R.S. Stringham, T.O. Passel, Pollution Prevention in Industrial Processes, J.J. Breen, M.J. Dellarco (Eds.), ACS Symposium Series Vol. 508, American Chemical Society, Washington, DC, 1992, p. 284. [12] H.M. Cheung, A. Bhatnangar, G. Jansen, Environ. Sci. Technol. 25 (1991) 1510. [13] C. Petrier, Y. Jiang, M.F. Lamy, Environ. Sci. Technol. 32 (1998) 1316. [14] J.D. Seymour, R.B. Gupta, Ind. Eng. Chem. Res. 36 (1997) 3453. [15] P. Kruus, R.C. Burk, M.H. Entezari, R. Otson, Ultrason. Sonochem. 4 (1997) 229. [16 ] C. Petrier, A. Francony, Ultrason. Sonochem. 4 (1997) 295. [17] A. de Visscher, H. van Langenhove, P. van Eenoo, Ultrason. Sonochem. 4 (1997) 145. [18] P.E. Wheat, M.A. Tumeo, Ultrason. Sonochem. 4 (1997) 55. [19] M.R. Hoffmann, I. Hua, R. Hochemer, Ultrason. Sonochem. 3 (1996) S163. [20] J.A. Leenheer, D.M. McKnight, E.M. Thurman, P. MacCarthy, in: R.C. Averett, J.A. Leenheer, D.M. McKnight, K.A. Thorn (Eds.), Humic Substances in the Suwannee River Georgia: Interactions Properties and Proposed Structures (1994) 195, US Geological Survey Water-supply Paper 2373. [21] X. Zhou, K. Mopper, Mar. Chem. 30 (1990) 71. [22] M.E. Lindsey, M.A. Tarr, Chemosphere (1999) submitted. [23] M.E. Lindsey, M.A. Tarr, Environ. Sci. Technol. (1999) submitted. [24] K. Hirai, Y. Nagata, Y. Maeda, Ultrason. Sonochem. 3 (1996) S205. [25] L. Ren, X.-D. Huang, B.J. McConkey, D.G. Dixon, B.M. Greenberg, Ecotoxicol. Environ. Saf. 28 (1994) 160. [26 ] J.E. Weinstein, J.T. Oris, J.T. Taylor, Aquatic Toxicol. 39 (1997) 1. [27] S.P. Zingg, M.E. Sigman, Photochem. Photobiol. 57 (1993) 453. [28] R. Dabestani, M. Melson, M.E. Sigman, Photochem. Photobiol. 64 (1996) 80. [29] R. Dabestani, K.J. Ellis, M.E. Sigman, J. Photochem. Photobiol. A 86 (1995) 231. [30] C. Walling, Acc. Chem. Res. 8 (1975) 125. [31] G.R. Aiken, P.A. Brown, T.I. Noyes, D.J. Pinckney, in: R.C. Averett, J.A. Leenheer, D.M. McKnight, K.A. Thorn (Eds.), Humic Substances in the Suwannee River Georgia: Interactions Properties and Proposed Structures (1994) 89, US Geological Survey Water-Supply Paper 2373. [32] Y.-P. Chin, W.J. Weber Jr., Environ. Sci. Technol. 23 (1989) 978. [33] P.G.-J. de Maagd, D.Th.E.M. ten Hulscher, H. van den Heuvel, A. Opperhuizen, D.T.H.M. Sijm, Environ. Toxicol. Chem. 17 (1998) 251. [34] M.A. Schlautman, J.J. Morgan, Environ. Sci. Technol. 27 (1993) 961. [35] A. de Visscher, P. van Eenoo, D. Drijvers, H. van Langenhove, J. Phys. Chem. 100 (1996) 11636.
