Ind. Eng. Chem. Res. 2007, 46, 7-14
7
Photocatalytic Degradation of Rhodamine Dyes with Nano-TiO2 T. Aarthi and Giridhar Madras* Department of Chemical Engineering, Indian Institute of Science, Bangalore 560012, India
The photocatalytic degradation of various Rhodamine dyes with different functional groups, namely, Rhodamine B (C28H31ClN2O3), Rhodamine 6G (C28H31ClN2O3), Rhodamine Blue (C28H32N2O3), and Rhodamine 6G perchlorate (C28H31ClN2O7) was investigated under UV irradiation. The kinetics of degradation was determined in the presence of two catalysts: commercial Degussa P-25 and TiO2 synthesized by the combustion solution method. The effect of organic solvents (ethanol and acetonitrile) and metal ions (Cu2+, Fe3+, Zn2+, and Al3+) on the photodegradation of Rhodamine B was investigated. The presence of solvents and metal ions significantly reduced the degradation rate. A detailed Langmuir-Hinshelwood kinetic model was developed to explain the effect of Cu2+ on the rate of photodegradation. It was shown that the degradation rate reduces primarily because of a decrease in electron concentration. Introduction Photocatalysis is an advanced oxidation process (AOP) that can be used for the destruction of pollutants in a simple and efficient manner. It is one of the most economic and promising industrial effluent treatment processes. Among photocatalytic materials, TiO2 has been extensively used in sterilization,1 sanitation, and remediation applications,2,3 as well as air purification and water treatment. TiO2 has proven to be an excellent photocatalyst material by which many organic substrates have been shown to be oxidatively (in some cases reductively) degraded and undergoes complete mineralization under UV exposure.4 A detailed analysis of electronic and charge-transfer processes occurring during heterogeneous photocatalysis on TiO2 has been discussed.5,6 In an aqueous environment, the holes created under UV irradiation are scavenged by the hydroxyl groups present on the surface, generating OH• radicals,7 which promote the oxidation of organics. This has been successfully employed in the mineralization of several hazardous chemicals such as dyes,5,6,8,9 phenols, haloaromatics, halogenated biphenyls, etc.1 The mode of synthesis of TiO2 influences the photocatalytic activity of the catalyst (i.e., band gap, bounded hydroxyl species, crystallinity, and particle size). The anatase-phase nano-titania (TiO2) prepared by the solution combustion method has been reported to have better photocatalytic activity compared to the commercial Degussa P-25 catalyst.10 Rhodamine dyes belong to a class of dyes called Xanthene dyes and are used as dye laser materials.11 Dye-containing industrial wastewater contains organic solvents, dissolved inorganic ions, and humic substances.8,12 Various studies report the retardation effect of inorganic salts on the photodegradation of pollutants, such as NaCl on phenol degradation;13 NaCl, Na2SO4, etc. on azo dyes;14 and transition metal ions such as Cu2+ and Fe3+ on several dyes.15 These constituents compete for the active sites of the catalyst, leading to reduced photodegradation of pollutants. Organic solvents are also encountered in dye wastewater and other industrial waste streams.8 Therefore, the understanding of the effect of solvents and metal ions on the degradation rate is important. The present study aims at exploring the photocatalytic activity of commercial TiO2 (Degussa P-25) and solution combustion* To whom correspondence should be addressed. Tel.: +91-802293-2321.Fax:+91-80-2360-0683.E-mail:
[email protected].
