Photocatalytic Degradation of Azo Dyes Using Microreactors: Mechanistic Study of its Effects on H 2 O 2 Addition

The photocatalytic reaction involved in TiO 2 photocatalysis was investigated using a microreactor coated with TiO 2 film on the glass plate attached on one side of the microreactor. It was confirmed that the effect of H 2 O 2 on the photocatalytic degradation efficiency of azo dyes (acid orange 7, acid red 151, and acid yellow 23) was dependent on the polymorphs (anatase and rutile) of TiO 2 coated on the glass plate of the UV-irradiated microreactor. Scavengers of holes (KI) and electrons (p-benzoquinone) were added to the solution of azo dyes, and their effects on the degradation efficiencies of the azo dye (acid orange 7) in the microreactor system were investigated. It was found that the electron scavengers of p-benzoquinone showed much larger effects on the photocatalytic degradation efficiency than the hole scavengers of KI. Based on these results, the mechanism of the photocatalytic degradation of the azo dyes in the presence of H 2 O 2 was proposed.


Introduction
Photocatalysis is known to degrade organic compounds, inactivate or remove viruses, and finds a wide range of applications [1][2][3][4]. Among them, TiO 2 is a widely used photocatalyst due to its low cost, non-toxicity, and good oxidizing nature [5,6]. TiO 2 exists as three polymorphs of anatase (tetragonal), rutile (tetragonal), and brookite (orthorhombic) [7]. Since it is difficult to prepare the brookite TiO 2 without anatase and rutile phases [8], the photocatalytic activities and stabilities are mainly investigated for anatase and rutile TiO 2 . Although rutile TiO 2 is the stable form and anatase is the metastable form, the latter has higher photocatalytic activity than the former type [9]. Recently, Zhang et al. [10] investigated the difference in the photocatalytic activities of the three polymorphs using density functional theory calculations and suggested that the differences were attributed to the longer lifetime of the photoexcited electrons and holes in anatase due to its indirect transition, while rutile and brookite showed direct transition. Also, the effective mass of photogenerated electrons in anatase was lowest, which eased the migration of the photoexcited electrons compared to those in rutile and brookite forms.
The effects of H 2 O 2 addition on the photocatalytic activities of TiO 2 have been extensively investigated [11,12] to study the enhancement in the degradation rate of organic compounds. The accelerated degradation rate of the organic compounds was attributed to the oxidation of OH radicals formed by the UV-light photolysis of H 2 O 2 and the photocatalytic reaction of H 2 O 2 with photogenerated electrons on the TiO 2 surfaces (i.e., H 2 O 2 + e -→ OH + OH -) in addition to the heterogeneous photocatalytic redox reactions on the surface. Li et al. [13] observed the photodecomposition of H 2 O 2 on the TiO 2 surface under visible light irradiation and detected OH radicals by the ESR spectroscopic technique. Based on these results, the reaction mechanism for the decomposition of H 2 O 2 on the TiO 2 surface by the photocatalytic reduction reaction, such as H 2 O 2 on TiO 2 + e -→ OH + OH -, was elucidated. Furthermore, Li et al. [13] investigated the mechanism of H 2 O 2 decomposition by monitoring the HO 2 radical diffused from the TiO 2 surfaces in the gas phase using the cavity ring-down method. The HO 2 radicals diffused from the TiO 2 surfaces were successfully detected in the gas phase. Also, it was found that the lifetime of the HO 2 radicals was dependent on the type of TiO 2 . Thus, it was concluded that the HO 2 radicals were generated by the photocatalytic reaction on the TiO 2 surface, not by the photolysis of H 2 O 2 in the gas phase.
The direct detection of OH radicals released by the photocatalytic reaction on TiO 2 was performed by Nosaka and co-workers using coumarin and terephthalate as the fluorescence probe of the radicals [14,15]. The effect of H 2 O 2 addition on the amount of OH radicals formed by the photocatalytic reaction was studied. The amount of OH radicals increased in rutile TiO 2 but decreased in the anatase form. Thus, the difference in the reactivity of H 2 O 2 for anatase and rutile surfaces caused the difference in the effect of H 2 O 2 addition on the amount of OH radicals generated by the photocatalytic reaction. Since the photocatalytic activity was dependent not only on the polymorphs of TiO 2 but also on other parameters such as particle diameter, the comparison of the photocatalytic activity of the three TiO 2 polymorphs in the film form was preferable.
Recently, microreactors have attracted much attention as the reactions can be regulated in more precise and controlled conditions than those in a batch reactor [16]. There were numerous studies on photocatalytic reactions using microreactors. The review articles on the photocatalytic reaction using microreactors are also available in the literature 17 [17]. In the present study, the effect of H 2 O 2 addition on the TiO 2 photocatalytic reactions using conventional microreactor systems was investigated, and the difference in the reactivity of H 2 O 2 for anatase and rutile surfaces was determined by measuring the degradation of azo dyes inside the microreactors. Since the photocatalytic microreactor system avoids the complexity arising from the particle dispersion or aggregation in the TiO 2 suspension during the photocatalytic reaction, it is a desirable reactor to investigate such polymorph dependency of the H 2 O 2 addition on the TiO 2 photocatalytic reaction.

