Exploring the interparticle electron transfer process in the
photocatalytic oxidation of 4-chlorophenol
Hsin-Hung Ou
a, Shang-Lien Lo
a,∗, Chung-Hsin Wu
baResearch Center for Environmental Pollution Prevention and Control Technology, Graduate Institute of Environmental Engineering,
National Taiwan University, 71 Chou-Shan Rd., Taipei 106, Taiwan, ROC
bDepartment of Environmental Engineering, Da-Yeh University, 112 Shan-Jiau Rd., Da-Tsuen, Chang-Hua 515, Taiwan, ROC Received 6 October 2005; received in revised form 15 March 2006; accepted 15 March 2006
Available online 16 May 2006
Abstract
This work aimed to investigate the interparticle electron transfer (IPET) process within the coupled-photocatalyst systems on the basis of the degradation of 4-chlorophenol (4-CP). TiO2, ZnO and SnO2 are used as the model photocatalysts owing to their increasing energy levels
which correspond to the IPET concept. In the single-photocatalyst tests, ZnO tests are associated with the highest degradation rate constants (0.347± 0.083 h−1at pH 7 and 0.453± 0.011 h−1at pH 11) and a better DOC reduction than in other single catalyst tests under given conditions. ZnO/SnO2coupled tests have constants of 0.612± 0.068 and 0.948 ± 0.069 h−1at pH 7 and 11, respectively. Additionally, the 4-CP prefers the
breakdown of chloride group in TiO2system while proceeding hydroxylation reaction in ZnO systems. Meanwhile, a phenomenonlogical model
coupled with the IPET effect was developed to explore the separation of photo-electrons and photo-holes within catalysts. Based on the model parameters, the recombination rate of photo-electrons and photo-holes in TiO2/SnO2and ZnO/SnO2systems is 20–45% lower than that obtained
by a respective single catalyst. Thus, coupled-photocatalyst tests, TiO2/SnO2and ZnO/SnO2efficiently suppress the recombination, particularly
for ZnO/SnO2tests at pH 11.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Photodegradation; Interparticle electron transfer process; Coupled-photocatalyst
1. Introduction
The problem of pollutants, particularly the serious environ-mental impact of their residues and the relatively low efficiency of the current remediation process, has led to several efforts to elucidate more efficient degradation alternatives. Many works have attempted to degrade numerous pollutants by advanced oxidation processes (AOPs)[1,2], rather than by other treatment processes. Heterogeneous photocatalysis has been emerged as an efficient method for purifying water and air [3,4]. In most cases, the most effective materials for photocatalytic applica-tions are nano-sized semiconductor oxides, which have been proven as excellent catalysts because of their highly reactive surfaces[5]. Such oxides include TiO2, ZnO and CdS, among
others. TiO2and ZnO have been extensively examined as
het-erogeneous semiconductor photocatalysts, primarily because of
∗Corresponding author. Fax: +886 2 23928830.
E-mail address:sllo@ntu.edu.tw(S.-L. Lo).
their high capacity for degrading toxic and recalcitrant chemical species via relatively simple and low-cost procedures[6,7].
However, the use of semiconductors as photocatalysts is mainly limited in the recombination of the generated photo-holes and photo-electrons. Photocatalysis involves the oxidation of a chemical by photo-holes from the semiconductor, so every recombination event involves the loss of holes that might oth-erwise have promoted degradation. Thus, the vectorial transfer of photogenerated electrons and holes between the valence and conduction bands of semiconductors is an important process in photocatalysis. Accordingly, a major focus of current photo-catalysis research is to improve the separation characteristics, to improve photo-efficiency. Attempts are made to solve the prob-lem of increasing the photo-efficiency, one of these methods is interparticle electron transfer (IPET) process.
The concept of IPET process is to exploit the distinction of the band gap between semiconductors with different energy poten-tials to drive the electrons transfer. That means the coupled semi-conductors with their corresponding conduction and valence bands can be used to achieve such phenomenon, increasing the
0304-3894/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jhazmat.2006.03.023
Fig. 1. Diagram depicting the redox potentials of the valence and conduc-tion bands and the band gap energies for various semiconductor particulates [8,30–33].
lifetime of charge carries and the efficiency of the interfacial charge transfer to the adsorbed substrate and ultimately enhanc-ing their photocatalytic performance considerably[8–11]. How-ever, an ambiguous point still needed to be clarified is the promotion in retarding recombination within IPET process. Therefore, it is essential to expound the IPET effect on the basis of some kinetic parameters, and which is the major concern in this study.
This study elucidate the IPET effect under three buffer con-ditions, pH 4, 7 and 11, using TiO2, ZnO and SnO2as model
photocatalysts because of the increase in their energy levels (Fig. 1). Attempts are made to compare the photocatalytic mechanism in the TiO2 and ZnO systems, by exploiting the
reduction of dissolved organic carbon and the ratios of released chloride ions to 4-CP concentration. Additionally, a notional model proposed in the study is used to ascertain whether IPET effect promote the separation of electrons and photo-holes.
