Pulse Laser Photolysis of Aqueous Ozone in the Microsecond Range

Apr 5, 2013 - Department of Chemistry, School of Science and Technology, Kwansei Gakuin University, Sanda, Hyogo 669-1337, Japan. § Department of Che...
0 downloads 8 Views 475KB Size
Download PDF
Article pubs.acs.org/ac

Pulse Laser Photolysis of Aqueous Ozone in the Microsecond Range Studied by Time-Resolved Far-Ultraviolet Absorption Spectroscopy Takeyoshi Goto,† Yusuke Morisawa,‡,§ Noboru Higashi,∥ Akifumi Ikehata,† and Yukihiro Ozaki*,‡ †

National Food Research Institute, National Agriculture and Food Research Organization (NARO), Tsukuba, Ibaraki 305-8642, Japan Department of Chemistry, School of Science and Technology, Kwansei Gakuin University, Sanda, Hyogo 669-1337, Japan § Department of Chemistry, School of Science and Engineering, Kinki University, Higashiosaka, Osaka 577-8502, Japan ∥ Kurabo Industries Ltd., Shimokida-cho, Neyagawa, Osaka 572-0823, Japan ‡

ABSTRACT: Chemical dynamics of an ozone (O3) pulsephotolytic reaction in aqueous solutions were studied with pump−probe transient far-ultraviolet (FUV) absorption spectroscopy. With a nanosecond pulse laser of 266 nm as pump light, transient spectra of O3 aqueous solutions (78−480 μM, pH 2.5−11.3) were acquired in the time range from −50 to 50 μs in the wavelength region from 190 to 225 nm. The measured transient spectra were linearly decomposed into the molar absorption coefficients and the concentration−time profiles of constituted chemical components with a multivariate curve resolution method. From the dependences of the time-averaged concentrations for 20 μs of the constituted chemicals on the initial concentration of O3, it was found that the transient spectra involve the decomposition of O3 and the formation of hydrogen peroxide (H2O2) and a third component that is assigned to hydroxyl radical (OH) or perhydroxyl radical (HO2). Furthermore, the pH dependence of the time-averaged concentration of the third components indicates that HO2 is more probable than OH as the third component. The time-averaged concentration ratio of each chemical component to the initial O3 concentration depends on the pH conditions from −0.95 to −0.60 for O3, 0.98 to 1.2 for H2O2, 0.002 to 0.29 for OH, and 0.012 to 0.069 for HO2.

A

reactivities of those radical scavengers are chemically specific, and O3 itself also reacts with the scavengers. Therefore, the experimental developments are still required to directly determine the chemical dynamics of the transient species in the O3 photolytic reaction. UV absorption spectroscopy is simple and practical to quantitatively and qualitatively analyze aqueous solutions. Liquid water is optically transparent above 200 nm, while most of the organic and inorganic molecules have relatively strong optical absorptions corresponding to various types of electronic transitions including σ, n, and π orbitals, and charge transfer (CT).21−23 The chemical species involved in the O3 photolytic reaction also show characteristic UV absorption spectra. The molar absorption coefficients (ε) of O3 and H2O2 were experimentally determined with stationary absorption spectroscopy, and those of the transient species, OH, ozonide radical ion (O3−), perhydroxyl radical (HO2), perhydroxyl ion (HO2−), superoxide ion (O2−), and hydrogen trioxide radical (HO3), were determined with pulse-radiolysis transient absorption spectroscopy.24−30

dvanced oxidation processes (AOP) using ozone (O3) with hydrogen peroxide (H2O2) and ultraviolet (UV) light irradiation have been widely used to remove organic and inorganic contaminants from aqueous solutions by oxidative destruction without any chemical residues.1−3 The studies of AOP have focused on the experimental conditions to efficiently generate hydroxyl radicals (OH) at the target sites, which are short-lived species of O3 decomposition process. This is because OH possesses high reactivity to any target compounds, while the reactivity of O3 is quite selective. Specifically, the optimum conditions have been investigated with the numerical simulations4−6 and the measurements of the decomposition rates of O3 and dissolved organic compounds under the various experimental factors: the types of the dissolved chemical compounds,7,8 pH and temperature,9 and the irradiation of UV light.10−14 The real-time monitoring of the chemical dynamics of the O3 photolytic reaction in aqueous solutions is not straightforward, because the transient species are short-lived (nanoseconds to milliseconds) and their concentrations are relatively low. The quantitative analyses of the transient species especially for OH have been carried out by monitoring the molecular states and the concentrations of radical scavengers, such as p-nitrosodimethylaniline,15,16 dimethyl sulfoxide (DMSO),17,18 and 5,5-dimethyl-1-pyrroline-N-oxide (DMPO).19,20 However, the © 2013 American Chemical Society

