Environ. Sci. Technol. 2000, 34, 1280-1285
Kinetics and Mechanism of Pentachlorophenol Degradation by Sonication, Ozonation, and Sonolytic Ozonation LINDA K. WEAVERS,† NOAH MALMSTADT, AND MICHAEL R. HOFFMANN* W. M. Keck Laboratories, California Institute of Technology, Pasadena, California 91125
The decomposition of pentachlorophenol (PCP) by sonication was investigated at two frequencies (20 and 500 kHz) and two concentrations (20 and 60 µM) to gain insight into the kinetics and mechanisms occurring at different frequencies. Results suggest the presence of parallel pyrolysis and OH• reaction pathways that are altered both as a function of frequency and PCP concentration. In addition, ozone was added to the system and compared to experiments of sonication and ozonation separately to explore the mechanism of sonolytic ozonation. The addition of ozone during sonication did not affect the first-order degradation constant for PCP compared to the linear combination of separate sonication and ozonation experiments; hence, the residual kinetic effect of the combined system was zero (kUS/O3 ) 0). Observed byproducts of sonication include the following: tetrachloro-obenzoquinone (o-chloranil), tetrachlorocatechol (TCC), oxalate, and chloride.
Introduction Pentachlorophenol (PCP) has been used extensively throughout the world as a wood preservative and general biocide (1). PCP residues are widespread in the environment. For example, PCP was detected in 80% of human urine specimens in an EPA study (1). Even though PCP is an EPA priority pollutant and a suspected carcinogen, its pyrolysis and combustion reaction products, polychlorodibenzodioxins (PCDDs) and polychlorodibenzofurans (PCDFs), are considerably more toxic than PCP (2). Ultrasonic irradiation of aqueous solutions has been shown to be effective for the in situ destruction of a variety of organic and inorganic contaminants (3-11). Sonolysis of an aqueous solution results in the formation and adiabatic collapse of bubbles generating local high temperatures and pressures and reactive free radicals in the bubble (12). Destruction of organic compounds occurs in the cavitation bubble itself or its interfacial sheath due to direct pyrolysis, hydroxylation, or radical reactions that results from the gasphase pyrolysis of H2O. Radicals escaping the cavitation bubble diffuse into solution and react near the cavity boundary. Secondary reactions also occur in the bulk aqueous * Corresponding author phone: (626)395-4391; fax: (626)395-3170; e-mail:
[email protected]. † Present address: Department of Civil and Environmental Engineering and Geodetic Science, The Ohio State University, Columbus, OH 43210; e-mail:
[email protected]. 1280
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 7, 2000
phase. For example, hydrogen peroxide has been observed to form in sonolytic systems. Sonication has been shown to be particularly effective with volatile and hydrophobic compounds, since they partition to the bubble or interfacial sheath rapidly (6). In a collapsing cavitation bubble, the thermolytic decomposition of ozone and subsequent OH• formation occurs as follows (13, 14):
O3 f O2 + O(3P) 3
O( P) + H2O f 2OH
(1) •
(2)
These decomposition reactions occur in the gas phase. The reaction products migrate to the interfacial sheath of the bubble where they subsequently react in the aqueous phase. Even for compounds that react quickly with O3 such as PCP (kO3 > 3 × 105 M-1 s-1) (15), the combination of O3 and ultrasound may be more effective since two OH• molecules are formed per O3 molecule consumed. Hydrogen peroxide production, which is enhanced during aqueous-phase sonolysis of ozone, is the result of increased OH• concentrations in the gas phase and in the interfacial region (13, 16). Furthermore, ultrasonic irradiation has been demonstrated to increase the mass transfer of ozone to solution, allowing more ozone to enter solution than in a nonirradiated system (17-19). The sonolytic degradation of PCP has been performed previously at both 20 and 500 kHz (20-22). Investigators have observed decomposition of PCP and subsequent formation of Cl-, CO, and CO2 (20). Pe´trier et al. found more rapid destruction at 500 kHz than 20 kHz and hypothesized a OH• reaction pathway (21). Gonze et al. used PCP to probe key geometric parameters of reactor design such as volumetric power input (22). However, byproduct formation and the mechanism of degradation have not been explored. The primary objective of this study was to investigate the degradation of PCP in several different reactor systems in order to explore the effect of power intensity and frequency and to determine the kinetics and the mechanism of the sonolytic degradation.
