Environ. Sci. Technol. 2001, 35, 3247-3251
Oxidation of 4-Chloro-3-methylphenol in Pressurized Hot Water/Supercritical Water with Potassium Persulfate as Oxidant J U H A N I K R O N H O L M , H E N R I M E T S A¨ L A¨ , KARI HARTONEN, AND MARJA-LIISA RIEKKOLA* Laboratory of Analytical Chemistry, Department of Chemistry, P.O. Box 55, FIN-00014 University of Helsinki, Finland
4-Chloro-3-methylphenol (c ) 2.0 mM), representing a model pollutant, was oxidized in pressurized hot water and in supercritical water in a continuous flow system. Potassium persulfate was used as oxidant in concentrations of 8.0 and 40.0 mM. Contact times (reaction times) were 3-59 s, temperatures 110-390 °C, and pressures 235-310 bar. A wide temperature range was tested to determine the range over which potassium persulfate can be used effectively. Good oxidation efficiencies for 4-chloro-3methylphenol were obtained at both oxidant concentrations and with short contact times at temperatures clearly under the critical temperature of water; total organic carbon content of the effluent was low under optimized conditions. Corrosion, measured as nickel and chromium concentrations of the effluent, was more severe at oxidant concentration of 40.0 mM. Sulfate was present in the effluent in high concentrations. Sulfate is the limiting factor in the use of potassium persulfate in wastewater treatment and requires further water treatment.
Introduction Phenolic compounds are often hazardous to the environment and human health. 4-Chloro-3-methylphenol is one of the phenolic compounds included in the priority pollutants list of the U.S. Environmental Protection Agency and in the European Union list 76/464. It appears, for example, in tannery wastewaters (1). Pressurized hot water oxidation (PHWO) and supercritical water oxidation (SCWO) are effective techniques for wastewater treatment. For example, various phenolic compounds have been effectively treated in this way (2-4). When oxidation takes place under supercritical conditions (Tc ) 374.2 °C, Pc) 221 bar for water), the technique is referred to as supercritical water oxidation; pressurized hot water oxidation takes place at temperatures below the critical temperature and at pressures high enough to keep water in liquid state. In the supercritical state the physicochemical properties of water change dramatically as the two phases, vapor and liquid, become indistinguishable. Organic compounds and gases become soluble in the reaction medium and inorganic compounds such as salts become insoluble (5-7). Water behaves like a dense gas in the supercritical * Corresponding author phone: +358-9-191 50268; fax: +3589-191 50253; e-mail:
[email protected]. 10.1021/es000275e CCC: $20.00 Published on Web 07/04/2001
2001 American Chemical Society
FIGURE 1. Equipment used in the study. Temperature was measured at A and B. state because of increased diffusivity and decreased viscosity and dielectric constant; at the same time, it resembles a liquid in its density and solvent properties (8-11). To some extent these favorable properties of water are met at temperatures lower than the critical temperature (PHWO). Potassium persulfate is an inorganic oxidant, which is effective under milder conditions than the hydrogen peroxide and oxygen commonly used as oxidants (12). If oxidation can be carried out at low temperatures, energy can be saved and the process is more economical. High oxidation efficiencies with potassium persulfate have been achieved in a number of studies, with and without the use of catalyst (13-15). Oxidation with persulfate does not proceed at a convenient rate at room temperature unless a catalyst is present. Ag(I), for example, increases the rate of oxidation of a wide variety of compounds, including substituted phenols (13, 16). Potassium persulfate is relatively inexpensive and easily and safely handled as its potassium salt, and the reaction is free of serious interferences, which makes it an attractive choice for oxidant. Potassium persulfate has also been used as oxidizing agent in determinations of dissolved organic carbon (17, 18). Work on the kinetics and mechanisms of the decomposition and of the oxidation reactions of persulfate has been reviewed by House (12) and Willmarth (19). It is generally believed that the decomposition of persulfate proceeds by thermal and photolytic decomposition. The sulfate (and hydrogen sulfate) released in water is a limiting factor when potassium persulfate is used in wastewater treatment. Because water is corrosive at high temperature and pressure, especially with oxidant added, metals such as Ni, Cr, and Mo, originating in the reactor material, may be found in the effluent (20). While they are not good for the environment, they may have a catalytic effect on the oxidation process. For example, Cr2O3 has proven to be an effective catalyst in the oxidation of phenol (21, 22). If the equipment is to be durable and resistant to corrosion, the reactor needs to be made of material such as Inconel or Hastelloy, which is more inert than ordinary stainless steel (23, 24). The behavior of heated aqueous potassium persulfate solution in narrow tubes and a pressure regulator has not been studied widely under various conditions. The primary aim of the present study was to find out if 4-chloro-3methylphenol can be oxidized safely and effectively over a wide range of temperatures in a continuous flow system. Temperatures clearly below the critical temperature of water were of primary interest.