Received: January 8, 2013 Accepted: April 5, 2013 Published: April 5, 2013 4500

dx.doi.org/10.1021/ac400056m | Anal. Chem. 2013, 85, 4500−4506

Analytical Chemistry

Article

The pump light is a fourth-harmonic-generation nanosecond pulse Nd:YAG laser (266 nm, Pro-350; Spectra Physics). The pulse repetition rate of the pump light is 10 Hz, and its power at the sample position is about 50 mJ/pulse. The probe and pump lights overlap at the sample position with a perpendicular geometry. The monochromator is set after the sample position to remove scattered light from the pump light. A probe light signal is taken in 1 ns interval in the range 50 μs before and after the pump laser irradiation by a digital oscilloscope (Wave Pro 715Zi; LeCroy) and is averaged 1000 times to improve the signal-to-noise ratio (SNR). The timing of collecting the probe light signal is synchronized with the pump pulse using a delay generator (DG 645, Stanford Research Systems). The full width at half-maximum of the signal response time from a pulse is about 40 ns. Sample Preparation and Spectroscopic Measurements. The sample solutions were mixtures of O3 aqueous solutions (78−480 μM) from a silent discharge O3 generator (RC2W-2-60M-Z, Roki Co, Ltd., Japan) using O2 gas (purity 99.9999%) and phosphoric buffer solutions (pH 2.5, 5.2, 7.3, 9.0, and 11.3, 10 mM). Those were cooled with a chiller to stabilize the solubility of O3 in aqueous solutions and continuously flowed (1 L/min) during the spectroscopic measurements. The pH values and temperature of the sample solutions were monitored with a pH meter (D-51, Horiba, Ltd., Japan). The sample volume in the optical cell made from fused silica is 5 × 1 × 30 mm3, and the optical path length is 5 mm. The transient absorption spectra of the sample solutions were measured in the range from 190 to 225 nm at 18.0 ± 1.0 °C. Transient absorbance is defined as −log(I/I0), where I0 is the time-averaged signal of the probe light intensity for 48 μs before the pump light irradiation and I is the signal of the probe light intensity in 1 ns interval after the pump light irradiation. Data Analysis. From the measured transient absorption spectra, the molar absorption coefficients and the concentration−time profiles of the constituted chemical components were analyzed by a MCR method using an alternating leastsquares (ALS) fitting.40,41 The MCR method for a spectroscopic analysis is based on a bilinear model of the Beer− Lambert law as given in eq 1:

We have developed the basic and applied studies of analyzing aqueous solutions with far-UV (FUV) absorption spectroscopy.22 By extending measurement wavelength to 145 nm with a very thin sample cell32 or an attenuated total reflection (ATR) method,33 the absorptions of solutes as well as solvent water can be employed for the analyses of the aqueous solutions. Because FUV absorption of liquid water depends on its hydrogen-bonding state, FUV spectra of aqueous solutions contain copious information about the aqueous solutions.23,34−37 The applied examples are monitoring the quality of the cleaning solutions for semiconductor wafer32 and quantitative analyses of ions in spring water.38 Recently, we have developed a pump−probe nanosecond time-resolved FUV spectrophotometer for the analyses of chemical dynamics of the transient species, and the photochemistry of phenol and tryptophan in aqueous solutions was analyzed from the transient FUV absorption spectra.39 In the present study, the transient FUV absorption spectra of O3 in aqueous solutions (78−480 μM, pH 2.5−11.3) were measured with a pump−probe spectrophotometer to directly determine the chemical dynamics of the O3 pulse-photolytic reaction. The measured transient spectra were decomposed into the molar absorptivities and the concentration−time profiles of each probable chemical component with a multivariate curve resolution (MCR) method. The calculated concentration−time profiles directly reveal that a third component, as which OH or HO2 is the most probable, is prevailing in the order of micromolar in the microsecond time scale, in addition to O3 and H2O2. The real-time monitoring of the concentrations of O3, H2O2, and the third component is a basis for controlling the AOP quality in washing lines.