Fabrication of the TiO 2 Film on a Microreactor
The microreactor purchased from Key Chem-Lumi Co. Ltd. was used for the study without any further modification. It has an internal channel width of 1.0 mm, depth of 0.5 mm, and length of 1100 mm. In order to avoid the disclosure of the channel of the microreactor, TiO 2 film was deposited on the inner side of the quartz window in the microreactor by the sol-gel calcination method, as shown in Figure 1(a). The calcination temperature for the TiO 2 sol on the glass substrate was set to 200 °C. The TiO 2 sols of the anatase form (AT-01) were purchased from Photo-Catalytic Materials Co. Ltd., and those of the rutile form (RA-06) were a generous gift from Taki Chemical Co. Ltd. The crystal structure of TiO 2 on the glass substrate was verified by measuring the XRD patterns of each TiO 2 film, as shown in Figure 2.

Experimental Procedures
The schematic figure of the experimental setup is given in Figure 1(b). The azo dye solution injected by the microsyringe flowed to the microreactor at a speed of about 0.1 µL/min. Meanwhile, a 150W Xe lamp (L2273, Hamamatsu Co. Ltd.) was used to irradiate the TiO 2 films on the quartz glass attached to the flowing dye solution in the microreactor. Upon ceasing the flow inside the microreactor, the azo dye solution was collected in a glass container and used for the evaluation of the degradation efficiency by measuring the absorbance of the azo dye solution. The absorbance of the azo dye solution was recorded by a spectrometer (Lambda 35, Perkin Elmer, Inc.). The degradation efficiency is defined by the following equation based on the ratio of the absorbance of the azo dye solution with and without the photocatalytic reactions: Degradation efficiency (%) = 100 -{(absorbance before reaction)/(absorbance after reaction) × 100}. In order to understand the effect of H 2 O 2 addition, a certain concentration of H 2 O 2 was mixed in the azo dye solution, which was injected into the microreactor using the microsyringe. In the present study, three types of azo dyes (acid orange 7, acid red 151, and acid yellow 23) were used without any further purification. In order to confirm that the films of anatase and rutile TiO 2 were prepared by the sols of AT-01 and RA-06, respectively, the UV absorption spectra of these films were recorded.
The results are shown in Figure 3. The absorption peaks of the anatase and rutile films were observed at 380 nm and 400 nm, respectively. It is well known that the rutile TiO 2 shows an absorption peak at a longer wavelength than anatase. The present results were consistent with the XRD patterns to confirm that the films of the sol AT-01 and sol RA-06 corresponded to the anatase and rutile forms, respectively.