Dissolved organic matter inhibition of sonochemical degradation of aqueous polycyclic aromatic hydrocarbons E. Taylor Jr., B.B. Cook, M.A. Tarr * Department of Chemistry, University of New Orleans, New Orleans, LA 70148-2820, USA Received 9 March 1999; received in revised form 20 March 1999
Abstract Sonochemical degradation of aqueous polycyclic aromatic hydrocarbons (PAHs) was found to be rapid in the absence of other dissolved compounds (k=0.006–0.015 s−1). In the presence of 20 mg C l−1 fulvic acid, first-order PAH degradation rate constants decreased from 2.3- to 3.7-fold. Similar results were obtained with added benzoic acid, a crude analog for fulvic acid. In natural waters, PAH degradation was almost completely inhibited. Analysis of the kinetic behavior and reaction products indicates that PAHs are most likely degraded through a radical cation mechanism. Hydroxyl radical appeared to play an insignificant role in the degradation. Inhibited degradation was probably the result of either altered cavitation processes or isolation of the PAH away from cavitation sites. © 1999 Elsevier Science B.V. All rights reserved. Keywords: Degradation; Dissolved organic matter; Fulvic acid; Polycyclic aromatic hydrocarbons; Sonochemistry
1. Introduction Sonochemistry has been demonstrated as a possible tool for degradation of aqueous pollutants. Cavitation during ultrasonic treatment of aqueous solutions results in the concentration of the sonic energy into a very small volume, thereby producing local areas of high energy. Within these localized high-energy regions, chemical transformations become quite probable. The temperature and pressure in the region of cavitation can reach 3000–5000 K and 500–10 000 atm. [1–4]. Three reactive zones exist in the region of cavitation: the gas phase, the gas–liquid interface, and the liquid directly surrounding the cavitation bubble [3]. Numerous reports indicate the formation of hydroxyl radical and hydrogen atoms in the vicinity of the cavitation site [1– 7]. Despite some controversy over the formation of hydrated electrons during sonication of neutral pH water, recent work indicates that hydrated electrons are not formed under these conditions [8]. Some sonochemical transformations that have been observed include degradation of nucleic acid bases [9] and degradation of aqueous carbon tetrachloride [10], 1,1,1-trichloroethane [11], methylene chloride [12], and * Corresponding author. Fax: +1-504-280-6860. E-mail address: [email protected] (M.A. Tarr)
trichloroethylene [12]. In addition, chloroaromatics [13– 15], phenols [14,16 ], substituted benzenes [17], and polycyclic aromatic hydrocarbons (PAHs) [18] have also been degraded by sonochemistry. These studies indicate the possibility of utilizing sonochemistry for degradation of pollutants, including chlorinated hydrocarbons, petroleum products and by-products, and light and dense non-aqueous phase liquids (LNAPLs and DNAPLs). Three mechanisms have been proposed for sonochemical degradation: (1) oxidation by hydroxyl radical, (2) pyrolytic decomposition, and (3) supercritical water oxidation [19]. Despite the extent of sonochemical degradation studies, little work has been conducted on the effects of matrix compounds present in the sample. The presence of matrix components, such as dissolved organic and inorganic matter, may dramatically alter the sonochemical processes. Since both industrial waste streams and contaminated waters (surface and groundwater) generally contain a wide array of both pollutant and nonpollutant chemicals, it is necessary to understand the effect of such matrix compounds before sonochemistry can be successfully utilized as a degradation tool. This study has assessed the effects of natural organic matter (NOM ) on the sonochemical degradation of hydrophobic pollutants. Three polycyclic aromatic hydrocarbons (see structures in Fig. 1) were degraded
1350-4177/99/$ – see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S1 3 5 0 -4 1 7 7 ( 9 9 ) 0 0 01 5 - 2
176
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
Fig. 1. Structures of anthracene (a), phenanthrene (b), pyrene (c), and a proposed average structure [20] of Suwannee River fulvic acid (d).
in pure water, in solutions of Suwannee River fulvic acid, in the presence of benzoic acid (a crude analog for fulvic acid), and in two natural surface waters. Fulvic acids are a class of naturally occurring polyphenolic compounds that are typically derived from plant lignins. Fulvic acids, as well as other classes of natural organic matter, are ubiquitous in surface water, ground water, soil, and sediment. An average structure for fulvic acid is given in Fig. 1.
