Environ. Sci. Technol. 2006, 40, 1616-1621
Photocatalytic Activity for Degradation of Nitrogen Oxides over Visible Light Responsive Titania-Based Photocatalysts Y U - M I N G L I N , * ,† Y A O - H S U A N T S E N G , † JIA-HUNG HUANG,† CHIH C. CHAO,† CHIEN-CHIH CHEN,‡ AND IKAI WANG‡ Center for Environmental, Safety and Health Technology Development, Industrial Technology Research Institute, Room 204, Building 32, 195, Section 4, Chung Hsing Road, Chutung, Hsinchu 310 Taiwan, and Department of Chemical Engineering, National Tsing-Hua University, Hsinchu 300 Taiwan
This study investigates photocatalytic degradation of nitrogen oxides over titania-based photocatalysts illuminated by ultraviolet and visible light. The TiO2 photocatalyst was synthesized in a sol-gel process using titanium butoxide as the precursor. After calcination between 150 and 300 °C, the synthesized TiO2 responded strongly to visible light photocatalytically degrading NOx, probably because of the existence of carbonaceous species that act as sensitizers. The optimum calcination temperature was found to be around 200 °C. Additionally, platinum ion-doped TiO2 was prepared by impregnation using Pt(NH3)4(NO3)2 as a dopant, which improved the photocatalytic activity that degraded NOx in the visible light region. The Pt ion was doped in oxide form at the surface of TiO2 and was expected to be responsible for sensitization. At an optimum calcination temperature of around 200 °C, the Pt ion-doped TiO2 exhibited higher activity in the further oxidation of NO2 to NO3clearly reducing NO2 selectivity. The TiO2 catalysts chemically prepared by either the sol-gel process or impregnation exhibited stronger activity than conventional TiO2 when illuminated under a fluorescent lamp. Rinsing with water was responsible for the restored reactivity of prepared TiO2 catalysts for NOx degradation.
Introduction Most nitrogen oxides (NOx) emitted in air are nitric oxide (NO) and nitric dioxide (NO2). As industrial pollutants, NOx emissions are generated mainly from combustion, in which oxygen oxidizes nitrogen to form NOx at very high temperature (1-3). Several catalytic approaches have been developed for controlling NOx, such as selective catalytic reduction with ammonia or hydrocarbon, catalytic decomposition of NO, and removal through NOx storage. A three-way catalyst is also an established technology for use in the catalytic reduction of NOx that is produced by gasoline engines (4, 5). All such technologies provide efficient methods for abatement of NOx emitted from stationary sources or vehicles. * Corresponding author phone: 886-3-5915439; fax: 886-35820016; e-mail:
[email protected]. † Industrial Technology Research Institute. ‡ National Tsing-Hua University. 1616
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 5, 2006
Photocatalytic oxidation was recently applied to remove atmospheric NOx, which is believed to cause acid rain and to be toxic to human health. Titanium dioxide (TiO2) photocatalyst was found to be able to oxidize low-level NOx (near 1 ppm or below) in air to HNO3 very rapidly (6, 7). As a most widely used photocatalyst, TiO2 has a relatively large band gap of 3.2 eV, corresponding to wavelengths of shorter than 388 nm. Restated, TiO2 has to be activated by a UV light source, which can use only 3-4% of the solar energy that reaches the earth (8, 9). From this perspective, TiO2 that can be used efficiently under both UV and visible light irradiation must be developed to increase its usefulness in a wide range of wavelengths. The irradiation of wavelengths from 300 to 700 nm may offer almost 50% of solar energy. TiO2 photocatalysts that respond to visible light have attracted much attention. The general approach is to modify TiO2 by creating intra-band gap states that are close to the conduction or valence band edges, and absorb visible light at sub-band gap energies of less than 3.2 eV (10). The most commonly employed method is doping with transition metal ions such as Cr, V, and others by ion-implantation or the sol-gel method. The substitution of metal ion for Ti4+ changes the electronic properties of TiO2, sufficiently reducing the energy band gap to absorb visible light (8, 9, 11). The substitution of the lattice O2- of the TiO2 with nitrogen ions has been reported to lead to the appearance of an absorption band in visible light regions (12). The newly formed intraband gap states were found to be sufficiently close to the conduction band edge. This nitrogen-doped TiO2 exhibits in both photodetoxification and super hydrophilic activity in visible light. Intra-band gap states can also be formed by lattice defects. Treating the TiO2 powder in the hydrogen plasma creates oxygen