synthesized TiO2 (CS TiO2) for the photodegradation of Rhodamine dyes. To compare the photocatalytic efficiency of these catalysts for dyes with similar structures but different functional groups, the degradation of Rhodamine dyes, namely, Rhodamine B (RB), Rhodamine Blue (RBL), Rhodamine 6G (R6G), and Rhodamine 6G perchlorate (R6GPC) was investigated. The effect of solvents like ethanol, acetonitrile, and the mixture of these solvents with water and metal ions (like Cu2+, Fe3+, Zn2+, and Al3+) on the rate of photocatalytic degradation of Rhodamine B in the presence of CS TiO2 has been investigated. A detailed Langmuir-Hinshelwood (LH) kinetic model to explain the photocatalytic degradation in the presence of metal ions is not available in the literature and has been proposed in this work. The photocatalytic degradation rate constants were determined by fitting the model to experimental data. Experimental Section Materials. The dyes, Rhodamine B (RB, C28H31N2O3Cl, CAS: 81-88-9), Rhodamine Blue (RBL, C28H32N2O3, CAS: 1326-03-0), Rhodamine 6G (R6G, C28H31 ClN2O3, CAS: 98938-8), the salts, cupric nitrate, ferric nitrate, aluminum nitrate, zinc nitrate, and nitric acid were purchased from S. D. Fine Chemicals (India). Rhodamine 6G perchlorate (R6GPC, C28H31ClN2O7, CAS: 13161-28-9) and sodium diethyldithiocarbamate were purchased from Sigma-Aldrich, U.S.A. Titanium isopropoxide (Lancaster Chemicals, U.K.) and glycine (Merck, India) were used in the preparation of catalyst. Double-distilled water was filtered through a Millipore membrane filter before use. Catalyst Preparation. The solution combustion method16 was used to prepare nanosized anatase TiO2. The precusor titanyl nitrate [TiO(NO3)2] and the fuel glycine (H2N-CH2-COOH) were used in this method. The precursor titanyl nitrate was synthesized as follows: titanyl hydroxide [TiO(OH)2] was obtained by the hydrolysis of titanium isopropoxide [Ti(i-OPr)4]. Titanyl nitrate was obtained by the reaction of titanyl hydroxide with nitric acid. In a typical combustion synthesis, a Pyrex dish (with a volume of 300 cm3) containing an aqueous redox mixture of stoichiometric amounts of titanyl nitrate and glycine in 30 mL of water was introduced into a muffle furnace that was preheated at 350 °C. The solution initially undergoes dehydration, and a spark appears at one corner, which spreads throughout the mass, finally yielding anatase titania. Thus, TiO2 was formed by the complete combustion of the titanyl-glycine
10.1021/ie060948n CCC: $37.00 © 2007 American Chemical Society Published on Web 12/05/2006
8
Ind. Eng. Chem. Res., Vol. 46, No. 1, 2007
redox mixture. The liberation of the large volumes of the gases leads to the high porosity and high surface area of the material. Catalyst Characterization. The catalyst has been characterized by various techniques such as XRD, transmission electron microscopy (TEM), Brunauer-Emmett-Teller (BET), thermogravimetric-differential thermal analysis (TG-DTA), X-ray photoelectron spectroscopy (XPS), IR, and UV spectroscopy. The X-ray diffraction (XRD) patterns of catalysts were recorded on a Siemens D-5005 diffractometer using Cu KR radiation with a scan rate of 2° min-1. The XRD pattern of combustionsynthesized TiO2 was recorded in 2θ range from 5 to 100°. The pattern can be indexed to the pure anatase phase of TiO2 with the space group of I41/amd. The data were then refined using the Fullprof-98 program. There was a good agreement between the calculated pattern and the observed pattern. The lattice parameter for TiO2 is a ) 3.7865(5) Å and c ) 9.5091(1) Å. The crystallite size was determined from the XRD pattern using the Sherrer formula and based on the full width half-maxima (fwhm) of X-ray diffraction pattern; the mean crystallite size is estimated to be 10 ( 2 nm. Transmission electron microscopy (TEM) of powders was carried out using a JEOL JEM-200CX transmission electron microscope operated at 200 kV. TEM studies also showed that the crystallites of TiO2 are homogeneous with the mean size of 8 ( 2 nm, which agrees well with the XRD measurements. The surface area of the catalyst was determined with standard BET apparatus (NOVA-1000, Quantachrome) and was 240 m2/g and higher than the surface area of commercial catalysts like Degussa P-25 (50 m2/g). Fourier transform infrared (FTIR) studies were carried out in the 4004000 cm-1 frequency range in the transmission mode (PerkinElmer, FTIR-Spectrum-1000) and showed higher surface hydroxyl content for the combustion-synthesized TiO2. The assynthesized TiO2 was subjected to thermogravimetric-differential thermal analysis (TG-DTA) (Perkin-Elmer, Pyris Diamond), which showed an 11% weight loss, indicating more surface hydroxyl groups. X-ray photoelectron spectra (XPS) of these materials were recorded with ESCA-3 Mark II spectrometer (VG Scientific Ltd., U.K.) using Al KR radiation (1486.6 eV). UVvis absorption spectra of TiO2 powders were obtained for the dry-pressed disk samples using a UV-vis spectrophotometer (GBC Cintra 40, Australia) between 270 and 800 nm range. The combustion-synthesized TiO2 