2. Materials and methods
2.1. Reagent and chemicals
Three semiconductor powders TiO2 (Degussa P25), ZnO
(Fluka) and SnO2 (RDH) were used as model photocatalysts.
4-CP (99%) was purchased from Aldrich and was of the best grade available; these agents were used without further treat-ment. Water was deionized and doubly distilled with Milli-Q.
2.2. Experimental apparatus and procedures
All photodegradation experiments were conducted in a batch reactor. The reaction mixtures were illuminated under a UV lamp (8 W Xenon lamp, Philips), which was placed approxi-mately 35 cm below the bottom of the glass plate at a controlled reaction temperature of 25◦C during the experimental period. The intensity of UV lamp is 4.32 mW cm−2measured by UV radiometer (MS-100, UVP) at the sampling position.
Prior to the photocatalytic experiments, the suspensions were prepared by mixing definite volumes (3 L) of solutions of the desired concentration of 4-CP (2× 10−4M) with various
sin-Table 1
Characteristic property of TiO2, ZnO and SnO2 Catalyst Maximum absorption
wavelength (nm)
Band gap (eV) BET (m2g−1)
TiO2 391.4 3.17 54.18
ZnO 424.9 2.92 4.56
SnO2 300.4 4.13 4.78
gle and coupled-photocatalysts, TiO2, ZnO, SnO2, TiO2/SnO2,
ZnO/SnO2and TiO2/ZnO (1.2 g L−1). Meanwhile, the coupled
catalysts are of physical mixing in the solution without any pretreatments. The portion of SnO2 is 50 wt.% for respective
coupled catalyst. That means both the concentrations of TiO2
and SnO2in the coupled catalyst (TiO2/SnO2) are 0.6 g L−1, the
proportion of SnO2in ZnO/SnO2coupled catalyst is the same
with the aforementioned description. The slurry was stirred mag-netically for 30 min in the dark to achieve adsorption equilibrium for the substrate on the photocatalytic system. Hydrochloric acid or sodium hydroxide was added to maintain the desired buffered conditions. When the photocatalytic experiment was initiated, the samples were withdrawn at different times. After they were centrifuged at 3000 rpm with a centrifuge (KUBOTA KN-70) and then filtered through a 0.45m filter membrane (Millipore), the samples proceeded to a series of analysis, the remaining of 4-CP (HPLC, Millipore Waters 600E with a Waters 486 detector), the quantification of released chloride ions (IC, Dionex model DX-120) and dissolved organic carbon (TOC, 1020A,OI Ana-lytical). And all the physical characteristics of the photocatalysts performed in the study were presented inTable 1.
3. Model development
Although the production of by-products such as hydro-quinone (HQ), benzohydro-quinone (BQ) and 4-chlorocatechol (4-CC) can potentially influence the degradation behavior of 4-CP, their impacts have been inclusive of the model development. The model primarily concerns the type of coupled-photocatalystic systems and the reaction conditions to survey the IPET effect. The simplified notional mechanism of photocatalysis is described as follows:
TiO2, ZnO + hv → e−+ h+ (1) e−+ SnO2→ e−·SnO2 (2) e−+ h+→ heat (3) e−+ O2,ads→ O2− (4) h++ H2Oads→ OH• + H+ (5) h++ Subads→ R• + H2O (6)
Eq. (1) shows the formation of holes and photo-electrons in the surface of TiO2 or ZnO on irradiation with
the valid UV light energy which is significantly affected by the optical and physical properties of irradiated catalysts. Eq.
(2) refers to the vectorial transfer of photo-electrons between two catalysts owing to the distinct energy potential within
the coupled systems. The reaction channel is so-called IPET process which can reduce the recombination opportunities of photo-holes and photo-electrons. Therefore, such a route theoretically provides an effective approach to conduce on the photocatalytic performance. Meanwhile, Eq. (3) represents the recombination rate of electrons and holes occurred on the surface of the irradiated photocatalysts. The reaction channel is always taken as the major limiting factor on photocatalytic efficiency resulting to the decline of the yields of photo-holes. Another pathway for photo-electrons is the trapping reaction by oxygen molecules, which has been confirmed as the major channel for photo-holes in the photocatalysis [12], such as shown in Eq. (4). Eqs.(5) and(6) demonstrate the channels for the photo-holes to react with the molecular water and target compound, respectively. This study envisages that the photo-holes are the prevailing oxidant species on the basis of some prior investigations[13–15]. That means there is a hypothesis,
k6KsubCsub>> k5KH2OCH2O, where the KH2O and CH2O are
adsorption constant and concentration of water molecules in the system. Regarding the validity of the hypothesis, some typical kinetics from literature provided solid evidences to support that. Sun and Bolton[16]even evaluated the formation rates of OH•((3.40± 0.13)×10−7M s−1) in which HCHO was chosen as the trapping reagent in the TiO2 suspensions. Ishibashi et
al.[15]quantified the quantum yields of hydroxyl radicals and photogenerated holes which were estimated to be 7× 10−5and 5.7× 10−2, respectively. This implies that oxidative reactions on TiO2photocatalyst occur mainly via photogenerated holes
instead of OH•. Accordingly, the intrinsic photocatalytic kinetic of organic substrate in aqueous slurry can be described as Eq.
