Anal. Chem. 2004, 76, 6359-6364
A Kinetic Method for HO2•/O2•- Determination in Advanced Oxidation Processes Bum Gun Kwon and Jai H. Lee*
Department of Environmental Science and Engineering, Gwangju Institute of Science and Technology, 1 Oryong-Dong, Buk-Gu, Gwangju 500-712, Korea
A new kinetic method is developed for the determination of hydroperoxyl radical (HO2•)/superoxide radical (O2•-) in aqueous solution, and the calibration using a kinetic half-life technique is also established for determining the concentration of HO2•/O2•- as produced in the UV/H2O2 process. This new method is based on the reduction of Fe3+-EDTA into Fe2+-EDTA by HO2•/O2•- and the wellknown Fenton-like reaction of H2O2 and Fe2+-EDTA to yield the hydroxyl radicals (OH•). Benzoic acid scavenges the OH radicals to produce hydroxybenzoic acids, which are analyzed by fluorescence detection (λex ) 320 nm; λem ) 400 nm). The limit of detection for the new method depends on the pH values, and it is determined as 3.22 × 10-11 M with signal-to-noise ratio of 2 at pH 5. In addition, the present technique has the advantage of using inexpensive and easily available nonenzymatic reagents that do not require the specific instrument and chemicals and of being insensitive to the moderate concentration of possible interferences often found in aqueous phase. The chemistry of hydroperoxyl radical (HO2•)/superoxide radical (O2•-) (pKa ) 4.8)1 has been studied in numerous advanced oxidation processes employing UV/H2O2,2 Fe2+/H2O2,3 UV/TiO2,4 O3,5 radiolysis (pulse and stopped-flow), and ionizing radiation (electron beam).6,7 The HO2•/O2•- has been considered as an intermediate and a propagator of the ensuing chain reactions of radicals8 during the photochemical9 and ozonation processes.10-12 The oxidation/reduction reactions of HO2•/O2•- have been considered to be indicative of a source of the OH radical as a strong oxidant in biological system.13,14 As a result, an understanding of * Corresponding author. Tel.: 82-62-970-2444. Fax: 82-62-970-2434. E-mail:
[email protected]. (1) Bielski, B. H. J.; Allen, A. O. J. Phys. Chem. 1977, 81, 1048-1050. (2) Bielski, B. H. J.; Cabelli, D. E.; Arudi, R. L.; Ross, A. B. J. Phys. Chem. Ref. Data 1985, 14, 1041-1100. (3) Walling, C. Acc. Chem. Rev. 1975, 8, 125-131. (4) Nosaka, Y.; Yamashita, Y.; Fukuyama, H. J. Phys. Chem. 1997, 101, 58225827. (5) von Sonntag, C.; Dowideit, P.; Fang, X.; Mertens, R.; Pan, X.; Schuchmann, M. N.; Schuchmann, H.-P. Water Sci. Technol. 1997, 35, 9-15. (6) Farhataziz and Ross, A. B. Nat. Stand. Ref. Data Ser., Nat. Bur. Stand. (U. S.) 1977, 59. (7) Schwarz, H. A. J. Chem. Educ. 1981, 58, 101-105. (8) Cabelli, D. E.; Bielski, B. H. J. J. Phys. Chem. 1983, 87, 1809-1812. (9) Legrini, O.; Oliveros, E.; Braun, A. M. Chem. Rev. 1993, 93, 671-698. (10) Staehelin, J.; Hoigne, J. Environ. Sci. Technol. 1982, 16, 676-681. (11) Bu ¨ hler, R. E.; Staehelin, J.; Hoigne, J. J. Phys. Chem. 1984, 88, 2560-2564. (12) Kang, J. W. Ph.D. Dissertation, University of California, Los Angeles, 1989. (13) Fridovich, I. Acc. Chem. Res. 1972, 5, 321-326. 10.1021/ac0493828 CCC: $27.50 Published on Web 10/01/2004
© 2004 American Chemical Society
