Iron-functionalized titanium dioxide on flexible glass fibers for photocatalysis of benzene, toluene, ethylbenzene, and o-xylene (BTEX) under visible- or ultraviolet-light irradiation

Abstract

Iron-functionalized titanium dioxide (TiO₂) composites with various Fe-to-Ti ratios were prepared on flexible glass fibers (GF-Fe-TiO₂) via a sol-gel method, followed by a dip-coating process. The photocatalytic ability of these composites in degrading selected volatile organic compounds (VOCs; benzene, toluene, ethylbenzene, and o-xylene [BTEX]) at indoor concentration levels was examined. The GF-Fe-TiO₂ composites were characterized using scanning electron microscopy, energy-dispersive X-ray elemental analysis, ultraviolet (UV)-visible spectroscopy, and X-ray diffraction. The GF-Fe-TiO₂ composites showed superior photocatalytic performance to that of a reference glass fiber–supported TiO₂ photocatalyst for the treatment of BTEX under visible light. However, this trend was reversed under UV irradiation. Specifically, the average BTEX photocatalytic efficiencies of the 0.01-GF-Fe-TiO₂ composite in a 3-hr visible-light photocatalytic process were 4%, 33%, 51%, and 74%, respectively. Conversely, the average BTEX photocatalytic efficiencies obtained for GF-TiO₂ were close to 0%, 5%, 16%, and 29%, respectively. These findings demonstrated that the GF-Fe-TiO₂ composites could be applied to photocatalytically purify BTEX, especially under visible-light exposure. Moreover, the GF-Fe-TiO₂ composites prepared with different Fe-to-Ti ratios displayed different BTEX photocatalytic decomposition efficiencies under visible or UV light, allowing for optimization of the Fe-to-Ti ratio (which was found to be 0.01).

Implications: The application of nanomaterials for air purification necessitates a supporting material to stabilize them while in contact with the treated air in the photocatalytic chamber. Glass fibers have an obvious advantage over other supporting materials mainly because of its flexibility, which makes it much easier to handle. However, the applications of glass fiber–supported, visible light–activated photocatalysts to the treatment of air pollutants are rarely reported in literature. Accordingly, this study aimed to investigate the applicability of glass fiber–supported Fe-TiO₂ for the purification of VOCs under visible- as well as UV-light exposure.

Introduction

Photocatalysis over titanium dioxide (TiO₂) has recently received increasing attention for its ability to degrade a range of environmental pollutants. There are several advantages associated with the usage of TiO₂, including low cost, high chemical and thermal stability, and good oxidation capacity (Gaya and Abdullah, 2008; Paz, 2010; Folli et al., 2012; Lan et al., 2013). However, environmental applications for stand-alone TiO₂ nanopowders have been limited by the low photocatalytic efficiency resulting from the high recombination effects of their charge carriers. TiO₂ nanopowders are also limited to photocatalytic activation under ultraviolet (UV) irradiation, because of the broad band gap of TiO₂ (Paola and García-López, 2012). Consequently, many surface modification techniques have been used for enhancing the performance of TiO₂ photocatalysts (Park et al., 2013). Doping of transition or noble metals into TiO₂ is a promising method for improving the photocatalytic performance in the treatment of environmental contaminants (Bouras et al., 2007; Vargas et al., 2012; Yu et al., 2013; Inturi et al., 2014). Iron (Fe)-functionalized TiO₂ composites (Fe-TiO₂) are among the most popular transition metal–functionalized TiO₂ photocatalysts. Their popularity is a result of the reduced electron-hole pair recombination and narrow band gaps relative to conventional TiO₂ photocatalysts. Consequently, Fe-TiO₂ has been effectively used for the degradation of a number of environmental pollutants under visible-light irradiation (Cui et al., 2009; Tieng et al., 2011; Sun et al., 2012; Vargas et al., 2012). Moreover, Fe is less expensive than noble metals such as Pd, Pt, Rh, Ag, and Au, which allows for its widespread usage.