(7). The detail of the similar model development discussed herein is also presented in the previous studies[5,17–19]:
dCsub
dt = −k6Ch+Csub,ads (7)
where the Ch+ and Csub,ads are the concentration of
photo-holes and the target compounds on the surface of the TiO2,
respectively. Prior to deduce the phenomenonlogical model, one need to suggest the photo-holes and photo-electrons obey the pseudo-steady-state assumption meaning for the results,
dCe−
dt = 0 and dCh+
dt = 0[17]. That indicates that the photo-holes
and photo-electrons may have constant concentration in an aqueous solution with constant dissolved oxygen concentration, just as shown in Eqs.(8)and(9):
dCe− dt = gave− (k3− k2)Ce−Ch+− k4Ce−CO2,ads≈ 0, Ce− = gave (k3− k2)Ch++ k4CO2KO2 (8) dCh+ dt = gave− (k3− k2)Ce−Ch+− k5Ch+CH2O,ads − k6Ch+Csub,ads≈ 0, Ch+ = gave
(k3− k2)Ce−+ k5CH2OKH2O+ k6CsubKsub
(9)
In the above expressions, Ce− and Ch+ represented the
con-centration of the photo-electrons and photo-holes generated in each test, respectively. KO2, and Ksubstand for the adsorption
constants of oxygen and target compound. kiis the reaction rate
for the respective aforementioned reactions. gaveis the average
generated rate of electron–hole pairs in the reaction systems which is dependent on the physical character of TiO2and ZnO
and the properties of UV light, but it is treated as a lumped parameter in this model. Among the parameters, CH2Oand CO2
are referred to the concentrations of molecular water and oxygen in the slurry systems. Substituting Eq.(8)into Eq.(9)will derive the quadratic expression for the concentration of the photo-holes adhered to the TiO2surface, just as shown in Eq.(10):
C2 h+ +
(k4CO2KO2k5CH2OKH2O+ k4CO2KO2k6CsubKsub)
(k3− k2)(k5CH2OKH2O+ k6CsubKsub)
Ch+
− gavek4CO2KO2
(k3− k2)(k5CH2OKH2O+ k6CsubKsub)
= 0 (10) Let, a= k4CO2KO2, b = k5CH2OKH2O, c = k6Ksub, σ =
(k3− k2), and then make the assumption which k6KsubCsub>> k5KHCH2Oindicating that the major portion of photo-holes react
with target compounds instead of the molecular water. That means the reaction rate between photo-holes and target com-pounds is higher than the rate of hydroxyl-radicals-producing which reflects that the photo-holes may be the predominant oxi-dant species corresponding to the assumption in the study. Such concepts have been held true in some early studies[13–15]. The solution to Eq.(10)is shown as Eq.(11)and then arrange as the following form by substituting it into Eq.(7), and the outcome is showed in Eq.(12):
Ch+ =
−(ab + acCsub)
+(ab+ acCsub)2+ 4σgave(ab+ acCsub)
2σ(b+ cCsub)
= −acCsub+
(acCsub)2+ 4σgave(acCsub)
2σ(cCsub)
= −a 2σ +
(acCsub)2+ 4σgave(acCsub)
2σ(cCsub) (11) dCsub dt = ac 2σCsub− ac 2σ C2 sub+ 4σgave ac Csub (12)
Then to expand the term,
C2
sub+ (4σgave/ac)Csub, as the
Taylor series expansion with third order which leads to the results shown below and then to arrange as the Eq.(13)followed by some simplifications: dθ dt = 180 α 2βθ − α1/2 β1/2γ−(1/2)θ 1/2−1 8 α3/2 β3/2γ 1/2θ3/2 + 1 128 α5/2 β5/2γ 3/2θ5/2− 1 1024 α7/2 β7/2γ 5/2θ7/2 (13) where θ= Csub C0 , τ = t tall, α = c, β = σ a, and γ = gCave0
Eq.(13)is used to fit the experimental data, and the values of kinetic parameters, α, β and γ are obtained to compare the distinction of chemical behavior among systems as the function of pH and the IPET process. Accordingly, one can elucidate the IPET effect on observation of the variance of β value while α is the observed degradation rate constant. γ is greatly dependent on the physical and optical properties of the irradiated photocat-alysts, TiO2and ZnO.