the reactions controlling concentration of HO2•/O2•- is necessary to elucidate its role in order to maximize the efficiency of the advanced oxidation processes. Nevertheless, the measurement of HO2•/O2•- has been so far limited in advanced oxidation processes. There are various methods available in the determination of HO2•/O2•-. The direct method includes the optical detection of the low UV absorbance as discussed by Bielski et al.2 Both HO2• and O2•- were found to show distinct absorption spectra with maximums at 225 and 245 nm, respectively. This method, however, is limited in practice due to the high sensitivity requirement as well as the spectrum overlapping as many chemical species absorb strongly at these wavelengths. In addition, HO2•/O2•- has characteristic electron spin resonance spectra, which can be detected only at very low temperatures and high HO2•/O2•- concentration.2 Moreover, the most commonly used indirect methods for HO2•/O2•- are the utilization of chemical indicators such as tetranitromethane,9,15 nitro blue tetrazolium,16 and cytochrome c,14,17,18 which form products with intense optical absorbance. These spectrophotometric methods, however, have suffered from low sensitivity. A recent method for the determination of O2•- is based on its decomposition with concomitant oxidation of a luminol solution to form a product that yields a chemiluminescence signal.4,19 Adding to the requirement of luminal purification, this technique exhibits negative interference from transition metals.20 To assess the activity of antioxidants on O2•-, several enzymatic methods using superoxide dismutase have also been developed. Among enzymatic methods, the most commonly used involves the inhibition of lipid peroxidation by antioxidants, which can be measured by the end products, adducts, or other indicators of oxidative reaction.13,21-23 However, enzymatic methods are disadvantageous as the enzymes are unstable and expensive. A sensitive and specific bioluminescence of protein polynoidin for O2•- has also been developed during the past decades and utilized limitedly in cell biology and medicine.24 Baker and Gebicki25 characterized (14) Halliwell, B. Biochem. Pharmacol. 1995, 49, 1341-1348. (15) Rabani, J.; Mulac, W. A.; Matheson, M. S. J. Phys. Chem. 1965, 69, 53-70. (16) Pasternack, R. F.; Halliwell, B. J. Am. Chem. Soc. 1979, 101, 1026-1031. (17) Fridovich, I. J. Biol. Chem. 1970, 245, 4053-4057. (18) Okado-Matsumoto, A.; Fridovich, I. Anal. Biochem. 2001, 298, 337-342. (19) Mere´nyi, G.; Lind, J. S. J. Am. Chem. Soc. 1980, 102, 5830-5835. (20) Ibusuki, T. Atmos. Environ. 1983, 17, 393-396 (21) Sawyer, D. T.; Valentine, J. S. Acc. Chem. Res. 1981, 14, 393-400. (22) Thomas, C. E.; Morehouse, L. A.; Aust, S. D. J. Biol. Chem. 1985, 260, 3275-3280. (23) Naguib, Y. M. Anal. Biochem. 1998, 265, 290-298.
Analytical Chemistry, Vol. 76, No. 21, November 1, 2004 6359
Figure 1. (a) Schematic diagram and calibration equipment for measuring HO2•/O2•-: A, peristaltic pump; B, UV photolysis; C, D1, D2, knotted tubing reactors (KTRs); E, porous tube debubbler; F, flow-through cell type fluorometer; G, A/D converter; H, computer; P1, P2, P3, P4, solution inlet ports. (b) UV photolysis system: quartz, volume 2.52 cm3, and length 80 cm.