shows two optical absorption thresholds at 570 and 467 nm that correspond to the band gap energies of 2.18 and 2.65 eV, respectively. Further details are presented elsewhere.10,16 Photochemical Reactor. The photochemical reactor employed in this study comprised a jacketed quartz (GE Type 219, which has a UV cutoff at about 230 nm) tube of 3.4 cm i.d., 4 cm o.d., and 21 cm length with an outer Pyrex glass reactor of 5.7 cm i.d. and 16 cm length. The UV light was provided by a 125 W high-pressure mercury vapor lamp (Philips, India) that was used after removal of the outer shell and placed inside the jacketed quartz tube. The ballast and capacitor were connected in series with the lamp to avoid fluctuations in the input supply. Water was circulated through the annulus of the quartz tube to avoid heating of the solution because of dissipative loss of UV energy. The solution was taken in the outer reactor and continuously stirred using a magnetic stirrer to ensure that the suspension of the catalyst was uniform during the course of the reaction. The lamp radiated predominantly at 365 nm, corresponding to an energy of 3.4 eV. Further details of the experimental setup can be found elsewhere.16,17 Degradation Experiments. During the degradation of each of the dyes, a known mass of the dye was dissolved in Millipore-
filtered double-distilled water and subjected to UV irradiation in the photochemical reactor described above with a catalyst loading of 1 g/L. The reactions were carried out at 40 °C, which was maintained by circulating water in the annulus of the jacketed quartz reactor. Samples were collected at regular intervals, filtered through Millipore membrane filters, and centrifuged to remove the catalyst particles prior to analysis. Sample Analysis. The molar absorptivity of the Rhodamine dyes employed is measurably high for even dilute solutions. Hence, photometric studies were carried out for the dyes used. The UV-vis spectrophotometer (Shimazdu, UV 2100) with quartz cuvettes was used for the determination of color intensity in the range of 190-700 nm. The λmax for RB and RBL are 553 and 665 nm, respectively, whereas the λmax for both R6G and R6GPC is 525 nm. The UV spectrum before and after degradation showed a decrease of peak corresponding to the dye, and the solution became colorless after catalytic degradation. Calibration based on the Beer-Lambert law was used to quantify the dye concentration. Results and Discussion The photocatalytic degradation of four different types of Rhodamine dyes was investigated in the absence of catalyst and in the presence of two catalysts: combustion-synthesized TiO2 (CS TiO2) and commercial TiO2 (Degussa P25). All the experiments were conducted at the natural pH of the dye with a catalyst concentration of 1 g L-1. After the addition of the catalyst, the dye solution was stirred for 30 min to ensure that the equilibrium adsorption/desorption of the dye on the catalyst was attained. The corresponding concentration of the dye (as measured by UV spectrophotometer and the concentration evaluated using Beer-Lambert’s law) was taken as the initial concentration of the dye for all the catalyzed reactions. Parts a-d of Figure 1 show the variation of concentration profile of various dyes, at different initial concentrations, in noncatalysed and TiO2 catalyzed (both Degussa TiO2 and CS TiO2) systems. The initial rates of the reaction were determined by extrapolating the tangent (based of the linear fit of the first four points) of the concentration profile back to initial conditions. The slopes calculated at the initial 3-6 points were nearly constant, indicating the accuracy of the initial rates reported in this study. The model is developed by following the photocatalytic mechanism, which is well-understood and reported.4-6,16 The positive holes and electrons generated by UV on photocatalyst are involved in the formation of the hydroxyl radicals. The reaction between the positive holes and the adsorbed water forms hydroxyl species. A series of steps is involved in the formation of OH• by electron pathway. The dye degrades by the attack of direct hole and hydroxyl species. The presence of Cu2+ reduces the concentration of e- because of the reduction of Cu2+ to Cu+ by e-. Consequently, this scavenging reduces the formation of OH• by the e- pathway (though the formation of OH• by the hole pathway remains unaffected). The kinetic model reported in this work includes the generation of OH• radicals via the electron pathway for the photocatalytic degradation of dyes, and it is an extension of the model reported in a previous study.16 In this work, we have included this pathway and the scavenging of electrons by metal to derive the rate equation in terms of dye and the concentration of retardant Cu2+. The detailed mechanism and kinetics are discussed in Appendix A.
Ind. Eng. Chem. Res., Vol. 46, No. 1, 2007 9
Figure 1. Concentration profiles of (a) Rhodamine B, (b) Rhodamine 6G, (c) Rhodamine 6G perchlorate, (d) Rhodamine Blue [legend for all panels: (O) without catalyst, (X) Degussa P25, (b) CS TiO2].