2. Experimental Purified water (NP) was obtained by further purification of distilled water with a Barnstead (Dubuque, IA) NanopureUV water treatment system. Natural water samples were collected from sites in Southeast Louisiana at Crawford Landing on the West Pearl River (PR) and from a small water body connecting Lake Pontchartrain and Lake Maurepas (LM ). All natural water samples were filtered using pre-combusted 0.5-mm glass fiber filters (Rundfilter MN, Machery-Nagel, Alberta, Canada) and were stored in the dark at 4°C. Suwannee River fulvic acid standard material was obtained from the International Humic Substances Society (http://www.ihss.gatech.edu/index.html ). Pyrene (99%), anthracene (99+%), phenanthrene (99.5+%), benzoic acid (99.5%), and p-hydroxybenzoic acid (99+%) were purchased from Aldrich. All chemicals were used as received. Aqueous PAH solutions were prepared in the following manner. First, a more concen-
trated (mM ) stock solution was prepared in hexane. A small aliquot of this solution was transferred to a clean, dry volumetric flask. The solvent was evaporated under a stream of nitrogen. NP water was added to the flask, and the solute was dissolved by low-energy (60 W ) sonication in a bath sonicator for ~30 min. Final PAH concentrations of the resulting aqueous solutions were in the 0.1–0.5 mM range and were well below the solubility limit for the PAH. Fulvic acid solutions were prepared by mixing aqueous fulvic acid solutions with the aqueous PAH solution. These solutions were prepared so that the PAH concentration remained constant for all sonicated samples, but the fulvic acid concentration was varied. Natural water solutions were prepared by mixing the aqueous PAH solution with an equal volume of the natural water. To avoid possible alteration of the natural organic matter, natural waters and fulvic acid solutions were not placed into the bath sonicator. All solutions were stored in the dark, and natural waters and fulvic acid solutions were kept refrigerated (4°C ). All samples were sonicated using an Ace Glass ( Vineland, NJ ) 600 W sonochemical apparatus operating at 20 kHz. A 0.5-inch-diameter titanium probe was used with the power supply set to maximum amplitude. Sonication was carried out with a 10 or 20% duty cycle (sonication on for 1 s, off for 9 s or on for 1 s, off for 4 s). The low duty cycle was used to avoid excess heating. Samples were contained in an all-glass, waterjacketed reaction vessel with the cooling water maintained at 20°C. The vessel was sealed by O-rings and glass stoppers during sonication. Solutions were air-
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
equilibrated, and were sonicated under a headspace of air. To avoid any photochemical reactions, all samples were kept in the dark during and after sonication. Samples were typically 50–100 ml in volume, and 1– 5-ml aliquots were removed for analysis at regular time intervals. The tip of the titanium probe was polished on a regular basis to maintain a smooth surface. PAH fluorescence was monitored with a PTI (Monmouth Junction, NJ ) Quantamaster spectrofluorometer. Anthracene and phenanthrene were excited at 251 nm, and pyrene was excited at 319 nm. The fluorescence emission intensity was measured at 377 nm for anthracene, 345 nm for phenanthrene, and 370 nm for pyrene. Two- or 5-nm excitation and emission slits were used. For all samples, background fluorescence spectra were acquired from solutions without PAH and were subtracted from the spectra of the PAHs. Due to the relatively short measurement times, no significant photobleaching of the PAHs was observed during fluorescence measurements. Some of the sonicated samples were analyzed by HPLC with fluorescence detection. A Varian 5000 chromatograph was used with a 25 cm×4.6 mm, 5-mm Spherisorb ODS-2 column (Alltech Associates, Deerfield, IL). The mobile phase was 90% methanol and 10% water at a flow rate of 1 ml min−1. Analytes were detected using a Dionex (Sunnyvale, CA) FD300 fluorescence detector. Samples were injected using a 100-ml loop. The hydroxyl radical was quantitated using benzoic acid as a chemical probe [21]. The formation of phydroxybenzoic acid was monitored by HPLC with absorbance detection (254 nm) [21,22]. The observed first-order rate constant for p-hydroxybenzoic acid formation was divided by the second-order rate constant for the hydroxyl radical–benzoic acid reaction to calculate the hydroxyl radical concentration.