■

EXPERIMENTAL SECTION Instrumentation. Figure 1 shows a schematic diagram of the experimental setup. The details of a pump−probe

A = CST + R

(1)

Figure 1. Block diagram of the experimental setup measuring pump− probe FUV absorption spectra of aqueous O3 solutions.

where A is the transient absorbance matrix (time × wavelength channels), C is the concentration−time profile matrix (time × number of the components (n) channels), S is the molar absorption coefficient matrix (wavelength × n channels), and R is the residual matrix that is not explained by the bilinear model. With the initial estimates of the S matrix from the references of the molar absorption coefficients of the constituted chemical components including O3, the ALS calculations determined the optimized C and S matrices with a non-negativity constraint for S matrix by using MCR-ALS toolbox for MATLAB software (2010b, Mathworks, Inc.).41

nanosecond time-resolved FUV absorption spectrophotometer were previously reported.39 Briefly, the spectrophotometer is composed of a laser-driven Xe lamp (EQ-99, Energetiq Technology) as a probe light source, a monochromator (2500 gr/mm, blazed at 150 nm, KV-200, Bunkoh-Keiki Co. Ltd., Japan), and a photomultiplier tube (PMT) equipped with a fused silica plate coated with a sodium salicylate film. The probe light path is continuously purged with N2 gas (8 L/min).

RESULTS AND DISCUSSION Figure 2 shows the time profiles of the transient absorbances of O3 in phosphoric buffer solutions at (a) pH 2.5, (b) 7.3, (c) 9.0, and (d) 11.3. Because that of O3 in pure water (pH 4.6−4.9) is similar to that in the pH 5.2 buffer solutions, the effects of the phosphate anions on the O3 pulse-photolytic reaction are limited. The time at 0 s corresponds to the laser irradiation timing. For all of the pH conditions, the transient absorbances decrease just after the pulse laser irradiation until 20 μs and

■

4501

dx.doi.org/10.1021/ac400056m | Anal. Chem. 2013, 85, 4500−4506

Analytical Chemistry

Article

Figure 2. Time profiles of the transient absorbance of O3 in aqueous solutions at (a) pH 2.5, (b) 7.3, (c) 9.0, and (d) 11.3. The nanosecond laser pulses are irradiated at 0 s. (e) Expanded view of (a) around 0 s.

Figure 3. Time-averaged absorbance of O3 in aqueous solutions at (a) pH 2.5, (b) 7.3, (c) 9.0, and (d) 11.3 for 20−40 μs.

become mostly stable between 20 and 50 μs. The decreases in the absorbances become larger and the time decay shorter, as

the wavelength is longer. For pH 2.5, the absorbances at 20 μs are −0.0143 and −0.0953 for 190 and 225 nm, respectively. 4502