a conversion of superoxide radical to the hydroxyl radical by superoxide-driven Fenton reactions in 60CO γ-radiation as measured by the formation of fluorescent hydroxylated derivatives from benzoate. However, this method has only focused on the conversion effect of superoxide radical by ferric ethylenediamineacetate (Fe3+-EDTA) as the catalyst, rather than the absolute concentration of superoxide radical. During the past decade, several methods specific for the detection of HO2•/O2•- have been developed, but the calibration and the standards for HO2•/O2•bring about an analytical challenge. Therefore, the requirement of a new method has emerged for the detection of HO2•/O2•taking the problems of existing methods into consideration. This paper presents a kinetic and an analytical method for the determination of HO2•/O2•-, which is based upon both Fe3+-EDTA reduction into Fe2+-EDTA by HO2•/O2•- 25-28 and the well-known Fenton-like reaction to yield the hydroxyl radical (OH•). The hydroxyl radical is scavenged by benzoic acid (BA) to form hydroxybenzoic acid (OHBA). The optimization and the calibration using a kinetic half-life method were established for (24) Colepicolo, P.; Camarero, V. C. P. C.; Nicolas, M. T.; Bassot, J.-M.; Karnovsky, M. L.; Hastings, J. W. Anal. Biochem. 1990, 184, 369-374. (25) Baker, M. S.; Gebicki, J. M. Arch. Biochem. Biophys. 1984, 234, 258-264. (26) McCord, J. M.; Day, E. D., Jr. FEBS Lett. 1978, 86, 139-142. (27) Gutteridge, J. M. C. FEBS Lett. 1982, 150, 454-458. (28) Gutteridge J. M. C. Free Radical Biol. Med. 1991, 11, 401-406.
6360 Analytical Chemistry, Vol. 76, No. 21, November 1, 2004
determining the concentration of HO2•/O2•- as produced in the UV/H2O2 process. EXPERIMENTAL SECTION Materials. Fe3+-EDTA, sulfuric acid, sodium hydroxide, BA, 3% hydrogen peroxide, m-hydroxybenzoic acid (m-OHBA), and o-hydroxybenzoic acid (o-OHBA) were of reagent grade and used without further purification. Interference studies were conducted by adding various amounts of methanol, acetone, ethanol, acetate, formaldehyde, oxalate, nitrite, chloride, bromate, bromide, phosphate, sulfite, bisulfite, thiosulfite, and nitrate to working solutions. Reagent-grade chemicals were used for all the experiments, and all chemicals were purchased from Sigma-Aldrich. All the solutions were made with high-purity water from a Millipore ultrapurification system (>18 MΩ cm). The solution pH was adjusted to the ranges between 3 and 10 with acetate buffer (Aldrich), phosphate buffer (Sigma), phosphoric acid buffer, and borate buffer (Baker & Adamson) along with H2SO4 and NaOH. The concentration of the stock H2O2 solution was determined by using a KMnO4 titration method prior to use. Working solution of H2O2 was prepared daily by diluting the H2O2 stock in proper level with distilled water. Apparatus and Procedures. A schematic diagram of the apparatus for HO2‚/O2•- measurement is shown in Figure 1a. All solutions were delivered by using a peristaltic pump (Ismatec Co.)