The stoichiometric balances for all the species using the mechanism and applying the quasi-steady-state assumption (QSSA) for all the intermediate species yields
-rD ) koh[D] + k1[D]
(
k3
1 + K2[D]
+
(
k4
1 + K2[D]
(
)
×
1 1 + K6[C2+]
))
(1)
Equation 1 gives the relation between the photocatalytic degradation rate, rD, and the concentrations of dye [D] and retardant metal [C2+] and indicates that the competitive adsorption of the metal lowers the rate of photocatalysis. The rate constants, k1k3 and k1k4, refer to the generation of OH• by hole and electron pathway, respectively. Taking the inverse of eq 1 and neglecting the quadratic term of [D],
(
)
(1 + K6[C2+]) 1 1 ) + K2 rD [D] (k0 + K6k7[C2+])
(2)
where k0 ) koh + k1k3 + k1k4 and k7 ) koh + k1k3 When no metal ion is present, [C2+] ) 0, the variation of the inverse of the initial rate is
(
)
1 1 1 ) + K2 r′D,0 k0 [D]
(3)
Thus, the inverse of initial rates, rD,0, varies linearly with the inverse of the initial concentration [D] for various dyes, as
shown in parts a-d of Figure 2. The values of k0 and K2 are obtained from linear regression. Control experiments without catalyst showed only a small decrease in dye concentration (as shown in the figures), with the initial rates of ∼0.003 mg L-1 min-1 for all dyes. Table 1 shows the estimated rate constants for the photodegradation of Rhodamine dyes in both catalyzed and noncatalyzed systems. The concentration of the dyes reduces only by 10-20% at very long times in the absence of catalysts, indicating that the rate of the noncatalytic reaction is very slow. Rhodamine B (RB) and Rhodamine 6G (R6G) have the same molecular weight with similar functional groups but different end groups. Rhodamine B, Rhodamine Blue (RBL), and R6G perchlorate (R6GPC) have similar structures, with one H of RBL substituted by Cl in RB and by perchlorate in R6GPC. In the case of catalytic degradation with Degussa P25, RB and R6G show similar adsorption, while the rate constant k0 and the rates are nearly four times higher in R6G compared to RB. In the case of catalytic degradation with CS TiO2, the degradation rates of both these dyes are similar. In the case of RBL and R6GPC, the degradation rates are significantly higher in the case of Degussa P25 compared to that of CS TiO2. From the results, it is found that the equilibrium adsorption coefficient for RB, R6G, and R6GPC is higher on CS TiO2 compared to that on Degussa catalyst. This is consistent with the higher surface area of CS TiO2 compared to that of Degussa. The initial photocatalytic rates for the degradation of RB and R6G in the presence of CS TiO2 were higher by 19.7 and 5.7 times, respectively, compared to that observed in the presence of Degussa P25. This could be attributed to the lower band gap value (2.18 and 2.6 eV for CS TiO2, 3.1 eV for Degussa catalyst),10 the higher hydroxyl content
10 Ind. Eng. Chem. Res., Vol. 46, No. 1, 2007
Figure 2. Variation of the initial rates and initial dye concentration for (a) Rhodamine B, (b) Rhodamine Blue, (c) Rhodamine 6G, (d) Rhodamine 6G perchlorate [legend for all panels: (O) without catalyst, (X) Degussa P25, (b) CS TiO2]. Table 1. Rate Constants for Photodegradation of Different Rhodamine Dyes in Catalyzed and Noncatalyzed System dye Rhodamine B Rhodamine Blue Rhodamine 6G Rhodamine 6GPC
catalyst no catalyst Degussa TiO2 CS TiO2 no catalyst Degussa TiO2 CS TiO2 no catalyst Degussa TiO2 CS TiO2 no catalyst Degussa TiO2 CS TiO2
k0 (min-1)
K2 (L mg-1)
0.0024 0.0146 0.0190 0.2827 0.3080 0.0027 dye adsorption very high 0.0072 0.0288 0.0029 0.0590 0.0210 0.3330 0.4410 0.0028 0.0363 0.0860 0.0163 0.3975
(as shown by TG and IR studies done previously),10 and the smaller size/higher surface area of CS TiO2 compared to that of Degussa TiO2 (240 m2/g for CS TiO2, 50 m2/g for Degussa TiO2).10 The enhancement of photocatalytic activity due to increased hydroxyl groups and decreased particle size (size quantization effect) has been reported earlier.16 The rates of degradation for 10 mg L-1 solutions of RB, R6G, and R6GPC, respectively, are 0.122, 0.487, and 0.195 mg L-1 min-1 and 0.692, 0.615, and 0.033 mg L-1 min-1 when catalyzed by Degussa P25 and CS TiO2, respectively. This indicates that the degradation is selective and depends on the dye and the catalyst. The industrial effluents often contain some amount of aromatic solvents. Therefore, the effect of solvents on the photodegradation using Degussa TiO2 has been well-studied.8,18 The effect of solvents on the photodegradation of RB in the presence of CS TiO2 was investigated. In the mixture of ethanol and water, when ethanol was 10% by volume, the rate decreased by 91.5%, and when ethanol was increased to 50% by volume,
Figure 3. Concentration profiles for the degradation of Rhodamine B in acetonitrile and ethanol in water with CS TiO2.