The application of nanomaterials for air purification necessitates the use of a supporting material to stabilize them while in contact with the treated air in a photocatalytic chamber (Bianchi et al., 2012). A variety of materials have been explored as supports for TiO₂ nanopowders for the purification of airborne pollutants. These include powdered activated carbon (Matos et al., 2010), activated carbon fibers (Jo et al., 2011), glass tubes (Jo and Kim, 2009), polymer materials (Kedem et al., 2009), Teflon (Bianchi et al., 2012), steel (Bianchi et al., 2012), and glass fibers (GFs) (Lin et al., 2013). GFs have an obvious advantage over other supporting materials because of their flexibility, making them much easier to handle (Panniello et al., 2012; Lin et al., 2013). Moreover, Lin et al. (2013) reported that GF-supported Fe-TiO₂ (GF-Fe-TiO₂), synthesized by a dip-coating sol-gel process, exhibited photocatalytic performance superior to that of the fiber-supported pure TiO₂ for the decomposition of phenol in an aqueous medium. However, the application of GF-Fe-TiO₂ to the treatment of airborne pollutants is relatively unknown. In fact, the light absorbance kinetics and reaction mechanisms of contaminants differ between the water-solid and air-solid boundaries (Henderson, 2011). Accordingly, extrapolation of the photocatalytic performance of GF-Fe-TiO₂ from solvent applications to gas applications may not be sound. Therefore, specific investigation of the application of GF-Fe-TiO₂ to the decomposition of airborne pollutants is required.

In this study, Fe-TiO₂ photocatalysts with different Fe-to-Ti ratios were synthesized using tetrabutyl titanate (TnBT) and iron(III) nitrate as the Ti and Fe sources, respectively. The photocatalysts were then dip-coated on to flexible GFs to prepare GF-Fe-TiO₂ composites, and their photocatalytic activity was determined for the degradation of low-concentration toxic air pollutants (benzene, toluene, ethylbenzene, and o-xylene [BTEX]). These pollutants were selected based on their prevalence in urban indoor and outdoor environments (Esplugues et al., 2010) as well as their associated health hazards (Ramírez et al., 2012). Atmospheric BTEX mainly originates from anthropogenic sources such as motor vehicle exhaust gases, combustion emissions, and a variety of industrial handling and storage processes (Leuchner and Rappenglück, 2010). In addition, there are many indoor BTEX sources, such as household products, paints, furniture, and building finishing materials (Shin and Jo, 2012). Additionally, a GF-TiO₂ photocatalyst was synthesized, and its photocatalytic activity was determined and compared with those of the GF-Fe-TiO₂ composites.

Methods

Synthesis and characterization of photocatalysts

GF-Fe-TiO₂ composites with different Fe-to-Ti ratios were prepared using a sol-gel procedure, followed by a dip-coating process. First, 17 mL of TnBT (Ti[O(CH₂)₃CH₃]₄; 97%; Aldrich, St. Louis, MO), 0.5 mL of nitric acid (98%; Aldrich), and 1.5 mL of deionized water were added sequentially to 136 mL of ethanol (99.9%; Aldrich). This was followed by the addition of either 0.001, 0.002, 0.01, 0.02, or 0.1 g of iron(III) nitrate (Fe(NO₃)₃; 99.99%; Aldrich) to the solution, resulting in Fe-TiO₂ composites with Fe-to-Ti ratios of 0.005, 0.01, 0.05, 0.1, and 0.5, respectively. A clean GF sheet (11.5 × 20 cm; C120-3K; Hyundai Fiber Co., Hanam, Korea) was subsequently dipped into the mixture for 10 min and then pulled out slowly at a rate of 1 cm min⁻¹. Following this, the coated GF was dried at room temperature for 3 hr. This coating procedure was repeated three times to obtain high Fe-TiO₂ loading of the samples. The final products were calcined in an electric furnace, in which the temperature was increased to 400 °C at a rate of 2 °C min⁻¹, and was then held constant for 1 hr before finally being cooled to room temperature over 24 hr. The GF-Fe-TiO₂ composites with Fe-to-Ti ratios of 0.005, 0.01, 0.05, 0.1, and 0.5 were designated as 0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂, respectively. In addition, a reference photocatalyst (0.05-GF-TiO₂) was prepared, following the same procedure used for the preparation of GF-Fe-TiO₂, but without the addition of Fe(NO₃)₃. The characteristics of the GF-Fe-TiO₂ and GF-TiO₂ photocatalysts were determined via field emission scanning electron microscopy (FE-SEM) equipped with an energy-dispersive X-ray (EDX) spectrometer (Hitachi S-4300 and EDX-350 FE-SEM; Tokyo, Japan), UV-visible spectroscopy (Varian CARY 5G; Santa Clara, CA), and X-ray diffraction (XRD; Rigaku D/max-2500 diffractometer; Tokyo, Japan). Pure GF was also investigated as a control substrate for FE-SEM characteristics.