Experimental Section Figure 1 shows a simplified view of the equipment. The reaction tube (l ) 0.57 m,V ) 1.0 mL, 1.5 mm i.d., 1/8 in. o.d.) inside the oven (Carlo Erba Fractovap model G1, Italy) and VOL. 35, NO. 15, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3247
other tubes between the pumps and precipitant collection were made of Inconel 600 (nickel alloy, Ni >72%, Cr ∼ 15.5%). The capillary (0.5 mm i.d., 1/16 in. o.d.) leading from the precipitant collector (stainless steel) to the pressure regulator (stainless steel, Jasco, Japan) was made of stainless steel, and the capillary (0.5 mm i.d., 1/16 in. o.d.) leading from the pressure regulator to the sample collection was made of Teflon. Two high-pressure pumps (LKB-2150 HPLC pump (Sweden) for the oxidant and Jasco PU-980 HPLC pump (Japan) for the organic model pollutant) were employed to deliver aqueous solutions to the reaction zone. The flow rates of the influents (oxidant and 4-chloro-3-methylphenol in distilled deionized water) were adjusted to 1.0-6.0 mL/min (total flow from two pumps) corresponding contact times 3-59 s in the reaction tube. Contact time (reaction time) was defined by the ratio of the reaction tube volume to the volumetric flow rate. The altered density and volume properties of heated and pressurized hot water and supercritical water were taken into consideration in the calculations of flow rate. The theoretical values for density and volumes at various temperatures and pressures were calculated by the NIST/ASME Steam Properties program (Formulation for General and Scientific Use, Standard Reference Database 10 Version 2.01., U.S.A.). The temperature range was 110-390 °C (each temperature measured at two points and the average was notified, see Figure 1) and the pressure range was 235-310 bar (high enough to keep water in liquid state under all conditions). An ice bath was used to cool the stream before it entered the settling tank type precipitant collector, where insoluble particles in the effluent were collected to prevent blockage of the pressure regulator. A solution containing the model contaminant 4-chloro3-methylphenol (c ) 2.0 mM, Fluka AG, Buchs SG, Switzerland, ∼99%) was treated by PHWO/SCWO with potassium persulfate (c ) 8.0 or 40.0 mM, E. Merck, Darmstadt, Germany, 99%) as oxidant. The concentrations of the model contaminant and oxidant in the reaction tube were half of the values given above because the streams were mixed 1:1 (v/v) in the T junction in front of the reaction zone (Figure 1). Two internal standards were used in sample analysis: 4-bromophenol (Fluka AG, Buchs SG, Switzerland, ∼99%) as internal standard one (ion m/z 156 used in quantification) and 1-bromobenzene (Sigma Chemical CO, USA) as internal standard two (ion m/z 172 used to control the efficiency of liquid-liquid extraction). Ion m/z 107 was used as quantification ion for 4-chloro-3-methylphenol. Dichloromethane (Lab-Scan, Analytical Sciences, Dublin, Ireland, HPLC grade) was used as solvent in the extraction of the organic pollutants from effluent. Each set of experiments was started at 25 °C where the reference samples were collected. Reference samples were used in the calculations of conversions for 4-chloro-3methylphenol: the relation of the peak area of the quantification ion for internal standard 1 to that of 4-chloro-3methylphenol under reference conditions (25 °C) was compared to the same relation under certain experimental (oxidation) condition. The flow rates of the pumps were kept constant, and the temperature was raised during one set of experiments. Other sets of experiments were carried out with different pump flow rates. For GC-MS and corrosion analysis, five effluent samples (A-E, 10 mL each) were collected parallel in test tubes under each set of conditions; for example,a total of five 10 mL samples was collected at 150 °C with flow rate of 2.0 mL/min. The temperature was allowed to settle 10-15 min, depending on the flow rate, before sample collection. For GC-MS analysis, the first internal standard (4bromophenol, Vinj ) 50 µL) was added in concentration 5.0 mg/L to three of the samples (A-C) and in concentration 1.0 3248