4. Results and discussions
4.1. Background experiments
The direct photolysis, evaporation of 4-CP free of catalysts in the solution and adsorption of 4-CP on catalysts, need to be per-formed to exclude the degradation of 4-CP caused by the ambient environment during photocatalysis course. Before that, the opti-mum amount of catalyst should be ascertained by conducting a series of tests.Fig. 2plots the obtained initial photodegrada-tion rate constants for different loading amount of TiO2 and
the results of background experiments under neutral condi-tions. Accordingly, the optimal dosage of TiO2was 1.2 g L−1
in all experiments. The results of the background experiments revealed that the disappearance of 4-CP caused by ambient con-ditions can be negligible, suggesting that the ambient condition had almost no effect on the degradation of 4-CP. Thus, the dis-appearance of 4-CP in the photocatalytic experiments could be definitely attributed to the photodegradation.
4.2. Effect of pH and IPET process
The observed rate constants of 4-CP, the values of α in
Table 2, at pH 7 for the TiO2/SnO2 and ZnO/SnO2 tests are
0.186± 0.070 and 0.612 ± 0.068 h−1, respectively. The corre-sponding values in the TiO2and ZnO tests are 0.127± 0.015
and 0.347± 0.083 h−1, indicating that the photocatalytic
per-Fig. 2. The effects of background experiments inclusive of the tests of the optimum adding amounts at its neutral condition ([4-CP]0= 200M, [TiO2] = 1.2 g L−1).
Table 2
Comparisons of observed degradation rate constants and model parameters for 4-CP degradation for each system
Tests Conditions (pH) Model parameters α (h−1) β (10−3) γ (10−3) TiO2 4 0.109± 0.015 2.257 ± 0.741 4.005 ± 0.304 7 0.127± 0.015 0.626 ± 0.049 18.98 ± 2.379 11 0.100± 0.012 0.419 ± 0.037 49.04 ± 6.742 TiO2/SnO2 4 0.174± 0.012 1.580 ± 0.152 0.152 ± 0.009 7 0.186± 0.070 0.427 ± 0.051 66.92 ± 2.882 11 0.150± 0.011 0.332 ± 0.028 60.53 ± 4.191 ZnO 7 0.347± 0.083 9.321 ± 0.881 260.4 ± 15.31 11 0.453± 0.011 4.660 ± 0.346 392.7 ± 17.96 ZnO/SnO2 7 0.612± 0.068 5.450 ± 0.032 255.1 ± 15.32 11 0.948± 0.069 3.494 ± 0.119 602.1 ± 31.17
formance of the coupled-photocatalyst tests is better than that of the single-photocatalyst tests, indirectly confirming the IPET effect. Also expected, the same results are observed at pH 11, at which the observed rate constants in the TiO2, TiO2/SnO2,
ZnO and ZnO2/SnO2 tests are 0.100± 0.012, 0.150 ± 0.011,
0.453± 0.011 and 0.948 ± 0.069 h−1, respectively.
Such finding is consistent with the concept of the IPET pro-cess. The TiO2/SnO2and ZnO/SnO2tests significantly differ in
redox energy levels and so exhibit better results concerning the IPET effect, especially for ZnO/SnO2tests in which the energy
gap is large (Fig. 1). Notably, TiO2/ZnO tests dose not deliver the
anticipated results, even though the IPET effect is exhibited. This event is associated with the small energy gap between TiO2and
ZnO. Accordingly, TiO2/ZnO tests may offer more
opportuni-ties for photo-electrons and photo-holes to recombine, reducing the IPET efficiency below that in the TiO2/SnO2and ZnO/SnO2
tests. Although the IPET effect is evident, it is stronger at pH 11 for ZnO systems but weaker at pH 11 for TiO2systems. The
results obtained at pH 11 agree with those obtained in most stud-ies, in which highly alkaline media have been shown to promote photocatalysis[8,20].
Notably, the catalytic capacity of TiO2follows the order pH
7 > pH 4 > pH 11. For TiO2 systems at pH 4 and 7, the
pre-dominant species in the solution are the undissociated 4-CP molecules, rather than the corresponding anions (pKa= 9.2).
Meanwhile, positively charged TiO2, had an acidic surface at
pH 4 and 7, which is an effective environment for adsorption, under which the affinity between TiO2and 4-CP is higher than
at pH 11, resulting in the higher degradation rate at pH 7. Also, Ku and Jung[21]found the surface charge of Degussa P25 in aqueous solution; the pH at the zero point of charge (pHZPC) was
determined to be 7.52. The aforementioned statements support the result related to the characteristics of TiO2.
Regarding the ZnO systems, the values of observed rate con-stant, α, are 0.612± 0.068 and 0.948 ± 0.069 h−1at pH 7 and 11, respectively. The point worthy to mention is that the results indi-cate alkaline conditions promote photocatalytic performance, which finding is in contrast with the prior expressions. Eqs.
