Environ. Sci. Technol. 2001, 35, 411-415
Enhanced Photocatalytic Activity of Nafion-Coated TiO2 M. S. VOHRA AND K. TANAKA* National Institute of Materials and Chemical Research, 1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan
Photocatalytic degradation (PCD) of aqueous paraquat was accelerated by the addition of either phosphate or sulfate salt. Attachment of these anions to the TiO2 surface possibly results in increased adsorption of the cationic paraquat species and in turn its photocatalysis rate. The same effect was obtained more consistently using the Nafion (an anionic polymer)-coated TiO2. Enhanced PCD of paraquat and some amine compounds was noted. However the anionic and neutral compounds were not affected significantly. Nafion proved to be stable against photocatalysis. It has been suggested that the degradation rate is larger for the cationic compounds with higher pKa. For a phenolparaquat-TiO2 system, paraquat degradation did not begin till near-complete phenol removal. Using the Nafioncoated TiO2, both phenol and paraquat degradations started simultaneously. Nevertheless, complete paraquat removal still took longer than phenol.
FIGURE 1. Schematic illustration of the photocatalytic reactor. 1, Pyrex glass tube; 2, black light lamp; 3, stirrer; 4, magnetic stirrer. as compared with its direct photolysis. Tennakone and Kottegoda also studied aqueous paraquat photocatalysis using TiO2 (10). Use of both UV and sunlight caused paraquat degradation. However, no paraquat removal was observed in the absence of TiO2. The present study reports that surface modification of TiO2 causes an increase in the photocatalytic degradation of paraquat and other cationic compounds.
Experimental Section Introduction It has been recognized that TiO2 is the most efficient photocatalyst for environmental applications as compared with the other semiconductors (1). However, its activity needs to be enhanced especially for water purification. Several methods have been proposed to increase the TiO2 efficiency, e.g., incorporation of an adsorbent such as activated carbon into TiO2 (2-4). Tanaka and Robledo showed that the adsorption of substrate onto TiO2 significantly affects its photocatalysis rate (5). The present study discusses an efficient method of increasing the substrate adsorption by modifying the TiO2 surface charge, which in turn improves the photocatalytic activity specifically for the cationic pollutants. It has been shown (6, 7) that modifications of TiO2 surface charge by the inorganic anions either reduced the photocatalytic activity (sulfate, phosphate, chloride) or had no influence (nitrate, perchlorate). However, we found that the addition of sulfate and phosphate species enhanced the photocatalytic degradation of paraquat (cationic species). These results led us to consider better modifications of the TiO2 surface. Paraquat, a widely used herbicide, was chosen as a representative cationic pollutant to evaluate the activity of modified TiO2. Paraquat is very toxic and can cause irreversible lung damage upon ingestion. Also, water and soil contamination at its point of application remains a concern. In the surface soils, paraquat may photodecompose in several weeks (8). However, paraquat in the subsurface environment may remain intact for several years. Recently, Moctezuma et al. studied the paraquat PCD employing TiO2 (9). The paraquat PCD rate was much higher * Corresponding author phone: +81-298-61-4563; fax: +81-29861-4563; e-mail:
