Sol-Gel Hydrothermal Synthesis and Visible Light Photocatalytic Degradation Performance of Fe/N Codoped TiO2 Catalysts

Using Ti(OC4H9)4 as a precursor, Fe(NO3)3⋅9H2O as the source of iron, and NH4NO3 as the source of nitrogen, an Fe/N codoped TiO2 catalyst was prepared using a sol-gel hydrothermal method. The as-prepared powders were characterized using X-ray powder diffraction, electron spectroscopy for chemical analysis, Fourier-transform infrared spectroscopy, and ultraviolet-visible spectrophotometry. Fe and N codoping resulted in decreased crystallite size and increased specific surface area. Results of the photocatalytic degradation of acid orange 7 (AO7) in a continuous-flow fluidized-bed reactor indicated that the maximum decolorization (more than 90%) of AO7 occurred with the Fe/N-TiO2 catalyst (dosage of 20 g/L) when a combination of visible light irradiation for 10 h HRT (hydraulic retention time), and a heterogeneous system was used. The AO7 degradation efficiency was considerably improved by increasing the hydraulic retention time from 2.5 to 10 h or by reducing the initial AO7 concentration from 300 to 100 mg/L. The reaction rate increased with the light intensity and the maximum value occurred at 35 mW/cm2; moreover, the efficiency of the AO7 degradation increased when the pH decreased with maximum efficiency at pH 3.


Introduction
Environmental pollution is a considerable concern in the modern world. An estimated 2% of dyes produced annually are discharged as effluents from manufacturing plants, whereas 10% of dyes are discharged from textile and related industries [1]. Effluents generated from textile manufacturing contain a variety of pollutants characterized by deep coloration, high oxygen demand, high pH, large amounts of suspended solids, and low or nonbiodegradability [2,3]. Many methods have been tested to remove dyes from industrial effluents, including biological processes, adsorption, and coagulation. However, these methods still generate a large amount of sludge or solid waste that requires further treatment.
Advanced oxidation processes are a suitable alternative to traditional methods for solving environmental problems caused by the discharge of textile-dyeing wastewater. Titanium dioxide (TiO 2 ) is a heterogeneous photocatalysts and TiO 2 based photocatalysis is a promising technique for wastewater treatment [4], especially for wastewater containing refractory organic compounds. However, the large band gap for highly oriented TiO 2 powders with pure anatase structure and rutile reactor (40/30 mm OD/ID; height = 500 mm) containing 20 g of Fe/N-TiO 2 and 5 L of a 200 mg/L AO7 aqueous solution. The photoreactor was open to the atmosphere, and the quartz reactor was surrounded by 14 light tubes. The visible light tubes were germicidal lamps with a wavelength of 419 nm (Sankyo Denki, Tokyo, Japan). The light power (approximately 8 mW/cm 2 ) in the center of the reactor in air was measured using a hand-held optical power meter (Model 840-C, Newport, Irvine, CA, USA). The photodegradation rates of AO7 solutions were determined by periodically measuring the absorbance at λ = 484 nm by using a Hach DR 4000 UV-visible spectrophotometer (Hach, Loveland, CO, USA).

Photocatalytic Experiments
The upflow fluidized-bed system is shown in Figure 1. The photocatalytic activities of Fe/N-TiO2 samples under visible light were evaluated based on the degradation rate of AO7 in a cylindrical quartz reactor (40/30 mm OD/ID; height = 500 mm) containing 20 g of Fe/N-TiO2 and 5 L of a 200 mg/L AO7 aqueous solution. The photoreactor was open to the atmosphere, and the quartz reactor was surrounded by 14 light tubes. The visible light tubes were germicidal lamps with a wavelength of 419 nm (Sankyo Denki, Tokyo, Japan). The light power (approximately 8 mW/cm 2 ) in the center of the reactor in air was measured using a hand-held optical power meter (Model 840-C, Newport, Irvine, CA, USA). The photodegradation rates of AO7 solutions were determined by periodically measuring the absorbance at λ= 484 nm by using a Hach DR 4000 UV-visible spectrophotometer (Hach, Loveland, CO, USA).  Figure 2a presents the XRD patterns of undoped (TiO2) and Fe/N-TiO2 particles as a function of the reaction temperature. Fe/N-TiO2 particles were readily indexed to the diffraction peaks of the anatase phase (JCPDS 04-002-2678) and exhibited the presence of an intense peak corresponding to the (101) plane. The major peaks observed corresponded to the (101), (004), (200), (105), and (204) planes of the anatase phase [14]. For 100, 150, 175, and 200 °C Fe/N-TiO2 particles, the crystallite sizes were 10.65, 10.79, 12.11, and 13.46 nm, respectively. Smaller crystallite sizes were obtained for the codoped samples, which indicated that the incorporation of Fe and N ions restricted the growth of TiO2 crystallite and prevented the transformation of anatase to rutile [15]. Deng et al. [16] also investigated the morphology of Fe-doped titania nanotubes synthesized using the sol-gel and hydrothermal methods. They found that the addition of Fe slowed the crystallization process and prevented the growth of crystallite TiO2. The crystallite size of Fe/N-TiO2 particles increased with the reaction temperature (Table 1).