Experimental Methods Materials and Reagents. All chemicals were of highest purity and used as received. Solutions were prepared with water purified by a Millipore Milli-Q UV Plus system (R ) 18.2 MΩ-cm). Sonochemical Reactors. The experimental setup consisted of a reactor surrounded by a self-contained water jacket, a constant gas flow, and a source of ultrasound. The temperature of the solution was maintained constant at approximately 30 °C with a recirculating water bath (Haake Co., model A81) flowing through the surrounding water jacket. Three different reaction systems were used for the destruction of PCP. Sonication experiments at 500 kHz were performed with an orthoreactor ultrasonic transducer (Undatim Ultrasonics) operating at 515 kHz. Two 20 kHz reactors were used: a direct immersion probe system (Sonifier cell disruptor 200, Branson) and a tube resonator (Telesonics, model SG-22-2000S) emitting ultrasound radially over its shaft. Characteristics of these different systems are tabulated in Table 1. The ultrasonic power inputs into the aqueous solution were determined by calorimetry which assumes that all power entering the sonicating solution is dissipated as heat (23). Note that heat transfer from the transducers to the sonicating solution is not taken into account, resulting in a 10.1021/es980795y CCC: $19.00
2000 American Chemical Society Published on Web 02/16/2000
TABLE 1. Reactor Characteristics reactor
freq (kHz)
reaction vol. (mL)
emitting area (cm2)
calorimetric power (W)
power intensity (W cm-2)
power density (W L-1)
gas flowrate (mL min-1)
orthoreactor tube res. probe
515 20 20
640 1750 50
25.5 376 1.20
48.3 466 66.5
1.89 1.24 55.8
75.5 266 1340
30 82 50
slight overestimation of the total amount of ultrasonic energy absorbed by solution (24). However, calorimetry has been shown to correlate with chemical reaction (25). Power density is calorimetric power per solution volume, whereas power intensity is the calorimetric power per energy input area. Turbulence from gas sparging was assumed to be sufficient for complete mixing in the 20 kHz reactors, while a magnetic stirrer was used in the 500 kHz reactor to ensure macroscale mixing. Oxygen gas was filtered through a Drierite and molecular sieve economy purifier (Alltech) and an activated charcoal hydrocarbon trap (Alltech) before flowing through an OREC Ozonator (Model O3V10-O) at 0.3 atm and 3 L min-1. The gas flowed into the reactors at the flowrate specified in Table 1 through a coarse fritted-glass diffuser at ambient pressure resulting in a saturated aqueous oxygen solution. In all experiments, PCP solutions were exposed to O2 and assumed to reach equilibrium with the O2 gas prior to the start of a kinetic experiment. In select experiments when O3 was used, the O3 gas phase concentration was determined to be approximately 0.7 wt%/wt by oxidation of indigo trisulfonic acid (26). Initial solutions containing either 20 or 60 µM of PCP were adjusted to pH 7.3 using 5 mM phosphate buffer, thus, all experiments were run with PCP in the deprotonated form (pKa ) 4.71) (15). Phosphate was chosen as a buffer due to its slow rate of reaction with •OH (kH2PO4- ) 2 × 104 M-1 s-1). To initiate a kinetic run, ultrasound and/or voltage from the ozonator was applied exposing the PCP solution to ultrasound and/or ozone gas. PCP losses by volatilization were assumed to be low due to the small Henry’s constant (0.079 Pa‚m3 mol-1) (15) and the fact that PCP is predominantly in the phenolate form at pH 7.3. Aliquots were collected at designated times and quenched with 100 mM thiosulfate to eliminate any residual ozone in solution. All samples were filtered before analysis with 0.2 µm Teflon syringe filters (Gelman). No loss of PCP to filters was observed in control experiments. Analysis. A Hewlett-Packard Series II 1090 HPLC was used to quantify PCP and aromatic intermediates. A 5 mm, 100 × 2.1 mm ODS Hypersil column (Hewlett-Packard) with an eluent gradient of aqueous phosphate buffer solution (pH ) 3) and methanol was used for analysis. Chloride and organic ions were detected by ion chromatography with electrical conductivity detection (Bio-LC, Dionex). An AS-11 column (Dionex) with NaOH and methanol as the mobile phase was used. The possible formation of dioxins was observed with a Hewlett-Packard gas chromatograph (HP 5890 series II, GC) with a mass-selective detector (HP 5989AMSD). A 0.25 mm × 30 m HP-1 chromatographic column (Hewlett-Packard) was used for GC separation. Samples of 4 mL were extracted into 1 mL of toluene to enhance detection.