with PTFE tubing (Cole-Parmer, i.d. 0.8 mm). The pump was composed of four inlet ports. The H2O2 solution was delivered through port 1 (P1, 0.42 mL/min) and was photolyzed by UV irradiation (B), which was equipped with a 4-W low-pressure Hg lamp (λmax ) 254 nm, Philips). The HO2•/O2•- stream produced from the photolysis of H2O2 in a quartz coil-type reactor (Figure 1b) was passed through a knotted tube reactor, KTR (C),29 which was able to control the concentration of HO2•/O2•-. The Fe3+-EDTA solution (P2, 0.23 mL/min) was mixed with BA solution (P3, 0.23 mL/min) in KTR (D1) and then joined with the HO2•/O2•--H2O2 stream. The reduction of Fe3+-EDTA by HO2•/O2•- resulted in the production of Fe2+-EDTA, which reacted further with residual H2O2. Fenton-like reaction of H2O2 and Fe2+-EDTA produced the OH radicals, which were then scavenged by BA to produce OHBA in KTR (D2). After 0.05 N NaOH (P4, 0.23 mL/min) was added to raise the pH level above 11, at which the fluorescence intensity of OHBA could be maintained at a maximum level.30 The mixed solution occasionally caused the formation of air bubbles in the effluent stream. The bubbles were then removed by incorporating a small piece (3 cm) of porous hydrophobic membrane tube (E) (Gore-Tex TA 001, Gore and Associates) prior to entry to the fluorometer (F) in order to prevent a noise signal by the air bubbles.31 The OHBA fluorescence was then measured with a fluorometer (Waters 474 model) equipped with a 16-µL flow-through cell using 320 nm (excitation)/400 nm (emission) with a slit width of 40 nm. The fluorescent signal was transferred to a data acquisition system, Auto-chrowin (Younglin Co.), consisting of an analog-todigital converter (G) with a personal computer (H). To calibrate the HO2•/O2•- measurement system, all working solutions were passed through the appropriate ports under UV lamp-off and the baselines were monitored. The H2O2 solution placed in UV photolysis (B) under UV lamp-on was photolyzed and then passed through the gradient length of KTR (C), which was controlled stepwise as 0, 1, 2, 3, and 4 m. In the absence of additives, HO2• and O2•- in each KTR (C) were disproportionated by self-reactions of R1-R3 according to the empirically observed pH-dependent rate constant, kobs:1,2,15
HO2• + HO2• f H2O2 + O2
(R1)
HO2• + O2•- f HO2- + O2
(R2)
O2•- + O2•- f O2 + O22-
(R3)
kobs ) {k1 + k2 (KHO2/[H+])}/(1 + KHO2/[H+])2
nism including the Fenton-like reaction. In this manner, the fluorescence signal of OHBA produced from the OH radical reaction with BA is proportional to a given HO2•/O2•- concentration. The half-life (t1/2) of the HO2•/O2•- was obtained by plotting a simple linear relationship of signal ratio versus reaction time. The half-life in a second-order reaction was inversely proportional to the initial concentration. Thus, the initial concentration of HO2•/O2•- could be kinetically calculated from the disproportionation reaction of HO2•/O2•- based on the half-life and calculated kobs at a given pH. RESULTS AND DISCUSSION Reaction Scheme. H2O2 is photodecomposed to produce two OH radicals by an ultraviolet absorption with a small molar extinction coefficient ( ) 19 M-1cm-1) at 254 nm. Most of the OH radicals formed under UV irradiation (R4) in B (Figure 1) react with residual H2O2 giving HO2• (R5):32,33
H2O2 + hν f 2OH•
(R4)
OH• + H2O2 f HO2• + H2O
(R5)
It is well known that hydroperoxyl radical is dependent upon the acid-base equilibrium:
HO2• T O2•- + H+
pKa ) 4.8
(R6)
Fe3+-EDTA may alter the reactions of (R1-R3).26-28 Fe3+-EDTA is reduced by HO2•/O2•- to produce Fe2+-EDTA and O2 with k7 ) 2 × 106 M-1 s-1.2, 34
Fe3+-EDTA + O2•- f Fe2+-EDTA + O2
(R7)
Fe2+-EDTA and H2O2 (k8 ) (2 ( 1) × 104 M-1 s-1)34 lead to the production of the OH radical and regeneration of Fe3+-EDTA in (R8).
Fe2+-EDTA + H2O2 f Fe3+-EDTA + OH- + OH• (R8) Then, the OH radical produces OHBA in the presence of BA with a nearly diffusion-controlled rate constant of k9 ) 4.3 × 109 M-1 s-1.35
OH• + BA f OHBA
(R9)
(I)
where kobs can be calculated using k1 ) (8.3 ( 0.7) × 105 M-1 s-1, k2 ) (9.76 ( 0.6) × 107 M-1 s-1, k3 < 0.3 M-1 s-1, and KHO2 ) 1.6 × 10-5 M-1 as recommended values.2 Stream containing HO2• and O2•- radicals produces the OH radicals by the mecha(29) Clark, G. D.; Hungerford, J. M.; Christian, G. D. Anal. Chem. 1989, 61, 973-979. (30) Lee, J. H.; Tang, I. N.; Weinstein-Lloyd, J. B. Anal. Chem. 1990, 62, 23812384. (31) Martine, G. B.; Cho, H. K.; Meyerhoff, M. E. Anal. Chem. 1984, 56, 26122613.
