Environ. Sci. Technol. 2005, 39, 5851-5855
Preliminary Investigation of the Supply of Chemical Species to an Aqueous Solution Using a Hydrogen-Oxygen Flame M I H O U C H I D A , * ,† T A K A H I R O S O G A B E , † TADAAKI IKOMA,‡ AND AKITSUGU OKUWAKI† Graduate School of Environmental Studies, Tohoku University, Sendai 980-8579, Japan, and Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan
A new method of supplying radical species to aqueous solutions using a hydrogen-oxygen flame is investigated. When a hydrogen-oxygen flame is directed on the surface of an aqueous solution, hydroxyl radicals (•OH) produced in the flame are extracted into the aqueous phase. The presence of •OH in the aqueous solution was confirmed by electron paramagnetic resonance with spin trapping using 5,5-dimethyl-1-pyrroline-N-oxide. The extraction of •OH into the aqueous solution was monitored using a quantitative analysis of hydrogen peroxide (H2O2). The effects of the hydrogen and oxygen gas flow rates, hydrogen/oxygen ratio, and atmosphere on H2O2 formation were studied. When the hydrogen-oxygen flame blew on a phosphate buffer solution (pH 6.7) under an Ar atmosphere, the concentration of H2O2 increased with the blowing time of the flame and the flow rate of hydrogen gas. Under air, nitrate and nitrite ions were formed in the aqueous phase in addition to H2O2, and the H2O2 concentration was lower than that under argon. The application of this new method to an aqueous solution of Cu(II)-ethylenediaminetetraacetic acid (EDTA) caused a remarkable decrease in the concentration of Cu(II)-EDTA and total organic carbon.
Introduction Recently, many novel, highly efficient chemical and physicochemical treatment methods have been developed for wastewater containing persistent organic pollutants. Advanced oxidation processes (AOPs) (1-4) for treating hazardous waste chemicals, such as phenols and pesticides, present in water and wastewater have been studied in detail. Examples of AOPs include the use of H2O2 with ultraviolet light to treat compounds, semiconductor photocatalysis, ozonolysis, and sequential or parallel combinations of different processes. AOPs include treatment methods that place an extremely high-energy load on the water, such as ultrasonic irradiation (sonolysis) (5), supercritical water * Corresponding author phone +81-22-229-1151; fax: +81-22228-0353; e-mail:
[email protected]. Present address: Department of Environmental Information Engineering, Tohoku Institute of Technology, 35-1, Yagiyama-Kasumicho, Sendai 982-8577, Japan. † Tohoku University, Graduate School of Environmental Studies. ‡ Tohoku University, Institute of Multidisciplinary Research for Advanced Materials. 10.1021/es048299z CCC: $30.25 Published on Web 06/28/2005
2005 American Chemical Society
oxidation (SCWO) (6), plasma processing (7), and γ-ray irradiation (8). The application of these methods to aqueous solutions produces highly active species, especially the primary oxidant •OH, which plays an important role in degrading organic contaminants in wastewater. Therefore, the efficiency of the degradation of organic pollutants in water depends on an effective supply of •OH, which is a powerful oxidant species in aqueous solution in AOPs. Various radicals are produced in a flame during the combustion of a fuel and an oxidizing agent. In a hydrogenoxygen flame, mainly •OH and •H are formed (9). When water vapor is present in the gas phase adjacent to a combustion flame, the water vapor reacts chemically and thermally with the flame (10); however, little is known about the chemical effects of the water vapor interaction with flames. Furthermore, in supercritical water, more dramatic flaming combustion can take place in addition to normal flameless oxidation (11). Under certain conditions, a steadily burning diffusion flame can appear spontaneously in supercritical water. When a flame blows against the surface of an aqueous solution, radical species produced in the flame are extracted into the aqueous phase. In solution, these radicals can react with dissolved organic compounds, resulting in oxidation or the addition of a hydroxyl group to the organic compounds (12, 13). Methods using a combustion flame can supply radical species from outside the solution, in contrast to methods such as ultrasonic or γ-ray irradiation that produce •OH via the decomposition of water under high-energy loading. Radical species can be supplied to the solution in a local, concentrated manner because the position of the flame can be changed. Moreover, the supply of radicals ceases when the flame is extinguished. This paper examines a new method of supplying radical species to aqueous solutions using a hydrogen-oxygen flame. We confirmed that •OH was supplied to a solution by blowing a hydrogen-oxygen flame against the surface of the solution. We also investigated the effects of the fuel flow rate and atmosphere on the concentrations of the chemical species that formed in the solution and the application to the decomposition of a chelating agent in solution, which is one application of AOPs (14).