Determination of photocatalytic efficiency

The photocatalytic efficiencies of the GF-Fe-TiO₂ photocatalysts with different Fe-to-Ti ratios and GF-TiO₂ were investigated for their decomposition of BTEX using a cylindrical-shaped photocatalytic reactor. Figure 1 shows a schematic diagram of the experimental system. The reactor was prepared using a Pyrex tube (length: 26.5 cm, inner diameter: 3.8 cm), containing a cylindrical-shaped daylight lamp (F8T5DL; Youngwha Lamp Co., Seoul, Korea) or a cylindrical-shaped black-light lamp (F8T5BL; Youngwha Lamp Co.) with a peak spectral intensity of 352 nm. The inside wall of the Pyrex tube was covered with either a GF-Fe-TiO₂ or a GF-TiO₂ sheet. Gas manufacturer-dried zero-grade air was supplied by an air cylinder and was allowed to flow through a carbon filter for repurification. The dried air stream was humidified by passing it through a humidification apparatus, which consisted of water-containing impingers. Standard gases were prepared by mixing the humidified air with vaporous BTEX. Pure liquid BTEX mixture was infused into a syringe, and then the syringe was loaded to an autoprogrammed syringe pump (KdScientific model 210; Holliston, MA) following its operating procedure. Vaporous BTEX standard gases were generated by injecting the liquid BTEX into a heated Pyrex bulb via the syringe pump. For experimental convenience, the input concentration of the model gas was fixed at 0.1 ppm for all target compounds, which was within typical residential indoor air quality concentration levels (Shin and Jo, 2013), whereas the air flow rate was fixed to 1 L min⁻¹ and maintained during the 3-hr photocatalytic process. The space velocity was 1.7 cm sec⁻¹, which was calculated by dividing the flow rate by the cross-sectional area of the reactor. The relative humidity was fixed at 45%, which is within a comfortable range for humans. Finally, the model gas was transferred into the photocatalytic reactor to determine the photocatalytic decomposition activity of the photocatalysts. The light intensities provided by the visible- and UV-light sources were 2.7 and 0.4 mW cm⁻², respectively, which were determined at the distance between the lamps and the inside wall of the cylindrical-shaped reactor. Air sampling was conducted every hour during the 3-hr photocatalytic process. Air samples were collected at upstream and downstream portions of the reactor using a stainless steel trap containing Tenax GC adsorbent. The gas-phase species adsorbed on the Tenax GC adsorbent trap were analyzed via a gas chromatography/mass spectrometer (GC/MS) (PerkinElmer Clarus SQ 8; Santa Clara, CA) combined with a thermal desorbing device (PerkinElmer ATD 350). The chemical species were qualitatively analyzed based on both their retention times and mass spectra (Wiley software library 275, Palisade Corporation, Ithaca, NY). Quantification of the chemical species was conducted using calibration curves obtained using five concentrations normalized to an internal standard chemical (fluorobenzene). The calibration curve was established using standard gases. For quality control, laboratory blank and spiked standard samples were analyzed. For each experimental day, one laboratory blank trap was analyzed to check for any trap contamination. To check the quantitative response, known standards were daily injected into a trap to transfer the target compounds to the analytical system through the thermal desorbing device. The entire experimental procedure was repeated three times in order to give reliable data, and the mean values are reported in this study.