Characterization of the N/Fe-TiO2 Samples
To determine whether codoping with Fe/N was successful, the surface of Fe/N-TiO2 composites was examined through ESCA. The ESCA spectra of Ti 2p in Fe/N-TiO2 shown in Figure 2(b) reveal that the Ti 2p₁/₂ and Ti 2p₃/₂ peaks at 464.2 and 458.5 eV, respectively, were in a favorable agreement with those previously observed for Ti 4+ [17].The presence of N in TiO2 particles was substantiated by the N 1s spectra and significant peaks around 400 eV, which can be attributed to the formation of anionic N in O−Ti−N linkages [18], whereas the iron peak (710 eV) was attributed to Fe 3+ , indicating the formation of Fe2O3 [19]. Saha and Tompkins [20] investigated N 1s ESCA spectra during the

Characterization of the N/Fe-TiO 2 Samples
Figure 2a presents the XRD patterns of undoped (TiO 2 ) and Fe/N-TiO 2 particles as a function of the reaction temperature. Fe/N-TiO 2 particles were readily indexed to the diffraction peaks of the anatase phase (JCPDS 04-002-2678) and exhibited the presence of an intense peak corresponding to the (101) plane. The major peaks observed corresponded to the (101), (004), (200), (105), and (204) planes of the anatase phase [14]. For 100, 150, 175, and 200 • C Fe/N-TiO 2 particles, the crystallite sizes were 10.65, 10.79, 12.11, and 13.46 nm, respectively. Smaller crystallite sizes were obtained for the codoped samples, which indicated that the incorporation of Fe and N ions restricted the growth of TiO 2 crystallite and prevented the transformation of anatase to rutile [15]. Deng et al. [16] also investigated the morphology of Fe-doped titania nanotubes synthesized using the sol-gel and hydrothermal methods. They found that the addition of Fe slowed the crystallization process and prevented the growth of crystallite TiO 2 . The crystallite size of Fe/N-TiO 2 particles increased with the reaction temperature (Table 1).
To determine whether codoping with Fe/N was successful, the surface of Fe/N-TiO 2 composites was examined through ESCA. The ESCA spectra of Ti 2p in Fe/N-TiO 2 shown in Figure 2(b) reveal that the Ti 2p 1/2 and Ti 2p 3/2 peaks at 464.2 and 458.5 eV, respectively, were in a favorable agreement with those previously observed for Ti 4+ [17].The presence of N in TiO 2 particles was substantiated by the N 1s spectra and significant peaks around 400 eV, which can be attributed to the formation of anionic N in O−Ti−N linkages [18], whereas the iron peak (710 eV) was attributed to Fe 3+ , indicating the formation of Fe 2 O 3 [19]. Saha and Tompkins [20] investigated N 1s ESCA spectra during the oxidation process of Ti-N and assigned the peaks at 400 eV to be molecularly chemisorbed γ−N 2 . Kim et al. [15] reported that the ionic radii of Fe 3+ (0.64 Å) and Ti 4+ (0.68 Å) are similar and that Fe 3+ can therefore be incorporated into the lattice of TiO 2 to form a Ti-O-Fe bond in Fe/N-TiO 2 . The results indicate that Fe is present in the form of Fe 3+ by replacing Ti 4+ in the doped photocatalyst, which may change the charge distribution of atoms on the photocatalyst surface, resulting in enhanced photocatalytic activity. By contrast, the decrease of Ti binding energy upon N-doping could be interpreted as the formation of O-Ti-N in the TiO 2 lattice [19], which indicates that nitrogen incorporation can successfully retard the charge recombination at the TiO 2 /dye/electrolyte interface. Additionally, the concentrations of Fe and N in Fe/N-TiO 2 determined using ESCA were 5.58 and 5.48 wt %, respectively, which were consistent with the theoretical expectation.