Results As shown in Figures 1 and 2, sonolysis of PCP under O2 initially followed first-order kinetics according to the following rate equation:
-
d[PCP] ) kobs[PCP] dt
(3)
FIGURE 1. First-order plot of [PCP] degradation by sonication at 20 kHz.
FIGURE 2. First-order plot of [PCP] degradation by sonication at 500 kHz.
TABLE 2. Pseudo-First-Order Rate Constants of 60 µM PCP Degradation Normalized to the 20 kHz Probe for Sonication in an Oxygen Saturated Solution experiment
[PCP]0 (µM)
k (min-1) × 10-2
knorm (min-1) × 10-2
20 kHz probe 20 kHz tube res. 20 kHz tube res. 500 kHz orthoreactor 500 kHz orthoreactor
60 60 20 60 20
2.1 ( 0.044 1.3 ( 0.042 3.2 ( 0.15 2.0 ( 0.044 2.6 ( 0.059
2.1 ( 0.044 6.5 ( 0.21 16 ( 0.78 35 ( 0.78 46 ( 1.1
However, the pseudo-first-order rate constants summarized in Table 2 indicate that at both 20 and 500 kHz a lower initial concentration of PCP resulted in more rapid rates. Pseudo-first-order rate constants for the degradation of 60 µM PCP by sonication, ozonation, and sonolytic ozonation are summarized in Table 3. Degradation by ozonolysis was rapid due to its fast second-order reaction rate with O3. Since ozone was continually bubbled into the reactor, PCP loss followed pseudo-first-order kinetics. Although the gas flow rate was carefully controlled and proportional to the volume of solution in both reactors, the geometries in each reactor VOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1281
TABLE 3. Pseudo First-Order Rate Constants for Degradation of 60 µM PCP under Various Conditions experiment
k20 kHz (min-1) × 10-2
k500 kHz (min-1) × 10-2
sonication w/O2: kUS ozonation: kO3 sonication w/O3: ktotal residual effect: kUS/O3
1.3 ( 0.042 42 ( 6.2 52 ( 11 8.4 ( 12
2.0 ( 0.04 31 ( 5.2 26 ( 3.4 -6.2 ( 6.2
ments. Ozonation and sonolytic ozonation of PCP resulted in the complete recovery of Cl- in 30 min with sonolytic ozonation experiments releasing Cl- more rapidly than ozonation experiments.
Discussion Kinetics of Sonolytic Decomposition. To compare the degradation rate constants of the different reactor systems, the pseudo-first-order rate constants were normalized to the degradation rate constant of the 20 kHz probe system by
knorm ) kobs
FIGURE 3. Degradation of 60 µM PCP and intermediates formed by sonication with O2 in the 20 kHz tube resonator.