177
3. Results and discussion Observation of the PAH fluorescence spectrum as a function of sonication time indicated a decrease in fluorescence intensity but no change in the spectral shape. The consistency of the fluorescence spectral shape indicated that there were no fluorescent interferences with the excitation and emission wavelengths used. We therefore used the fluorescence intensity at the emission wavelength maximum as a measure of PAH concentration. Fluorescence data for degradation of phenanthrene in pure water are presented in Fig. 2. Anthracene and pyrene gave similar results. All three PAHs studied exhibited pseudo-first-order loss during sonolysis in pure water. Fig. 3 illustrates the first-order behavior observed for phenanthrene degradation in pure water. All the PAHs degraded fairly rapidly in pure water, with firstorder rate constants ranging from 0.006 to 0.015 s−1 (see Table 1). These rate constants are expressed in terms of actual sonication time and do not include the ‘off ’ period of the duty cycle. In order to prevent volatilization losses in our experiments, a sealed reaction vessel was used. Any volatilization of PAH would have resulted in deposition of the PAH back to the solution or onto the glassware. Postsonication extraction of the reaction vessel with acetone indicated no loss of PAH onto the reaction vessel. Therefore, it was assumed that volatilization losses were negligible, and all observed losses of PAH were due to chemical transformation. This result is reasonable, based on the low volatility of the PAHs used and the sealed design of the reaction vessel. For all three PAHs studied, the addition of fulvic acid resulted in a decrease in the PAH degradation rate constant. Fig. 4 illustrates the observed first-order rate constants as a function of fulvic acid concentration. Phenanthrene showed decreasing rate constants as the
Fig. 2. Fluorescence spectra of phenanthrene during sonication. From top to bottom, traces are after 0, 1, 2, 3, and 4 min of sonication.
178
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
Fig. 3. Observed first-order degradation of phenanthrene during sonication. A log plot with regression data is shown in the inset.
fulvic acid concentration increased. Pyrene and anthracene showed an initial drop in rate constant upon addition of a small amount of fulvic acid, but showed little additional change in rate constant with increasing fulvic acid concentration. The observations for pyrene and anthracene cannot be explained by a simple reactive transient scavenging model. During sonication, reactive transients, such as hydroxyl radicals, are formed. These transients may then act as oxidants for the degradation of pollutants present. However, scavenging of reactive transients would result in a continuously decreasing rate constant with added scavenger (fulvic acid in this case), with the rate constant eventually approaching zero. We have modeled a simple scavenging system with a single reactive transient using numerical methods. A fixed reactive transient production rate and a fixed background scavenging rate were used. The relative rate constants of the added scavenger and the PAH were varied. The model was then adjusted for each PAH to approximate the change in observed k at the lowest level of added FA. Although this model is only qualitative, it yields valuable information about the expected shape of the k vs. [FA] curves for simple scavenging behavior. The predicted behavior for reactive transient scavenging
is illustrated in Fig. 4 along with the experimental observations. None of the three compounds studied showed experimental curves that matched those predicted for scavenging behavior. Using this simple model, no conditions could be found that reproduced the experimental results. Therefore, the changes in k observed upon addition of FA cannot be simply due to scavenging of a reactive transient by the FA, and some other mechanism(s) must be responsible. One important reactive transient present in sonochemical systems is the hydroxyl radical. The concentration of hydroxyl radical in these systems was measured using benzoic acid as a radical trap [21,22]. Results indicated that the hydroxyl radical concentration present is not sufficient to explain the observed PAH degradation kinetics. For pyrene, after 3 min of sonication, ~60% of the pyrene had been degraded. Based on the measured hydroxyl radical concentration in this system (~4×10−16 M ) and a pyrene-hydroxyl radical rate constant of ~1×1010 [23], reaction with hydroxyl radical accounts for <1% loss of pyrene. Similar results were obtained for the other PAHs, clearly indicating that the hydroxyl radical is not playing a major role in the sonochemical degradation of these compounds.