(5) and(6)demonstrate a competition between the molecular water and the target compound for photo-holes. Subsequently,
Eq.(5)theoretically dominates under alkaline conditions. The degradation efficiency in acid condition theoretically exceeds that in alkaline condition. The hypothesis of k6KsubCsub>> k5KHCH2O gives a good account of why the opposite events
occur. Namely, Eq.(5)is just a minor channel for photo-holes to proceed even in the alkaline conditions.
The results also reveal that ZnO is a poor photocatalyst in the oxidative degradation of 4-CP at pH 4, because it corrodes in aqueous acidic media, which finding was consistent with that of an earlier literature[22]. Also as expected, SnO2 is not to
be inactivated because of the low overlap between the UV emis-sion spectrum (the maximum emisemis-sion wavelength: 365 nm) and adsorption spectrum of SnO2(the maximum adsorption
wave-length: 300.4 nm). Interestingly, the activity pattern of 4-CP is ZnO > HO2> SnO2at pH 7 and 11. ZnO photocatalyst has an
activity that is 1.9 times higher than that of TiO2, which is the
reference photocatalyst at pH 7 and almost three times that of TiO2at pH 11. ZnO is concluded to be a good photocatalyst
of the degradation of 4-CP in alkali aqueous media, suggesting that ZnO is highly active in 4-CP degradation. The same results have been published in the previous researches.[23,24].
When TiO2or ZnO are coupled with SnO2, the oxidation of
4-CP is faster than that in TiO2and ZnO single-tests, respectively,
reflecting the charge transfer between TiO2or ZnO and SnO2
[8]. The phenomenon indicates that TiO2and ZnO are active in
4-CP oxidation but SnO2is not. However, SnO2may be a charge
carrier in the IPET process while coupled with TiO2and ZnO.
Several studies have involved analogous physical experiments and have found the same trend in the degradation of compounds using coupled-photocatalyst[9–11,25].
4.3. Mineralization and release of chloride ions
Many studies have reported the appearance of by-products in the solution, while 4-CP was treated photocatalytically; these are dominantly hydroquinone (HQ), benzoquinone (BQ) and 4-chlorocatechol (4-CCT) [8,26–29]. In this work, the reaction pathway in the TiO2 and ZnO systems is surveyed from the
released chloride ions and the extent of mineralization. How-ever, attempts to study the scenario of DOC (dissolved organic carbon) degradation have been inconclusive, because the results have been widely scattered. The ratio of released chloride ions to degraded 4-CP ([Cl−]formed/[4-CP]degraded) and the degree of
mineralization are plotted asFigs. 3–5.
Obviously, the initial conversions of 4-CP to chloride ions in ZnO systems are lower than unity regardless of the condi-tions (Figs. 4and 5). This fact clearly indicates that the initial reaction causes the significant collapse of phenyl groups instead of chloride groups. Therefore, the predominant intermediates in ZnO systems are probably chlorinated organic matters. Despite the better performance of ZnO systems in 4-CP degradation, most of the corresponding by-products are still highly toxic. In contrast, the ratio of the released chloride ions to degraded 4-CP is lower in ZnO systems than in TiO2 systems, in which
it is approximately unity after 50 min. Subsequently, the initial reactions in the TiO2systems probably involve the breaking off
of the chlorinated function group. That means the by-products
Fig. 3. The scenario of ratios of released chloride ions to the degraded 4-CP inclusive of the reduction of DOC at pH 4 ([Cl− ]formed/[4-CP]degraded) ([4-CP]0= 200M, [DOC]0= 14.4 ppm, [TiO2] = 1.2 g L−1, [TiO2/SnO2] = 0.6/0.6 g L−1).
produced in TiO2system prefer HQ and BQ instead of 4-CC.
Stafford et al.[26]have indicated the increased TiO2loading
(0.025∼1.000 g L−1) significantly decreases the 4-CC
concen-tration. This observation corresponded to the aforementioned inference, where the TiO2loading in our study was 1.2 g L−1.
Alike phenomenon has also been reported by Alekabi and Ser-pone[27]and Alsayyed et al.[28]. Regarding the by-products produced in ZnO system, the majority of the by-products may be 4-CC[29], which is the predominant reason leading to the ratios, the released chloride ions to the degraded 4-CP, are lower than unity.
Notably, the disappearance of DOC shows that decarboxyla-tion is an initial step in ZnO photo-degradadecarboxyla-tion, since the initial rate (−d[DOC]/dt) was not zero which is not showed in the
Fig. 4. The scenario of ratios of released chloride ions to the degraded 4-CP inclusive of the reduction of DOC at pH 7 ([Cl−]formed/[4-CP]degraded) ([4-CP]0= 200M, [DOC]0= 14.4 ppm, [TiO2] = [ZnO] = 1.2 g L−1, [TiO2/SnO2] = [ZnO/SnO2] = [TiO2/ZnO] = 0.6/0.6 g L−1).