FIGURE 4. Degradation of 60 µM PCP and intermediates formed by sonication with O2 at 500 kHz. were different. However, this did not significantly affect the pseudo-first-order reaction rate of O3 with PCP as shown in Table 3. Sonolytic ozonation of PCP also followed a pseudofirst-order rate law. PCP destruction and byproduct formation by sonication in the 20 kHz tube resonator and 500 kHz orthoreactor is shown in Figures 3 and 4. Aromatic intermediates, tetrachloro-o-benzoquinone (o-chloranil), and tetrachlorocatechol (TCC) were formed and destroyed. In addition, chloride was recovered after 3 h as represented by the chloride concentration divided by the number of chlorine atoms per PCP molecule. Organic acids such as succinic, maleic, formic, acetic, glycolic, and fumaric acid were not detected. However, oxalic acid was formed. Other unidentified intermediates were detected by ion chromatography; however, they were formed and destroyed over the 4-h sonication in both reactors. The formation of octachlorodibenzodioxin as an intermediate, as followed with GC-MS, did not occur. In addition, intermediates were followed for ozonation and sonication combined with ozonation. Aromatic intermediates, which were formed during sonication alone, were not detected in ozonation and sonolytic ozonation experi1282
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 7, 2000
1 PD20kHz probe PDreactor
(4)
where knorm is the rate constant normalized for ultrasonic power density under specified conditions, kobs is the observed rate constant in the reactor under specified conditions, and PD is the power input per volume of solution sonicated (i.e., power density) as measured by calorimetry. Normalized rate constants are the degradation rate constant normalized to the same energy input per volume of solution degraded in each reactor. Although other factors such as power intensity (see below) and reactor geometry and composition (due to ultrasonic wave patterns in solution) play a role in degradation efficiencies, normalization to energy input per volume provides a rough basis for comparison. Values in Table 2 demonstrate that the 20 kHz probe reactor has the slowest normalized degradation rate constant at [PCP]0 ) 60 µM. knorm is a factor of 3.1 greater in the 20 kHz tube resonator than in the 20 kHz probe. Similarly, knorm in the 500 kHz reactor is a factor of 5.4 greater than in the 20 kHz tube resonator at 60 µM PCP. The tube resonator has similar power intensity and power density compared to the 500 kHz system, whereas the 20 kHz probe utilizes considerably more energy per volume and the area of energy input is much smaller. For PCP destruction, Pe´trier and co-workers observed an enhancement factor of 4.8 at 500 kHz over 20 kHz (21). This increase is similar to the relative enhancement observed between the tube resonator and the 500 kHz orthoreactor. The reactors used in experiments by Pe´trier et al. (11) have the same area for emitting ultrasound as well as the same power density. In addition, other studies of the frequency effect with organic compounds have observed enhancement at 500 kHz over 20 kHz (4, 11, 27, 28). The significant difference between the probe and tube resonator demonstrates the large effect of power intensity. Hua et al. have also demonstrated the effect of intensity with a nearfield acoustical processor (29). They observed a maximum rate constant (k) at an intensity of 1.2 W cm-2. Beyond this power intensity, k decreased. Our present study confirms their observations. For the same frequency, significantly lower degradation rates were observed at an intensity of 55.8 W cm-2 compared to 1.24 W cm-2. The effect of concentration on the sonochemical degradation rate of both polar and apolar substrates has been observed previously (7, 8, 11, 19, 30, 31). For hydrophobic volatile compounds an elaborate model was developed which took into account the higher specific heat of organic compounds (typically 120 J mol-1 K-1) compared to water vapor (36 J mol-1 K-1) and air (30 J mol-1 K-1). As the specific heat is increased in the vapor, the temperature rise upon collapse is decreased, thus smaller rate constants are observed at higher concentrations due to lower gas-phase temperatures reached for pyrolysis reactions (30). In an investigation of p-nitrophenol degradation, decrease in the first-order rate constant with concentration was observed (7). Loss of p-nitrophenol was assumed to decay by parallel zero- and first-order rates corresponding to OH• attack and liquid-phase pyrolysis. Serpone and co-workers (31) also observed a decrease in the first-order degradation