Materials 2018, 11, x FOR PEER REVIEW 4 of 9 oxidation process of Ti-N and assigned the peaks at 400 eV to be molecularly chemisorbed N2. Kim et al. [15] reported that the ionic radii of Fe 3+ (0.64 Å ) and Ti 4+ (0.68 Å ) are similar and that Fe 3+ can therefore be incorporated into the lattice of TiO2 to form a Ti-O-Fe bond in Fe/N-TiO2. The results indicate that Fe is present in the form of Fe 3+ by replacing Ti 4+ in the doped photocatalyst, which may change the charge distribution of atoms on the photocatalyst surface, resulting in enhanced photocatalytic activity. By contrast, the decrease of Ti binding energy upon N-doping could be interpreted as the formation of O-Ti-N in the TiO2 lattice [19], which indicates that nitrogen incorporation can successfully retard the charge recombination at the TiO2/dye/electrolyte interface. Additionally, the concentrations of Fe and N in Fe/N-TiO2 determined using ESCA were 5.58 and 5.48 wt %, respectively, which were consistent with the theoretical expectation.  The calculation of the band gap of materials can be conducted using the following formulation: absorption coefficient (a) and the incident photon energy (hν) can be written as a = Bi·(hν−Eg) 2 /hν, where Bi is the absorption constant for indirect transitions, hν is the photon energy, and Eg is the band gap energy [21]. Plots of (ahν) 1/2 versus hν from the spectral data are presented in Figure 3a, which shows the UV-visible spectra of the undoped (TiO2) and Fe/N-TiO2 particles from 250 to 700 nm. Samples A-E exhibited typical UV-visible spectra for semiconductor materials with a band gap absorption onset at 465, 388, 464, 452, and 485 nm, which corresponded to energy bandgaps at 2.67, 3.20, 2.67, 2.74, and 2.55 eV, respectively. These results demonstrate that the absorption of doped TiO2 in the visible light region is significantly enhanced compared with that of undoped TiO2, which in turn may considerably increase the photocatalytic activity of TiO2 under visible light irradiation. Fourier-transform infrared (FT-IR) spectrum of the Fe/N-TiO2 prepared using the sol-gel hydrothermal method at 150 °C and the undoped TiO2 over the 400-4000 cm −1 range are shown in Figure 3(b). The strong absorption at 3442 and 1640 cm −1 were assigned to the stretching vibration and the bending vibration of OH, respectively, originating from water adsorbed on the samples' surface [15]. The peaks around 1090 cm −1 were attributed to the N atoms embedded in the TiO2 network. In addition, the small peak observed at 570 cm −1 indicates Fe-O-Ti vibrations [22]. No  The calculation of the band gap of materials can be conducted using the following formulation: absorption coefficient (a) and the incident photon energy (hν) can be written as a = Bi·(hν−E g ) 2 /hν, where Bi is the absorption constant for indirect transitions, hν is the photon energy, and E g is the band gap energy [21]. Plots of (ahν) 1/2 versus hν from the spectral data are presented in Figure 3a, which shows the UV-visible spectra of the undoped (TiO 2 ) and Fe/N-TiO 2 particles from 250 to 700 nm. Samples A-E exhibited typical UV-visible spectra for semiconductor materials with a band gap absorption onset at 465, 388, 464, 452, and 485 nm, which corresponded to energy bandgaps at 2.67, 3.20, 2.67, 2.74, and 2.55 eV, respectively. These results demonstrate that the absorption of doped TiO 2 in the visible light region is significantly enhanced compared with that of undoped TiO 2 , which in turn may considerably increase the photocatalytic activity of TiO 2 under visible light irradiation. Fourier-transform infrared (FT-IR) spectrum of the Fe/N-TiO 2 prepared using the sol-gel hydrothermal method at 150 • C and the undoped TiO 2 over the 400-4000 cm −1 range are shown in Figure 3b. The strong absorption at 