Preliminary Experiment: Measurement of the Fluorescence of Rhodamine 6G in the Microreactor
Before measuring the photocatalytic degradation efficiencies of the azo dyes in the photocatalytic microreactor by monitoring the change in their absorbance after the completion of the photocatalytic reaction, the in-situ measurement of the degradation process was carried out using the real-time fluorescence probe techniques. For the real-time fluorescence probe techniques, the quartz glass capillary was connected to the end of the flow channel in the microreactor instead of collecting the dye solutions that flowed through them. Further, the fluorescence of Rhodamine G in the glass capillary was recorded. The schematic figure of this experimental setup is shown in Figure 4. In the present experiment, Rhodamine G was used due to its strong fluorescence upon excitation with a 550 nm light. The wavelength of irradiated light was chosen by dispersing and selecting the wavelength of a metal halide lamp using a monochromator. The time evolution of the fluorescence intensity of Rhodamine G is given in Figure 5. The temporal decay of the fluorescence intensity was observed after the initiation of the irradiation of the Xe lamp. The time of the decay was found to be dependent on the flow rate, as shown in Figure 5. Since the Xe lamp irradiated the whole area of the quartz glass attached to the top of the microreactor, the dye solution of Rhodamine G continuously decomposed as far as the UV light irradiated the flowed azo dye solution in the microreactor due to the photocatalytic reaction of the TiO 2 film within the system. Since the dye solution at the entrance of the microreactor suffered the TiO 2 photocatalytic reactions for a longer time, the fluorescence intensity proportional to the concentration of Rhodamine G continued to decay until the solution reached the end of the microreactor.  Considering the whole length of the reaction flow in the microreactor as L (m) and the time duration taken to achieve constant fluorescence intensity as t 0 (sec), the flow rate is expected to be proportional to the value of L/t 0 . Since the length of the flow in the microreactor L (m) is constant, the flow rate is expected to be proportional to the reciprocal to the time taken to achieve constant fluorescence intensity (that is, t 0 (sec)). The reciprocal of the time taken to achieve constant fluorescence intensity of rhodamine G versus flow rate was plotted, as shown in Figure 6.

Figure 6
Relationship between the reciprocal of the time taken to achieve constant fluorescence intensity and flow rate of the microreactor.
As shown in Figure 6, a linear plot was obtained between the reciprocal of the time taken to achieve constant fluorescence intensity of rhodamine G and flow rate, suggesting that the decay of the fluorescence intensity was attributed to the photocatalytic reaction inside the microreactor caused by the irradiation of the microreactor with the Xe lamp. It was also found that the fluorescence intensity was constant in the absence of UV light irradiation on the Rhodamine 6G solution, suggesting no or very little adsorption of the dye on the wall of the microreactor or quartz glass during the flow process.
Thus, the azo dye solution at the end of the flow channel in the microreactor was irradiated with Xe lamp to evaluate the degradation efficiency of the photocatalysis in the following experiments.

Effects of the Addition of H 2 O 2 : Comparison between Anatase and Rutile TiO 2
Since the degradation process in the microreactor by the occurrence of photocatalytic reaction was confirmed in the present system, the effects of H 2 O 2 on the photocatalytic degradation efficiency were investigated. In order to perform this experiment, a certain amount of H 2 O 2 was mixed with the azo dye solution and injected into the microreactor. The azo dye with an acid orange 7 was used for the test. The results are presented in Figure 7. In the anatase TiO 2 film (AT-01, Photo-Catalytic Materials), the degradation efficiency monotonically decreased with the increasing concentration of H 2 O 2 , as shown in Figure 7 [14] reported that the anatase TiO 2 form released less amount of OH radicals upon the addition of H 2 O 2 , while the rutile form showed enhanced OH radical formation. The present results are consistent with the reported experimental observations. Harir et al. [11] performed a detailed kinetic study of the H 2 O 2 /TiO 2 system using commercially available TiO 2 (P25, Deggusa) powders. Also, the effect of H 2 O 2 on photocatalytic activity was investigated and reported to be maximum at around 10 mM of H 2 O 2 . The AT-01 and RA-06 sols were mixed to develop this mixed-phase film. The results are presented in Figure 8. Irreversible to the anatase-rutile ratios of TiO 2 , degradation efficiency, The photocatalytic activity was maximum at around 10 mM of H 2 O 2 , which was consistent with the results of Harir et al. [11], suggesting that the effect of H 2 O 2 on the degradation efficiency was attributed to the crystal forms. In other words, the rutile form in the mixed crystal phase of TiO 2 played a key role in the effect of H 2 O 2 addition. The irreversible morphology of the TiO 2 was film or powders. To further confirm the effect of H 2 O 2 concentration on the degradation efficiency of the azo dyes, two different azo dyes of acid red 151 and acid yellow 23 were chosen to perform studies on the anatase and rutile TiO 2 films, respectively. The results are illustrated in Figure 9 and Figure 10. Although the acid orange 7 azo dye was replaced by acid red 151 and acid yellow 23, the effect of H 2 O 2 on the degradation efficiency by anatase and rutile was similar to the trend observed in the former. This suggested that the effect of H 2 O 2 on the degradation of the azo dyes within the TiO 2 /H 2 O 2 system was different for different crystal structures of TiO 2 (anatase or rutile form).