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 15, 2001
mg/L (Vinj ) 20 µL) to the fourth sample (D). All samples were extracted with dichloromethane (3 × 3.5 mL). Samples A-C were concentrated to 1.5 mL in a Pierce Reacti-Therm heating module (U.S.A., T ) 30 °C, nitrogen evaporation) and the second internal standard (1-bromobenzene, Vinj ) 50 µL) in concentration 5.0 mg/L was added. Samples A-C were used in determining the removal efficiency of 4-chloro3-methylphenol, and the fourth sample (D) was concentrated to 0.2 mL for determination of the products formed during oxidations. Compounds left in the effluent were investigated with a Hewlett-Packard model 5890 gas chromatograph equipped with a model 5989A mass spectrometer (U.S.A.). The software used in the computer connected to the GC-MS was Hewlett-Packard ChemStation version B.02.05. The software included a library of mass spectra (Wiley), which was used in identifying the oxidation products. A library match >90 (100 is the maximum) was considered as the minimum requirement for the positive identification of the intermediates. All MS analyses were carried out in SCAN mode by electron impact ionization (EI, 70 V). Samples were injected in splitless mode (Vinj ) 2.0 µL). The analytical column of the gas chromatograph was a 20.0-m NB 351 (HNU Nordion, Espoo, Finland) with 0.2 mm i.d. and 0.2 µm film thickness. A 0.25 mm i.d. retention gap (BGB Analytik AG, Rothenfluh, Switzerland, 1.0-2.0 m) deactivated with DPTMDS (1,2-diphenyl-1,1,3,3-tetramethyldisilazane) was connected to the analytical column with a press fit connector (BGB Analytik AG, Rothenfluh, Switzerland). Corrosion was measured as nickel and chromium concentrations in the effluent (sample E) with a Perkin-Elmer 272 flame atomic absorption spectrophotometer (AAS, made in the U.S.A.). Nickel(II) sulfate (E. Merck, Darmstadt, Germany, >99%) was used to obtain the calibration curve for Ni and potassium dichromate (E. Merck, Darmstadt, Germany, analytical grade) to obtain the calibration curve for Cr. The wavelengths used in the measurements were 232.0 and 357.9 nm, respectively. The acidity of the effluent was measured with a Jenway 3030 pH meter (England). For total organic carbon (TOC) analysis, three effluent samples (10 mL each) were collected under each set of conditions, and a Shimadzu TOC-5000 analyzer (U.S.A.) was employed to measure total organic carbon left in the effluent. Samples for TOC measurements were collected separately from other analytical methods. Total organic carbon was measured by subtracting the inorganic carbon (IC) from the total carbon (TC) in the sample. Standard method SFS-EN 1484 was employed in the analysis. For sulfate analysis (25), three samples (10 mL each, acidified with 2 drops of concentrated hydrogen chloride) were collected under each set of conditions as described before. The procedure for sulfate analysis was carried out separately from GC-MS, corrosion, and TOC analysis. Potassium chloride (c ) 50 g/L, V ) 1.0 mL, Riedel-deHae¨n, Seelze, Germany, >99.5%) was added to prevent ionization interference. An accurately measured amount of barium chloride (c(BaCl2‚2H2O) ) 3.558 g/L, V ) 10.0 mL, E. Merck, Darmstadt, Germany, >99%) was added to the samples, the volume of each sample was made to up 25.0 mL with distilled deionized water, and finally each sample was shaken vigorously for 2 min. The unreacted excess of barium was determined indirectly by AAS. Na2SO4 solution (c ) 1.479 g/L, E. Merck, Darmstadt, Germany, >99%) was used as standard stock solution for preparing the calibration standards.