Experimental Section Apparatus. The reaction equipment is shown in Figure 1. The reactor consists of a cylindrical 304 stainless steel vessel with an inner volume of 1.7 L and a quartz window for observing the flame. The flow rates of the fuel and oxidizing agent, i.e., hydrogen and oxygen gas, were controlled between 300 and 800 mL min-1, and the mixing ratio of the gases (H2:O2) ranged from 2:1 to 1:1. A hydrogen-oxygen flame was produced using a microtorch (Nissan Tanaka 191C). The inner diameter of the burner nozzle was 1 mm. The distance from the solution to the tip of the burner nozzle was 50 mm. Under all the experimental conditions, the flame visible to the naked eye was about 50 mm long. The gas phase was ventilated with argon gas at a flow rate of 150 mL min-1 during the reaction to flush any residual hydrogen gas from the reactor. To detect radicals in the gas phase, an optical fiber was inserted in the side of the reactor vessel. To prevent the residual hydrogen gas from igniting and exploding, a thermosensor was set near the burner nozzle. If the flame were to go out, the sensor would detect the abrupt temperature drop and shut off the supply of hydrogen gas immediately with a nitrogen-gas-driven valve. Furthermore, a hydrogen gas detector and an acryl resin shield were also VOL. 39, NO. 15, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5851
FIGURE 2. Emission spectra of a hydrogen-oxygen flame in an argon atmosphere. observe EPR signals by a phase-sensitive detection method. To analysis the EPR spectrum, computer simulation was carried out. FIGURE 1. Setup of a combustion flame-water reactor system equipped with units to protect against explosion. used. Reactions under open air were carried out using a 1-L glass beaker as the reactor. Experimental Procedure. In each experiment, the reactions were carried out with 500 mL of phosphate buffer solution (4 mM Na2HPO4-KH2PO4, pH 6.7), except in the experiment that examined the degradation of Cu(II)ethylenediaminetetraacetic acid (EDTA). The reaction vessel was immersed in a water bath connected to a cooling water circulator to control the temperature of the solution. The reaction vessel was sealed and purged with argon gas at a flow rate of 50 mL min-1 for 30 min before the reaction. After the tip of the microtorch was ignited and it was installed in the reaction vessel, the gas phase in the reactor was ventilated with argon gas. The reaction was started after controlling the flow rates of hydrogen and oxygen and moving the burner nozzle to the desired position. At regular intervals during the experiment, 10-mL aliquots were withdrawn from the solution using a glass syringe. Analytical Methods. Emission spectra (between 250 and 850 nm wavelength) to monitor the formation of radicals were obtained using a fiber optic spectrometer (K-MAC Spectra View 2000). The distance between the visible flame and the tip of the fiber probe was about 5 mm. Measurements were made at the top of the burner nozzle and at 20 and 0 mm above the surface of the solution. The pH of the solution was measured with a pH meter. Hydrogen peroxide was quantified spectrophotometrically (15) using Ti(VI)-(2-((5bromo-2-pyridyl)azo)-5-[N-propyl-N-(3-sulfopropyl)amino]phenol) (PAPS, Dojindo Co.). Nitrate and nitrite ions and Cu(II)-EDTA were determined using ion chromatography (DIONEX QIC). Total organic carbon (TOC) was measured using a total carbon analyzer (Shimadzu TOC5000). The presence of •OH in the solution was confirmed by a spin-trapping method using a nonvolatile nitrone trap, 5,5dimethyl-1-pyrroline-N-oxide (DMPO), and electron paramagnetic resonance (EPR). The •OH detection experiment used 40 mM phosphate buffer solution containing 25 mM DMPO. The sample solution was transferred into a flat quartz cell (50 × 10 × 1 mm3) for EPR measurement under an argon atmosphere. EPR measurements were carried out at room temperature using an X-band spectrometer (Bruker ESP380E). The microwave frequency and magnetic field were monitored using a frequency counter (Anritsu MF2414B) and an NMR gauss meter (Bruker ER035M), respectively. The microwave frequency was 9.7733 GHz. A modulation frequency of 100 kHz and a modulation amplitude of 0.1 mT were used to 5852
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 15, 2005