3442 and 1640 cm −1 were assigned to the stretching vibration and the bending vibration of OH, respectively, originating from water adsorbed on the samples' surface [15]. The peaks around 1090 cm −1 were attributed to the N atoms embedded in the TiO 2 network. In addition, the small peak observed at 570 cm −1 indicates Fe-O-Ti vibrations [22]. No absorption peak for Fe-N stretching was observed, indicating that Fe did not substitute for Ti at sites where N atoms substituted for O atoms. absorption peak for Fe-N stretching was observed, indicating that Fe did not substitute for Ti at sites where N atoms substituted for O atoms. The optimal synthesis temperature of Fe/N-TiO2 was determined from batch experiments. Figure 4a shows the photocatalytic AO7 degradation curves for Fe/N-TiO2 catalysts synthesized at different temperatures (see Table 1). The photocatalytic activity evolved as follows: Fe/N-TiO2 (150 °C) > Fe/N-TiO2 (175 °C) > Fe/N-TiO2 (200 °C) > Fe/N-TiO2 (100 °C) > undoped TiO2. Fe/N TiO2 (150 °C) exhibited the highest photocatalytic activity and led to 95.2% AO7 degradation in 5 hr. In addition, Figure 4b plots ln(C/C0) versus time obtained by assuming first-order kinetics for the degradation reaction. C and C0 are the AO7 concentrations at time t and initial concentration, respectively. The plots were almost linear, indicating that the reactions followed pseudo first-order kinetics. The first-order degradation rate constants (k) for Fe/N-TiO2 (150 °C), Fe/N-TiO2 (175 °C), Fe/N-TiO2 (200 °C), Fe/N-TiO2 (100 °C), and undoped TiO2 catalysts were 5.64 × 10 −2 , 4.57 × 10 −2 , 2.23 × 10 −2 , 1.36 × 10 −2 , and 8.53 × 10 −1 min −1 , respectively. This suggests that codoping of Fe and N narrows the TiO2 band gap. Cong et al. [23] reported that the overlap of the Ti-d orbital of TiO2 and the doped The optimal synthesis temperature of Fe/N-TiO 2 was determined from batch experiments. Figure 4a shows the photocatalytic AO7 degradation curves for Fe/N-TiO 2 catalysts synthesized at different temperatures (see Table 1). The photocatalytic activity evolved as follows: Fe/N-TiO 2 (150 • C) > Fe/N-TiO 2 (175 • C) > Fe/N-TiO 2 (200 • C) > Fe/N-TiO 2 (100 • C) > undoped TiO 2 . Fe/N TiO 2 (150 • C) exhibited the highest photocatalytic activity and led to 95.2% AO7 degradation in 5 h. In addition, Figure 4b plots ln(C/C 0 ) versus time obtained by assuming first-order kinetics for the degradation reaction. C and C 0 are the AO7 concentrations at time t and initial concentration, respectively. The plots were almost linear, indicating that the reactions followed pseudo first-order kinetics. The first-order degradation rate constants (k) for Fe/N-TiO 2 (150 • C), Fe/N-TiO 2 (175 • C), Fe/N-TiO 2 (200 • C), Fe/N-TiO 2 (100 • C), and undoped TiO 2 catalysts were 5.64 × 10 −2 , 4.57 × 10 −2 , 2.23 × 10 −2 , 1.36 × 10 −2 , and 8.53 × 10 −1 min −1 , respectively. This suggests that codoping of Fe and N narrows the TiO 2 band gap. Cong et al. [23] reported that the overlap of the Ti-d orbital of TiO 2 and the doped metal d orbital leads to a narrowing of the TiO 2 band gap in TiO 2 implanted with metal ions, allowing the absorption of visible light. Therefore, N and Fe were incorporated into the TiO 2 framework, narrowing the band gap of TiO 2 to 2.67 eV (Table 1) and causing a large red shift, which in turn caused a much narrower band gap and greatly improved photocatalytic activity. By contrast, it inhibits the recombination of photogenerated electrons and holes. Fe ions with a suitable concentration can trap photogenerated electrons, which enhances the utilization efficiency of the photogenerated electron and hole [24]. Consequently, under these experimental conditions, Fe/N-TiO 2 (150 • C) was optimal for AO7 removal after 5 h of visible light irradiation time.