vacancies, and the new intraband gap states are formed close to conduction band edge in the TiO2 band structure. This sub-band gap energy is reduced to less than 2.5 eV, enabling visible light to be absorbed (7). Sensitization by electron transfer from an excited donor, such as a dye or a metal complex, has been well documented in the field of photoelectrochemistry. However, it cannot be used in photocatalytic detoxification, because the sensitizer is degraded in the presence of oxygen (10). Titania that contains platinum (IV), rhodium (III), and gold (III) chlorides either in the bulk or only at the surface has been reported to exhibit photocatalytic activity in visible light in aqueous suspension, degrading organic pollutants (13-15). The injection of electrons from the sensitizer to TiO2 improves photocatalytic activity in visible light. Carbon-containing TiO2 photocatalyst prepared by the sol-gel process is reported to be able to photodegrade p-chloraphenol in visible light. The highly condensed, carbonaceous species formed during calcination is considered to be responsible for the photosensitiztion. The catalysts also exhibit good long-term stability, despite the carbonaceous nature of the sensitizing species (16). In this study, TiO2-based photocatalysts were synthesized using a sol-gel process to investigate the activity in response to visible light and the consequent degradation of nitrogen oxides. Bulk TiO2 was doped with platinum ions in a sol-gel process in which Pt(NH3)4(NO3)2 was used as a dopant. Additionally, platinum ions were added as the oxide to the surface of TiO2 by the impregnation of aqueous Pt(NH3)4(NO3)2 solution, followed by calcination in air. The platinum oxide may behave as a sensitizer, inducing the photocatalytic activity of TiO2 in the visible light region. The rates of removal of NOx and the selectivity of NO2 during photodegradation 10.1021/es051007p CCC: $33.50
2006 American Chemical Society Published on Web 02/02/2006
FIGURE 1. Schematic continuous flow reaction system for photocatalytic degradation of NOx. over various TiO2-based catalysts were studied in ultraviolet and visible light. Moreover, the long-term stability of the prepared photocatalyst was examined.
Experimental Section Synthesis of TiO2 that Responds to Visible Light. The TiO2 sample was prepared using an acid-catalyzed sol-gel process. In a typical procedure, 50 mmol of titanium butoxide, Ti(O-nBu)4, was added into a solution of 90 mL of anhydrate ethanol and 20 mL of deionized (DI) water in a 250 mL flask. Following complete dissolution, 4 mL of nitric acid was added to the solution to catalyze the hydrolysis and condensation reactions. After agitation at 500 rpm for 3 h, the precipitated titanium hydroxide was dried at 120 °C for 10 h and then calcined at an elevated temperature for 10 h. The calcination temperature ranged from 150 to 600 °C. When bulk TiO2 was doped with platinum ions, a tetra-amine platinum nitrate, Pt(NH3)4(NO3)2, solution was added into the matrix before nitric acid was added, to yield the desired atomic ratio of Pt/Ti (such as 1/100, for example). The surface of TiO2 was doped with platinum ions by impregnation. Either commercially available TiO2 powders (Hombikat UV100, Sachtleben Chemie) or those as-synthesized by the sol-gel process, could be used as the raw material for impregnation. TiO2 (10 g) was added to 100 mL of Pt(NH3)4(NO3)2 solution and stirred at 500 rpm for 1 h. The amount of Pt in solution was determined from the desired Pt/Ti atomic ratio. The mixture was dried at 120 °C for 4 h and then calcined at an elevated temperature, from 150 to 600 °C for 10 h. Following calcination, the nitrate and ammonia groups would be removed and the TiO2 surface was doped using Pt ions as an oxide. The color of the sample was between yellow and black, depending on the calcination temperature. X-ray diffractometry (XRD) with Cu KR radiation (Scintag XRD3000) was employed to analyze the crystal structure of the synthesized products. The crystallite size was determined from the broadening of the peaks using Scherrer’s equation. A diffuse-reflectance scanning spectrophotometer (Shimadzu, UV-2450) was employed to obtain the UV-visible absorption spectra of the powders. The reflectance data were converted to the absorbance values, F(R), based on the Kubelka-Munk theory. Degradation of NOx over TiO2 Photocatalyst. As shown in Figure 1, a continuous flow reaction system, which consists