Figure 1. Schematic design of experimental setup.

Results and Discussion

Characteristics of GF-Fe-TiO₂ composites with different Fe-to-Ti ratios

The characteristics of pure GF (FE-SEM analysis only), a reference GF-TiO₂ photocatalyst, and GF-Fe-TiO₂ composites with different Fe-to-Ti ratios were examined by FE-SEM, EDX, XRD, and UV-visible analyses. Figure 2 illustrates the FE-SEM images of the pure GF, reference GF-TiO₂, and five Fe-TiO₂ composites (0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂). There was a noticeable difference between images of the pure GF and photocatalyst-covered GFs. The results confirmed that, although there were some cracks in the coatings, GF was successfully coated with TiO₂ and Fe-TiO₂ for the GF-TiO₂ and GF-Fe-TiO₂ composites, respectively. Similarly, Panniello et al. (2012) observed the occurrence of certain cracks during the coating processes of nanostructured TiO₂ on silica fibers prepared using a sol-gel process. The gaps between neighboring GFs may allow light to penetrate deep into the photocatalyst-coated GF nets. Moreover, the EDX images of all the GF-Fe-TiO₂ composites displayed bands corresponding to O, Ti, Fe, Si, and Pt, whereas GF-TiO₂ did not exhibit any band corresponding to Fe (Figure 3). The bands attributed to O and Ti were ascribed to TiO₂, whereas the peak corresponding to Fe was attributed to Fe embedded into the TiO₂ lattice. This result is consistent with that of Azad and Hershey (2010), who reported that the EDX spectra of electrospun Fe-TiO₂ nanofibers revealed the presence of Fe atoms. The Pt bands were ascribed to the coating of GFs with Pt during a pretreatment step. Additionally, the Fe-to-Ti ratios observed in the EDX images of the GF-Fe-TiO₂ composites (apart from 0.05-GF-Fe-TiO₂) were consistent with their calculated values, indicating precise synthesis of the composites (Table 1). Even for 0.05-GF-Fe-TiO₂, the observed Fe-to-Ti ratio (0.06) differed slightly from the calculated ratio (0.05).

Figure 2. Scanning electron microscopy of pure GF, a reference photocatalyst (GF-TiO₂), and GF-Fe-TiO₂ composites with different Fe-to-Ti ratios (0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂).

Figure 3. Energy-dispersive X-ray images of a reference photocatalyst (GF-TiO₂) and GF-Fe-TiO₂ composites with different Fe-to-Ti ratios (0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂).

Table 1. Surface characteristics of a reference photocatalysts (GF-TiO₂) and GF-Fe-TiO₂ composites with various Fe-to-Ti ratios

Type	Ratio of Anatase Crystal^a	Diameter of Anatase Phase (nm)	Ratio of Fe to Ti
GF-TiO2	1.0	21.5^b (26.9)^c	0
0.005-GF-Fe-TiO2	1.0	20.9 (25.9)	0.01
0.01-GF-Fe-TiO2	1.0	21.7 (27.1)	0.01
0.05-GF-Fe-TiO2	1.0	20.8 (25.7)	0.06
0.1-GF-Fe-TiO2	1.0	21.0 (26.1)	0.1
0.5-GF-Fe-TiO2	1.0	21.0 (26.0)	0.5

Notes: ^aCalculated ratio of anatase crystal using the Spurr equation. ^bDiameter read from the XRD results. ^cCalculated diameter using the Scherrer formula.