Table 1 First-order rate constants (s−1) for PAH sonochemical degradation (all values ×10−3)
Anthracene Phenanthrene Pyrene
NP
PR
LM
FA (10 mg C l−1)
FA (20 mg C l−1)
15±6 (7)a 5.9±0.3 (3) 6±3 (7)
b b 2.3 (1)
b b 1 (1)
5.72±0.03 (2) 3.4±0.3 (2) NDc
4.1 (1) 1.8 (1) 2.58±0.06 (2)
a The error is ±1SD. Parenthetical values indicate the number of measurements. b Little or no degradation observed; did not follow first order kinetics. c ND, not determined.
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
179
(a)
(b)
(c)
Fig. 4. Observed first-order rate constants for anthracene (a), phenanthrene (b), and pyrene (c) sonochemical degradation as a function of added Suwannee river fulvic acid concentration. The dashed lines indicate the predicted behavior for a single reactive transient which is scavenged by the added fulvic acid. Error bars indicate one standard deviation. Error was not determined for the points at 21 mg C l−1. For anthracene (a), the error bars for the point at 10 mg C l−1 are smaller than the marker.
180
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
Previously reported studies on halocarbons have also revealed that hydroxyl radical concentrations were not sufficient to explain degradation rates and therefore concluded that pyrolysis was the likely mechanism [24]. Using HPLC with fluorescence detection, a single fluorescent product was observed for the sonochemical degradation of pyrene in pure water. This product was also degraded under continued sonication to yield no additional fluorescent products. No fluorescent products were observed for anthracene or phenanthrene. However, it is likely that all three PAHs have oxygenated products that likely further degrade under continued sonication. These intermediate degradation products are probably very important since oxygenated PAHs are often more toxic and carcinogenic than the parent PAH [18,25,26 ]. The pyrene fluorescent product has not yet been identified; however, it does have the same retention time as the photochemical degradation product of aqueous pyrene. Given the selectivity of fluorescence detection and the coincident retention times, there is little doubt that the sonochemical and photochemical degradation products are the same compound. The shorter retention time of this product indicates that it is likely more polar than pyrene. Based on these observations, it is postulated that the product is a dihydroxypyrene, a dihydrodihydroxypyrene, a pyrenedione, or another similar oxygenated pyrene. A phenanthrene-diol has been previously observed upon sonication of phenanthrene [18], and similar oxygenated compounds have been previously reported from photochemical degradation of PAHs [27– 29]. Furthermore, since both sonochemistry and photochemistry result in the same pyrene degradation product, it is likely that the product is formed by a similar mechanism. A plausible mechanism involves the formation of the pyrene radical cation, followed by reaction with water or molecular oxygen (R. Dabestani, personal communication). Degradation of pyrene by hydroxyl radicals (generated by Fenton’s reagent [22,30]) did not produce any observable fluorescent products. The above evidence indicates that the hydroxyl radical is not an important species and suggests a sonochemical mechanism involving formation of PAH radical cation species via high-temperature pyrolysis. Such reactions would require the PAH to be in very close proximity to the cavitation site or even within the cavitation bubble. The presence of FA results in a significant inhibition of the degradation of PAHs in aqueous solution. In the presence of ~20 mg C l−1 fulvic acid (~50 mM FA based on the average MW for Suwannee River fulvic acid of ~800 g mol−1 [31]), the first-order rate constant for anthracene dropped by a factor of 3.7, the rate constant for phenanthrene decreased by a factor of 3.3, and the rate constant for pyrene decreased by a factor of 2.3. Significant decreases in PAH rate constants were observed at fulvic acid concentrations as low as ~1 mM.