Results and Discussion Formation of Hydroxyl Radicals. Emission spectra of the hydrogen-oxygen flame were measured at various feed rates of fuel gas in an argon atmosphere. Measurements were made at the top of the burner nozzle and at 20 and 0 mm above the surface of the solution. The emission spectra near the position of the burner nozzle at various fuel feed rates are shown in Figure 2. When the ratio of hydrogen gas in the fuel gas was increased, the peaks caused by the emission of hydroxyl radicals at about 280 and 310 nm increased in intensity. This suggests that the number of hydroxyl radicals produced in the flame increases when the ratio of hydrogen gas in the fuel gas is increased. At fuel gas feed rates of 800 mL min-1 for H2 and 400 mL min-1 for O2, a peak presumed to be that of H2O was observed at about 590 nm. This suggests that the high feed rate of hydrogen gas in the fuel gas increases the combustion speed of the flame (16) and various radical reactions occur simultaneously. Therefore, the amount of H2O produced might increase. The emission spectrum at 20 mm above the surface of the solution was of similar intensity to that at the position of the burner nozzle. By contrast, at the solution surface, no peaks attributable to the hydroxyl radical were observed. This shows that the formation of chemical species in a combustion flame occurs mainly in and around the visible combustion flame and that the concentrations of the chemical species decrease near the surface of the solution where the flame is invisible. Figure 3 shows the EPR spectrum of an aqueous solution of 40 mM DMPO and 4 mM phosphate buffer exposed to a hydrogen-oxygen flame for 60 min in an argon atmosphere and the simulated one. With the analysis of observed spectrum, the hyperfine constants aN ) 1.49 mT, aH ) 1.49 mT, and g value 2.0054 were estimated, and the simulated spectrum with these parameters agreed well with the observed. These parameters coincided closely with those of DMPO-OH, suggesting the generation of hydroxyl radical in the solution, which forms the DMPO-OH adduct with DMPO. This implies that •OH formed in the hydrogenoxygen flame is transferred to the solution. Reaction under Argon Gas. H2O2 forms via the dimerization of •OH as the main active species in various AOPs. Therefore, the amount of H2O2 formed in the solution should indicate the amount of •OH supplied to the solution (17). By use of the amount of H2O2 formed in the solution as an indicator, the effects of the flow rate of fuel gas and the ratio of H2/O2 on the supply of •OH from the flame to the solution were examined.
FIGURE 5. Effects of the hydrogen and oxygen flow rates to a flame in air on the H2O2 concentration in phosphate buffer solution.
FIGURE 3. EPR spectra of DMPO-OH adduct (a) generated in an aqueous solution of phosphate buffer exposed to a hydrogenoxygen flame and (b) the simulated one.
FIGURE 4. Effects of the rates of hydrogen and oxygen flow to a flame in an argon atmosphere on the H2O2 concentration in phosphate buffer solution. Chloride and carbonate ions can act as radical scavengers in aqueous solution (1). Consequently, the effect of phosphate ion on the formation of hydroxyl radicals, i.e., the formation of H2O2, in the solution was examined. The time course of the H2O2 concentration in the solution under argon was analyzed at three different phosphate buffer concentrations (0, 1, and 4 mM) and two different initial solution pH values (pH 5.2 and 6.7). Under all of the experimental conditions, the H2O2 concentration increased with time and the H2O2 concentration rarely changed in each condition. In 4 mM phosphate buffer (pH 6.7), the basal condition in this study, phosphate ions had minimal effects as radical scavengers in the solution. Figure 4 shows the effects of the hydrogen and oxygen flow rates for the flame under an argon atmosphere on the H2O2 concentration in phosphate buffer solution. The concentration of H2O2 that formed in the solution increased with the blowing time of the combustion flame. This suggests that the hydrogen-oxygen flame is the source of the hydroxyl radicals that form the H2O2. When the flow
rate of hydrogen was increased with a fixed oxygen flow, the H2O2 concentration increased remarkably. That is, when the H2/O2 ratio and the total flow rate of fuel gas were increased, the H2O2 concentration increased. Therefore, the amount of H2O2 formed in the solution depends heavily on the flow rate of hydrogen in the fuel gas. In a hydrogen-oxygen flame, increasing the ratio of hydrogen in the fuel increases the burning velocity of the fuel gas (17). This suggests that, under an argon atmosphere and with a high ratio of hydrogen in the fuel gas, i.e., a high burning velocity of the fuel gas, the amount of hydroxyl radical produced in the flame increases, so that the amount of H2O2 formed in the solution also increases. Reaction under Air. The nature of the gas phase surrounding a flame should affect the formation of the chemical species in the gas phase in the neighborhood of the flame. For example, if nitrogen is present in the gas phase, nitrogen oxides and their radicals might be produced in and around the flame. To examine the effect of the gas phase on the formation of chemical species in the solution, we conducted the experiments described above under air. Figure 5 shows the effects of the hydrogen and oxygen flow rates for the flame under air on the H2O2 concentration in the phosphate buffer. As under an argon atmosphere, the H2O2 concentration in the solution increased with the blowing time of the combustion flame against the solution. At a fixed oxygen flow rate, the H2O2 concentration in the solution was less dependent on the rate of