dx.doi.org/10.1021/ac400056m | Anal. Chem. 2013, 85, 4500−4506

Analytical Chemistry

Article

The time decay constants with single exponential (τ1) are 3.8 and 1.0 μs for 190 and 225 nm, respectively. For a closer look at around 0 s (Figure 2e), the slight increases of the absorbances by 0.008 for 190−195 nm were observed at 0.21 μs. As pH increases, the decreases of the absorbances become small especially in the short wavelengths. The absorbances of 190 nm at 20 μs are −0.007 and 7 × 10−4 for pH 7.3 and 9.0, respectively. For pH 11.3, the SNR of the absorbance−time profile is smaller than the other conditions, and the transient absorbances around 0 s highly fluctuate between 0.01 and −0.03 until 2 μs. These indicate (1) the lower O3 concentration at the measurement position due to the self-decomposition reaction of O3 with OH− after mixing the O3 solution with the alkaline buffer solution and (2) the increase in the scattered light of the pump laser due to the O2 bubbles. Because the absorbances after 20 μs are relatively stable for all of the pH conditions, the time-averaged absorbances between 20 and 40 μs are plotted versus the wavelength to distinguish the spectral patterns with pH. Figure 3 shows timeaveraged absorption spectra of O3 in aqueous solutions for (a) pH 2.5, (b) 7.3, (c) 9.0, and (d) 11.3. For all of the pH conditions, the time-averaged absorbances decrease in the longer wavelength region and linearly respond to the initial concentrations of O3. The absorbances around 192.5 nm increase from −0.079 for pH 2.5 ([O3]ini = 137 μM) through −0.044 for pH 7.3 ([O3]ini = 154 μM) to 0.002 for pH 11.3 ([O3]ini = 162 μM), while the absorbances above 205 nm do not significantly change with the pH conditions. As a result, the shoulder bands appear around 205 nm for pH 7.3 and 9.0, and there are the isosbestic points at 192.5 and 201.5 nm for pH 9.0 and 11.3, respectively. Those indicate that more than two chemical components are involved in the observed spectral patterns. From the studies of stationary O3 photolysis, the decreases of the absorbances in the longer wavelength side are mainly ascribed to the photodecomposition of O3, and the increases in the shorter wavelength side are ascribed to the formation of H2O2.10−14,42 To reveal the chemical dynamics of the O3 pulse-photolytic reaction, the measured transient spectrum matrix (A, time × wavelength channels) was decomposed into the molar absorption coefficient matrix (S, wavelength × n channels) and the concentration−time profile matrix (C, time × n channels) by assuming that the number of the chemical components (n) is three. As the initial ε values of S matrices for the fitting, the ε values of the chemical components involved in the O3 photolytic reaction in an aqueous solution (O3, H2O2, OH, O3−, HO2, O2−, and HO2−) were taken from the references.26−30,43−45 Figure 4 shows the ε spectra of the constituted chemical components in the measured wavelength region. Those were experimentally determined with the transient absorption spectra from the pulse-radiolysis of aqueous solutions. The ε values of some chemical species, such as HO3, in the present measured wavelength could not be found from any references. The number of the components was determined as three, because the elements of the residual absorbance matrix (R) by considering three components are completely random patterns along the time and wavelength directions and their absolute values are smaller than 0.002. This result indicates that the contribution from other than the O3 decomposition and the H2O2 formation can be described as one component from the linear decomposition of the measured spectra. To distinguish which of the transient species (OH, O3−, HO2, O2−, and HO2−) is the most probable as the third

Figure 4. Molar absorption coefficients (ε) of the chemical species involved in the photolysis reactions of O3 in aqueous solution in the measured wavelength region from the references.

component, the MCR calculations were carried out in a way O3 and H2O2 were always selected as the first and second components, and one of the transient species was selected as the third component. The S and C matrices then were calculated using each third component. To determine the most probable chemical species as the third component, the following conditions were applied to the calculated C matrices: (1) the concentration changes of H2O2 and the third component are not negative along the time, and (2) the concentration change of the third component increases as the initial concentration of O3 is higher. From the calculated C matrices, the probable third species that satisfy the above conditions are OH and HO2. The concentration change of O3− becomes negative upon the pulse laser irradiation, and that of H2O2 becomes negative for HO2−. For O2−, the concentration change becomes smaller as the initial concentration of O3 is higher. The chemical dynamics of the O3 pulse-photolytic reaction in aqueous solutions are directly revealed from the decomposed C and S matrices. Figure 5 shows the C and S matrices calculated from the time profiles of the transient absorbances at the condition ([O3]ini = 449 μM, pH 5.2) in the cases of OH ((a) and (c)) and HO2 ((b) and (d)) as the third component. The concentration−time profiles of the three chemical components are similar between the OH and HO2 cases. Upon the pulse laser irradiation, the O 3 concentration instantaneously decreases by 53% (from 449 to 240 μM) for the first 0.3 μs and by 80% (to 89 μM) until 20 μs, and then it becomes almost stable after 20 μs. The H2O2 concentration instantaneously increases to 550 μM for the first 0.3 μs and slightly increases further to 610 μM around 2 μs, and then it decreases to 460 μM around 20 μs and becomes stable after 20 μs. The OH concentration increases to about 75 μM until 5 μs and gradually decreases to 20 μM until 20 μs, and then it remains stable after 20 μs. The HO2 concentration increases to 26 μM until 2 μs and gradually decreases to 22 μM until 40 μs. The SNR of the concentration−time profiles for the case of OH is smaller than that for the case of HO2. For the S matrices, the calculated ε spectra of O3 and H2O2 agree with the references for both the OH and the HO2 cases. For the third components, the calculated spectrum of OH agrees with the reference spectrum, while that of HO2 fluctuates around the reference spectrum. Therefore, the differences between the measured and calculated 4503