of a round-shaped Pyrex glass vessel (φ × H, 14 cm × 3.6 cm), was used to conduct the degradation of NOx. A sample dish (φ × H, 9 cm × 0.6 cm) located inside the vessel contained the TiO2 powder in the experiment. A mini fan was attached to the inside of the glass vessel to improve the mixing effect of the reaction mixture, if necessary. Before it was placed in the vessel, the TiO2 sample was pretreated by irradiation with UV light (1 mW/cm2 intensity) for 10 h, followed by rinsing with 200 mL of DI water to remove any contaminants on its surface. A 100 W mercury arc lamp (high pressure, Oriel model 6281) equipped with a filter to irradiate only light of the desired wavelength, with a full width at halfmaximum of 10 nm, was the source of light that was incident on the TiO2 sample. The allowed wavelengths through the filters of model 56531, 56541, 56551, 54341, and 56561 (Oriel Instruments) were 365, 400, 435, 500, and 546 nm, respectively, and the intensity of the light was 1 mW/cm2. Additionally, laser emitting diodes (LEDs) provided light that was blue (430-530 nm), green (470-570 nm), and red (580670 nm), covering a wide visible range. The spectrum of the light from the light source was obtained using a spectrophotometer (Ocean Optics, USB2000), and was used to determine the intensity. The temperature rise due to UV or visible light irradiation was not observed in the Pyrex vessel because adequate distance was kept from the lamp to the vessel. NOx was degraded at room temperature using an air stream that contained 1.0 ppm NO as feedstock. The NO gas was provided from a cylinder containing 100 ppm NO (N2 balance, from San Fu Chemical Co.) and diluted by a separate air stream. Two mass flow controllers (MFC) (Brooks 5850E) were used to control the relative humidity in the feeding stream. An air stream, controlled by MFC 2, brought in saturated water vapor from a pair of homemade humidifiers to the reaction system; it was mixed with a separate air stream via MFC 1, to adjust the relative humidity. The humidified air stream was then mixed with the stream that contained NO (via MFC 3) to form the feeding stream. The reaction gas in the feeding stream passed through the vessel that contained TiO2 powder (0.3 g) at a flow rate of 1 L/min. The reaction of NO with air was not observed when performing a control experiment with or without light in the absence of the catalyst. The NO and NO2 concentrations were continuously monitored using an on-line chemiluminescent NOx analyzer VOL. 40, NO. 5, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1617
(Eco Physics, CLD 700 AL) in gas-phase analysis. The removal rate of NOx (µmol/h) is defined as ([NO]inlet - [NO] - [NO2]) × Qinlet, where [NO]inlet is the concentration of NO in the feeding stream, [NO] and [NO2] are the concentrations of NO and NO2 in the outlet stream, and Qinlet is the feeding flow rate. In solid-phase analysis, NO3- and NO2- adsorbed onto the surface of the TiO2 sample were easily rinsed in DI water. The water that contained the anions was then analyzed using an ion chromatograph (DIONEX, DX-120) to determine the concentrations of those ions. To examine the mass balance on the reaction system, for example, 1.96 µmol NOx was removed according to the gas-phase analysis, during the operation of 2 h on stream. Meanwhile, 1.90 µmol NO3and 0.03 µmol NO2- were obtained from the solid-phase analysis. The amount of NOx removed determined from the gas-phase analysis was consistent with that determined from the solid-phase analysis, so the mass balance on the reaction system was favorable. Regeneration of TiO2 Photocatalyst. In the regeneration experiments, a nonwoven fabric (with a PE/coPET structure and density of 50 g/cm2) sheet was used as the substrate to carry the photocatlyst. TiO2 powder with colloidal silica was added to DI water to form a slurry of 15 wt % TiO2, 1 wt % SiO2, and the rest water. The nonwoven fabric sheet was immersed in the slurry for 1 h and then squeezed using a rotator presser to remove the excess material. The sheet was dried at 100 °C for 4 h to fix the TiO2 sample. Therefore, a specimen that contained 20 mg of TiO2 per cm2 of surface area was obtained. The specimen was then placed in a rectangular vessel (L × W × H, 20 cm × 5 cm × 0.7 cm) instead of the round vessel, as presented in Figure 1, to degrade NOx in light from light sources with wavelengths of 365 and 500 nm. After reaction had proceeded for 4 h on stream, the specimen was removed and rinsed in 100 mL of DI water for 1 h to eliminate the adsorbed anions from the surface of TiO2. The specimen was then dried at 80 °C for 1 h to restore the active sites. The activity associated with the degradation of NOx was measured again over the regenerated specimen. The procedure was repeated several times. The results obtained throughout the operation cycle of degradation and regeneration yield information on the long-term stability of the TiO2 photocatalyst.
