Abstract

The simultaneous production of molecular hydrogen (H2) and degradation of rhodamine B (RhB) was successfully achieved using TiO2 modified with platinum and nafion (Pt/TiO2/Nf) under visible light  nm). Pt/TiO2/Nf exhibited high activity for H2 production in the presence of RhB and EDTA as a photosensitizer (also an organic dye pollutant) and an electron donor, respectively. However, the activity of TiO2 modified with either platinum or nafion for H2 production was negligible under the same experimental conditions. The negatively charged nafion layer enhances the adsorption of cationic RhB and pulls protons, a source of hydrogen, to the surface of TiO2 through electrostatic attraction. On the other hand, platinum deposits on TiO2 can act as an electron sink and a temporary electron reservoir for the reduction of protons. With the production of H2, RhB was gradually degraded through -deethylation, which was confirmed by the spectral blue shift of the maximum absorption wavelength from 556 to 499 nm (corresponding to the of rhodamine 110). With Pt/TiO2/Nf employed at M (0.6 mol), approximately 70 mol of H2 was produced and RhB and its intermediates were completely removed over a 12 h period. A detailed reaction mechanism was discussed.

1. Introduction

Dye-sensitized TiO2 systems have been studied extensively for the remediation of environmental pollutants and the production of hydrogen under visible light [115]. Both applications are based on photo-induced electron transfer from the dyes to electron acceptors (e.g., pollutants and protons) through the TiO2 conduction band (CB). Although the photonic efficiency of this system is high, expensive synthetic dyes (e.g., Ru-based [1, 36, 10], (metal) porphyrin [5, 7], and engineered organic dyes [2, 8, 1115]) have been primarily used. In addition, most dye-sensitized TiO2 systems suffer from the instability of the dye because dyes can be degraded after electron injection from the excited dye to the TiO2 CB. The high price and low stability of dyes limit the practical applications of dye-sensitized TiO2 systems.

Organic dyes are one of the most serious pollutants in the aquatic environment due to their high production volumes from industry, toxicity, and low biodegradability. The annual global production of organic dyes is approximately one million tons, and a significant amount (10–15%) of wastewater containing dyes is discharged to the surface water without any treatment [16, 17]. In addition, some organic dyes are carcinogenic to humans and negatively affect aquatic organisms by interfering with their metabolic processes [1618]. However, the biological treatments widely employed in water treatment are usually inefficient for the degradation of organic dyes [19].

Recently, bifunctional TiO2 photocatalysts have been developed, particularly for simultaneous hydrogen production and pollutant degradation. This bifunctionality was achieved through the surface modification of TiO2 with metal nanoparticles or two different components (i.e., anions and metal nanoparticles) [2025]. However, the previously developed bifunctional TiO2 photocatalysts worked only under UV light, limiting their practical applications because UV light accounts for only 3% of natural sunlight at ground level. Visible active photocatalysts can be used to achieve the simultaneous hydrogen production and pollutant degradation under visible light [26]. Another approach to addressing this challenge would be to use the organic dyes to be treated as a photosensitizer for hydrogen production instead of expensive synthetic dyes, while simultaneously degrading these organic dyes under visible light (44% of natural sunlight).

In this work, we successfully achieved the simultaneous production of hydrogen and degradation of rhodamine B (RhB, an azo dye that accounts for more than 65% of total dye production [16]) using TiO2 modified with platinum and nafion (Pt/TiO2/Nf) under visible light. The effects of various experimental parameters (e.g., initial pH (), RhB concentration, and EDTA concentration) on H2 production were investigated. In addition, a detailed reaction mechanism for the simultaneous production of hydrogen and degradation of RhB was suggested.

2. Experimental

2.1. Materials and Reagents

The materials and reagents were used as received without further purification. These substances include TiO2 powder (Hombikat, UV-100), rhodamine B (RhB, C28H31ClN2O3, Aldrich), chloroplatinic acid (H2PtCl6, Aldrich), methanol (CH4O, J. T. Baker), nafion perfluorinated resin solution (5 wt% solution in a mixture of alcohol and water, Aldrich), and ethylenediaminetetraacetic acid disodium salt dihydrate (EDTA, C10H14N2Na2O8·2H2O, Aldrich). Ultrapure deionized water (18.3 MΩ cm) prepared by a water purification system (Barnstead) was used.