Materials 2018, 11, x FOR PEER REVIEW 6 of 9 metal d orbital leads to a narrowing of the TiO2 band gap in TiO2 implanted with metal ions, allowing the absorption of visible light. Therefore, N and Fe were incorporated into the TiO2 framework, narrowing the band gap of TiO2 to 2.67 eV (Table 1) and causing a large red shift, which in turn caused a much narrower band gap and greatly improved photocatalytic activity. By contrast, it inhibits the recombination of photogenerated electrons and holes. Fe ions with a suitable concentration can trap photogenerated electrons, which enhances the utilization efficiency of the photogenerated electron and hole [24]. Consequently, under these experimental conditions, Fe/N-TiO2 (150 °C) was optimal for AO7 removal after 5 h of visible light irradiation time.

Degradation of AO7 in a Continuous-Flow Fluidized-Bed System
The optimal Fe/N-TiO2 (150 °C) catalyst was selected for photocatalytic activity tests of the degradation of AO7. The effect of the initial AO7 concentration on the photocatalytic degradation efficiency was examined for concentrations ranging from 100 to 300 mg/L with an Fe/N-TiO2 (150 °C) dosage of 20 g/L, a hydraulic retention time (HRT) of 10 h, a pH of 3, and a visible light intensity of 35 mW/cm 2 . Figure 5a shows the AO7 removal efficiency and observed rate constant (Kobs) as a function of the initial AO7 concentration at a pH of 3 and with an HRT of 10 hr. The degradation rate of AO7 decreased when the initial AO7 concentration increased. The number of photons decreased because of the decreasing intensity of the visible light, leading to a decrease in the formation of hydroxyl radicals, which ultimately reduced AO7 removal efficiency [25]. Moreover, the reaction rate also increased when the visible light intensity increased, and the maximum rate was reached for an irradiation of 35 mW/cm 2 , as illustrated in Figure 5b. This indicates that the rate of photons per unit area of catalyst powder increased with the light intensity [26], and there was a corresponding increase in photocatalytic degradation rate of AO7.

Degradation of AO7 in a Continuous-Flow Fluidized-Bed System
The optimal Fe/N-TiO 2 (150 • C) catalyst was selected for photocatalytic activity tests of the degradation of AO7. The effect of the initial AO7 concentration on the photocatalytic degradation efficiency was examined for concentrations ranging from 100 to 300 mg/L with an Fe/N-TiO 2 (150 • C) dosage of 20 g/L, a hydraulic retention time (HRT) of 10 h, a pH of 3, and a visible light intensity of 35 mW/cm 2 . Figure 5a shows the AO7 removal efficiency and observed rate constant (K obs ) as a function of the initial AO7 concentration at a pH of 3 and with an HRT of 10 h. The degradation rate of AO7 decreased when the initial AO7 concentration increased. The number of photons decreased because of the decreasing intensity of the visible light, leading to a decrease in the formation of hydroxyl radicals, which ultimately reduced AO7 removal efficiency [25]. Moreover, the reaction rate also increased when the visible light intensity increased, and the maximum rate was reached for an irradiation of 35 mW/cm 2 , as illustrated in Figure 5b. This indicates that the rate of photons per unit area of catalyst powder increased with the light intensity [26], and there was a corresponding increase in photocatalytic degradation rate of AO7.