Results and Discussion. Effect of Calcination Temperature on Characteristics of Photocatalysts during Sol-Gel Synthesis. The TiO2 samples prepared by the sol-gel process had crystal structures that were dominated by anatase, as observed by XRD analysis. Figure 2 shows that anatase was the main phase at a calcination temperature of below 500 °C. When the samples were calcined at 600 °C, the dominant crystal phase was rutile. At calcination temperatures of between 200 and 500 °C, a minor brookite phase was also detected. The prepared TiO2 photocatalysts exhibited the differential absorbance of visible light (λ > 400 nm), depending on the calcination temperature. As shown in Figure 3, the visible light absorbance increased with the calcination temperature up to 200 °C, and then decreased as the temperature increased further. At a calcination temperature of above 300 °C, the absorbance profile was similar to that of conventional TiO2 (such as UV100, for example) and visible light absorbance disappeared. As reported in the literature (16), the presence of carbonaceous species might cause TiO2 photocatalyst prepared by the sol-gel process to respond to visible light. The carbonaceous species are burnt out by calcination at elevated temperature. Thermo-gravimetric analysis (TGA) indicated that the weight loss of the prepared TiO2 sample depended on the calcination temperature. Table 1 shows that the weight loss of the sample markedly decreases as the calcination temperature rose above 300 °C, indicating that 1618
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 5, 2006
FIGURE 2. XRD patterns of sol-gel-synthesized TiO2 calcined at various temperatures.
FIGURE 3. UV-vis absorption spectra of sol-gel synthesized TiO2 calcined at various temperatures.
TABLE 1. Weight Loss of Sol-Gel Synthesized Photocatalysts, Depending on Calcination Temperaturea calcination temp. (°C) as-syn. 150 200 250 300 400 500 600 weight loss (%)
14
13
8
8
4
1
0
0
a
Operating conditions of TGA: flow rate of air, 30 mL/min; heating rate, 5 °C/min; end point, 800 °C; initial weight of the sample, 0.01 g.
the carbonaceous species may have been thoroughly eliminated by burning, reducing the absorbance of visible light. The relationship between the photocatalytic activity of the prepared TiO2 samples and calcination temperature in the degradation of NOx was further examined. As shown in Table 2, the removal rates of NOx over photocatalysts calcined at different temperatures were measured under blue, green, and red light from LED sources. The sample calcined at 200 °C exhibited the strongest activity for NOx degradation in visible light. The activity in visible light clearly decreases as the calcination temperature increased above 200 °C, which is consistent with the measured visible light absorbance. The optimum calcination temperature appeared to be 200 °C, which was close to the value reported in the literature (16), 250 °C. Below 200 °C, the carbonaceous species did not become highly dense coke and did not act as sensitizers that induced a response to visible light. At a higher calcination temperature, the carbonaceous species were burnt out and so the ability to absorb visible light was eliminated. Therefore,
TABLE 2. Removal Rate of NOx (µmol/h) over Sol-Gel Synthesized Photocatalysts Calcined at Various Temperatures in Visible Light from LEDsa
TABLE 3. Removal Rate of NOx (µmol/h) over Pt Ion-Doped TiO2 Calcined at Various Temperatures in Visible Light from LEDsa
calcination temperature (°C)
calcination temperature (°C)
irradiation
150
200
250
300
400
500
600
irradiation
UV100
150
200
300
400
500
600
blue LED green LED red LED
1.71 1.62 0.73
1.74 1.71 1.42
1.35 1.35 0.71
0.86 0.73 0.32
0.37 0.24 0.10
0.34 0.17 0.07
0.24 0.12 0.05
blue LED green LED red LED
0.39 0.17 0.05
1.59 1.32 0.34
2.08 1.96 1.47
1.22 0.83 0.49
1.15 0.86 0.54
0.98 0.66 0.49
0.29 0.20 0.12
a Catalyst loading, 0.5 g; irradiation intensity, 1 mW/cm2; relative humidity, 50%; feeding concentration of NO, 1 ppm; feeding flow rate, 1 L/min; reaction temperature, rt; time on stream, 2 h.
a Catalyst loading, 0.5 g; irradiation intensity, 1 mW/cm2; relative humidity, 50%; feeding concentration of NO, 1 ppm; feeding flow rate, 1 L/min; reaction temperature, rt; time on stream, 2 h.