Results and Discussion The purpose of our work was to study the oxidation efficiency of 4-chloro-3-methylphenol with two concentrations of potassium persulfate and a wide range of temperatures. The conversions (in percentage) of 4-chloro-3-methylphenol are averages of three experiments performed under identical
TABLE 1. Conversion of 4-Chloro-3-methylphenol under Various Conditions with Potassium Persulfate Concentration of 40.0 mM
FIGURE 2. Conversion of 4-chloro-3-methylphenol at 110-180 °C and 235-280 bar with potassium persulfate concentration of 40.0 mM. Also TOC removal (in %) is presented by plots. Exact information on TOC removal can be found in Table 3.
T (°C)
t (s)
P (bar)
conversion (%)
RSD (%)
200 200 200 200 200 200 250 250 250 250 250 250 350 350 350 350 350 385 385 385 385 385 385
8.7 13.1 17.5 26.3 34.9 52.5 8.2 12.2 16.3 24.5 32.5 49.0 6.3 9.3 12.6 18.6 25.2 3.2 4.8 9.7 10.2 12.9 18.3
260 255 250 285 235 255 270 240 250 270 252 275 280 245 280 250 280 240 285 290 230 285 275
31.3 96.1 99.5 99.5 99.8 99.8 99.7 99.5 99.9 99.9 99.6 99.8 99.8 99.4 99.5 99.4 98.9 99.2 99.0 99.2 99.1 99.1 98.8
0.02 0.20 0.01 0.03 0.05 0.08 0.02 0.02 0.04 0.13 0.02 0.01 0.05 0.03 0.01 0.10 0.06 0.02 0.04 0.08 0.06 0.10
TABLE 2. Conversion of 4-Chloro-3-methylphenol under Various Conditions with Potassium Persulfate Concentration of 8.0 mM FIGURE 3. Conversion of 4-chloro-3-methylphenol at 110-250 °C and 245-285 bar with potassium persulfate concentration of 8.0 mM. conditions. The relative standard deviations (RSDs) for the conversions were 0.01-260.4%. RSDs were clearly highest at very low conversions and lowest at high conversions. For reasons of clarity, the RSDs are not marked as error bars in Figures 2 and 3. For the same reason, curves are not drawn beyond 99% conversion of 4-chloro-3-methylphenol (see Figures 2 and 3). Oxidation results at temperatures g200 °C (oxidant concentration 40.0 mM) and g300 °C (oxidant concentration 8.0 mM) are marked in Tables 1 and 2 to draw attention to the small variations in conversion percentages. Also, the low values of RSDs at high conversions can be observed in the tables. The limit of detection for 4-chloro3-methylphenol was 5.0 ng (S/N ) 2). Accordingly, if the compound was not detected in the analysis, >99.9 was assumed for the conversion percentage. Conversion of 4-chloro-3-methylphenol obtained by GCMS analysis was studied at various temperatures with potassium persulfate concentrations of 40.0 and 8.0 mM (Figures 2 and 3, Tables 1 and 2). Contact times (reaction times) were calculated as explained in Experimental Section. Excellent conversions were obtained at temperatures clearly under 200 °C. The good oxidation power of potassium persulfate was observed in our previous studies, too (26, 27). As can be seen, the conversion increased with temperature and contact time (Figures 2 and 3). When the concentration of the oxidant was decreased from 40.0 mM (Figure 2) to 8.0 mM (Figure 3), the oxidation efficiency decreased, too. Nevertheless, even with potassium persulfate concentration of 8.0 mM, excellent conversions were obtained at low temperatures and short reaction times; for example, over 99% conversion was obtained at 150 °C and 37 s. In another study, Goulden et al. (28) found the oxidation efficiency of nicotinic acid to increase markedly with potassium persulfate concentration. With potassium persulfate as oxidant, good
T (°C)
t (s)
P (bar)
conversion (%)
RSD (%)
300 300 300 300 300 350 350 350 350 350 385 385 385
7.4 14.8 22.1 29.7 44.5 6.3 12.6 18.7 25.5 37.9 8.7 9.3 11.1
278 265 246 285 280 283 270 260 300 280 270 250 260
99.6 99.3 99.7 99.3 97.7 99.5 99.0 97.2 99.0 96.5 98.3 96.2 98.5
0.03 0.51 0.02 0.08 0.25 0.01 0.07 0.12 0.03 0.11 0.28 0.06