dx.doi.org/10.1021/ac400056m | Anal. Chem. 2013, 85, 4500−4506

Analytical Chemistry

Article

Figure 5. Calculated concentration−time profiles and molar absorption coefficients (ε) of the constituted chemical components of O3 pulsephotolytic reaction from the transient FUV absorption spectra in the cases of OH ((a) and (c)) and HO2 ((b) and (d)) as the third component and the ε values from the references. The concentrations of OH and HO2 were magnified by 3 and 5 times for clarity.

absorbances appear mainly in the C matrix for the case of OH and in the S matrix for the case of HO2. Because the calculated S matrices mostly agree with the ε values of O3, H2O2, OH, and HO2 from the references, that of the third component (OH or HO2) in the whole measured wavelength region (190−225 nm) can be determined from the following compromise solution of S3: A3 = A exp − S1C1 − S2C2

(2)

S3 = (C3 TC3)−1C3TA 3

(3)

Aexp is the measured transient absorbance matrix, Cn is nth row of the calculated C matrix in the previous paragraph, and the columns of S1 and S2 are extended to 190 nm with the ε values of O3 and H2O2 from the references. As the transient absorbance matrix A (time × wavelength channels) is not a square matrix, the calculated ε matrix S3 is a compromise solution. Figure 6a shows the calculated ε spectrum of each third component in the region from 190 to 225 nm. The increases in the ε values appear below 205 nm for both the cases of OH and HO2. For the case of HO2, the ε value below 205 nm increases as the initial concentration of O3 is higher. That may reflect the scattered probe light primarily due to O2 bubbles in the sample solution. For the case of OH, the increase of the ε value below 205 nm is mostly identical for all of the initial concentration of O3. Figure 6b shows the OH spectra calculated in this study and reported from the pulse-radiolysis transient absorption spectroscopy.26,27 The OH spectrum reported by Pageberg et al. indicates the presence of an absorption band below 210 nm. However, the theoretical calculation of electronic transition of hydrated OH with a coupled cluster method shows that the maximum oscillator strength is located around 230 nm, which is attributed to the charge transfer transition of hemihydrogenbonded OH to the surrounding water molecules.46 Therefore, the increase of the calculated ε values of OH below 205 nm

Figure 6. (a) Calculated molar absorption coefficients of OH and HO2 extended to the shorter wavelength to 190 nm using the calculated C matrices and the reference ε values of the corresponding components. (b) Comparisons of the ε values of OH calculated in the present experiment with the reported values.