Fe3+-EDTA, however, may compete with BA for the OH radicals. To minimize the scavenging of OH radicals by Fe3+-EDTA, which is characterized by a second-order rate constant of k10 ) 1.1 × (32) Kochany, J.; Bolton, J. R. Environ. Sci. Technol. 1992, 26, 262-265. (33) Stefan, M. I.; Hoy, A. R.; Bolton, J. R. Environ. Sci. Technol. 1996, 30, 23822390. (34) Bull, C.; McClune, G. J.; Fee, J. A. J. Am. Chem. Soc. 1983, 105, 5, 52905300. (35) Buxton, G. V.; Greenstock, C. L.; Helman W. P.; Ross, A. B. J. Phys. Chem. Ref. Data 1988, 17. (36) Amichai, O.; Czapski, G.; Treinin, A. Isr. J. Chem. 1969, 7, 351-359. (37) Zepp, R. G.; Hoigne, J.; Bader, H. Environ. Sci. Technol. 1987, 21, 443450.
Analytical Chemistry, Vol. 76, No. 21, November 1, 2004
6361
Figure 2. Dependence of the fluorescence signal on buffer solutions: borate buffer (pH 8.75) and acetate buffer (pH 4.90), [H2O2] ) 4 mM, [Fe3+-EDTA] ) 20 µM, [BA] ) 1 mM, [NaOH] ) 0.05 N, and 2.5-min reaction time.
109 M-1 s-1,35 a low concentration of Fe3+-EDTA to keep the ratio of k9[BA] to k10[Fe3+-EDTA] larger than 195 was maintained. In other words, ∼99.5% of the OH radicals will react with benzoic acid and other competition reactions for OH radicals would be negligible. Reaction Conditions. To determine the optimum reaction conditions for the analysis of HO2•/O2•-, the effects of various buffers and of their concentrations, the reaction time for Fentonlike reaction, the optimum concentration of Fe3+-EDTA, and the fluorescence intensities of OHBA have been investigated. Solutions of BA and Fe3+-EDTA are stable at room temperature for at least 2 weeks as long as their pH values are kept below neutral. As kobs values for self-decay of HO2• and O2•- are variant to pH levels in aqueous solutions,2 pH levels for the reactions of HO2• and O2•- need to be adjusted accurately to the pH ranges of 2-9.5 using various buffers as well as H2SO4 and NaOH. The fluorescence intensity of OHBA was decreased with increasing concentration of buffer. As shown in Figure 2, the fluorescence intensities of OHBA were almost invariable in lower concentration ranges of borate (pH 8.75) and acetate (pH 4.9) buffers. But they were rapidly decreased in the buffer concentration ranges of 0.6-1.5% (v/v) and were gradually decreased thereafter. Similar results were observed for phosphate (pH 6-8) and phosphoric buffers (pH 2-3), but their intensities were generally about two times higher than those of acetate and borate buffers. These results can be attributed to the buffering ions as scavengers of OH• in accordance with their known rates of reaction with OH• from 104 to 107 M-1 s-1.35 Thus, the volume percentage of buffer was adjusted below 0.6% to eliminate the scavenging effects of the OH radicals by buffer ions. The dependence of the fluorescence intensity on reaction time for reactions of (R7-R9) is investigated in [H2O2] ) 4 mM, [BA] ) 1 mM, [Fe3+-EDTA] ) 20 µM, and pH 6.11 as shown in Figure 3. The reaction time was controlled by varying length of KTR (D2) at a constant solution flow rate. As expected from kinetic considerations (above R7-R9), the fluorescence intensity increased rapidly with increasing reaction time up to 80 s and became constant thereafter. Thus, all subsequent experiments were performed with 2.5-min () 150 s) reaction time. 6362 Analytical Chemistry, Vol. 76, No. 21, November 1, 2004
Figure 3. Dependence of the fluorescence signal on the reaction time: pH 6.11, [H2O2] ) 4 mM, [BA] ) 1 mM, [Fe3+-EDTA] ) 20 µM, and [NaOH] ) 0.05 N.