2.2. Catalyst Preparation

Platinum (Pt) nanoparticles were deposited onto the surface of TiO2 using a photodeposition method [27]. An aqueous TiO2 suspension (0.5 g/L, 500 mL) containing chloroplatinic acid as a Pt precursor (100 μM) and methanol as an electron donor (1 M) was irradiated with a 200 W mercury lamp for 30 min. Next, the Pt-deposited TiO2 (Pt/TiO2) powder was collected by filtration through a 0.45 μm PVDF disc filter (Pall), washed with distilled water, and dried in an oven at 70°C. The typical Pt loading on TiO2 was estimated to be ca. 3 wt% by measuring the concentration of unused chloroplatinic acid remaining in the filtrate solution after the photodeposition using inductively coupled plasma atomic emission spectroscopy (ICP-AES, Thermo Jarrell Ash Corp.). To obtain Pt/TiO2/Nf (TiO2 modified with Pt and nafion), an aliquot of nafion solution (0.1 mL) was added to the Pt/TiO2 powder (0.1 g), mixed well, and dried at room temperature overnight [28].

2.3. Photocatalysis

The catalyst powder was dispersed in distilled water by sonication for 30 s in an ultrasonic cleaning bath. An aliquot of the RhB and EDTA stock solution was subsequently added to the suspension to yield the desired initial concentration. The initial pH () of the suspension was adjusted with HClO4 solution. The total volume of the suspension was 30 mL. Prior to visible light irradiation, N2 gas (99.9%) was purged through the suspension for 1 h to remove dissolved oxygen, and then the reactor was sealed with a rubber septum. A 300 W Xe arc lamp (Oriel) was used as a light source. Light was passed through a 10 cm IR water filter and a cutoff filter ( nm), and then the filtered light was focused onto a cylindrical glass reactor with a quartz window.

2.4. Analysis

The amount of photogenerated molecular hydrogen (H2) in the headspace of the reactor was analyzed using a gas chromatograph (GC, HP6890A) equipped with a thermal conductivity detector and a 5 Å molecular sieve column. Sample aliquots were withdrawn from the visible-light-irradiated reactor and filtered through a 0.45 μm PTFE syringe filter (Millipore) to remove catalyst particles prior to the analysis of RhB. The color disappearance of RhB and its maximum absorption wavelength () shift resulting from the stepwise -deethylation [29] were monitored using a UV-visible spectrophotometer (Shimadzu UV-2401PC). It should be noted that RhB (colored) is transformed to leuco RhB (colorless) through the addition of H (formed from the reduction of H+) in the absence of oxygen, but leuco RhB is dehydrogenated back to RhB in the presence of oxygen [30]. To exclude the color disappearance through the formation of leuco RhB, the absorption spectrum of samples was measured after exposure to air. The concentrations of RhB adsorbed on the catalyst surface were calculated by subtracting the equilibrated concentrations (with catalyst, after 30 min in the dark) from the initial concentrations (without catalyst). The concentration of RhB was determined by measuring the absorbance at 556 nm.

3. Results and Discussion

3.1. Effect of Nafion Coating on H2 Production

RhB, a representative azo dye pollutant, was selected as a photosensitizer, and the production of H2 in the presence of EDTA (electron donor) was compared between Pt/TiO2 and Pt/TiO2/Nf (Figure 1(a)). The production of H2 in the suspension of Pt/TiO2 was very low but was markedly enhanced by the nafion coating of Pt/TiO2.

Under visible light, RhB adsorbed on the TiO2 surface is excited (see, reaction (1)), and then electrons are transferred from the excited RhB to protons (or water molecules) through the TiO2 CB and Pt nanoparticles (see, reactions (2)–(4)): In these processes, Pt nanoparticles enhance the interfacial electron transfer from the TiO2 CB to protons as an electron sink (Schottky-barrier electron trapping) and a temporary electron reservoir that enables the two-electron reduction of protons (see, reactions (3) and (4)) [32, 33]. In the absence of Pt nanoparticles (i.e., in the cases of TiO2 and TiO2/Nf), the production of H2 was negligible (data not shown). The coating of nafion, an anionic perfluorinated polymer with a sulfonate group (), changes the surface charge of TiO2 ( [9]) from positive to negative under acidic conditions [27, 28, 34]. Therefore, the adsorption of cationic RhB on the TiO2 surface should be enhanced, which accelerates the electron transfer from the excited RhB to the TiO2 CB and, eventually, the production of H2.