Fig. 5. The scenario of ratios of released chloride ions the degraded 4-CP inclusive of the reduction of DOC at pH 11 ([Cl−]formed/[4-CP]degraded) ([4-CP]0= 200M, [DOC]0= 14.4 ppm, [TiO2] = [ZnO] = 1.2 g L−1, [TiO2/ SnO2] = [ZnO/SnO2] = [TiO2/ZnO] = 0.6/0.6 g L−1).
material. The TiO2system reveals that the initial step was the
breaking off the chlorinated function group as described above. It must rather be ascribed to the difference between the chemical and physical properties of the catalysts.
In summary, this section integrated the patterns ofFigs. 3–5to infer the simplified mechanism of 4-CP degradation in the TiO2
and ZnO systems. In both systems, DOC degradation occurs to a lesser extent than that of 4-CP, suggesting the presence of other organic compounds generated from 4-CP degradation. In the ZnO systems, the predominant intermediates are the chlorinated compounds while hydroxylation intermediates are significant in the TiO2systems. In fact, the appearance of chloride ions does
not balance the disappearance of 4-CP for each test. This fact clearly indicates that in a heterogeneous system, 4-CP
degra-dation occurs by a mechanism that is responsible for releasing chloride ions into the solution and by a mechanism in which that release dose not occur.
4.4. Kinetic parameters within IPET effect
The model parameters are obtained by fitting the temporal degradation rate of 4-CP to the polynomial equation.Table 2
lists the model parameters for each set of data. Data used herein must be analyzed by estimating the observed degradation rate constants and thus introduce additional uncertainty. Therefore, some statistical parameters are used to support the validity of the model in the work. The plot of residuals for this analysis is shown inFigs. 6–8for each system at its corresponding state, which indicate a random scatter with no discernible pattern about
εi= 0. The fit of the model to the experimental data is good
(r2> 0.9), sequentially, and all of the t-ratios of each parameter are exceed four, the error variances are all below some critical value. Accordingly, the proposed model is effective in determin-ing the parameters, β and γ.
The same electrode was used for the entire sets of experi-ments, so that gave was theoretically a constant as the reaction
proceeds in a given photocatalystic system. Therefore, γ, incor-porated in the terms gave, depends significantly on physical and
optical properties of catalysts. The γ values in the ZnO system normally exceed those in the TiO2 system (Table 2),
indicat-ing that ZnO powders accept lower irradiation, but all of the ZnO systems exhibit better photocatalytic performances. The phenomenon is consistent with the aforementioned result that ZnO is highly photo-active to 4-CP degradation. Meanwhile, the results also present that the γ values are explicitly higher in the coupled tests than in the single tests in the same sys-tems (Table 2). Such a trend delivers that more catalysts can be irradiated in single tests rather than in coupled tests during the irradiation course. Perhaps, the carrier catalyst, SnO2, has the
Fig. 7. The residuals resulting from the fitting analysis of the rate estimation for TiO2 and ZnO systems at pH 7 ([TiO2] = [ZnO] = 1.2 g L−1, [TiO2/SnO2] = [ZnO/SnO2] = [TiO2/ZnO] = 0.6/0.6 g L−1).
shelter effects, so less irradiated catalyst is present in the coupled tests. Accordingly, such an event clarifies definitely the factor dominating the photocatalytic efficiency is the recombination opportunities rather than the amount of the irradiated photons.
Although our investigation does not address the vectorial transfer rate within the IPET process, the IPET effect could still be checked on variance of β values for each test. Namely,
β is governed mainly by the synergy of the IPET effect and the
recombination reaction. β values are smaller in the coupled tests than in the single tests, revealing that IPET effect provides an effective approach to retard the recombination opportunities of generated photo-electrons and photo-holes. ZnO and ZnO/SnO2
tests at pH 11 yield β values of (4.660± 0.346)×10−3 and (3.494± 0.119)×10−3, respectively, suggesting that the
Fig. 8. The residuals resulting from the fitting analysis of the rate estimation for TiO2and ZnO systems at pH 11 ([TiO2] = [ZnO] = 1.2 g L−1, [TiO2/SnO2] = [ZnO/SnO2] = [TiO2/ZnO] = 0.6/0.6 g L−1).
decrease in the recombination rate is around 25± 3%. In contrast to TiO2and TiO2/SnO2tests under the same condition,
in which the recombination performance is almost 21± 0.3% lower. Therefore, the ZnO/SnO2tests provide a more reliable
channel to retard the recombination reaction. The same ten-dency can be observed at pH 7. These results reflect that the larger energy gap causes the more efficient separation of pairs of photo-electrons and photo-holes (Fig. 1). Notably, the variance of β values also remarkably reveals that the recombination rate is directly related to pH which seems to be responsible for the suppression of the recombination. However, the mechanism remains as a matter to be discussed further. Briefly, the results reveal that TiO2/SnO2, ZnO/SnO2 and TiO2/ZnO can
facilitate the separation of photo-electrons and photo-holes. The promotion extent of the IPET effect follows the order: ZnO/SnO2> TiO2/SnO2> TiO2/ZnO, which corresponds to the
corresponding energy gap. In summary, the efficiency of the photocatalysis is strongly attributed to the suppression of the recombination of photo-electrons and photo-holes, rather than the amount of the generated photons. Additionally, the results demonstrated herein constitute reliable evidences that the IPET effect apparently provide an effective means to improve photocatalysis.