FIGURE 5. Formation of oxalate from 60 µM PCP destruction by sonication with O2, ozonation, and sonolytic ozonation in the 20 kHz tube resonator (a) and 500 kHz orthoreactor (b). rate with increasing chlorophenol concentration and compared the phenomena to Langmuirian saturation type kinetics observed in solid-gas systems. In addition, they proposed an OH radical pathway of reaction at the bubble interface. Finally, in the degradation of ionic species such as phenolate and HS-, zero-order degradation rates are observed to increase with increasing initial substrate concentration (8, 11). Although PCP is in the deprotonated form (pKa ) 4.7) (15) at the pH of these experiments, it is hydrophobic in the deprotonated form with an ion-pair octanol-water partition coefficient, log Kip, of 2.68 (32). Therefore, it is expected to accumulate at the gas-liquid interface although it will not volatilize and enter the cavitation bubble. Lower frequency cavitation is expected to produce fewer but more violent bubble collapses than higher frequency cavitation (33). However, greater H2O2 generation at high frequency is attributed to more acoustic cycles per time. Therefore, it is assumed that lower frequency may be favorable if higher temperatures are preferred over the generation of free radicals (33) although investigations of the effect of frequency have consistently demonstrated more rapid degradation rates at 500 over 20 kHz. In this study, we compared the effect of concentration at 500 kHz and observed an increase of 30 ( 4% in the rate constant with a decrease in the concentration from 60 to 20 µM. However, at 20 kHz the increase in k was substantially higher (140 ( 14%) with a decrease in concentration from 60 to 20 µM. To understand this phenomena fully a larger study should be undertaken; however, a possible explanation follows. In comparing the diffusivity of a characteristic aromatic compound in water (DL ) 1.12 × 10-5 cm2 s-1) (34) and the corresponding mean weighted travel distance (35) during one growth and collapse cycle of a cavitation bubble
〈x〉 ) 2
( ) D Lt π
0.5
(5)
where 〈x〉 is the mean weighted travel distance from its origin and t is the time traveled; a 5-fold higher travel distance occurs at 20 kHz over 500 kHz (267 nm vs 53 nm) if convection is neglected. As a rough estimate, assuming all of the PCP in the shell surrounding the bubble with thickness 〈x〉 reaches the bubble interface with a thickness of 10 nm, 160, and 32 µM, excess PCP accumulates in the 20 and 500 kHz bubbles, respectively. Again a 5-fold higher excess accumulation of PCP occurs with a 20 kHz bubble over a 500 kHz bubble. The shorter growth and collapse cycle of a 500 kHz bubble results in a smaller flux of PCP to the interface; hence, less PCP is available to react per bubble cycle. Note that differences in bubble density may occur as a function of frequency, which will affect the rates of degradation but not relative changes with concentration. As shown by De Visscher et al. (30) the higher Cp of organic compounds reduces the internal temperature of the collapsing bubble. In the case of PCP, Cp ) 201 J mol-1 K-1 (36). Although it will not volatilize into the bubble, PCP will lower the temperature of the surrounding interfacial sheath resulting in less liquid phase thermolysis reactions with higher concentration and less penetration of heat from the bubble interface into the bulk solution. In addition, as the bubble collapses asymmetrically the surrounding liquid that becomes entrained in the bubble vaporizes and again lower temperatures are predicted at higher PCP concentrations due to high Cp values. The resulting lower collapse temperatures result in lower OH• production rates (37). Finally, surface tension is reduced as the surface excess of PCP at the bubble surface increases. In contrasting the two frequencies, for the same bulk concentration, there is less accumulation of PCP at the bubble interface at 500 kHz. This results in a higher fraction of reactions occurring via the OH radical pathway and in a lesser influence of concentration on the overall rate of degradation. Moreover, less surface excess of PCP at the interface results in a smaller effect on surface tension and hence a smaller subsequent change in the cavitation dynamics. The parallel pathways involving liquid-phase pyrolysis and OH radical attack of PCP at the interface can be reflected in the following overall rate expression:
-
d[PCP] ) kpyr[PCP] + kOH[PCP][OH•] dt
(6)
At the interface, [OH•] is constant and has been estimated to be 4 × 10-3 M in the interfacial region of a cavitation bubble based on OH• reaction with iodide (38). Based on above estimates of [PCP] at the interface [OH•] is assumed to be larger than [PCP] at the concentration range studied. PCP in the interface is expected to be insufficient to scavenge most of the OH•; therefore, the overall rate is
-
d[PCP] ) kpyr[PCP] + k′OH[PCP] dt
(7)
where k′OH ) kOH[OH•]. Depending on the steady-state [OH•], the pyrolysis pathway may be more effective at low concentrations due to the higher temperatures in the bubble. At higher concentrations of [PCP], as shown for gas-phase pyrolysis reactions (30), the pyrolysis rate is the decreased. We expect this to occur with PCP in the interfacial region as well, due to decreased thermolysis reactions in the bubble interface, injection of liquid droplets into the gas phase during the collapse of a bubble cloud, and reduced surface tension of a bubble in the presence of PCP. A larger reduction in kobs at 20 kHz compared to 500 kHz indicates that k′OH is altered less by [PCP] than kpyr and that k′OH is dominant over kpyr at 500 kHz. This apparent change in the pathway with concentration has been observed with p-nitrophenol (7). AlVOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1283
though we observed first-order kinetics at both concentrations investigated, our results suggest the existence of parallel pathways corresponding to OH• attack and liquid-phase pyrolysis. Kinetics of Decomposition by Sonolytic Ozonation. PCP reacts very rapidly with both O3 and OH• (kO3 > 3 × 105 M-1 s-1 at pH 2, kOH ) 3.7 × 109 M-1 s-1) (39, 40). A Hatta number is useful in determining situations when mass transfer limitations occur as well as understanding the zone where reactions take place in a gas-liquid system. It is particularly important in the kinetics of ozonation because it may be used to determine the relative effectiveness of both direct and decomposition reactions of ozone with a solute. For a second-order irreversible reaction of ozone with PCP it is defined as
Ha )
(kPCP-O3‚DO3‚C0PCP)0.5 kL
(8)
where kPCP-O3, kL, DO3, and C0PCP are the second-order rate constants of PCP + O3, liquid-phase mass transfer coefficient of O3, O3 diffusivity in water, and the initial PCP concentration, respectively (41). A value of Ha < 0.3 is indicative of ozone accumulating in the bulk water, whereas a value of Ha > 0.3 is indicative of ozone disappearing in the gas-liquid film layer (42). Using a conservative value of kL and the reported reaction rate at pH 2, we calculate Ha ) 0.76. Therefore, because of its rapid reaction rate with O3, PCP will react at the gas-liquid interface, and negligible amounts of O3 will be present in solution. Adding the independently obtained rate constants for sonication (kUS) and ozonation (kO3) and comparing it to the combined process of sonolytic ozonation (ktotal), the residual kinetic effect of the combined process (kUS/O3) is within experimental error, at both frequencies. This assumes that the observed first-order degradation rate constant for sonolytic ozonation is a combination of separate processes for sonication, ozonation, and the residual kinetic effect of the combined system (kobs ) ktotal ) kUS + kO3 + kUS/O3). Similar results were obtained using a PCP concentration of 20 µM, although, in this case, errors were larger due to the detection limit of the analytical equipment, and the shorter time frame for sampling. This indicates that with PCP, the sonication and ozonation processes occurring with sonolytic ozonation do not interact either in a chemical or physical sense. The apparent lack of benefit or detriment of the combined process compared to the individual processes is particular to compounds with a large reaction rate constant with O3. Furthermore, the results are consistent with the concept that O3 first dissolves into solution and then diffuses into a cavitation bubble where it undergoes thermolytic decomposition (19). Mechanisms of PCP Degradation. Less oxalate was formed by sonolytic ozonation at both frequencies. At 500 kHz, ozonation produced a 7-fold larger accumulation of [oxalate] compared to sonolytic ozonation. The difference at 20 kHz was less significant with a factor of 1.5 more oxalate formed by ozonation than by sonolytic ozonation. At pH values above 5, H2O2 accelerates the decomposition of O3 (the peroxone process) resulting in bulk phase OH• formation as follows (42):