Figure 1(b) shows the adsorption of RhB on the surfaces of Pt/TiO2 and Pt/TiO2/Nf under the same conditions ([RhB] = 20 μM and ). The amount of RhB adsorbed on Pt/TiO2 was very low (0.7 μM, 3.5%) because the positively charged surface of Pt/TiO2 at repels the cationic RhB molecules. Under this condition, the interfacial electron transfer from the excited RhB to the Pt/TiO2 CB should be limited. On the other hand, the adsorption of RhB on the surface of Pt/TiO2/Nf was significant (18.0 μM, 90.0%) because the negatively charged surface induced by nafion attracts cationic RhB molecules. This result can help to explain why the production of H2 is markedly enhanced by the nafion coating.

3.2. Effect of Various Parameters on H2 Production

In dye-sensitized TiO2 systems, the solution pH has a significant effect on the production of H2 because the adsorption kinetics of dyes on the TiO2 surface strongly depends on the pH-dependent surface charge of TiO2. Figure 2 shows the production of H2 and the adsorption of RhB in the suspension of Pt/TiO2 and Pt/TiO2/Nf as a function of . The pH-dependent H2 production trends for Pt/TiO2 and Pt/TiO2/Nf were opposite: as the increased, H2 production gradually increased for Pt/TiO2 but decreased for Pt/TiO2/Nf (Figure 2(a)). However, it should be noted that the H2 production in the suspension of Pt/TiO2/Nf was much higher than that in the suspension of Pt/TiO2 over the whole pH range of 2.3 to 5.0.

The surface charge of Pt/TiO2 is strongly positive at because most of the surface hydroxyl groups are protonated. Under this condition, the adsorption of cationic RhB on TiO2 should be inhibited, which results in the negligible production of H2. However, the number of surface hydroxyl groups that are not protonated increases as the pH increases. This makes the surface charge of TiO2 less positive and enables the adsorption of cationic RhB on TiO2 (see, reaction (5)) and H2 production. As expected, the adsorption of RhB on the surface of Pt/TiO2 was negligible at but was clearly observed at , albeit in small amounts (Figure 2(b)): ( [35]) (the surface charge of TiO2 is determined by assuming that the surface Ti having +4 formal charge is located at the octahedral site surrounded by five lattice oxygen atoms and one surface group).

On the other hand, the surface charge of Pt/TiO2/Nf is highly negative, even at acidic pH, because the anionic sulfonate groups () in the nafion layer outnumber the protonated surface hydroxyl groups of TiO2 (>) [27, 28]. Therefore, sufficient RhB to have little effect on H2 production can be adsorbed on the surface of Pt/TiO2/Nf at both and 5.0 (Figure 2(b)). Under this condition, the decrease in the production of H2 with increasing pH implies the existence of another mechanism for H2 production with Pt/TiO2/Nf. It has been reported that the concentration of protons in the nafion layer is much higher than that in the aqueous bulk phase due to the electrostatic attraction between the positively charged protons and anionic sulfonate groups [36]. Because the electron transfer from Pt nanoparticles to protons is kinetically more favorable than that to undissociated water molecules [ M−1 s−1 and  M−1 s−1] [37], the locally concentrated protons within the nafion layer provide good conditions for H2 production. At lower pH, more protons can be trapped within the nafion layer, which enhances the electron transfer from Pt nanoparticles to protons. Therefore, the production of H2 with Pt/TiO2/Nf increases as the pH decreases, in contrast to the case of Pt/TiO2. Overall, the nafion-enhanced H2 production in the RhB-sensitized Pt/TiO2/Nf system is ascribed to two factors. First, the negatively charged nafion layer enhances the adsorption of cationic RhB. Second, protons, which are more favorably reduced to H2 than undissociated water molecules, are concentrated within the nafion layer.

Figure 3 shows the effect of RhB and EDTA concentrations on the production of H2 in the suspension of Pt/TiO2/Nf. The production of H2 was negligible in the absence of either RhB or EDTA, which clearly indicates that RhB and EDTA act as a photosensitizer and an electron donor in the RhB-sensitized Pt/TiO2/Nf system, respectively. The production of H2 rapidly increased with increasing RhB and EDTA concentrations and then saturated at [RhB] = 10 μM and [EDTA] = 2 mM.