5. Conclusions
In the single-photocatalyst test, ZnO exhibited the high-est catalytic activity and a better degree of reduction of DOC than other single catalyst tests at pH 7 and 11. The observed rate constants of the coupled-catalyst systems, TiO2/SnO2and
ZnO/SnO2, are higher than that of TiO2 and ZnO
approxi-mately by a factor of 1.5–3. Also, the reductions of DOC for TiO2/SnO2 and ZnO/SnO2 tests are 8–20% higher than that
of the single-photocatalyst system. One recommended course of action related to this study is to assess the recombina-tion rate within IPET process. A comparison of the model parameters revealed significant trends that can be qualitatively related to the experimental conditions and test systems. Sub-stantially, this study concludes that the IPET effect on the photocatalysis dose deliver the anticipated advantages based on the disappearance rate of 4-CP. The results demonstrate that the recombination rate is prone to decline in systems with a large energy gap, which dose correspond to the concept of IPET process.
Acknowledgement
The authors would like to thank the National Science Council of the Republic of China for financially supporting this research under Contract No. NSC 93-2622-E-264-004-CC3.
References
[1] M. Lewandowski, D.F. Ollis, A two-site model simulating apparent deac-tivation during photocatalytic oxidation of aromatics on titanium dioxide (TiO2), Appl. Catal. B Environ. 43 (2003) 309–327.
[2] A. Sobczynski, L. Duczmal, W. Zmudzinski, Phenol destruction by pho-tocatalysis on TiO2: an attempt to solve the reaction mechanism, J. Mol. Catal. A Chem. 213 (2004) 225–230.
[3] K. Chiang, R. Amal, T. Tran, Photocatalytic oxidation oxidation of cyanide: kinetic and mechanistic studies, J. Mol. Catal. A Chem. 193 (2003) 285–297.
[4] P.B. Amama, K. Itoh, M. Murabayashi, Photocatalytic degradation of trichloroethylene in dry and humid atmospheres: role of gas-phase reac-tions, J. Mol. Catal. A Chem. 217 (2004) 109–115.
[5] C.B. Almquist, P. Biswas, Role of synthesis method and particle size of nanostructured TiO2 on its photoactivity, J. Catal. 212 (2002) 145– 156.
[6] J.C. Lee, M.S. Kim, B.W. Kim, Removal of paraquat dissolved in a pho-toreactor with TiO2immobilized on the glass-tubes of UV lamps, Water Res. 36 (2002) 1776–1782.
[7] A.D. Paola, E. Garcia-Lopez, S. Ikeda, G. Marci, B. Ohtani, L. Palmisano, Photocatalytic degradation of organic compounds in aqueous system by transition metal doped poly crystalline TiO2, Catal. Today 78 (2002) 87–93.
[8] N. Serpone, P. Maruthamuthu, P. Pichat, E. Pelizzatti, H. Hidaka, Exploiting the interpartical electron transfer process in the photocatalysed oxidation of phenol, 2-chlorophenol and pentachlorophenol: chemical evidence for electron and hole transfer between coupled semiconductors, J. Photochem. Photobiol. A Chem. 85 (1995) 247–255.
[9] C. Wang, J. Zhao, X. Wang, B. Mai, G. Sheng, P. Peng, J. Fu, Preparation, characterization and photocata;ytic activity of nano-sized ZnO/SnO2 coupled-photocatalysts, Appl. Catal. B Environ. 39 (2002) 269–279.
[10] S.C. Lo, C.F. Lin, C.H. Wu, P.H. Hsieh, Capability of coupled CdSe/TiO2 for photocatalytic degradation of 4-chlorophenol, J. Hazard. Mater. B114 (2004) 183–190.
[11] C.H. Wu, Comparison of azo dye degradation efficiency using UV/single semiconductor and UV/coupled semiconductor systems, Chemosphere 57 (2004) 601–608.
[12] M.R. Hoffmann, S.T. Martin, W. Choi, W. Bahnemann, Environmen-tal applications of semiconductor photocaEnvironmen-talysis, Chem. Rev. 95 (1995) 9–96.
[13] R.B. Draper, M.A. Fox, Titanium dioxide photosensitized reactions stud-ied by diffuse reflectance flash photolysis in aqueous suspensions of TiO2 power, Langmuir 6 (1990) 1396–1402.