TABLE 4. Effect of Calcination Temperature on the Calculated Crystallite Size of Pt Ion-doped TiO2 calcination temp. (°C) (UV100) crystallite size (nm)
9.6
150
200
300
400
500
600
10.0 10.3 13.6 18.5 22.2 24.4
TABLE 5. Effect of Oxidation State of Pt on the Removal Rate of NOx (µmol/h) in Monochromatic UV and Visible Lighta wavelength, nm
FIGURE 4. UV-vis absorption spectra of Pt ion-doped TiO2 (by impregnation on UV100) calcined at various temperatures. an optimum calcination temperature of 200 °C was applied in the preparation of a TiO2 that responded to visible light by the sol-gel process. Response of Pt Ion-Doped TiO2 to Visible Light. Transition metal ion-doped TiO2 has been extensively studied with a view to creating a sub-band gap that absorbs visible light, particularly using ion-implantation (8, 9). Additionally, vanadium or cobalt ion-doped TiO2 photocatalyst that is responsive to visible light has been prepared by the sol-gel process (11, 17). In this work, a platinum ion-doped TiO2 was chemically prepared by impregnation using Pt(NH3)4(NO3)2 as a dopant. Following impregnation, the TiO2 sample had the same anatase crystal structure as the undoped TiO2, UV100, according to XRD analysis. Its absorbance of visible light was found to be slightly higher than that of the undoped TiO2. As shown in Figure 4, the extent of absorption increased with the calcination temperature up to 500 °C, but was drastically lower when the calcination temperature was 600 °C. The photocatalytic activity of the impregnated TiO2 sample in degrading NOx was examined with regard to its dependence on calcination temperature. Table 3 presents the removal rates of NOx over Pt ion-doped TiO2 calcined at different temperatures were measured in blue, green, and red light from LEDs. The sample calcined at 200 °C exhibited the strongest activity for NOx degradation in the visible light region. This result is not consistent with the result for the absorption of visible light that the sample calcined at 500 °C exhibits the highest absorbance, as presented in Figure 3. This result may be explained by the increase in the size of the crystals with the temperature of calcination. XRD analysis indicates that the crystals of TiO2 gradually aggregated at a calcination temperature of over 300 °C. Table 4 shows that the calculated crystallite size was as high as 22 nm at a calcination temperature of 500 °C, the value of which is about double the original size (9.6 nm). The growth of the crystals
photocatalyst
365
400
435
500
546
Pt ion (impregnated)/UV100 Pt0 (photodeposited)/UV100 Pt0 (H2 reduced)/UV100
0.99 0.98 0.96
0.94 0.86 0.86
0.93 0.69 0.64
0.84 0.07 0.12
0.83 0.05 0.05
a Catalyst loading, 0.3 g; irradiation intensity, 1 mW/cm2; relative humidity, 50%; feeding concentration of NO, 1 ppm; feeding flow rate, 1 L/min; reaction temperature, rt.
may suppress the photocatalytic activity associated with the degradation of NOx, despite the stronger absorption of light. The Pt species that impregnated the surface of TiO2 were readily oxidized during calcination in air. The analyses of temperature programming reduction (TPR) and X-ray photoelectron spectroscopy (XPS) revealed that the formation of platinum oxide although the exact oxidation state of Pt ions was not thus determined. The oxidation state of the Pt ions might affect the activity of the modified TiO2 sample in response to visible light. Pt metal (zero oxidation state) dispersed on the surface of TiO2 was prepared by photodeposition to elucidate the influence of oxidation state (18). The activity associated with the degradation of NOx was measured under monochromatic irradiation with wavelengths of 365, 400, 435, 500, and 546 nm. Table 5 shows that the photodeposited Pt/TiO2 had much poorer activity than Pt ion-doped TiO2 prepared by impregnation, particularly at wavelengths of longer than 500 nm. The same behavior was observed when the platinum oxide at the surface of TiO2 was reduced at 200 °C in a 10% hydrogen stream (with nitrogen as the carrier gas) before the activity was measured. This result indicates that the high oxidation state of Pt ions used to dope the surface of the TiO2 might be responsible for the activity in response to visible light. Degradation of NOx on TiO2 Photocatalysts that Respond to Visible Light. The TiO2 photocatalysts clearly responded to visible light when chemically prepared by either the solgel process or impregnation. Combining the sol-gel process with the impregnation method offers better photocatalytic activity for degrading NOx. Table 6 shows the experimental results obtained following various combinations of approaches to preparation. Three commercially available photocatalysts (UV100, ST01, and P25) were used as reference for the degradation of NOx. The activities of photocatalyst irradiated at λ ) 365 and 400 nm were similar in all cases. However, the activities of the three commercial photocatalysts VOL. 40, NO. 5, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1619