no photodegradation was observed. Similarly, the rate decreased by 76, 96, and 100% when acetonitrile in the mixture of water and acetonitrile was increased from 10, 20, and 50% by volume, respectively (Figure 3). The detrimental effect of solvents on the photodegradation is attributed to the lesser solvation of excited electrons in the organic solvents compared to the aqueous solutions.19 The lesser solvation of excited electrons leads to the increased possibility of recombination with the positive holes leading to reduction in photocatalysis rates. Trace amounts of metal ions are present in polluted industrial water.11 When the concentration of the salt was 100 µM, the presence of metal ions like Cu2+, Al3+, Fe3+, and Zn2+ reduced the initial rates of photocatalytic degradation of RB in the presence of CS TiO2, by 81.5, 81, 70, and 76%, respectively
Ind. Eng. Chem. Res., Vol. 46, No. 1, 2007 11
sodium diethyldithiocarbamate. The method of analysis is discussed elsewhere.21 The variation of the initial rates of photodegradation of RB and R6G and the amount of Cu2+ adsorbed per unit weight of the catalyst, at various concentrations of Cu2+ in the solution, is illustrated in Figure 6. As the concentration of Cu2+ in the solution increases from 0 to 24 mg L-1, the amount of adsorbed Cu2+ increases and, correspondingly, there is a significant decrease in the initial degradation rate of photocatalysis. This metal retards the photodegradation rate for both dyes investigated, which is in agreement with the experimental results reported previously by Chen et al.15 This illustrates that the adsorption of Cu2+ correlates with the degradation rates of RB and R6G, and retardation is due to Cu2+ ions, which scavenge the electrons required for the formation of hydroxyl radicals. Figure 4. Concentration profiles for the degradation of Rhodamine B in the presence of various salts with CS TiO2.
(see Figure 4). In a previous study,15 it was shown that the presence of metal ions, Cu2+and Zn2+, reduced the rates for the degradation of Malachite Green by 76 and 66%, respectively. When such ions are present, the surface of TiO2 gets modified because of the adsorption of cations or anions.15 The interfacial charge-transfer processes, which play a major role in photodegradation, depend on the surface characteristics of the TiO2 particles. The effect of transition metal ions on the TiO2 assisted photodegradation has been studied experimentally in detail previously.15 The detrimental effect of metal ions on photodegradation rates was shown to be due to the suppression of OH• radicals due to trapping of the conduction band electrons by the adsorbed metal ions. An equation based on LH kinetics for the retardation effect of sodium chloride on phenol photodegradation with TiO2 catalyst has been previously reported.13 However, the mechanistic pathway that results in the LH equation has not been elucidated. Experiments were conducted at constant initial concentration of the dye Rhodamine B (8.2 mg L-1) and Rhodamine 6G (6 mg L-1) individually, at various initial concentrations of the salt copper nitrate, varying from 1.2 to 242 mg L-1. Figures 5 and 6 shows the variation of degradation profile and degradation rate at various concentrations of Cu2+. To estimate the amount of copper adsorbed, assuming that the adsorption of copper is not affected by the nature of the dye, the solution containing the copper salt was stirred along with the dye and 1 g L-1 of the catalyst CS TiO2 for 30 min. The concentration of Cu2+ in the solution after the removal of catalyst particles by centrifugation was determined by the spectrophotometric method using
Figure 6. Effect of Cu2+ concentration on the initial rate of photocatalytic degradation rate of Rhodamine B (C0 ) 8.2 mg L-1) and Rhodamine 6G (C0 ) 6 mg L-1). The figure also shows the variation of adsorbed Cu2+with Cu2+ concentration in solution.