hydrogen flow than under argon. In particular, at respective H2 and O2 flow rates of 800 and 400 mL min-1, the H2O2 concentration was similar to that at flow rates of 600 and 400 mL min-1, respectively. By contrast, when the rate of oxygen flow was varied at a fixed flow of hydrogen gas, the H2O2 concentration increased remarkably with the rate of oxygen flow, suggesting that the amount of hydroxyl radical supplied to the solution is proportional to the oxygen in the fuel gas under air. Under the experimental conditions, less H2O2 formed under air than under argon in all cases. The concentrations of nitrate and nitrite ions were quantified in order to examine the effects of the nitrogen oxides and their radicals formed by the reaction of the flame with air. No ammonia or ammonium ions were detected under the experimental conditions. Figure 6 shows the time courses of the concentrations of nitrate and nitrite ions formed under various flow rates of fuel gas for the combustion flame. As with the H2O2 concentration, the concentrations of nitrate and nitrite ions formed in the solution increased with the blowing time of the flame against the solution. For all the experimental conditions under air, the nitrite ion VOL. 39, NO. 15, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5853
FIGURE 7. Effects of the hydrogen and oxygen flow rates in an argon atmosphere on the H2O2 concentration in a solution of 0.1 mM Cu(II)-EDTA and 4 mM phosphate buffer.
FIGURE 6. Effects of the hydrogen and oxygen flow rates for a flame in air on the (a) NO2- and (b) NO3- concentrations in phosphate buffer solution. concentration was higher than that of nitrate ions under the same conditions. When varying the oxygen flow rate at a fixed rate of hydrogen flow, the H2O2 concentration was highest at H2 and O2 flow rates of 600 mL min-1 each. By contrast, the concentrations of both nitrate and nitrite ions were the lowest under the same conditions. The H2O2 concentration at respective H2 and O2 flow rates of 800 and 400 mL min-1 was similar to that at flows of 600 and 400 mL min-1, respectively. By contrast, the concentrations of both nitrate and nitrite ions at the former flow rates were more than twice those at the latter flow rates. By comparison of the results at respective H2 and O2 flow rates of 600 and 400 mL min-1 vs 600 and 300 mL min-1, the nitrate ion concentration was higher at the former flow rates, while the nitrite concentration was higher at the latter. Therefore, nitrite ion is readily oxidized to nitrate ion at higher oxygen flow rates. As occurred under an argon atmosphere, the burning velocity of the combustion flame under air increased with the ratio of hydrogen gas in the fuel gas. Consequently, the reactivity of radicals in the flame with nitrogen or oxygen in air increases, which increases the amounts of nitrogen oxides formed. Simultaneously, because more hydroxyl radicals react with the components in the gas phase, fewer hydroxyl radicals are supplied to the solution. Therefore, in contrast to an argon atmosphere, because nitrogen oxides and their radicals are formed by the reaction of nitrogen or oxygen in air with the radicals in the flame, the decrease in the amount of hydroxyl radicals supplied to the solution leads to a decrease in the amount of H2O2 formed in the solution. Decomposition of Cu(II)-EDTA. To examine the possible application of this method as an AOP, a solution of 0.1 mM Cu(II)-EDTA and 4 mM phosphate buffer was reacted using this method under an argon atmosphere. The H2O2 concentrations in the solution at various rates of hydrogen and 5854
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 15, 2005
FIGURE 8. Time course of the residual percentages of (a) Cu(II)EDTA and (b) TOC in a solution of 0.1 mM Cu(II)-EDTA and 4 mM phosphate buffer. oxygen flow are shown in Figure 7. Under the experimental conditions, at different flow rates of the combustion gases (H2 mL min-1:O2 mL min-1; 800:400, 600:600, and 600:300), the H2O2 concentration increased with time but was essentially independent of the flow rate of the combustion gases. In particular, the concentration of H2O2 that formed in the Cu(II)-EDTA solution was markedly lower compared with a control solution under an argon atmosphere. This suggests that the hydroxyl radical supplied to the solution was consumed in the decomposition of the Cu(II)-EDTA complex. The residual percentages of both Cu(II)-EDTA and TOC (Figure 8) decreased when the total flow rate of the combustion gases was small or the ratio of hydrogen in the fuel gas was high. At respective H2 and O2 flows of 800 and 400 mL min-1, the residual percentages of Cu(II)-EDTA and TOC
decreased by 20 and 55% at 60 min, respectively. This shows that at a high total flow rate of combustion gases or at a high ratio of hydrogen gas in the fuel, the amount of hydroxyl radical formed in the flame increases, which increases the supply to the solution; consequently, organic substances in the solution are decomposed effectively.