TABLE 6. Removal Rate of NOx (µmol/h) over Various Photocatalysts in Light from Various Sourcesa wavelength, nm catalyst
365
400
435
500
546
fluorescent lampb
mercury lampc
Cs Csp Csi Cui UV 100 ST01 P25
1.05 0.93 0.96 0.99 0.92 0.93 0.96
0.88 0.88 0.96 0.94 0.81 0.93 0.86
0.86 0.86 0.93 0.93 0.61 0.54 0.61
0.61 0.47 0.88 0.84 0 0 0
0.51 0.47 0.85 0.83 0 0 0
0.91 0.85 1.00 0.97 0.47 0.45 0.42
0.99 0.98 1.02 1.08 1.00 0.98 0.99
a Catalyst loading, 0.3 g; irradiation intensity, 1 mW/cm2; relative humidity, 50%; feeding concentration of NO, 1 ppm; feeding flow rate, 1 L/min; reaction temperature, rt; time on stream, 2 h. b Range of wavelengths, 400-670 nm; irradiation intensity, 1.2 mW/cm2 (calculated from the spectral analysis). c Range of wavelengths, 350-600 nm; irradiation intensity, 1.5 mW/cm2 (calculated from the spectral analysis).
in the visible light region (λ ) 435-546 nm) were much lower than those of the synthesized photocatalysts. Moreover, NOx was not photodegraded by commercial catalysts when irradiated at λ ) 500 and 546 nm. As shown in Table 6, the photocatalyst prepared by the sol-gel process followed by impregnation by Pt(NH3)4(NO3)2 as a dopant, hereafter Csi, exhibited the highest activity for the degradation of NOx in visible light. The conventional TiO2 photocatalyst (UV100) modified by impregnation of its surface with dopant Pt ion, Cui, also exhibited good activity. The impregnation method enhanced activity in response to visible light, especially at wavelengths of longer than 500 nm, more efficiently than the sol-gel process with Pt ion doped in the bulk phase (Csp). The catalyst prepared by the sol-gel process only (Cs) exhibited lower activity than that prepared by impregnation (either Csi or Cui), but stronger activity than that prepared by doping the bulk phase with Pt ions (Csp). Although the catalyst Csp was dark gray and absorbed visible light more effectively, it exhibited a lower photocatalytic activity, especially when irradiated with light with a wavelength of 500 or 546 nm, because only a small fraction of the Pt ions could be dispersed on the surface of TiO2. The results indicate that doping with Pt ions may favor the activity of the photocatalyst in visible light, and particularly the activity of the photocatalyst prepared by the impregnation method. Fluorescent and mercury lamps were used to simulate indoor and outdoor illumination conditions, respectively. Table 6 presents the experimental results thus obtained. All the light emitted from the fluorescent lamp had a wavelength between 400 and 670 nm, except for a very small peak at 365 nm. The removal rates of NOx over the prepared photocatalysts were almost double those over commercial catalysts under fluorescent irradiation, indicating that the indoor lighting increases the photocatalytic efficiency of a photocatalyst that responds to visible light. When illuminated under a mercury lamp, which emitted light with wavelengths between 350 and 600 nm, the activities of the synthesized catalysts were almost the same as those of catalysts illuminated under a fluorescent lamp, indicating that the photocatalyst that responds to visible light provides almost the same reactivity under indoor and outdoor conditions. Therefore, the photocatalysts that respond to visible light may extend the application in destruction of NOx from outdoor condition to indoor condition. Selectivity of NO2 during Degradation of NOx. The photocatalytic oxidation of NO over TiO2 catalyst involves a series of oxidation steps, NO(g) f NO2(g) f HNO3(ads). Although 1620
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 5, 2006