oxidation power was achieved at clearly lower temperatures and shorter contact times than normally with hydrogen peroxide or oxygen as oxidant. Oxidation efficiencies of hydrogen peroxide and potassium persulfate were compared in our previous study and potassium persulfate was clearly more efficient (27). Effect of pressure (in the range used in this study) on conversion was considered neglible. With potassium persulfate concentration of 40.0 mM and temperatures > 250 °C, over 99% conversion was obtained in less than 10 s (Table 1). With oxidant concentration of 8.0 mM, similar results were obtained at >300 °C (Table 2). Conversions of at least 99.9% were obtained only under a few conditions, perhaps because persulfate itself decomposes at a higher rate at increased temperature, and the useful reaction time is limited by this decomposition (27). Thus, the optimum conditions should be selected by paying attention to temperature and reaction time together. The results for the TOC studies with oxidant concentration of 40.0 mM are presented in Figure 2 and Table 3. Over 95% removal of TOC was obtained at 200 and 250 °C with reaction times 35.0 and 32.8 s, respectively. What is remarkable is that good conversions of 4-chloro-3-methylphenol were achieved at temperatures and reaction times that were not VOL. 35, NO. 15, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3249
TABLE 3. TOC Removal under Various Conditions with Potassium Persulfate Concentration of 40 mM T (°C) 120 150 200 250 390
t (s)
P (bar)
TOC removal%
RSD (%)
18.9 37.9 18.5 37.0 17.5 35.0 16.3 32.8 9.6 16.3
251 253 258 282 259 265 253 302 310 280
6.1 57.4 23.9 72.1 71.3 95.7 77.0 95.4 57.9 93.3
47.9 8.5 3.9 4.0 2.9 0.5 1.8 0.3 3.3 0.4
FIGURE 4. Relative abundance of the reaction product (bis(2ethylhexyl) ester of 1,2-benzenedicarboxylic acid) at 110-200 °C and 235-280 bar with potassium persulfate concentration of 40.0 mM. sufficient for effective TOC removal (Tables 1 and 3). Evidently 4-chloro-3-methylphenol was decomposed to carbon dioxide and water via intermediates. Bis(2-ethylhexyl) ester of 1,2-benzenedicarboxylic acid was overall the most abundant reaction intermediate, and it was identified tentatively (library match of mass spectra ∼95). Figure 4 shows its abundance relative to that of internal standard 1 under various conditions with potassium persulfate concentration of 40.0 mM. It can be observed that there is a clear maximum or systematic trend in the concentration of bis(2-ethylhexyl) ester of 1,2-benzenedicarboxylic acid under each condition. For example, with about 20 s reaction time and 150 °C, the conversion of 4-chloro-3-methylphenol was about 80% (Figure 2), and the amount of bis(2-ethylhexyl) ester of 1,2-benzenedicarboxylic acid was at its highest (Figure 4). The TOC removal under similar conditions (T ) 150 °C, t ) 18.5 s) was only 23.9% (Table 3). At temperatures g 250 °C, the intermediate was detected in either a very low amount or not at all. 3-Methylphenol and 2-methyl-1,4-benzenediol were also identified tentatively as reaction intermediates (library match of mass spectra ∼95). They were present in low abundances under some conditions and detected in either very low amounts or not at all at temperatures g 250 °C. Some other intermediates were seen as well under some conditions, mainly at temperatures < 250 °C, but in trace amounts and with poor identification. No further efforts were made to identify them. Figure 5 shows the results of the pH studies with potassium persulfate concentration of 40.0 mM. The pH of the effluent can be seen to decrease with an increase in reaction time and temperature. Presumably this was mainly due to the formation of acidic intermediates (for example, low molecular weight organic acids), hydrogen chloride, and hydrogen sulfate during oxidation. However, at temperatures from 300 to 385 °C the pH remained more or less constant (about 1.7) until a slight increase occurred at longest reaction times. 3250
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 15, 2001
FIGURE 5. pH of the effluent at 110-385 °C and 235-290 bar with potassium persulfate concentration of 40.0 mM.