Figure 7 illustrates the model fit with respect to the experimental data at various concentrations of Cu2+. On the basis of the experiments in the absence of Cu2+, k0 and K2 are obtained. Thus, only K6 and k7 are unknowns. By nonlinear regression of experimentally obtained 1/rD at various [C2+], the values of K6 and k7 are determined. k7, k0, and (k0-k7) refer to the contribution of hole alone, hole plus electron, and electron alone toward degradation. The constants are tabulated in Table 2. It is interesting to compare the values of k0-k7 and k0, which indicate that the percentages of degradation of RB and R6G, contributed from OH• generated via electron pathway, are 96.8%
Figure 5. Effect of Cu2+ concentration on photocatalytic degradation of (a) Rhodamine B, (b) Rhodamine 6G (the numbers on the figures indicate Cu2+ concentration in µM).
12 Ind. Eng. Chem. Res., Vol. 46, No. 1, 2007 kr
Recombination: h+ VB + e 98 heat
(A.2)
Hydroxyl radical formation: ktr
\ z S-OH• (a) hole pathway: S-OH- + h+ VB y k′ tr
(A.3a)
ktr
S-H2O + h+ \ z S-OH• + H+ VB y k′tr (A.3b) Ke1
(b) electron pathway: e- + O2 98 O-• 2
(A.4a) Ke2
+ O-• 2 + e + 2H 98 H2O2
Figure 7. Variation of inverse of initial rate of degradation with Cu2+ concentration for the degradation of Rhodamine B (C0 ) 8.2 mg L-1) and Rhodamine 6G (C0 ) 6 mg L-1). The insert shows the variation for SRB (C0 ) 11.6 mg L-1) based on the experimental data of Chen et al.15
Ke3
+ 2O-• 2 + 2H 98 O2 + H2O2
rate constants
RB
R6G
k0, min-1 K2, L mg-1 k7, min-1 K6, L mg-1 k0-k7, min-1
0.283 0.308 0.009 0.165 0.274
0.333 0.441 0.008 0.143 0.325
0.0034 0 0.0001 0.6700 0.0033
and 97.5%, respectively. In order to verify this conclusion, nonlinear regression of the experimental data of Chen et al.15 corresponding to the degradation of dye sulfo Rhodamine B (SRB) was modeled. The model shows that the contribution from electron is 95.6% (Table 2). This also indicates that the rate of OH• produced by electrons is much higher than that produced by hole-hydroxyl species interaction. Therefore, in the presence of copper, the degradation rate reduces primarily because of a decrease in electron concentration. This is similar to the case of degradation in the presence of organic solvents, wherein the degradation rate reduces because of a decrease in electron concentration. This model thus provides insights into the detailed mechanism and indicates whether the degradation of the dyes occurs primarily because of hydroxyl radicals generated by holes or electron. Conclusions The photocatalytic degradation of various Rhodamine dyes was investigated with two catalysts. The photocatalytic activity of solution combustion synthesized TiO2 (CS TiO2) was considerably higher than that of Degussa P-25 for degrading Rhodamine B and Rhodamine 6G. In the case of Rhodamine Blue and Rhodamine 6G perchlorate, the dye degraded faster in the presence of Degussa P25 compared to that of CS TiO2. The degradation rate was significantly reduced in the presence of various metal ions and solvents. A detailed kinetic model was proposed to explain the retardation of the degradation rate of the dyes in the presence of copper. It was shown that the degradation of RB and R6G primarily occurred by the contribution from OH• generated via the electron pathway and the reduction of the degradation rate in the presence of copper was primarily due to a decrease in electron concentration. Acknowledgment The authors thank the Department of Science and Technology for financial support. Appendix A ke
8 h+ + eElectron-hole generation: TiO2 9 hV VB