Figure 4. Dependence of the fluorescence signal on Fe3+-EDTA concentration: pH 6.11, [H2O2] ) 4 mM, [BA] ) 1 mM, [NaOH] ) 0.05 N, and 2.5-min reaction time.
It is desirable to keep a minimum Fe3+-EDTA concentration in order to determine the optimum reaction conditions. The dependence of the fluorescence intensity on Fe3+-EDTA indicates that the optimum Fe3+-EDTA concentration lies between 20 and 100 µM as shown in Figure 4. Considering the kinetic data and the presence of 1 mM BA, it is estimated that ∼0.5% of the OH radicals reacts with 20 µM Fe3+-EDTA and ∼2.5% of the OH radicals also reacts with 100 µM Fe3+-EDTA. In this study, 20 µM Fe3+-EDTA was used, while Baker et al.25 used 30 µM Fe3+-EDTA to produce the OH radicals in their study. Finally, conditions for optimum detection of HO2•/O2•- were determined. Based on the reaction scheme for the analysis of HO2•/O2•- and the empirically observed fluorescence signal, all subsequent experiments were performed with 1 mM BA, 20 µM Fe3+-EDTA, 4 mM H2O2, 0.05 N NaOH, and 2.5-min reaction time. In addition, the concentrations of HO2•/O2•- were widely investigated in the pH ranges of 2-9.5. Kinetic Method: Calibration and Sensitivity. The basic principle for calibration is based on the half-life measurement of HO2•/O2•- decay. The rate of second-order reaction mainly given by the reactions R1-R3 is
-
d[HO2•/O2•-] ) kobs[HO2•/O2•-]2 dt
(II)
The solution of eq II is
kobst )
[HO2•/O2•-]o - [HO2•/O2•-]t [HO2•/O2•-]o[HO2•/O2•-]t
(III)
Since [HO2•/O2•-]t1/2 is equal to [HO2•/O2•-]o/2 at the half-life (t1/2), eq III becomes
[HO2•/O2•-]o ) 1/kobst1/2
(IV)
In the case of the second-order reaction, the initial concentration of HO2•/O2•- is inversely proportional to the half-life. Thus, the initial concentration of HO2•/O2•- can readily be determined from the half-life of HO2•/O2•- decay in the aqueous solution with calculated kobs at the given pH. The half-life of HO2•/O2•- decay is based on each length of KTR (C) at a constant flow rate of the HO2•/O2•- stream. Since the decay of HO2•/O2•- gives a kobs (eq I), the concentrations of HO2•/O2•- can be expected to decrease with increasing length of KTR, i.e., the reaction time, which are stepwise varied as 0, 1, 2, 3, and 4 m. The signal ratio (SR) is defined as (Ao - An)/(AoAn), where Ao is signal height of fluorescent OHBA at KTR 0 m and An is signal heights at KTR 1, 2, 3, and 4 m, respectively. By assuming a linear proportionality between concentration of HO2•/O2•- and the corresponding fluorescence signal, a plot of the SR versus length of KTR (D or reaction time) at given pH gives a straight line, as shown in Figure 5. From the linear relationship as expected we can derive as following equations:
Figure 5. Signal ratio versus length of KTR with straight line: [H2O2] ) 4 mM, [BA] ) 1 mM, [Fe3+-EDTA] ) 20 µM, [NaOH] ) 0.05 N, and 2.5-min reaction time.