[email protected]. 10.1021/es001238q CCC: $20.00 Published on Web 12/09/2000
2001 American Chemical Society
All the chemicals used were of reagent grade. TiO2 (Fujititan TP-2) used in the present work was of anatase form, and its specific surface area is 17.3 m2/g (11). Paraquat (1,1-dimethyl4,4-bipyridinium dichloride) was purchased from Dr. Ehrenstorfer GmbH, Germany. Asulam (methysulfanilyl carbamate), ethylamine, trimethylamine, morpholine, and phenol were obtained from Wako Pure Chemical Industries Ltd. Chloride, nitrate, sulfate, and phosphate species were added as sodium salts. Nafion solution [5 wt % solution, a product of DuPont, SE-5112, copolymer of perfluorosulfonic acid and poly(tetrafluoroethylene)] was purchased from Aldrich Chemical Co. A Pyrex-glass batch type reactor (6 cm i.d. and 26 cm length) was used for the photocatalysis experiments. Figure 1 shows the experimental setup. For each experiment, 500 mL of 10-4 M substrate solution was transferred to the reactor, and 2 g of TiO2 was added; a magnetic stirrer was employed for continuous mixing. A 6-W black light lamp was used for illumination. The pH adjustments were made using 1 M HClO4. Samples were taken at several time intervals and filtered using 0.2-µm pore size Millipore filters. The first three drops of the filtrate were rejected, and the rest of it was collected and analyzed. Paraquat was analyzed employing a Shimadzu LC-10AD HPLC equipped with SCL-10A system controller and SPD-10A UV-Vis detector. The HPLC column was ICcation-SW TSK gel from the Tosoh. The eluent consisted of 4:1 (v:v) of 0.2 M NaH2PO4‚2H2O (its pH was adjusted to 3 using H3PO4) and CH3CN, respectively (12). Phenol was analyzed using the same HPLC unit with an ODS column. Ethylamine, trimethylamine, and morpholine were monitored using an ion chromatograph equipped with a JASCO 880 PU pump and a Shodex CD-4 conductometer. Nitrate and ammonium ions were detected using the same ion chromatograph unit. The columns were Shodex IC-613 for VOL. 35, NO. 2, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
411
FIGURE 2. Photocatalytic degradation of paraquat in the presence of phosphate, sulfate, nitrate, and chloride (10-3 mol L-1). O, No anion present; 4, phosphate; 0, sulfate; ×, nitrate; b, chloride at (a) pH 5.5 and (b) pH 3.0. nitrate and IC Y-521 for the ammonium ions. Acetic and formic acids were determined using a Yokogawa ion chromatograph (IC 7000 series II) equipped with a background suppressor. The Nafion-coated TiO2 samples were prepared by adding the desired amount of Nafion solution (5 wt %) to 2 g of TiO2 along with an appropriate amount of methanol for proper mixing and to ensure a homogeneous coating of Nafion onto TiO2. After being mixed manually, the mixture was dried overnight at room temperature. To determine the CO2 evolution rate, a 34-mL Pyrex glass vial covered with an air-tight plastic cap was used (Mininert Valve SC-100, GL Sciences Inc. Japan). A total of 75 mg of the Nafion-coated TiO2 (2.5-67.5 mg of Nafion/g of TiO2) suspended in 25 mL of a 5 × 10-4 M paraquat solution was illuminated by a 500-W super-high-pressure mercury lamp. Headspace CO2 was quantified using a Shimadzu 6AM gas chromatograph. In the NH4+ desorption experiment, for both TiO2-only and Nafion-coated TiO2 samples at illumination time >20 h, the suspension pH was adjusted to below 3 (2 and 4 mM HNO3). The difference in the NH4+ concentration (if any) before and after HNO3 addition showed adsorption.