The fact that H2 was not produced in the absence of EDTA (electron donor) indicates that the electron transfer from the TiO2 CB to RhB•+ (i.e., charge recombination, reaction (6)) is much faster than that from the TiO2 CB to Pt nanoparticles (see, reaction (3)), which is consistent with other dye-sensitized TiO2 systems using expensive synthetic dyes for H2 production [3, 7]. However, EDTA can prevent the charge recombination process by regenerating RhB•+ to RhB (i.e., electron transfer from EDTA to RhB•+, reaction (7)), which enables the electron transfer from the TiO2 CB to Pt nanoparticles and, eventually, the production of H2:

3.3. Mechanism of RhB Degradation

The stability of dye molecules is an important parameter for the evaluation of dye-sensitized TiO2 systems using expensive synthetic dyes. On the other hand, the degradation of dye during the course of H2 production should increase the utility of dye-sensitized TiO2 systems if organic dye pollutants are used as a photosensitizer. This is conceptually similar to bifunctional photocatalysis (i.e., the simultaneous production of hydrogen and degradation of pollutants) [2026].

Figure 4 shows the absorption spectral change (i.e., degradation) of RhB in the suspension of Pt/TiO2 and Pt/TiO2/Nf as a function of irradiation time. In accordance with the absence of H2 production with Pt/TiO2 at (Figure 2(a)), the degradation of RhB was negligible (Figure 4(a)). On the other hand, with Pt/TiO2/Nf, the intensity of the RhB absorption spectrum decreased and its position shifted to shorter wavelengths with irradiation time (Figure 4(b)). This result clearly indicates that RhB is degraded with H2 production in the suspension of Pt/TiO2/Nf.

Some dyes including RhB can be degraded through two pathways: the cleavage of the chromophoric ring and -de-alkylation [29, 38, 39]. Between these pathways, the cleavage of the chromophoric ring is not favored under our experimental conditions (i.e., in the absence of oxygen) because it is primarily initiated by the reaction between reactive oxygen species (e.g., and OH) and RhB•+ in the bulk phase [29, 38]. On the other hand, it has been reported that the -deethylation proceeds through the electron transfer from RhB to the TiO2 CB and subsequent hydrolysis of RhB•+ [29], leading to a blue shift of the maximum absorption wavelength () because -de-ethylated RhB intermediates exhibit at shorter wavelengths than RhB ( nm) [29, 38]. With Pt/TiO2/Nf, the shifted from 556 to 502 nm after 4 h of visible light irradiation. This significant spectral shift of () clearly indicates that the degradation of RhB is primarily initiated by -deethylation.

RhB was not degraded in the absence of EDTA (data not shown), which implies that EDTA•+, not the hydrolysis of RhB•+, plays a critical role in the -deethylation of RhB in our system. This seems to be because the -deethylation through the hydrolysis of RhB•+ is much slower than the reduction of RhB•+ by the TiO2 CB electron (see, reaction (6)) in the absence of oxygen. In this situation, the -deethylation of RhB should be due to the oxidation of RhB (or more preferentially RhB•+) by EDTA•+ (see, reaction (8)) generated from the oxidation of EDTA by RhB•+ (see, reaction (7)): It has been reported that EDTA•+ can oxidize -tetramethyl--phenylenediamine (TMPD), which has electron-donating alkyl groups [40]. Likewise, EDTA•+ could react with the ethyl group of RhB•+ to generate -de-ethylated RhB intermediates. It should be noted that RhB is regenerated when RhB•+ accepts one electron from the TiO2 CB or EDTA (see, reactions (6) and (7)). However, -de-ethylated RhB intermediates generated by further oxidation of RhB•+ (see, reaction (8)) should not be regenerated to RhB and undergo further degradation.

Figure 5(a) shows the spectral shift of depending on [EDTA] and . In all cases, shifted from 556 to 499 nm and then remained constant. Rh-110 (the fully -de-ethylated form of RhB) exhibits an absorption maximum at 499 nm [28]. Therefore, the generation of Rh-110 further confirms that RhB is degraded through -deethylation in the suspension of Pt/TiO2/Nf with EDTA.