SR ) slope × D + intercept
Figure 6. Linear plots for fluorescence intensity and concentrations of HO2•/O2•-: pH 6.12, [H2O2] ) 4 mM, [BA] ) 1 mM, [Fe3+-EDTA] ) 20 µM, [NaOH] ) 0.05 N, and 2.5-min reaction time.
(V)
The SRt1/2 becomes identical with 1/Ao at half-life and then we can derive the following equation:
SRt1/2 ) slope × Dt1/2 + intercept
(V′)
where SRt1/2 is the signal ratio at t1/2 and Dt1/2 is the length of the KTR (m) at t1/2. Since the constant flow rate for KTR is 42 s/m,
t1/2(s) ) Dt1/2(m) × 42(s/m)
(VI)
Consequently, a given concentration of HO2•/O2•- was kinetically calculated from eq IV, based on the half-life (t1/2) and calculated kobs at a given pH. The limit of detection of HO2•/O2•- is determined as 3.22 × 10-11 M with signal-to-noise ratio of 2 at pH 5. The relative standard deviation for five replicate measurements is 5.3% at 1.02 × 10-9 M HO2•/O2•-. In addition, the fluorescence intensity using various H2O2 photolysis time is linear for HO2•/O2•- concentrations between 4.90 × 10-10 and 1.16 × 10-7 M at pH 6.12 as shown in Figure 6. SRt1/2, Dt1/2, half-life (t1/2), kobs, concentration of HO2•/O2•-, and limit of detection for a given pH are listed in Table 1. The HO2•/O2•- concentration and the limit of detection were the lowest
at pH 5.0, with increases gradually on either side of this value (see Supporting Information for plots). These shapes were due to the kobs values on the disproportionation reactions of (R1-R3) and the pH dependences of the equilibrium between HO2• and O2•-. Three possible cases depending upon pH values are distinguished. First, the rate for reaction R1 with k1 ) (8.3 ( 0.7) × 105 M-1 s-1 between two HO2• radicals (HO2• + HO2•) is found most significant at lower pH range, i.e., pH 7), where HO2• is negligible and its decay rate is very slow. Hence, there is relatively low loss of O2•- in these ranges and its concentration is relatively high. Third, at near pKa () 4.8) value, the rate for reaction R2 with k2 ) (9.76 ( 0.6) × 107 M-1 s-1 between HO2• and O2•- is dominant and fast. Hence, the initial concentration of coexisting HO2• and O2•- is much lower than both acidic and basic pH ranges. These results are consistent with the kinetic data by Bielski et al.2 Possible Interferences. Interference studies were performed for the various compounds because they have often been found in wastewater, drinking water, and the atmosphere. The interferAnalytical Chemistry, Vol. 76, No. 21, November 1, 2004
6363
Table 1. Summary of the Results for the New Kinetic Method in the Determination of HO2•/O2•-a run no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
pH 9.26 9.20 9.10 8.92 8.67 8.5 8.12 7.99 7.51 6.95 6.11 5.00 4.15 3.25 2.12
SRt1/2 4.19 4.44 4.48 4.83 6.49 8.20 5.76 5.85 8.77 6.29 4.63 8.93 5.814 11.79 11.76
Dt1/2 (m) 4.26 2.99 3.08 2.82 1.84 2.19 0.99 1.31 1.01 0.78 1.07 1.02 0.64 1.13 1.04
t1/2 (s)
kobs (M-1s-1)b
178.90 125.66 129.33 118.44 77.30 92.07 41.53 55.10 42.34 32.69 44.96 42.67 26.80 47.66 43.58
3.33 × 3.82 × 103 4.82 × 103 7.29 × 103 1.30 × 104 1.92 × 104 4.59 × 104 6.20 × 104 1.87 × 105 6.71 × 105 4.28 × 106 2.31 × 107 1.51 × 107 3.39 × 106 1.03 × 106 103
[HO2•/O2•]o (M) 10-6
1.68 × 2.08 × 10-6 1.61 × 10-6 1.16 × 10-6 9.98 × 10-7 5.67 × 10-7 5.24 × 10-7 2.93 × 10-7 1.27 × 10-7 4.56 × 10-8 5.19 × 10-9 1.02 × 10-9 2.46 × 10-9 6.18 × 10-9 2.23 × 10-8
LOD (M) 1.41 × 10-8 1.85 × 10-8 1.44 × 10-8 1.12 × 10-8 1.30 × 10-8 9.29 × 10-9 6.04 × 10-9 3.43 × 10-9 2.22 × 10-9 5.74 × 10-10 9.53 × 10-11 3.22 × 10-11 3.79 × 10-11 8.03 × 10-11 5.37E × 10-10
a Conditions: [H O ] ) 4 mM, [BA] ) 1 mM, [Fe3+-EDTA] ) 20 µM, [NaOH] ) 0.05 N, and 2.5-min Reaction Time 2 2 k2(KHO2/[H+])}/(1 + KHO2/[H+])2 (ref 2).