The XRD images of the reference GF-TiO₂ photocatalyst and five Fe-TiO₂ composites (0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂) are shown in Figure 4. Both the GF-Fe-TiO₂ composites and GF-TiO₂ displayed only anatase peaks, with a main peak at 2θ = 25.3°. This was different to the Degussa P25 TiO₂ photocatalyst, which revealed both anatase and rutile crystal phase peaks (Jo and Kim, 2009). However, our study is consistent with other reports (Cui et al., 2009; Tieng et al., 2011), which observed only anatase peaks in their XRD spectra. In addition, no Fe-matched peaks were observed in Figure 4. Liu et al. (2012) reported a Fe-matched peak at 2θ = 36.9° in the XRD spectra of Fe-TiO₂ composites synthesized with an Fe-to-Ti ratio of 0.03. However, they did not observe any Fe-matched peaks for Fe-TiO₂ composite prepared with a lower Fe-to-Ti ratio (0.01). Therefore, the low amounts of Fe ions embedded into TiO₂ in the GF-Fe-TiO₂ composites might be undetectable by the instrument used in this study owing to a limited detection capacity.

Figure 4. X-ray diffraction patterns of a reference photocatalyst (GF-TiO₂) and GF-Fe-TiO₂ composites with different Fe-to-Ti ratios (0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂).

The ratios of anatase crystal and the sizes of crystal phases were calculated using the Spurr (Spur and Myers, 1957) and Scherrer (Cullity and Atock, 2001) equations, respectively. The Spurr equation states that C_A = 1/(1 + 1.265I_R/I_A), where C_A indicates the content of anatase in a catalyst, and I_A and I_R indicate the intensities of major XRD peaks for the anatase and rutile phases, respectively. Using the Scherrer formula, D = kλ/β cos θ, where k indicates the shape coefficient (0.9), λ indicates the wavelength (Cu Ka = 0.15405 nm), β indicates the full width at half maximum of the most intense peak (in radians), and θ indicates the peak location. As shown in Table 1, both the reference GF-TiO₂ and the GF-Fe-TiO₂ composites revealed only an anatase crystal phase. The diameters of TiO₂ and Fe-TiO₂ (calculated using the Scherrer formula) were comparable to the diameters determined from their corresponding XRD patterns. In addition, the calculated diameter of TiO₂ (21.5 nm) was similar to the calculated diameters of the Fe-TiO₂ composites (20.9−21.7 nm). These results indicated that Fe³⁺ ions did not significantly influence the diameter of the photocatalysts. In contrast, Cui et al. (2009) reported that the diameter of a laboratory-made unmodified TiO₂ photocatalyst (12.0 nm) was smaller than the diameters of Fe-TiO₂ composites with Fe-to-Ti ratios of 0.001, 0.005, 0.010, and 0.015 (13.4, 12.6, 16.4, and 12.3 nm, respectively). The differences between the two studies were attributed to the different synthetic routes used in preparing the photocatalysts. Cui et al. (2009) used a thermal hydrolysis method, followed by a calcination treatment, whereas we used dop-coating, followed by calcination.

The UV-visible absorbance spectra of the samples (GF-Fe-TiO₂ composites and GF-TiO₂) are illustrated in Figure 5. The reference GF-TiO₂ photocatalyst displayed an absorption edge at λ ≈ 410 nm, which was consistent with that reported by other researchers (Jo and Kim, 2009; Sun et al., 2012). However, the absorption spectrum of the GF-Fe-TiO₂ composite was substantially shifted toward the visible-light region. This observation was in good agreement with the results reported by Sun et al. (2012). Moreover, the GF-Fe-TiO₂ composites showed larger red shifts with increasing amounts of Fe impregnated into TiO₂. These results were attributed to rapid charge transfer between TiO₂ and Fe resulting from the electron levels created by the embedded Fe; this charge transfer could cause impurities within the band gap range of TiO₂ (Cui et al., 2010). Therefore, the Fe-TiO₂ composites could be used under both visible- and UV-light irradiation.

Figure 5. UV-visible spectra of a reference photocatalyst (GF-TiO₂) and GF-Fe-TiO₂ composites with different Fe-to-Ti ratios (0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂).