FIGURE 5. Concentrations of NO, NO2, and NOx versus wavelength of light during the degradation of NO over catalyst Cs. Catalyst loading, 0.3 g; irradiation intensity, 1 mW/cm2; relative humidity, 50%; feeding concentration of NO, 1 ppm; feeding flow rate, 1 L/min; reaction temperature, rt; time on stream, 2 h. nitrite ions (NO2-) are formed as an intermediate during the reaction, their rate of formation is negligible by comparison with the rate of formation of nitrate ions (NO3-). The NO3adsorbed on the TiO2 surface is easily eliminated by washing with DI water (3, 6). Before the adsorbed species were thus eliminated, the photocatalyst was deactivated by the storage of nitrate ions and the coverage of active sites on the surface of the catalyst (19). According to the aforementioned reaction scheme, the degradation of NOx (NO + NO2) involves two consecutive oxidation steps, the oxidation of NO to NO2 followed by the further oxidation of NO2 to NO3-. The measured concentrations of NO and NO2 in the reaction system are influenced by the oxidizing capacity of the catalyst and the photoenergy of the light. Figure 5 shows the concentration profiles of NO, NO2, and NOx, depending on the wavelength of irradiation used to degrade NOx over the catalyst Cs. The figure shows that the NO2 concentration gradually increases with the wavelength of the irradiation. Meanwhile, the concentration of NO remains invariant, indicating that a higher photoenergy, and thus a shorter wavelength, are required to activate the further oxidation of NO2 to NO3-. Therefore, the increase of NO2 production in visible light is mainly responsible for the decrease in the removal rate of NOx. NO2 is an undesired intermediate in the consecutive photooxidation of NO because the toxicity of NO2 exceeds that of NO (threshold concentration of NO2 ) 3 ppm, while that of NO ) 25 ppm (20). The selectivity of NO2 is defined as [NO2]/[NO]inlet, to elucidate the effect of the species of photocatalyst and the wavelength of light on the formation of NO2, where [NO2] is the concentration of NO2 produced and [NO]inlet is the concentration of NO in the feeding stream. As shown in Figure 6, the selectivity of NO2 over catalyst Cs clearly increased with the wavelength of the radiation. The selectivities of NO2 over catalysts Csp, Csi, and Cui were measured to be lower than that over Cs. At a constant loading of Pt ions, the selectivity of NO2 over both Csi and Cui was much lower than that over Csp. For example, when irradiated at λ ) 500 nm, the selectivity of NO2 over either Csi or Cui was around 2%, markedly lower than the 10% selectivity over Csp. The experimental results indicate that the doping of Pt ions at the surface of TiO2 enhanced the photocatalytic activity in the oxidation of NO2 to NO3-, and suppressed the production of NO2 during the degradation of NOx.
Acknowledgments We thank the Ministry of Economic Affairs of the Republic of China, Taiwan, for financially supporting this research.
Literature Cited
FIGURE 6. Selectivity of NO2 versus wavelength of light over various photocatalysts. Catalyst loading, 0.3 g; irradiation intensity, 1 mW/ cm2; relative humidity, 50%; feeding concentration of NO, 1 ppm; feeding flow rate, 1 L/min; reaction temperature, rt; time on stream, 2 h.
TABLE 7. Influence of Regeneration Cycle Time on the Removal Rate of NOx (µmol/h) Over the Sheet that Contained Various Photocatalysts in Monochromatic UV and Visible Lighta UV100
Csi
Cui
cycle time
365 nm
500 nm
365 nm
500 nm
365 nm
500 nm
0 (fresh) 1 2 3
2.93 2.83 2.80 2.80
0.03 0.03 0 0
3.00 2.93 2.83 2.88
2.28 2.20 2.13 2.13
2.95 2.93 2.95 2.78
2.20 2.13 2.10 2.05
a Area of sheet, 50 cm2; catalyst loading, 20 mg/cm2; relative humidity, 60%; feeding flow rate, 3 L/min; feeding concentration of NO, 1 ppm; reaction temperature, rt; irradiation intensity, 1 mW/cm2; time on stream, 4 h.
Regeneration of Photocatalyst. The reaction scheme of the photooxidation of NO reveals that the photocatalyst is deactivated after long-term operation because the NO3- ions adsorbed onto the surface of the catalyst covers the active sites. However, the adsorbed NO3- ions are easily rinsed away with DI water, enabling the catalyst to be regenerated. Table 7 presents the removal rates of NOx during the cycle of degradation and regeneration. The experiments were conducted in light with wavelengths of λ ) 365 and 500 nm. Three photocatalysts (UV100, Csi, and Cui) were coated on a nonwoven fabric sheet to measure their activity. The experimental results showed that over 90% of the photocatalytic activity was recovered after three cycles of degradation and regeneration, indicating the Pt ion-doped TiO2 prepared by impregnation was highly stable for degrading NOx. The overall results of this work indicate that titania-based photocatalysts that are responsive to either UV or to visible light exhibit high reactivity and good stability and can be simply regenerated. These characteristics indicate the great potential of these photocatalysts to be used in air purification, to promote cleanliness and energy-saving.