ence studies were considered in two aspects: interferences by increasing signal intensity due to impurities and by decreasing signal intensity from the competitive reactions of impurities for OH radicals. For the first aspect of interference studies, the possible interferences were evaluated by adding various amounts of each compound through P1 instead of H2O2 to solutions of 20 µM Fe3+-EDTA, 1 mM BA, and 0.05 N NaOH. Threshold concentration of nitrate was 1 mM (see Supporting Information for plots). The irradiation of these oxide ions in the fluorescence detector with excitation around 320 nm may often result in a primary photochemical process. For example, the possible photolysis of nitrate in its long-wavelength absorption band (maximum 302 nm) produced the OH radical in water:36,37
NO3- + hv f NO2- + O•-
(R10)
O•- + H2O f OH• + OH-
(R11)
The OH radical produces OHBA in the presence of BA from reaction R9. However, methanol, acetone, ethanol, acetate, formaldehyde, oxalate, nitrite, bromate, bromide, phosphate, and chloride did not show any positive interferences. For the second aspect of interference studies, all compounds mentioned above may compete with benzoic acid for the OH radicals formed in the Fenton-like reaction. Threshold concentrations of possible interfering compounds are estimated by using kinetic data and listed (see Supporting Information). When the concentrations of the possible interfering compounds are higher than threshold concentrations relative to 1 mM BA, the resulting fluorescence intensity of OHBA can be decreased.
6364
Analytical Chemistry, Vol. 76, No. 21, November 1, 2004
b
kobs ) {k1 +
CONCLUSIONS We have developed and demonstrated the use of a new kinetic method as an analytical tool for the measurement of HO2•/O2•-. Since this method for HO2•/O2•- shows high sensitivity and allows a simple calibration system, it can contribute significantly to understanding the basic functions of HO2•/O2•- in advanced oxidation processes. Moreover, the present technique has the advantage of using inexpensive and easily available nonenzymatic reagents and of being insensitive to the moderate concentration of possible interferences often found in aqueous phase. ACKNOWLEDGMENT This work was supported by Grant 1999-2-309-007-3 from the Basic Research Program of the Korea Science & Engineering Foundation and in part by the Korea Science and Engineering Foundation (KOSEF) through the Advanced Environmental Monitoring Research Center (ADEMRC) at Gwangju Institute of Science and Technology and the Brain Korea 21 Project, Ministry of Education and Human Resources Development (MOE) Korea. SUPPORTING INFORMATION AVAILABLE The pH dependences of the HO2•/O2•- concentration, the limit of detection, and kobs values, interferences by increasing signal intensity due to impurities, and threshold concentrations of possible interfering compounds estimated by using kinetic data. This material is available free of charge via the Internet at http://pubs.acs.org.
Received for review April 25, 2004. Accepted August 23, 2004. AC0493828