FIGURE 6. Nickel concentration of the effluent at 110-385 °C and 235-290 bar with potassium persulfate concentration of 40.0 mM.
FIGURE 7. Chromium concentration of the effluent at 110-385 °C and 235-290 bar with potassium persulfate concentration of 40.0 mM. The shape of the curves obtained with potassium persulfate concentration of 8.0 mM was similar to the shape obtained with concentration of 40.0 mM. With 8.0 mM, the minimum pH value was 2.3 (data not shown). Lowering of the pH with decomposition of persulfate has also been observed in other studies (28). The maximum oxidation rate constant, for example, for nicotinic acid, occurred in the pH range 3-6. Corrosion (nickel and chromium concentration in the effluent) was studied under various conditions. Figures 6 and 7 show the corrosion results with the potassium persulfate concentration of 40.0 mM. The concentration of nickel in the effluent was considerably higher than that of chromium, reflecting the ratios of the metals in Inconel 600. Corrosion was mostly increased with temperature and reaction time, but there were some exceptions. For example, there were clear drops (valleys) in nickel and chromium concentrations at increased reaction time. Metal concentrations of the effluent were low for all reaction times at temperatures e 170 °C. The curves of the metal concentrations for potassium persulfate concentration of 8.0 mM
below the critical temperature of water. Corrosion was considerably stronger at high (40.0 mM) than at low (8.0 mM) concentrations of potassium persulfate. High sulfate concentration in the effluent is clearly a negative factor for the use of potassium persulfate in wastewater treatment.
Acknowledgments This project was supported by the Maj & Tor Nessling Foundation. We thank Lic. Phil. Pentti Jyske and Ph.D. Pirkko Tilus for scientific help.
Literature Cited FIGURE 8. Sulfate (and hydrogen sulfate) concentration of the effluent at 120-390 °C and 230-260 bar with potassium persulfate concentration of 40.0 mM. followed the same trend as for concentration of 40.0 mM. However, the maximum metal concentrations were only 15.0 mg/L for nickel and 1.2 mg/L for chromium, occurring at 385 and 300 °C, respectively. Both these values are clearly lower than those for oxidant concentration of 40.0 mM, showing that the corrosion with potassium persulfate as oxidant is strongly dependent on oxidant concentration. We consider that the origin of nickel and chromium was the heated Inconel tube, because no corrosion was observed with reference samples (T ) 25 °C) and no other parts of the equipment were heated during the oxidation process. Sulfate and hydrogen sulfate concentration of the effluent was studied with potassium persulfate concentration of 40.0 mM (Figure 8). The maximum concentration of sulfate (about 3500 mg/L) was achieved at 200 and 390 °C with sufficient reaction times. The maximum sulfate concentration was also calculated theoretically. If it is assumed that the sulfur in potassium persulfate (c ) 40 mM) is totally converted to sulfate (or hydrogen sulfate), the concentration of sulfate in the effluent would be about 3850 mg/L, which is close to the measured value. It can be approximated that, with oxidant concentration of 8.0 mM, the sulfate concentration of the effluent would be about 20% of the value