(A.1)
(A.4c)
Ke4
H2O2 + e- 98 OH• + OH-
Table 2. Rate Constants for Photodegradation of Dyes in the Presence of [Cu2+] SRB15
(A.4b)
(A.4d)
Mobile species adsorption: kD k (a) dye: S + D y\ z S - D, KD ) k′D k′ D
D
kOH
(A.5a) kOH k′OH (A.5b)
(b) hydroxyl species: S + OH• y\z S-OH•, KOH ) k′OH
kC
(c) Cu2+: S + C2+ y\z S-C2+, KC ) k′C
kC (A.5c) k′C
ke5
Reduction of copper: e- + C2+ 98 C+
(A.6a)
ke6
e- + S-C2+ 98 S-C+ (A.6b) It was found in an earlier study that the reduction of Cu2+ to copper metal, Cu0, is very difficult even under UV radiation.20
Dye degradation: (a) Direct holes attack: Khr,a
+ S-D + h+ VB 98 S-D° f end species Khr,b
+ D + h+ VB 98 D° f end species
(A.7a) (A.7b)
(b) Hydroxy radicals attack: Kra
S-D + S-OH• 98 S + S-D′0 f end species Krb
D + S-OH• 98 S + D′0 f end species Krc
S-D + OH• 98 S-D′0 f end species Krd
D + OH• 98 D′0 f end species
(A.8a) (A.8b) (A.8c) (A.8d)
On the basis of the mechanism, the balance for the total hydroxyl species is
Ind. Eng. Chem. Res., Vol. 46, No. 1, 2007 13
d([OH•] + [S-OH•]) • ) ktr[S-OH•][h+ VB] - k′tr[S-OH ] + dt • + ktr[S-H2O][h+ VB] - k′tr[S-OH ][H ] kra[S-D][S-OH•] - krb[D][S-OH•] - krc[S-D][OH•] krd[D][OH•] + ke4[H2O2][e-] (A.9)
K2 )
kraKD[S][S] + krb[S] + krcKD[S] + krd
k3 )
The bracketed term refers to the concentration of the species. The electron concentration balance is
d[e-] )• ) ke[S] - krc[e-][h+ VB] - ke1[e ][O2] - ke2[e ][O2 ] × dt [H+]2 - ke4[H2O2][e-] - ke5[e-][C2+] - ke6[e-][S-C2+] (A.10) Similarly, we can write the balance for the concentration of positive holes, H2O2, O2-•, H+, etc. The rate of disappearance of dye can be written as
ktr{[S-OH-] + [S-H2O]}[h+ VB]
[S-OH•] ) KOH[S][OH•]
(A.12)
[S-D] ) KD[S][D]
(A.13)
[S-C2+] ) KC[S][C2+]
(A.14)
By applying the quasi-steady-state assumption (QSSA) for the hydroxyl radicals, electrons, and holes, we get
k4.0 )
[e-] )
ke[S] 2+ 2+ kr[h+ VB] + ke5[C ] + ke6Kc[S][C ]
(A.16)
Since [h+ VB] is invariant, eq A.11 can be written as
-rD ) koh[D] + k1[D][OH•]
(A.17)
koh ) {khr,aKD[S] + khr,b}[h+ VB]
(A.17a)
where
k1 ) kraKDKOH[S]2 + krbKOH[S] + krcKD[S] + krd
(A.17b)
From eq A.15,
[OH•] )
k3 1 + K2[D]
+
k4,0[e-] 1 + K2[D]
(A.18)
(A.18c)
KOHk′tr[S]{1 + [H+]}
-rD ) koh[D] + k1[D]
(
k3
1 + K2[D]
+
k4,0[e-] 1 + K2[D]
)
(A.19)
From eq A.16
[e-] )
k5
(A.20)
1 + K6[C2+]
where k5 and K6 are given as
k5 )
ke[S]
(A.20a)
kr[h+ VB]
ke5 + ke6Kc[S]
K6 )
kr[h+ VB]
(A.20b)
By substituting eq A.20 into eq A.19, we get
-rD ) koh[D] + k1[D]
[OH ] ) [ktr{[S-OH-] + [S-H2O]}[h+ VB] + ke4[H2O2][e ]]/
In the above equation, the species concentrations that vary with time are the dye and electron concentrations. Since the rate of recombination is faster than that of any other trapping steps,
ke4[H2O2]
By substituting eq A.18 in eq A.17, we get
•
[KOH{k′tr[S]{1 + [H+]} + kraKD[S][D][S] + krb[D][S]} + krcKD[S][D] + krd[D]] (A.15)
(A.18b)
KOHk′tr[S]{1 + [H+]}
+ • -rD ) khr,a[S-D][h+ VB] + khr,b[D][hVB] + kra[S-D][S-OH ] + • • • krb[D][S-OH ] + krc[S-D][OH ] + krd[D][OH ] (A.11)
From the adsorption equilibrium, the surface concentrations of the hydroxyl radical, dye, and copper are given, respectively, as
(A.18a)
KOHk′tr[S]{1 + [H+]}
(
k3
1 + K2[D]
(
k4
+
)(
1 1 + K2[D] 1 + K6[C2+]
))
(A.21)