4504

dx.doi.org/10.1021/ac400056m | Anal. Chem. 2013, 85, 4500−4506

Analytical Chemistry

Article

inflection points around the pKa values are observed in the stationary concentrations of nearly all of the transient compounds that is reported from the numerical simulation of the O3 decomposition kinetics employing the various proton concentrations.5 For both O3 (Figure 7a) and H2O2 (Figure 7b), the tendencies of the changes with pH calculated by employing OH or HO2 as the third component are nearly identical. The proportionality factor for O3 (Δ[O3]/Δ[O3]ini) is −0.95 at pH 2.5, which indicates that the decomposed O3 concentration (Δ[O3]) linearly depends on the increase in the initial concentration of O3 (Δ[O3]ini). The proportionality factor of O3 becomes smaller from −0.95 to −0.60 as pH increases from 2.5 to 11.3 with two inflection points around pH 5 and 9. That of H2O2 (Δ[H2O2]/Δ[O3]ini) fluctuates around 1.2 along pH with two inflection points. That of OH (Δ[OH]/ Δ[O3]ini) decreases from 0.29 at pH 2.5 to 0.002 around pH 5.2−9.0 and increases to 0.31 at pH 11.3. That of HO2 (Δ[HO2]/Δ[O3]ini) decreases from 0.069 at pH 2.5 to 0.012 at pH 11.3 without any apparent inflection points. The pH dependence of the proportionality factor of each third component in Figure 7c and the rate constants of the elementary reactions indicate that HO2 is more probable than OH as the third component. The proportionality factor of OH shows that the OH formation becomes unfavorable as the pH value increases from 2.5 to 9.0, which does not correspond to the kinetic studies of self-decomposition and UV decomposition of O3.5,13 The HO2 formation also becomes unfavorable as the pH value increases, which is accordance with the pKa value of HO2 (pKa = 4.8). Also, the lifetime of HO2 is much longer than that of OH in the O3 photolytic reaction. The reaction rate constants between the third component and O3 are 9.0 × 105 M−1 s−1 for OH and lower than 104 M−1 s−1 for HO2.31,48 The reaction rate constants of the self-radical recombination are 5.5 × 109 and 8 × 105 M−1 s−1 for OH and HO2, respectively.47

might also be derived from the scattering of the probe light or the absorption of the other transient components. The pH dependence (pH 2.5−11.3) of the concentration change ratio of each chemical component to the initial O3 concentration (78−480 μM) was examined from the timeaveraged concentration between 20 and 40 μs of the calculated C matrices. The time-averaged values were employed, because (1) the signal fluctuations due to the scattered light of the pump laser around the 0−5 μs range could not be completely removed from the measured transient spectra and (2) the rapid changes of the concentration−time profile are mostly finished before 20 μs for all of the conditions. First, the linear dependences of the time-averaged concentration changes between 20 and 40 μs of O3, H2O2, OH, and HO2 on the initial concentration of O3 were calculated for each condition, and then the proportionality factors (Δ[O3]/Δ[O3]ini, Δ[H2O2]/Δ[O3]ini, Δ[OH]/Δ[O3]ini, and Δ[HO2]/Δ[O3]ini) were plotted versus pH in Figure 7a−c. There are two cases for Δ[O3]/Δ[O3]ini and Δ[H2O2]/Δ[O3]ini in which OH or HO2 is employed as the third component.

■

CONCLUSION In the present work, the chemical dynamics of a pulsephotolytic reaction of O3 aqueous solutions (78−480 μM, pH 2.5−11.3) have been directly revealed with pump−probe transient FUV absorption spectroscopy. The transient absorption spectra in microsecond time scale were decomposed into the molar absorption coefficients and the concentration− time profiles of the constituted chemical components via the MCR method. From the investigation of the residual absorbances with the bilinear decomposition of the measured spectra, three chemical components are sufficient to account for the measured transient absorption spectra. The first and second components are definitely assigned to O3 and H2O2. From the calculated concentration−time profiles, the instantaneous decomposition of O3 and formation of H2O2 upon the laser irradiation occur below 2 μs, and the concentration changes of the two components after 20 μs are relatively stable until 50 μs. The probable third component is OH or HO2, which is determined from the dependence of the concentration of the third component on the initial concentration of O3. The instantaneous increase of the concentration of OH or HO2 also occurs below 2 μs. The pH dependence of the proportionality factor of each third component and the rate constants of the elementary reactions indicate that HO2 is more probable than OH as the third component of the O3 pulse-photolytic reaction. The time-averaged concentration ratio of each chemical component to the initial O3 concentration depends on the

Figure 7. pH dependence of the concentration change ratios of OH and HO2 to the O3 initial concentrations (Δ[HO2]/Δ[O3]ini and Δ[OH]/Δ[O3]ini). The error bars are derived from the linear fitting of Δ[HO2] or Δ[OH] versus Δ[O3]ini.