This added concentration of FA is comparable to the pyrene and phenanthrene concentrations used (0.25 mM ). Even trace amounts (mM ) of matrix species are sufficient to dramatically reduce the degradation efficiency of trace pollutants. The observed changes in rate constant upon addition of fulvic acid could be due to (1) changes in the cavitation process (caused for example by altered surface tension) or (2) sequestering of the PAH away from cavitation sites. Addition of salt has reportedly enhanced the sonochemical degradation of aromatic compounds by increasing their residence in or at the surface of cavitation bubbles [14]. In the current study, the matrix components may have removed the PAHs from the site of cavitation. Binding of the PAHs to FA may have this effect. However, even at the highest concentration of fulvic acid used, the PAHs were predicted to be less than 30% bound to FA based on reported and calculated K values [32–34]. This degree of binding to FA is not oc sufficient to explain the roughly threefold decrease in rate constants observed. Furthermore, the shape of the k vs. [FA] curves (Fig. 4) does not follow the shape expected if binding to FA were the primary mechanism for decreased reactivity. Partitioning of the fulvic acid across the air–water interface of the cavitation bubble could produce a barrier for entry of the PAH into the interior of the bubble, thereby minimizing the PAH reactivity. Alternatively, the presence of non-pollutant organic compounds in the gas phase within cavitation bubbles may lower the temperature upon bubble collapse [35], resulting in decreased PAH oxidation. Benzoic acid was also added to solutions of anthracene prior to sonication. Benzoic acid was used because it is a known hydroxyl radical scavenger and since it is an aromatic carboxylic acid, somewhat representative of the groups present in fulvic acid. The rate constant for anthracene degradation as a function of benzoic acid concentration is depicted in Fig. 5. This plot has the same shape as the analogous plots for fulvic acid. As stated above, the shape of these curves cannot be explained by a scavenging model. Addition of a free radical scavenger would result in decreasing rate constants that would approach zero as the scavenger concentration increased. With the addition of either benzoic acid or fulvic acid, both anthracene and pyrene showed initial decreases in rate constant followed by fairly constant non-zero rate constants with further increases in benzoic acid or fulvic acid concentration. Furthermore, the molarity of BA was significantly higher than FA because BA has a much lower molecular weight (at 21 mg C l−1, [BA]=172 mM and [FA]#26 mM ). Despite this large difference in molarity, the addition of BA or FA gave similar results. These observations indicate that scavenging processes are not the primary cause of decreased rate constants. The most likely expla-
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
181
Fig. 5. Observed first-order rate constants for anthracene as a function of added benzoic acid concentration. Error was determined only for the 0 mg C l−1 point and is reported as one standard deviation.