Effects of the Addition of Scavengers: Discussion of the Mechanism of Anatase and Rutile TiO 2 under the Microenvironmental Conditions
The photocatalytic degradation efficiencies of the three azo dyes by the UV-irradiation of anatase, rutile, and mixed TiO 2 in the microreactor were presented. The effects of H 2 O 2 concentration on the degradation efficiency of TiO 2 photocatalysis in the microreactor between anatase and rutile were observed. Although there are still some debates on the relative role of holes and OH radicals on the degradation of azo dyes, several authors confirmed the significance of OH radicals on the degradation process [18,19]. Since the OH radicals are formed by two pathways, namely water oxidation and reduction of H 2 O 2 produced by the multielectron reduction of oxygen, it is important to identify these radicals contributing to the degradation of azo dyes. Therefore, this study attempted to identify the mechanism of the formation of the OH radicals attributing to the degradation process, which was investigated by the addition of scavengers.
Based on the previous studies, benzoquinone (hereafter, BQ) [20] and KI [21] were used as the electron and hole (OH radical) scavengers, respectively, and acid orange 7 was chosen as the representative azo dye solution in the present work. The results are presented in Figure 11 and Figure 12. It was observed that the electron scavengers of BQ showed relatively larger effects compared to the slight effects of KI on the photocatalytic degradation efficiency of acid orange 7. Since the OH radicals were formed by the reduction reaction of O 2 via H 2 O 2 , the influence of the addition of BQ on the degradation efficiency of the azo dye (acid orange 7) in the presence of H 2 O 2 was also investigated.  The experimental results are presented in Figure 13. It was observed that the influence of BQ on the reduction of the photocatalytic degradation efficiency was very large in the presence of 10 mM of H 2 O 2 . When the concentration of BQ rose to 10 mM, the degradation efficiency was reduced to zero. The significance of the OH radicals formed by the reduction reaction by H 2 O 2 generated by the reduction of O 2 on the degradation mechanism was not explained in the literature. Thus, the results of the current study suggested the important role of the OH radicals formed by the reduction reaction. Finally, the mechanism of the scavengers and OH-radical formation by the photocatalytic reaction is illustrated in Figure 14. Thus, the difference in the photocatalytic degradation efficiency of TiO 2 was attributed to the different reactivity of the anatase and rutile forms in the reduction reaction. Nosaka and co-workers [14,22] reported that the different structures of the peroxy species of the TiO 2 surfaces in the anatase and rutile forms contributed to the difference in reactivity, as illustrated in Figure 15. The present results on the different photocatalytic degradation efficiencies upon the H 2 O 2 addition might be attributed to the same factor, as reported by Nosaka et al. [22,23], Hayashi et al. [24] and Murakami et al. [25,26] in the literature. Further studies are required to clarify their significance in the degradation mechanism.

Figure 15
Schematic representation of the peroxy species on the anatase and rutile TiO 2 surfaces, and the proposed mechanism for the formation of OH from the peroxy species on the TiO 2 surface [14].

Conclusions
The degradation efficiency of three azo dyes in the photocatalytic microreactor system was studied along with the effect of H 2 O 2 on the degradation process. It was found that the photocatalytic degradation efficiency of acid orange 7 within the TiO 2 photocatalytic microreactor system was monotonically decreased in the presence of the anatase form of TiO 2 . However, the degradation efficiency was enhanced in the presence of the rutile form of TiO 2 up to the H 2 O 2 concentration of 10 mM and then decreased with the increase in the concentration of H 2 O 2 . The other azo dyes, such as acid red 151 and acid yellow 23, also showed similar trends. In order to investigate the effect of H 2 O 2 on the photocatalytic degradation efficiency of the anatase and rutile forms, scavengers of holes and electrons were added in the photocatalytic microreactor systems. From these experiments, the importance of the OH radicals formed by the reduction of O 2 was suggested. Thus, it was speculated that the difference in the effect of H 2 O 2 on the photocatalytic degradation efficiency was probably attributed to the different reactivity and reduction reactions of H 2 O 2 on the anatase and rutile TiO 2 surfaces.