Results and Discussion Effect of Inorganic Anions. The effect of 10-3 M chloride, nitrate, sulfate, and phosphate onto the paraquat photocatalytic degradation (PCD) rate was investigated at pH 5.5 and pH 3.0 (Figure 2). Chloride and nitrate had almost no influence on the paraquat disappearance (both at pH 5.5 and at pH 3.0), while sulfate slightly increased the disappearance rate (at pH 3.0) and phosphate enhanced it considerably (at both pH 5.5 and pH 3.0). In contrast to these findings, chloride, sulfate, and phosphate have been reported to suppress the photocatalysis rate of several contaminants including amine compounds (6, 7). The adsorption of anions to the TiO2 surface possibly caused a reduced activity. Boehm et al. (13, 14) noted that sulfate and phosphate were adsorbed to the TiO2 surface from their acidic solutions, while chloride and nitrate were only slightly adsorbed. Electrophoretic measurements of the present TiO2 sample showed an isoelectric point of 4.0; it was positively charged at pH 3.0 and negatively charged at pH >4.0. This indicates that the sulfate species would adsorb more strongly onto TiO2 at pH 3.0 than at pH 5.5. Hence more paraquat is attracted at pH 3.0. Unlike sulfate, phosphate accelerated the photocatalysis rate both at pH 5.5 and at pH 3.0 (Figure 1); however, the effect at pH 3.0 was smaller than at pH 5.5. According to 412
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 2, 2001
Hadijivanov et al. (15) and Boehm et al. (13, 14), the phosphate species adsorbed to TiO2 at pH 7 is H2PO4-. Using the dissociation constants for the phosphate system (16), at pH 5.5, 97.8% H2PO4- and 2.2% HPO42- species were calculated. At pH 3, 88.2% H2PO4- and 11.8% H3PO4 species were calculated. Such a change in the phosphate speciation possibly causes reduced paraquat adsorption to the TiO2 surface at pH 3 and hence a lower photocatalysis rate result. In the previous studies (6, 7), the negative effect of anions on the PCD rate was attributed to their blocking of the active surface sites and competitive reactions with the active oxidizing species. However, such trends were not noted in the present study. Instead, it was found that altering the surface charge of TiO2 using anions enhances the photocatalysis rate of cationic pollutants. It should be noted that Abdullah et al. (6) quantified the CO2 formation to evaluate the substrate degradation, whereas we measured the disappearance rate and TOC removal. Unlike the degradation rate, TOC removal was not affected by the addition of phosphate. Considering that most of the intermediates from aromatic compounds are not cationic (17, 18), their attachment onto TiO2 will not be increased by the addition of phosphate. For example, the maximum 1.6 × 10-6 and 5.8 × 10-6 M of acetic acid and formic acid, respectively, were detected in the presence of phosphate. Since they are only a part of formed organic acids (several unidentified peaks were observed), significant portion of the intermediates can be anionic species. TiO2 was also retrieved from the TiO2-phosphate experiment and used for several more paraquat photocatalysis runs (without phosphate addition). A gradual decrease in the PCD rate was noted. This indicated a slow desorption of the surface phosphate species. A similar slow phosphate desorption process has been reported elsewhere (15). Modification of TiO2 by Nafion. Previously, it was demonstrated that a change in TiO2 surface charge by phosphate adsorption enhanced the paraquat degradation. However, phosphate detachment after repeated use of TiO2 necessitates a better modification, i.e., an adsorbent that does not detach and is also resistant to photocatalytic degradation. The Nafion polymer met these requirements. Nafion is an anion-exchange resin with negative sulfonate groups. It was also found to be resistant to photocatalysis. Paraquat photocatalysis was investigated using the different Nafion-coated TiO2 samples (Figure 3). Before illumination, the paraquat-TiO2 suspension was stirred for at least 2 h to ensure proper suspension of Nafion-coated TiO2. The initial paraquat adsorption (before illumination)
FIGURE 3. Photocatalytic degradation of paraquat on Nafion-coated TiO2. Plain TiO2: O. Nafion-TiO2: b, 2.25; 4, 4.5; 0, 45; and ×, 67.5 mg/g TiO2.