nations for the observed behavior are either (1) a change in cavitation processes due to altered surface tension or other solvent properties or (2) sequestering of the PAH away from the cavitation site. This latter explanation could involve partitioning of the benzoic acid or fulvic acid across the air–water interface of the initially formed cavitation bubble, thereby excluding other molecules from access to high-temperature regions occurring upon collapse of the cavitation bubble. A previous study has indicated competitive effects when an aqueous mixture of chlorobenzene and 4-chlorophenol was sonicated [13]. This report indicated that chlorophenol degradation was inhibited until chlorobenzene was degraded to a low concentration. Chlorobenzene was believed to be degraded by pyrolysis within the cavitation bubble, whereas chlorophenol was believed to be degraded by hydroxyl radicals outside the bubble. Although the effects observed in the previous report are likely different than those observed here, the previous work does illustrate that the presence of matrix compounds can dramatically affect pollutant degradation rates. Although the above data clearly demonstrate that natural organic matter significantly reduces the sonochemical degradation efficiency of PAHs, two natural water samples were also studied in order to extend these results to real-world samples. In addition to NOM content, these natural waters contained dissolved inor-
ganic matter that may also interfere with degradation by altering the cavitation process or scavenging reactive transients. Table 2 presents some compositional data for the natural waters used. In contrast to the rapid loss of PAH in pure water sonications, degradation rates were significantly retarded in the natural water samples. Fig. 6 illustrates the PAH concentrations as a function of sonication time in pure water and the natural waters. Due to background fluorescence from the natural waters, increased noise in the PAH concentration measurements was observed despite background correction. This result was caused by slight changes in the NOM fluorescence upon sonication. In pure water, phenanthrene was almost completely degraded after 5 min of sonication, whereas in the natural waters, little or no loss of phenanthrene was observed. Similar results were observed for anthracene and pyrene, as depicted in Fig. 6. First-order rate constants for each PAH in pure and natural water are presented in Table 1. The natural water matrices resulted in significant to nearly complete inhibition of the PAH degradation. The natural waters used showed much greater inhibition of PAH degradation than was observed for the fulvic acid studies. It is clearly illustrated that the ability to degrade PAHs by sonochemistry in natural waters is severely limited by matrix compounds.
Table 2 DOC and Na content of natural waters used
4. Conclusions
Pearl River Lake Maurepas
DOC (mg C l−1)
Na (ppm)
15 61
6.5 293
The results of this study indicate that sonochemical degradation of hydrophobic pollutants can be significantly inhibited by other dissolved species. Three possible explanations have been hypothesized: (1) matrix
182
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
(a)
(b)
(c) Fig. 6. PAH concentration as a function of sonication time for NP water (&), PR water ($), and LM water (+).
E. Taylor Jr. et al. / Ultrasonics Sonochemistry 6 (1999) 175–183
compounds scavenge reactive transients preventing PAH degradation, (2) the presence of additional dissolved species alters the cavitation process, and (3) dissolved matrix species may inhibit PAH access to cavitation sites. The third explanation is currently believed to be the most likely. Hydroxyl radicals, commonly present during sonication of aqueous solutions, were not present in sufficient quantities to account for the degree of PAH degradation. This result, along with the observation of a polar product that appears to be the same as the photochemical product, suggest that pyrolysis and/or combustiontype reactions proceeding via a PAH radical cation may be predominant in the sonolysis of PAHs in aqueous solution. The utility of sonochemical treatment for the degradation of aqueous hydrophobic pollutants is likely severely limited by the presence of dissolved matrix compounds. Although the current study focused on natural organic matter, similar results are expected for matrix compounds in industrial waste streams.
Acknowledgements This work was supported by the Louisiana Board of Regents under grant LEQSF(1998-01)-RD-A-34 and by the Academy of Applied Sciences Research and Engineering Apprenticeship Program (subgrant 705).
References [1] K.S. Suslick, D.A. Hammerton, R.E. Cline Jr., J. Am. Chem. Soc. 108 (1986) 5641. [2] K.L. Suslick, Science 247 (1990) 1439. [3] P. Riesz, D. Berdahl, C.L. Christman, Environ. Hlth Persp. 64 (1985) 233. [4] V. Misik, N. Miyoshi, P. Riesz, J. Phys. Chem. 99 (1995) 3605. [5] K. Makino, M.M. Mossoba, P. Riesz, J. Phys. Chem. 87 (1983) 1369. [6 ] K. Makino, M.M. Mossoba, P. Riesz, J. Am. Chem. Soc. 104 (1982) 3537. [7] P. Riesz, T. Kondo, Free Rad. Biol. Med. 13 (1992) 247. [8] V. Misik, P. Riesz, J. Phys. Chem. A 101 (1997) 1441.