H2O2 T HO2- + H+
pKa ) 11.6
O3 + HO2- f OH• + O2•- + O2
(9) (10)
H2O2 is a known byproduct of sonolysis, and the rate of H2O2 formation at 500 kHz has been shown to be a factor of 6.2 greater than that at 20 kHz (21). Therefore, less oxalate at 500 1284
9
kHz is suggestive of HO2- reacting with O3 and subsequent OH• formation. O-Chloranil and TCC byproducts formed at 60 µM PCP are due to OH• attack on the aromatic ring. OH• reacts with the ring by either e- abstraction (eq 11) or OH• addition (eq 12) as follows (43):
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 7, 2000
At temperatures greater than 400 K, OH• reacts with phenols and aromatics mainly by H-atom abstraction (44). However, based on the products observed, the overall reaction appears to proceed via OH• addition (eq 12). Furthermore, OH• addition is expected to occur at the ortho and para positions due to resonance stabilization provided by the phenoxyl substituent with the ortho position favored based on statistical considerations. o-Chloranil (eq 14) and the semiquinone radical is then formed by oxidation as shown in eqs 13-15:
The semiquinone radical disproportionates to form ochloranil and TCC. Intermediates with substitution in the para position were not detected, possibly due to their expected formation in smaller quantities. Aromatic ring opening is expected to occur by decomposition of o-chloranil yielding organic acids such as oxalate and chlorinated organic acids. In the sonochemical experiments, formation and loss of peaks by ion chromatography not associated with nonchlorinated organic acids occurred as o-chloranil was decreased and Cl- was increased. In the hydroxylation of PCP by photolysis, Wong and Crosby (45) observed similar
aromatic intermediates to this study as well as dichloromaleic acid. This suggests that these peaks are chlorinated organic acids including dichloromaleic acid. Pyrolysis products of volatile compounds have been observed in sonolysis (3, 46); however, octachlorodibenzodioxin (OCDD), a known product of PCP combustion, was not detected. Although they were not detected, liquid-phase pyrolysis byproducts such as OCDD may have been formed. The sonication of ethylbenzene produced a wide range of pyrolysis products that were detected only by delicate extraction techniques (3). The apparent absence of pyrolysis products could also indicate that the OH• radical pathways predominate over liquid-phase pyrolysis.
Acknowledgments The authors wish to thank Anne Johansen and Peter Green for help with analysis of intermediates. Financial support provided by Defense Advanced Research Projects Agency (DARPA), Office of Naval Research (ONR), the Electrical Power Research Institute (EPRI), and the Department of Energy (DOE) is gratefully acknowledged.
Literature Cited (1) Crosby, D. G. Pure Appl. Chem. 1981, 53, 1051. (2) Narang, A. S.; Swami, K.; Narang, R. S.; Eadon, G. A. Chemosphere 1991, 22, 1029. (3) De Visscher, A.; Van Langenhove, H.; Van Eenoo, P. Ultrasonics Sonochem. 1997, 4, 145. (4) Drijvers, D.; De Baets, R.; De Visscher, A.; Van Langenhove, H. Ultrasonics Sonochem. 1996, 3, S83. (5) Francony, A.; Pe´trier, C. Ultrasonics Sonochem. 1996, 3, S77. (6) Hua, I.; Hoffmann, M. R. Environ. Sci. Technol. 1996, 30, 864. (7) Kontronarou, A.; Mills, G.; Hoffmann, M. R. J. Phys. Chem. 1991, 95, 3630. (8) Kontronarou, A.; Mills, G.; Hoffmann, M. R. Environ. Sci. Technol. 1992, 26, 2420. (9) Kontronarou, A.; Mills, G.; Hoffmann, M. R. Environ. Sci. Technol. 1992, 26, 1460. (10) Okouchi, S.; Nojima, O.; Arai, T. Water Sci. Technol. 1992, 26, 2053. (11) Pe´trier, C.; Lamy, M.-F.; Francony, A.; Benahcene, A.; David, B. J. Phys. Chem. 1994, 98, 10514. (12) Leighton, T. G. The Acoustic Bubble; Academic Press: London, 1994. (13) Hart, E. J.; Henglein, A. J. Phys. Chem. 1986, 90, 3061. (14) Hart, E. J.; Henglein, A. J. Phys. Chem. 1985, 89, 4342. (15) Mackay, D.; Shiu, W. Y.; Ma, K. C. Illustrated Handbook of Physical-Chemical Properties and Environmental Fate for Organic Chemicals; Lewis Publishers: Boca Raton, FL, 1995; Vol. IV, Oxygen, Nitrogen and Sulfur Containing Compounds. (16) Barbier, P. F.; Pe´trier, C. J. Adv. Oxid. Technol. 1996, 1, 154. (17) Dahi, E. Water Res. 1976, 10, 677. (18) Olson, T. M.; Barbier, P. F. Water Res. 1994, 28, 1383. (19) Weavers, L. K.; Hoffmann, M. R. Environ. Sci. Technol. 1998, 32, 3941.