[14] E.R. Carraway, A.J. Hoffman, M.R. Hoffman, Photocatalytic oxidation of organic acids on quantum-sized semiconductor colloids, Environ. Sci. Technol. 28 (1994) 1994.
[15] K.I. Ishibashi, A. Fujishima, T. Watanabe, K. Hashimoto, Quantum yields of active oxidative species formed on TiO2photocatalyst, J. Photochem. Photobiol. A Chem. 134 (2000) 139–142.
[16] L. Sun, J.R. Bolton, Determination of quantum yield for the photochemical generation of hydroxyl radicals in TiO2suspensions, J. Phys. Chem. 100 (1996) 4127–4134.
[17] H. Gerischer, Photocatalysis in aqueous solution with small TiO2particles and the dependence of the quantum yield on particle size and light intensity, Electrochim. Acta 40 (1995) 1277–1281.
[18] C.B. Almquist, P. Biswas, A mechanistic approach to modeling the effect of dissolved oxygen in photo-oxidation reactions on titanium dioxide in aqueous systems, Chem. Eng. Sci. 56 (2001) 3421–3430.
[19] Y. Meng, X. Huang, Y. Wu, X. Wang, Y. Qian, Kinetic study and model-ing on photocatalytic degradation of para-chlorobenzoate at different light intensities, Environ. Pollut. 117 (2002) 307–313.
[20] R.A. Doong, C.H. Chen, R.A. Maithreela, S.M. Chang, The influence of Ph and Cadmium sulfide on the sulfide on the photocatalytic degradation of 2-chlorophenol in titanium dioxide suspensions, Water Res. 35 (2001) 2873–2880.
[21] Y. Ku, I.L. Jung, Photocatalytic reduction of Cr(VI) in aqueous solutions by irradiation with the presence of titanium dioxide, Water Res. 35 (2001) 135–142.
[22] A.A. Khodja, T. Sehili, J.F. Pilichowski, P. Boule, Photocatalytic degra-dation of 2-Phenyphenol on TiO2 and ZnO in aqueous suspensions, J. Photochem. Photobiol. A Chem. 141 (2001) 231–239.
[23] C.A.K. Gouvea, F. Wypych, S.G. Moraes, N. Duran, N. Nagata, P. Peralta-Zamora, Semiconductors-assisted photocatalytic degradation of reactive dyes in aqueous solution, Chemosphere 40 (2000) 433–440.
[24] B. Dindar, S. Icli, Unusual photoreactivity of zinc oxide irradiated by con-centrated sunlight, J. Photochem. Photobiol. A Chem. 140 (2001) 263–268. [25] C. Hu, Y. Tang, J.C. Yu, P.K. Wong, Photocatalytic degradation of cation blue X-GRL adsorbed on TiO2/SiO2photocatalyst, Appl. Catal. B Environ. 40 (2003) 131–140.
[26] U. Stafford, K.A. Gray, P.V. Kamat, Photocatalytic degradation of 4-chlorophenol: the effects of varying TiO2 concentration and light wave-length, J. Catal. 167 (1997) 25–32.
[27] H. Alekabi, N. Serpone, Kinetic studies in heterogeneous photocatalysis. 1. photocatalytic degradation of chlorinated phenols in aqueous solutions over TiO2supported on a glass matrix, J. Phys. Chem. 92 (1988) 5726–5731. [28] G. Alsayyed, J.C. D’Oliveira, P. Pichat, Semiconductor-sensitized
pho-todegradation of 4-CP in water, J. Photochem. Photobiol. A Chem. 58 (1991) 99–114.
[29] T. Sehili, P. Boule, J. Lemaire, Photocatalysed transformation of chloroaro-matic derivatives on zinc oxide IH: chlorophenols, J. Photochem. Photobiol. A Chem. 50 (1989) 117–127.
[30] M. Fujii, T. Kawai, S. Kawai, Photocatalytic activity and the energy levels of electrons in a semiconductor particle under irradiation, Chem. Phys. Lett. 106 (1984) 517–522.
[31] M.F. Finlayson, B.L. Wheeler, N. Kakuta, K.H. Park, A.J. Bard, A. Cam-pion, M.A. Fox, S.E. Webber, J.M. White, Determination of flat-band position of CdS crystals, films, and powders by photocurrent and impedance techniques. Photoredox reaction mediated by intragap states, J. Phys. Chem. 89 (1985) 5676–5681.
[32] J.R. White, A.J. Bard, Electrochemical investigation of photocatalysis at CdS suspensions in the presence of Methylviologen, J. Phys. Chem. 93 (1985) 1947–1954.
[33] G. Redmond, A. O’Keefe, C. Burgess, C. MacHale, D. Fitzmaurice, Spec-troscopic determination of the flatband potential of transparent nanocrys-talline ZnO film, J. Phys. Chem. 97 (1993) 11081–11086.