(1) Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change; John Wiley and Sons: New York, 1998. (2) Cooper, C. D.; Alley, F. C. Air Pollution Control: A Design Approach; Waveland Press: Prospect Heights, IL, 1994. (3) Devahasdin, S.; Fan, C.; Li, K.; Chen, D. H. TiO2 photocatalytic oxidation of nitric oxide: transient behavior and reaction kinetics. J. Photochem. Photobiol. A: Chem. 2003, 156, 161170. (4) Latta, C. A.; Weston, R. F. Methods for reducing NOx emissions. Plant Eng. 1998, 52, 105-112. (5) Delahay, G.; Coq, B. Pollution abatement using zeolites: state of the art and further needs. In Zeolites for Cleaner Technologies; Guisnet, M., Gilson, J., Eds.; Imperial College Press: London, 2002. (6) Ibusuki, T.; Takeuchi, K. Removal of low concentration nitrogen oxides through photoassisted heterogeneous catalysis. J. Mol. Catal. 1994, 88, 93-102. (7) Nakamura, I.; Negishi, N.; Kutsuna, S.; Ihara, T.; Sugihara, S.; Takeuchi, K. Role of oxygen vacancy in the plasma-treated TiO2 photocatalyst with visible light activity for NO removal. J. Mol. Catal. A: Chem. 2000, 161, 205-212. (8) Anpo, M.; Takeuchi, M. The design and development of highly reactive titanium oxide photocatalysts operating under visible light irradiation. J. Catal. 2003, 216, 505-516. (9) Yamashita, H.; Anpo, M. Application of an ion beam technique for the design of visible light-sensitive, highly efficient and highly selective photocatalysts: ion-implantation and ionized cluster beam methods. Catal. Surv. Asia 2004, 8, 35-45. (10) Kisch, H.; Macyk, W. Visible-light photocatalysis by modified titania. ChemPhysChem 2002, 3, 399-400. (11) Wu, J. C. S.; Chen, C. H. A visible-light response vanadiumdoped titania nanocatalyst by sol-gel method. J. Photochem. Photobiol. A: Chem. 2004, 163, 509-515. (12) Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Visiblelight photocatalysis in nitrogen-doped titanium oxides. Science 2001, 293, 269-271. (13) Zang, L.; Lange, C.; Abraham, I.; Storck, S.; Maier, W. F.; Kisch, H. Amorphous microporous titania modified with platinum(IV) ChloridesA new type of hybrid photocatalyst for visible light detoxification. J. Phys. Chem. B 1998, 102, 1076510771. (14) Macyk, W.; Kisch, H. Photosensitization of crystalline and amorphous titanium dioxide by platinum(IV) chloride surface complexes. Chem. Eur. J. 2001, 7, 1862-1867. (15) Zang, L.; Macyk, W.; Lange, C.; Maier, W. F.; Antonius, C.; Meissner, D.; Kisch, H. Visible-light detoxification and charge generation by transition metal chloride modified titania. Chem. Eur. J. 2000, 6, 379-384. (16) Lettmann, C.; Hildenbrand, K.; Kisch, H.; Macyk, W.; Maier, W. F. Visible light photodegradation of 4-chlorophenol with a cokecontaining titanium dioxide photocatalyst. Appl. Catal. B: Environ. 2001, 32, 215-227. (17) Iwasaki, M.; Hara, M.; Kawada, H.; Tada, H.; Ito, S. Cobalt iondoped TiO2 photocatalyst response to visible light. J. Colloid Interface Sci. 2000, 224, 202-204. (18) Sano, T.; Kutsuna, S.; Negishi, N.; Takeuchi, K. Effect of Pdphotodeposition over TiO2 on product delectivity in photocatalytic degradation of vinyl chloride monomer. J. Mol. Catal. A: Chem. 2002, 189, 263-270. (19) Negishi, N.; Takeuchi. K.; Ibusuki, T. Surface structure of the TiO2 film photocatalyst. J. Mater. Sci. 1998, 33, 5789-5794. (20) Lewis, R. J. Sax’s Dangerous Properties of Industrial Materials, 9th ed.; Van Nostrand Reinhold: New York, 1996; Vols. 1-3.
Received for review May 29, 2005. Revised manuscript received October 18, 2005. Accepted December 14, 2005. ES051007P
VOL. 40, NO. 5, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1621