where k4 ) k4,0k5. Equation A.21 gives the relation between the photocatalytic degradation rate and the concentrations of dye and retardant metal. The rate constant influenced by radiation is ke (eq A.1). Because k5 depends on ke (eq A.20a), the final rate expression also depends on radiation. Literature Cited (1) Fujishima, A.; Rao, T.N.; Tryk, D.A. Titanium Dioxide Photocatalysis. J. Photochem. Photobiol., C: Photochem. ReV. 2001, 1, 1. (2) Fujishima, A.; Hashimoto; Watanabe, T. Fundamentals of TiO2 photocatalysis, first ed.; BKC, Inc.: Herndon, VA, May 1999. (3) Winkler, J. Nano-scaled titanium dioxidesProperties and use in coatings with special functionality. Macromol. Symp. 2002, 187, 317. (4) Fox, M. A.; Dulay, M. T. Heterogeneous Photocatalysis. Chem. ReV. 1993, 93, 341. (5) Martin, S. T.; Herrmann, H.; Hoffmann, M. R. Time resolved microwave conductivity. Part 1. TiO2 photoreactivity and size quantization. J. Chem. Soc., Faraday Trans. 1994, 90, 3315. (6) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Environmental applications of semiconductor photocatalysis. Chem. ReV. 1995, 95, 69. (7) Gnaser, H.; Huber, B.; Ziegler, C. Nanocrystalline TiO2 for photocatalysis. Encycl. Nanosci. Nanotechnol. 2004, 6, 505-535. (8) Konstantinou, I. K.; Albanis, T. A. TiO2-assisted photocatalytic degradation of azo dyes in aqueous solution: Kinetic and mechanistic investigationssA review. Appl. Catal., B 2004, 49, 1.
14 Ind. Eng. Chem. Res., Vol. 46, No. 1, 2007 (9) Wang, Y. Solar photocatalytic degradation of eight commercial dyes in TiO2 suspension. Water. Res. 2000, 34, 990. (10) Sivalingam, G.; Nagaveni, K.; Hegde, M. S.; Madras, G. Photocatalytic degradation of various dyes by combustion synthesized nano anatase TiO2. Appl. Catal., B 2003, 45, 23. (11) Qu, P.; Zhao, J.; Shen, T.; Hidaka, H. TiO2 assisted photodegradation of dyes: A study of two competitive processes in the degradation of RB in an aqueous TiO2 colloidal solution. J. Mol. Catal., A 1998, 129, 257. (12) Muzzarelli, R. A. A.; Weckx, M.; Sigon, O. F. Removal of trace metal ions from industrial waters, nuclear effluents and drinking water, with the aid of cross-linked N-carboxymethyl chitosan. Carbohydr. Polym. 1989, 11, 293. (13) Azevedo, E. B.; Aquino, Neto, F. R.; Dezotti, M. TiO2-photocatalyzed degradation of phenol in saline media: lumped kinetics, intermediates, and acute toxicity. Appl. Catal., B 2004, 54, 165. (14) Dong, Y.; Chen, J.; Li, C.; Zhu, H. Decoloration of three azo dyes in water by photocatalysis of Fe(III)-oxalate complexes/H2O2 in the presence of inorganic salts. Dyes Pigm. 2006, in press. (15) Chen, C.; Li, X.; Ma, W.; Zhao, J. Effect of transition metal ions on the TiO2-assisted photodegradation of dyes under visible irradiation: A probe for the interfacial electron transfer process and reaction mechanism. J. Phys. Chem. B 2002, 106, 318.
(16) Nagaveni, K.; Sivalingam, G.; Hedge, M. S.; Madras, G. Solar photocatalytic degradation of dyes: High activity of combustion synthesized nano TiO2. Appl. Catal., B 2004, 48, 83. (17) Priya, M. H.; Madras, G. Kinetics of TiO2 catalyzed ultrasonic degradation of Rhodamine dyes. Ind. Eng. Chem. Res. 2006, 45, 482. (18) Anpo, M. Photocatalysis on small particle TiO2 catalysts, reaction intermediates and reaction mechanisms. Res. Chem. Intermed. 1989, 11, 67. (19) Epling, G. A.; Lin, C. Investigation of retardation effects on the titanium dioxide photodegradation system. Chemosphere 2002, 46, 937. (20) Hermann, J. M.; Disdier, J.; Pichat, J. Photoassisted platinum deposition on TiO2 powder using various platinum complexes. J. Phys. Chem. B 1986, 90, 6028. (21) Claassen, A.; Bastings, L. The photometric determination of copper by extraction as diethyldithiocarabamatesInterferences and their elimination. Fresenius’ J. Anal. Chem. 1956, 153, 30.
ReceiVed for reView July 21, 2006 ReVised manuscript receiVed October 13, 2006 Accepted October 21, 2006 IE060948N