The concentration changes of the constituted chemical components upon the O3 pulse-photolysis in the microsecond time scale depend on the pH conditions as shown in Figure 7. Although the tendencies of the changes with pH are different among the chemical components, there are two inflection points around pH 5 and 9 for O3, H2O2, and OH. Those correspond to the protonation states of HO2 and HO3 (pKa = 4.8 and 8.2) in the elementary reactions of O3 photolysis.41 Also, the protonation of HO3 might be related to the exceptionally large error bars observed for pH 9.0. The 4505

dx.doi.org/10.1021/ac400056m | Anal. Chem. 2013, 85, 4500−4506

Analytical Chemistry

Article

pH conditions from −0.95 to −0.60 for O3, 0.98 to 1.2 for H2O2, 0.002 to 0.29 for OH, and 0.012 to 0.069 for HO2. The examination of the chemical dynamics around 0 s was limited at this moment, because the signal fluctuations due to the scattered light of the pump light could not be completely removed. Certainly, the signals just after the pulse laser irradiation are crucial, and the improvement of the experimental setup is currently in progress to reveal the chemical dynamics of O3 pulse-photolytic reaction in further. Finally, a compact FUV spectrophotometer using optical filters is currently being developed for real-time monitoring of the concentrations of O3, H2O2, and the third component.

■

(22) Ozaki, Y.; Morisawa, Y.; Ikehata, A.; Higashi, N. Appl. Spectrosc. 2012, 1, 1−25. (23) Ikehata, A.; Higashi, N.; Ozaki, Y. J. Chem. Phys. 2008, 129, 234510. (24) Bahnemann, D.; Hart, E. J. J. Phys. Chem. 1982, 86, 252−255. (25) Sehested, K.; Holcman, J.; Bjergbakke, E.; Hart, E. J. J. Phys. Chem. 1982, 86, 2066−2069. (26) Czapski, G.; Bielski, B. H. J. Radiat. Phys. Chem. 1993, 41, 503− 505. (27) Pagsberg, P.; Christensen, H.; Rabani, J.; Nilsson, G.; Fenger, J.; Nielsen, S. O. J. Phys. Chem. 1969, 73, 1029−1038. (28) Nielsen, S. O.; Michael, B. D.; Hart, E. J. J. Phys. Chem. 1976, 80, 2482−2488. (29) Sehested, K.; Holcman, J.; Bjergbakke, E.; Hart, E. J. J. Phys. Chem. 1982, 86, 2066−2069. (30) Buehler, R. E.; Staehelin, J.; Hoigne, J. J. Phys. Chem. 1984, 88, 2560−2564. (31) Sehested, K.; Holcman, J.; Bjergbakke, E.; Hart, E. J. J. Phys. Chem. 1984, 88, 4144−4147. (32) Higashi, N.; Ikehata, A.; Kariyama, N.; Ozaki, Y. Appl. Spectrosc. 2008, 62, 1022−1027. (33) Higashi, N.; Ikehata, A.; Ozaki, Y. Rev. Sci. Instrum. 2007, 78, 103107. (34) Higashi, N.; Yokota, H.; Hiraki, S.; Ozaki, Y. Anal. Chem. 2005, 77, 2272−2277. (35) Goto, T.; Ikehata, A.; Morisawa, Y.; Higashi, N.; Ozaki, Y. Phys. Chem. Chem. Phys. 2012, 14, 8097−8104. (36) Goto, T.; Ikehata, A.; Morisawa, Y.; Higashi, N.; Ozaki, Y. Inorg. Chem. 2012, 51, 10650−10656. (37) Ikehata, A.; Mitsuoka, M.; Morisawa, Y.; Kariyama, N.; Higashi, N.; Ozaki, Y. J. Phys. Chem. A 2010, 114, 8319−8322. (38) Mitsuoka, M.; Shinzawa, H.; Morisawa, Y.; Kariyama, N.; Higashi, N.; Tsuboi, M.; Ozaki, Y. Anal. Sci. 2011, 27, 177−182. (39) Morisawa, Y.; Higashi, N.; Takaba, K.; Kariyama, N.; Goto, T.; Ikehata, A.; Ozaki, Y. Rev. Sci. Instrum. 2012, 83, 073103. (40) Malinowski, E. R. Factor Analysis in Chemistry, 3rd ed.; John Wiley and Sons, Inc.: New York, 2002. (41) Jaumot, J.; Gargallo, R.; de Juan, A.; Tauler, R. Chemom. Intell. Lab. Syst. 2005, 76, 101−110. (42) Taube, H. Trans. Faraday Soc. 1957, 53, 656−665. (43) Hart, E. J.; Sehested, K.; Holoman, J. Anal. Chem. 1983, 55, 46− 49. (44) Czapski, G. J. Phys. Chem. 1964, 68, 1169−1177. (45) Active Oxygen in Chemistry, 1st ed.; Foote, C. S., Valentine, J. S., Greenberg, A., Liebman, J. F., Eds.; Springer: New York, 1995. (46) Chipman, D. M. J. Phys. Chem. A 2008, 112, 13372−13381. (47) Bezbarua, B. K.; Reckhow, D. A. Ozone: Sci. Eng. 2004, 26, 345− 357. (48) Mizuno, T.; Tsuno, H.; Yamada, H. Ozone: Sci. Eng. 2007, 29, 55−63.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the System Development Program for Advanced Measurement and Analysis (Program-S) of Japan Science and Technology Agency (JST).