Photocatalytic decomposition efficiency

The BTEX photocatalytic decomposition efficiencies of the reference TiO₂ photocatalyst and as-prepared GF-Fe-TiO₂ composites were measured under visible- or UV-light irradiation after 3-hr adsorption processes in the absence of irradiation. In the control experiments (using noncoated GF), the decomposition of BTEX was negligible under both visible and UV light (0% to ≤1%). In addition, the effect of adsorption of BTEX on the surface of the GF-Fe-TiO₂ was also negligible under both visible- and UV-light irradiation. Figure 6 illustrates the time-series BTEX photocatalytic efficiencies of the GF-Fe-TiO₂ composites and GF-TiO₂ under visible-light activation conditions. Point (0,0) indicates that the photocatalytic efficiencies of the GF-Fe-TiO₂ composites were nearly zero prior to activation by the light sources. All of the GF-Fe-TiO₂ composites showed higher photocatalytic efficiencies for BTEX relative to the reference photocatalyst, GF-TiO₂. Specifically, the average BTEX photocatalytic efficiencies of the 0.01-GF-Fe-TiO₂ composite over a 3-hr photocatalytic process were 4%, 33%, 51%, and 74%, respectively. Moreover, those obtained for GF-TiO₂ were close to 0%, 5%, 16%, and 29%, respectively. In a study of aqueous 4-nitrophenol, Zhao et al. (2010) found that the photocatalytic efficiencies of Fe-TiO₂ composites (prepared by embedding Fe(NO₃)₃ into the anatase phase of TiO₂) were higher than those of the pure TiO₂. The improved photocatalytic activity of the GF-Fe-TiO₂ composites over GF-TiO₂ was ascribed to the Fe³⁺ ions. It was found that Fe³⁺ would serve as electron and hole trap (Fe³⁺ + e⁻ → Fe²⁺ and Fe³⁺ + h⁺ → Fe⁴⁺), thereby decreasing charge carrier recombination and elevating the photocatalytic activity of the composites (Zhao et al., 2010; Sun et al., 2012). The presence of Fe³⁺ ions is further supported by the EDX spectrum (Figure 1). Fe³⁺ ions could easily embed in the crystal lattice of TiO₂ owing to similar diameters of Fe³⁺ (0.64 Å) and Ti⁴⁺ (0.68 Å) ions (Lu et al., 2011). The GF-Fe-TiO₂ composites showed lower photocatalytic activity for benzene than for the other target compounds. This was ascribed to the lower photocatalytic oxidation rate of benzene compared with other target chemicals (Strini et al., 2005).

Figure 6. Photocatalytic efficiencies (PEs; %) of benzene, toluene, ethylbenzene, and o-xylene as determined via a reference photocatalyst (GF-TiO₂) and GF-Fe-TiO₂ composites with different Fe-to-Ti ratios (0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂) under visible-light irradiation.

Figure 6 also shows that the BTEX photocatalytic efficiencies for the GF-Fe-TiO₂ composites (under visible-light exposure) increased as the Fe-to-Ti ratio increased from 0.005 to 0.010. This was also due to the positive effect of Fe³⁺ ions as described previously. However, the photocatalytic efficiencies decreased gradually as the Fe-to-Ti ratio increased from 0.010 to 0.50. This was attributed to excess Fe³⁺ ions, which might serve as centers for charge carrier recombination (Lu et al., 2011). Excess Fe³⁺ ions could also produce Fe(OH)²⁺, which could interfere with the absorption of photons over the wavelength range 290−400 nm, and subsequently lower photocatalytic performance (Zhao et al., 2010). Additionally, it is noteworthy that, although their peak areas were not high enough to be quantified, there were some peaks for compounds other than the target aromatics on the GC chromatogram. This could possibly indicate the formation of by-products. Our preliminary tests also demonstrated that there was no generation of benzene during the photocatalysis of individual compounds of toluene, ethylbenzene, and o-xylene.