FIGURE 4. Effect of the Nafion amount on CO2 evolution rate from the degradation of paraquat. O, initial 60 min; 4, initial 180 min. increased with an increase in the Nafion amount (Figure 3). For example, using 67.5 mg of Nafion/g of TiO2, approximately 87% paraquat adsorption resulted at time zero (Figure 3): adsorption of paraquat was confirmed by retrieving TiO2, desorbing paraquat in 0.1 N HCl, and measuring the UV spectrum. At a higher Nafion volume, it was difficult to ascertain whether the paraquat disappearance results from PCD or adsorption. Therefore, the determination of an optimum Nafion amount was not possible. Considering this, CO2 formation was quantified (Figure 4). Experiments were conducted using 5 × 10-4 M paraquat solution and different Nafion-coated TiO2 samples; other details are given in the Experimental Section. The CO2 results helped to find the optimum Nafion amount for paraquat mineralization. Figure 4 illustrates that initially the CO2 evolution rate increases with an increase in the Nafion amount. This is followed by a plateau (or maximum). However, any further increase in the Nafion amount caused a decrease in the CO2 formation rate. It seems that too much Nafion retards the degradation either by blocking the active sites (on TiO2) or scattering the incident light. The highest PCD rate was observed at about 2.3-4.5 mg of Nafion/g of TiO2. Also at 4.5 mg of Nafion the initial paraquat adsorption onto TiO2 was around 10%, which is much smaller as compared with the other Nafion amounts used (Figure 3) and may not affect the evaluation of the photocatalytic acitivity much. Therefore 4.5 mg of Nafion/g of TiO2 was used for all further investigations. Nafion product information shows an ion-exchange capacity of 0.9 mequiv/g. Therefore, the total ion-exchange
FIGURE 5. TOC removal and NH4+ formation from the degradation of paraquat on Nafion-coated TiO2 (4.5 mg/g TiO2). TOC: O, on plain TiO2; 4, on Nafion-coated TiO2. NH4+: b, on plain TiO2; 2, on Nafioncoated TiO2. capacity of 2 g of TiO2 covered with 9 mg of Nafion (0.2 mL of Nafion/2 g of TiO2) is 0.81 × 10-5 equiv or 0.4 × 10-5 mol (2 equiv/mol of paraquat). Paraquat adsorption before illumination was also almost 0.5 × 10-5 mol (10% from 500 mL of 10-4 M solution, Figure 3). This suggests that most of the sulfonate groups on Nafion-coated TiO2 are utilized for paraquat attachment. Paraquat PCD experiments conducted at pH 3 and pH 6 did not show any marked differences. This indicates that both the sulfonate (group of the Nafion) and the paraquat dissociation are not significantly influenced by such a change in the pH. The initial TOC removal (up to 2 h) was quite rapid, both for the Nafion-coated and for the plain TiO2. However after 2 h, the TOC removal rate decreased. Also, comparing the Nafion-coated and plain TiO2 results (Figure 5), the TOC removal rate was slightly larger for the Nafion-coated TiO2 before 2 h. However, after 2 h this order was reversed (Figure 5). Major reaction intermediates produced after 2 h may be anionic species. Consequently, their adsorption and hence PCD on Nafion-coated TiO2 may not be favorable. This may cause an overall slower TOC removal for the Nafion-coated TiO2. As mentioned later, a slightly negative effect on the acetic and formic acid degradation was noted. Results for the plain TiO2 study after 20 h of illumination showed 95% original N conversion to NH4+ (Figure 5). Similarly results for the Nafion-coated TiO2 (4.5 mg/g of TiO2 after 20 h illumination) also revealed approximately 70% NH4+ production. However, the NH4+ concentration (after 20 h) decreased with an increase in the Nafion amount (Figure 6). It can be suggested that part of the NH4+ attaches to the surface Nafion species; the percent adsorption increases with an increase in the surface Nafion amount. The adsorption was confirmed by measuring NH4+ desorption at lower pH (see Experimental Section). For TiO2 coated by 67.5 mg of Nafion, NH4+ adsorption was 25% of the total SO3- group; it was assumed that a part of nitrogen formed NO3- (8 × 10-6 M). On the other hand, five experiments conducted using the same Nafion-TiO2 each time showed an increasing PCD NH4+ concentration. Saturation of the TiO2 surface sites with repeated use may leave fewer sites for NH4+ adsorption. This in turn may cause a gradual increase in the aqueous NH4+ concentration. In summary, CO2 and NH4+ were the major mineralization products (Figures 4 and 5). Nitrate, if detected, was present in small amounts (