The -deethylation rate, which is proportional to the rate, was greatly dependent on both [EDTA] and . The -deethylation at [EDTA] = 4 mM was slower than that at [EDTA] = 0.4 mM, as the reaction of RhB•+ with EDTA (regeneration of RhB, reaction (7)) becomes more favored than that with EDTA•+ (-deethylation of RhB, reaction (8)) as the concentration of EDTA increases. On the other hand, the -deethylation at was faster than that at at the same [EDTA], which is related to the pH-dependent speciation of EDTA. At , EDTA primarily exists as H2EDTA2−, which can be repelled from the negatively charged surface of TiO2 induced by the nafion (Figure 5(b)). On the other hand, H4EDTA and H3EDTA, which can more favorably approach the negatively charged surface, are the main species at (Figure 5(b)). Therefore, the concentration of EDTA within the nafion layer at should be lower than that at despite the concentration of EDTA added being the same. A lower concentration of EDTA within the nafion layer at higher pH reduces the regeneration of RhB and therefore enhances the degradation of RhB. The lower production of H2 at higher pH (see Figure 2(a)) should also be related to the faster degradation of RhB at higher pH (see Figure 5(a)). The degradation of RhB reduces the number of RhB molecules adsorbed on the surface of Pt/TiO2/Nf and eventually decreases the production of H2. Overall reactions occurring in the suspension of Pt/TiO2/Nf with RhB and EDTA under visible light are illustrated in Scheme 1.

3.4. Long-Term Experiment

In the suspension of Pt/TiO2/Nf, H2 was continuously produced up to 12 h at [RhB] = 20 μM (0.6 μmol), [EDTA] = 4 mM, and , although the RhB (i.e., absorbance at  nm) was completely removed within 4 h (Figure 6). This result implies that not only RhB but also the intermediates generated from the degradation of RhB can act as a photosensitizer. However, the rate of H2 production gradually decreased (9.0 μmol/h for 0–3 h, 6.4 μmol/h for 3–6 h, 3.6 μmol/h for 6–9 h, and 2.3 μmol/h for 9–12 h) as RhB and its intermediates were degraded (Figure 6(a)). The degradation of Rh-110 ( nm) resulting from the complete -deethylation of RhB proceeded (i.e., the absorbance at  nm continuously decreased) after 4 h and was completed after 12 h (Figure 6(b)). In accordance with the complete degradation of RhB and its intermediates after 12 h, the production of H2 stopped after 12 h.

Using Pt/TiO2/Nf under visible light, approximately 70 μmol of H2 was produced and 20 μM (0.6 μmol) of RhB and its intermediates were completely degraded over a 12 h period without the use of external energy or chemical oxidants (Figure 6). Although high concentrations of EDTA (millimolar levels) are required in the RhB-sensitized Pt/TiO2/Nf system for the simultaneous production of H2 and degradation of RhB, many industrial wastewaters (e.g., pulp and paper, textile, and cosmetics wastewaters) contain high concentrations of EDTA [41]. In addition, EDTA can be easily degraded through conventional biological treatment, unlike RhB [42, 43]. Therefore, the application of Pt/TiO2/Nf for the simultaneous production of H2 and degradation of organic dye pollutants could become practicable by using industrial wastewaters containing high concentrations of EDTA.

4. Conclusions

The present study introduces a new strategy for visible-light-induced bifunctional photocatalysis (i.e., the simultaneous production of hydrogen and degradation of pollutants) using TiO2 modified with platinum and nafion (Pt/TiO2/Nf) and organic dye pollutants. In the presence of RhB (as both a photosensitizer and an organic dye pollutant) and EDTA (as an electron donor), Pt/TiO2/Nf exhibited considerably higher activity for H2 production than Pt/TiO2. This is ascribed to the negatively charged surface of TiO2 induced by the nafion, which enhances the adsorption of cationic RhB and pulls protons, a source of hydrogen, to the surface of TiO2. In addition, the degradation of RhB was accompanied by the concurrent production of H2. The intensity of the RhB absorption spectrum decreased and its position shifted to shorter wavelengths, which indicates that RhB is primarily degraded through -deethylation. EDTA•+, which is generated from the oxidation of EDTA by RhB•+, is found to be involved in the degradation mechanism, as RhB was not degraded in the absence of EDTA. Rh-110, a fully -de-ethylated form of RhB, was further degraded and the production of H2 continued until RhB and its intermediates were completely degraded. Based on its high efficiency, bifunctionality, and visible light activity, this organic dye pollutant-sensitized TiO2 system using Pt/TiO2/Nf can be proposed as a viable photocatalytic system for the simultaneous production of hydrogen and degradation of organic dye pollutants.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgments

This research was supported by Basic Science Research Program (no. 2013R1A1A1007312 and no. 2012R1A2A2A01004517), Framework of International Cooperation Program (no. 2013K2A1A2052901), Korea Center for Artificial Photosynthesis (no. 2012M1A2A2671779) through the National Research Foundation of Korea (NRF), and DGIST research program (14-EN-01) funded by the Ministry of Science, ICT & Future Planning. Jungwon Kim is also grateful for the Hallym University Research Fund (HRF-G-2013-3).