Figure 7 illustrates the BTEX photocatalytic efficiencies of the Fe-TiO₂ composites exposed to UV light. Unlike the visible-light conditions, the BTEX photocatalytic efficiencies of GF-Fe-TiO₂ composites were lower than those obtained for GF-TiO₂. Specifically, the average BTEX photocatalytic efficiencies of the 0.5-GF-Fe-TiO₂ composite were 15%, 63%, 79%, and 86%, respectively. Conversely, those obtained for GF-TiO₂ were 68%, 90%, 96%, and 97%, respectively. Moreover, the GF-Fe-TiO₂ composites displayed different BTEX photocatalytic efficiencies (under visible and UV light) between the various Fe-to-Ti ratios. The BTEX photocatalytic efficiencies of the GF-Fe-TiO₂ composites decreased as the Fe-to-Ti ratios increased. These results were also ascribed to Fe ions acting as a recombination center for the charge carriers generated under the UV-light conditions (Cui et al., 2010).

Figure 7. Photocatalytic efficiencies (PEs; %) of benzene, toluene, ethylbenzene, and o-xylene as determined via a reference photocatalyst (GF-TiO₂) and GF-Fe-TiO₂ composites with different Fe-to-Ti ratios (0.005-GF-Fe-TiO₂, 0.01-GF-Fe-TiO₂, 0.05-GF-Fe-TiO_2, 0.1-GF-Fe-TiO₂, and 0.5-GF-Fe-TiO₂) under UV irradiation.

Conclusion

Flexible glass fiber–supported Fe-TiO₂ composites were synthesized using TnBT and Fe(NO₃)₃ as Ti and Fe sources, respectively. The Fe-TiO₂ composites were applied to the photocatalytic treatment of BTEX at indoor concentration levels. FE-SEM results demonstrated that the GF was coated with TiO₂ and Fe-TiO₂ for the GF-TiO₂ and Fe-TiO₂ composites, respectively. EDX spectroscopy of the GF-Fe-TiO₂ composites demonstrated that Fe ions were embedded into the crystal lattice of TiO₂. Furthermore, EDX spectroscopy showed that the desired compositions of the Fe-TiO₂ composites were achieved during synthesis. UV-visible spectroscopy demonstrated that the GF-Fe-TiO₂ composites could be used under both visible- and UV-light conditions. In addition, the visible light–activated GF-Fe-TiO₂ composites displayed a photocatalytic performance superior to that of the reference GF-TiO₂ photocatalyst for the treatment of BTEX. However, the results were reversed for the UV-activated GF-Fe-TiO₂ composites. Because visible-light sources consume a lower energy compared with UV-light sources, when considering energy efficiency, use of GF-Fe-TiO₂ that shows a higher photocatalytic activity with a low-energy visible-light source is preferred. In addition, the BTEX photocatalytic decomposition efficiencies for the GF-Fe-TiO₂ composites varied with Fe-to-Ti ratio, suggesting that a Fe-to-Ti ratio of 0.01 was optimal.

Acknowledgment

The authors would like to thank Mr. Joon Yeob Lee, who is a Ph.D. student in the Department of Environmental Engineering, Kyungpook National University, for his experimental works.

Funding

This work was supported by the Rural Development Administration Korea (PJ00923504) and the National Research Foundation of Korea (NRF) grant funded through GCRC-SOP (no. 2011-0030013).

Additional information

Funding

This work was supported by the Rural Development Administration Korea (PJ00923504) and the National Research Foundation of Korea (NRF) grant funded through GCRC-SOP (no. 2011-0030013).

Notes on contributors

Sung-Bong Yang

Sung-Bong Yang is a professor at the Department of Chemistry, University of Ulsan, Ulsan, Republic of Korea.

Ho-Hwan Chun

Ho-Hwan Chun is a professor at the Department of Naval Architecture and Ocean Engineering, Pusan National University, Busan, Republic of Korea.

Rajesh J. Tayade

Rajesh J. Tayade is a senior technical officer at the Discipline of Inorganic Materials and Catalysis, Central Salt and Marine Chemicals Research Institute, Council of Scientific and Industrial Research (CSIR-CSMCRI), Gujarat, India.

Wan-Kuen Jo

Wan-Kuen Jo is a professor at the Department of Environmental Engineering of Kyungpook National University, Daegu, Republic of Korea.