Acknowledgments The authors thank the staff of the mechanical factory of the Department of Molecular Chemical Engineering, Tohoku University, for fabricating the apparatus.
Literature Cited (1) Parsons, S. Advanced Oxidation Processes for Water and Wastewater treatment; IWA Publishing: London, 2004. (2) Glaze, W. H.; Kang, J.-W.; Chapin, D. H. The chemistry of water treatment processes involving ozone, hydrogen peroxide and ultraviolet radiation. Ozone: Sci. Eng. 1987, 9, 335-352. (3) Legrini, O.; Oliveros, E.; Braun, A. M. Photochemical Processes for Water Treatment. Chem. Rev. 1993, 93, 671-698. (4) Neyens, E.; Baeyens, J. A revies of classic Fenton’s peroxidation as an advanced oxidation technique. J. Hazard. Mater. 2003, B98, 33-50. (5) Adewuyi, Y. G. Sonochemistry: Environmental Science and Engineering Applications. Ind. Eng. Chem. Res. 2001, 40, 46814715. (6) Savage, P. E. Organic Chemical Reactions in Supercritical Water. Chem. Rev. 1999, 99, 603-621. (7) Clements, J. S.; Sato, M.; Davis, R. H. Preliminary Investigation of Prebreakdown Phenomena and Chemical Reactions Using a Pulsed High-Voltage Discharge in Water. IEEE Trans. Ind. Appl. 1987, IA-23, 224-235. (8) Getoff, N. Radiation-induced degradation of water pollutantsstate of the art. Radiat. Phys. Chem. 1996, 47, 581-593.
(9) Charton, M.; Gaydon, A. G. Excitation of spectra of OH in hydrogen flames and its relation to excess concentrations of free atoms. Proc. R. Soc. London, Ser. A 1958, 245, 84-92. (10) Richard, J.; Garo, J. P.; Souil, J. M.; Vantelon, J. P.; Knorre, V. G. Chemical and physical effects of water vapor addition on diffusion flames. Fire Safety J. 2003, 38, 569-587. (11) Schilling, W.; Franck, E. U. Combustion and Diffusion Flames at High Pressures to 2000 bar. Ber. Bunsen-Ges. Phys. Chem. 1988, 92, 631-636. (12) Cleaver, C. S.; Blosser, L. G.; Coffman, D. D. Syntheses by Freeradical Reactions. IX. Use of Free Radicals from Flames. J. Am. Chem. Soc. 1959, 81, 1120-1126. (13) Nomoto, S.; Shimoyama, A.; Shiraishi, S.; Sahara, D. Underflame Oxidation of Amines and Amino Acids in an Aqueous Solution. Biosci. Biotech. Biochem. 1996, 60, 1851-1855. (14) SillanpA¨ A¨ , M.; Pirkanniemi, K. Recent developments in chelate degradation. Environ. Technol. 2001, 22, 791-801. (15) Matsubara, C.; Kudo, K.; Kawashita, T.; Takamura, K. Spectrophotometric Determination of Hydrogen Peroxide with Titanium 2-((5-Bromopyridyl)azo)-5-(N-propyl-N-sulfopropylamino)phenol Reagent and Its Application to the Determination of Serum Glucose Using Glucose Oxidase. Anal. Chem. 1985, 57, 1107-1109. (16) Iki, N.; Furutani, H.; Hama, J.; Liu, F.; Takahashi, S.; Kurata, O. Burning Velocity of Stoichiometric Hydrogen-Oxygen-Steam Mixture. J. Gas Turbine Soc. Jpn. 1997, 25, 85-92. (17) Nam, S.-N.; Han, S.-K.; Kang, J.-W.; C. Heechul Kinetics and mechanisms of the sonolytic destruction of nonvolatile organic compounds: investigation of the sonochemical reaction zone using several OH• monitoring techniques. Ultrason. Sonochem. 2003, 10, 139-147.
Received for review November 1, 2004. Revised manuscript received May 30, 2005. Accepted June 1, 2005. ES048299Z
VOL. 39, NO. 15, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5855