(20) Pe´trier, C.; Micolle, M.; Merlin, G.; Luche, J. L.; Reverdy, G. Environ. Sci. Technol. 1992, 26, 1639. (21) Pe´trier, C.; David, B.; Laguian, S. Chemosphere 1996, 32, 1709. (22) Gonze, E.; Gonthier, Y.; Boldo, P.; Bernis, A. Can. J. Chem. Eng. 1997, 75, 245. (23) Mason, T. J.; Lorimer, J. P. Sonochemistry: Theory, Applications and Uses of Ultrasound in Chemistry; Ellis Horwood Ltd: Chichester, 1988. (24) Sostaric, J. Z.; Mulvaney, P.; Grieser, F. J. Chem. Soc., Faraday Trans. 1995, 91, 2843. (25) Ratoarinoro; Contamine, F.; Wilhelm, A. M.; Berlan, J.; Delmas, H. Ultrasonics Sonochem. 1995, 2, S43. (26) Bader, H.; Hoigne´, J. Water Res. 1981, 15, 449. (27) Hua, I.; Hoffmann, M. R. Environ. Sci. Technol. 1997, 31, 2237. (28) Weavers, L. K.; Ling, F. H.; Hoffmann, M. R. Environ. Sci. Technol. 1998, 32, 2727. (29) Hua, I.; Ho¨chemer, R. H.; Hoffmann, M. R. Environ. Sci. Technol. 1995, 29, 2790. (30) De Visscher, A.; Van Eenoo, P.; Drijvers, D.; Van Langenhove, H. J. Phys. Chem. 1996, 100, 11636. (31) Serpone, N.; Terzian, R.; Hidaka, H.; Pelizzetti, E. J. Phys. Chem. 1994, 98, 2634. (32) Jafvert, C. T.; Westall, J. C.; Grieder, E.; Schwarzenbach, R. P. Environ. Sci. Technol. 1990, 24, 1795. (33) Crum, L. A. Ultrasonics Sonochem. 1995, 2, S147. (34) Perry, R. H.; Green, D. Perry’s Chemical Engineers’ Handbook; McGraw-Hill: New York, 1984. (35) Atkins, P. Physical Chemistry, 5th ed.; W. H. Freeman and Co.: New York, 1994. (36) NIST Chemistry WebBook: NIST Standard Reference Database, Number 69; Mallard, W. G., Ed.; 1998; http://webbook.nist.gov/ chemistry/. (37) Colussi, A. J.; Weavers, L. K.; Hoffmann, M. R. J. Phys. Chem. A 1998, 102, 6927. (38) Gutie´rrez, M.; Henglein, A.; Iban ˜ ez, F. J. Phys. Chem. 1991, 95, 6044. (39) Neta, P.; Huie, R. E.; Ross, A. B. J. Phys. Chem. Ref. Data 1988, 17, 1027. (40) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data 1988, 17, 513. (41) Beltra´n, F. J. Ozone Sci. Eng. 1995, 17, 163. (42) Staehelin, J.; Hoigne´, J. Environ. Sci. Technol. 1982, 16, 676. (43) Terzian, R.; Serpone, N.; Barton Draper, R.; Fox, M. A.; Pelizzetti, E. Langmuir 1991, 7, 3081. (44) He, Y. Z.; Mallard, W. G.; Tsang, W. J. Phys. Chem. 1988, 92, 2196. (45) Wong, A. S.; Crosby, D. G. In Pentachlorophenol: Chemistry, Pharmacology, and Environmental Toxicology; Rao, K. R., Ed.; Plenum Press: New York, 1978; p 19. (46) Hart, E. J.; Fischer, C. H.; Henglein, A. J. Phys. Chem. 1990, 94, 284.
Received for review August 4, 1998. Revised manuscript received April 23, 1999. Accepted December 16, 1999. ES980795Y
VOL. 34, NO. 7, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1285