obtained with 40.0 mM. In wastewater treatment, the sulfate could be removed by precipitation, biological, or ion exchange methods. Hazards related to the equipment were not evident in any of the experiments. Reaction tubes and the pressure regulator were not blocked at any temperature despite the large amounts of metal (corrosion) and sulfate sometimes present in the effluent. In addition, the procedure was easy to carry out in continuous flow mode. A precipitant collection (see Figure 1) may have contributed to this though no significant amounts of insoluble substances were found in the precipitant collector and no efforts were made to analyze them. To conclude, 4-chloro-3-methylphenol was effectively oxidized with potassium persulfate in a continuous flow system at short reaction times and at temperatures clearly
(1) Castillo, M.; Puig, G.; Barcelo, D J. Chromatogr. A 1997, 778, 301. (2) Martino, C. J.; Savage. P. E. Ind. Eng. Chem. Res. 1997, 36, 1391. (3) Koo, M.; Lee, W. K.; Lee, C. H. Chem. Eng. Sci. 1997, 52, 1201. (4) Martino, C. J.; Savage. P. E. Ind. Eng. Chem. Res. 1999, 38, 1784. (5) Connolly, J. F. J. Chem. Eng. Data 1966, 11, 13. (6) Japas, M. L.; Franck, E. U. Ber. Bunsen-Ges. Phys. Chem. 1985, 89, 1268. (7) Armellini, F. J.; Tester, J. W. J. Supercritical Fluids 1994, 7, 147. (8) Franck, E. U. Pure Appl. Chem. 1970, 24, 13. (9) Dudziak, K. H.; Franck, E. U. Ber. Bunsen-Ges. Phys. Chem. 1966, 70, 1120. (10) Kalinichev, A. G.; Bass, J. D. Chem. Phys. Lett. 1994, 231, 301. (11) Uematsu, M.; Franck, E. U. J. Phys. Chem. Ref. Data 1960, 9, 1291. (12) House, D. A. Chem. Rev. 1962, 62, 185. (13) Salem, M. A. Monatsh. Chem. 2000, 131, 117. (14) Beattie, J. K.; De Martin, J. A.; Kennedy, B. J. Aust. J. Chem. 1994, 47, 1859. (15) Ram. N.; Sidhu, K. S. Indian J. Chem. 1978, 16A, 195. (16) Srivastava, S. P.; Shukla, A. K. Indian J. Chem. 1977, 15A, 603. (17) Menzel, D. W.; Corwin, N. Limnol. Oceanog. 1965, 10, 280. (18) Wilson, R. F. Limnol. Oceanog. 1961, 6, 259. (19) Willmarth, W. K.; Haim, A. Mechanism of Oxidation of Peroxydisulphate in Peroxide Reaction Mechanism; Edwards, J. O., Ed.; John Wiley & Sons: New York, 1961; pp 175-225. (20) Mitton, D. B.; Orzalli, J. C.; Latanasion, R. M. Innovations in Supercritical Fluids: Science and Technology; ACS Symp. Ser. 608; American Chemical Society: Washington, DC, 1995; pp 327-337. (21) Ding, Z.-Y.; Aki, S. N. V. K.; Abraham, M. A. Environ. Sci. Technol. 1995, 29, 2748. (22) Krajnc, M.; Levec, J. Appl. Catal. B 1994, 3, L101. (23) Downey, K. W.; Snow, R. H.; Hazlebeck, D. A.; Roberts, A. J., Innovations in Supercritical Fluids: Science and Technology; ACS Symp. Ser. 608; American Chemical Society: Washington, DC, 1995; pp 313. (24) Alekseev, A. B.; Averin, S. A.; Geferova, M. N.; Kondratev, V. P.; Shikhalev, V. S. J. Nuclear Mater. 1996, 233-237, 1367. (25) Dunk, R.; Mostyn, R. A.; Hoare, H. C. At. Absorp. Newsl. 1969, 8, 79. (26) Kronholm, J.; Riekkola, M.-L. Environ. Sci. Technol. 1999, 33, 2095. (27) Kronholm, J.; Jyske, P.; Riekkola, M.-L. Ind. Eng. Chem. Res. 2000, 39, 2207. (28) Goulden, D.; Anthony, D. H. J. Anal. Chem. 1978, 50, 953.
Received for review November 20, 2000. Revised manuscript received April 9, 2001. Accepted May 7, 2001. ES000275E
VOL. 35, NO. 15, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3251