183
[9] J.R. McKee, C.L. Christman, W.D. O’Brien, S.Y. Wang, Biochemistry 16 (1977) 4651. [10] P.K. Chendke, H.S. Fogler, J. Phys. Chem. 87 (1983) 1362. [11] M.S. Toy, R.S. Stringham, T.O. Passel, Pollution Prevention in Industrial Processes, J.J. Breen, M.J. Dellarco (Eds.), ACS Symposium Series Vol. 508, American Chemical Society, Washington, DC, 1992, p. 284. [12] H.M. Cheung, A. Bhatnangar, G. Jansen, Environ. Sci. Technol. 25 (1991) 1510. [13] C. Petrier, Y. Jiang, M.F. Lamy, Environ. Sci. Technol. 32 (1998) 1316. [14] J.D. Seymour, R.B. Gupta, Ind. Eng. Chem. Res. 36 (1997) 3453. [15] P. Kruus, R.C. Burk, M.H. Entezari, R. Otson, Ultrason. Sonochem. 4 (1997) 229. [16 ] C. Petrier, A. Francony, Ultrason. Sonochem. 4 (1997) 295. [17] A. de Visscher, H. van Langenhove, P. van Eenoo, Ultrason. Sonochem. 4 (1997) 145. [18] P.E. Wheat, M.A. Tumeo, Ultrason. Sonochem. 4 (1997) 55. [19] M.R. Hoffmann, I. Hua, R. Hochemer, Ultrason. Sonochem. 3 (1996) S163. [20] J.A. Leenheer, D.M. McKnight, E.M. Thurman, P. MacCarthy, in: R.C. Averett, J.A. Leenheer, D.M. McKnight, K.A. Thorn (Eds.), Humic Substances in the Suwannee River Georgia: Interactions Properties and Proposed Structures (1994) 195, US Geological Survey Water-supply Paper 2373. [21] X. Zhou, K. Mopper, Mar. Chem. 30 (1990) 71. [22] M.E. Lindsey, M.A. Tarr, Chemosphere (1999) submitted. [23] M.E. Lindsey, M.A. Tarr, Environ. Sci. Technol. (1999) submitted. [24] K. Hirai, Y. Nagata, Y. Maeda, Ultrason. Sonochem. 3 (1996) S205. [25] L. Ren, X.-D. Huang, B.J. McConkey, D.G. Dixon, B.M. Greenberg, Ecotoxicol. Environ. Saf. 28 (1994) 160. [26 ] J.E. Weinstein, J.T. Oris, J.T. Taylor, Aquatic Toxicol. 39 (1997) 1. [27] S.P. Zingg, M.E. Sigman, Photochem. Photobiol. 57 (1993) 453. [28] R. Dabestani, M. Melson, M.E. Sigman, Photochem. Photobiol. 64 (1996) 80. [29] R. Dabestani, K.J. Ellis, M.E. Sigman, J. Photochem. Photobiol. A 86 (1995) 231. [30] C. Walling, Acc. Chem. Res. 8 (1975) 125. [31] G.R. Aiken, P.A. Brown, T.I. Noyes, D.J. Pinckney, in: R.C. Averett, J.A. Leenheer, D.M. McKnight, K.A. Thorn (Eds.), Humic Substances in the Suwannee River Georgia: Interactions Properties and Proposed Structures (1994) 89, US Geological Survey Water-Supply Paper 2373. [32] Y.-P. Chin, W.J. Weber Jr., Environ. Sci. Technol. 23 (1989) 978. [33] P.G.-J. de Maagd, D.Th.E.M. ten Hulscher, H. van den Heuvel, A. Opperhuizen, D.T.H.M. Sijm, Environ. Toxicol. Chem. 17 (1998) 251. [34] M.A. Schlautman, J.J. Morgan, Environ. Sci. Technol. 27 (1993) 961. [35] A. de Visscher, P. van Eenoo, D. Drijvers, H. van Langenhove, J. Phys. Chem. 100 (1996) 11636.