■

REFERENCES

(1) Peyton, G. R.; Glaze, W. H. Environ. Sci. Technol. 1988, 22, 761− 767. (2) Meunier, B.; Sorokin, A. Acc. Chem. Res. 1997, 30, 470−476. (3) Huang, C. P.; Dong, C.; Tang, Z. Waste Manage. 1993, 13, 361− 377. (4) Chelkowska, K.; Grasso, D.; Fábián, I.; Gordon, G. Ozone: Sci. Eng. 1992, 14, 33−49. (5) Ignatiev, A. N.; Pryakhin, A. N.; Lunin, V. V. Russ. Chem. Bull. 2009, 57, 1172−1178. (6) Westerhoff, P.; Song, R.; Amy, G.; Minear, R. Ozone: Sci. Eng. 1997, 19, 55−73. (7) Wu, J. J.; Yang, J.-S.; Muruganandham, M. Ind. Eng. Chem. Res. 2008, 47, 1820−1827. (8) Nemes, A.; Fábián, I.; van Eldik, R. J. Phys. Chem. A 2000, 104, 7995−8000. (9) Elovitz, M. S.; Gunten, U.; von Kaiser, H.-P. Ozone: Sci. Eng. 2000, 22, 123−150. (10) Gurol, M. D.; Akata, A. AIChE J. 1996, 42, 3283−3292. (11) Garoma, T.; Gurol, M. D. Environ. Sci. Technol. 2005, 39, 7964− 7969. (12) Glaze, W. H.; Kang, J.-W.; Chapin, D. H. Ozone: Sci. Eng. 1987, 9, 335−352. (13) Wittmann, G.; Horváth, I.; Dombi, A. Ozone: Sci. Eng. 2002, 24, 281−291. (14) Soo, O. B.; Suk Kim, K.; Gu Kang, M.; Je Oh, H.; Kang, J.-W. Ozone: Sci. Eng. 2005, 27, 421−430. (15) Bors, W.; Michel, C.; Saran, M. Eur. J. Biochem. 1979, 95, 621− 627. (16) Muff, J.; Bennedsen, L. R.; Søgaard, E. G. J. Appl. Electrochem. 2011, 41, 599−607. (17) Sahni, M.; Locke, B. R. Ind. Eng. Chem. Res. 2006, 45, 5819− 5825. (18) Lee, Y.; Lee, C.; Yoon, J. Water Res. 2004, 38, 2579−88. (19) Lloyd, R. V; Hanna, P. M.; Mason, R. P. Free Radical Biol. Med. 1997, 22, 885−888. (20) Mikuni, T.; Tatsuta, M.; Kamachi, M. Cancer Res. 1985, 45, 6442−6445. (21) Herzberg, G. Molecular Spectra and Molecular Structure III: Electronic Spectra and Electronic Structure of Polyatomic Molecules, 2nd ed.; Van Nostrand: New York, 1966. 4506

dx.doi.org/10.1021/ac400056m | Anal. Chem. 2013, 85, 4500−4506