To study the effect of pH on degradation efficiency, experiments were performed under visible light at pH values from 3 to 10 with constant concentrations of AO7 and Fe/N-TiO 2 (150 • C) catalyst. The results in Figure 6a indicate that the photodegradation efficiency for AO7 increased as the pH decreased, with maximum efficiency (88%) at pH 3. The degradation rates for the continuous-flow photoreactor evolved as follows: pH 3 > pH 7 > pH 10. In addition, increasing the HRT from 2.5 to 10 h increased the AO7 removal efficiency from 32% to 88% at pH 3. Explaining the effect of pH on the dye photodegradation efficiency is difficult because of the multiple roles of H + ions, but pH change is related to the charge in the functionalized surface of the solid catalyst according to the following reactions [27]: TiOH + H + ←→ TiOH 2 + , pH < pH ζ (1) because of the decreasing intensity of the visible light, leading to a decrease in the formation of hydroxyl radicals, which ultimately reduced AO7 removal efficiency [25]. Moreover, the reaction rate also increased when the visible light intensity increased, and the maximum rate was reached for an irradiation of 35 mW/cm 2 , as illustrated in Figure 5b. This indicates that the rate of photons per unit area of catalyst powder increased with the light intensity [26], and there was a corresponding increase in photocatalytic degradation rate of AO7. According to Equation (1), when TiO 2 is suspended in an acidic solution (pH < point of zero charge, pH ζ ), the surface charge of TiO 2 becomes positive. Conversely, when TiO 2 is suspended in a basic solution (pH > pH ζ ), the surface charge becomes negative, as shown in Equation (2). Figure 6b shows that pH ζ for the Fe/N-TiO 2 was 6. Therefore, the surface of the catalyst was positively charged at pH < 6 and negatively charged at pH > 6. AO7 is an anionic dye and was negatively charged under the experimental conditions used because of the SO 3 2− groups. Therefore, electrostatic interactions between the Fe/N-TiO 2 catalysts and the sulfonate groups resulted in adsorption at pH < 6 and enhanced degradation efficiency. Conversely, adsorption of AO7 onto Fe/N-TiO 2 surfaces was weak at pH > 6 because of Coulombic repulsion between the negatively charged Fe/N-TiO 2 and the AO7 molecules. Therefore, the degradation efficiency decreased. To study the effect of pH on degradation efficiency, experiments were performed under visible light at pH values from 3 to 10 with constant concentrations of AO7 and Fe/N-TiO2 (150 °C) catalyst. The results in Figure 6a indicate that the photodegradation efficiency for AO7 increased as the pH decreased, with maximum efficiency (88%) at pH 3. The degradation rates for the continuous-flow photoreactor evolved as follows: pH 3 > pH 7 > pH 10. In addition, increasing the HRT from 2.5 to 10 hr increased the AO7 removal efficiency from 32% to 88% at pH 3. Explaining the effect of pH on the dye photodegradation efficiency is difficult because of the multiple roles of H + ions, but pH change is related to the charge in the functionalized surface of the solid catalyst according to the following reactions [27]: TiOH + H + ←→ TiOH2 + , pH ＜ pH (1) According to Equation (1), when TiO2 is suspended in an acidic solution (pH < point of zero charge, pH), the surface charge of TiO2 becomes positive. Conversely, when TiO2 is suspended in a basic solution (pH > pH), the surface charge becomes negative, as shown in Equation (2). Figure 6b shows that pH for the Fe/N-TiO2 was 6. Therefore, the surface of the catalyst was positively charged at pH < 6 and negatively charged at pH > 6. AO7 is an anionic dye and was negatively charged under the experimental conditions used because of the SO3 2− groups. Therefore, electrostatic interactions between the Fe/N-TiO2 catalysts and the sulfonate groups resulted in adsorption at pH < 6 and enhanced degradation efficiency. Conversely, adsorption of AO7 onto Fe/N-TiO2 surfaces was weak at pH > 6 because of Coulombic repulsion between the negatively charged Fe/N-TiO2 and the AO7 molecules. Therefore, the degradation efficiency decreased.

Conclusions
Fe/N-TiO2 catalysts were synthesized using a combination of sol-gel and hydrothermal processes. The average size and distribution of the Fe/N-TiO2 particles synthesized was approximately 10-15 nm. The average size of the particles synthesized increased with the reaction temperature, and the absorption edge of Fe/N-TiO2 catalysts was red-shifted toward 480 nm. The Fe/N-TiO2 photocatalyst exhibited favorable photocatalytic activity for the degradation of AO7 in a continuous-flow fluidized-bed system under visible light. The experimental results revealed that the optimal dosage of Fe/N-TiO2 was 20 g/L, and that AO7 degradation efficiency was substantially improved by increasing HRT from 2.5 to 10 hr or by reducing initial AO7 concentration from 300 to

Conclusions
Fe/N-TiO 2 catalysts were synthesized using a combination of sol-gel and hydrothermal processes. The average size and distribution of the Fe/N-TiO 2 particles synthesized was approximately 10-15 nm. The average size of the particles synthesized increased with the reaction temperature, and the absorption edge of Fe/N-TiO 2 catalysts was red-shifted toward 480 nm. The Fe/N-TiO 2 photocatalyst exhibited favorable photocatalytic activity for the degradation of AO7 in a continuous-flow fluidized-bed system under visible light. The experimental results revealed that the optimal dosage of Fe/N-TiO 2 was 20 g/L, and that AO7 degradation efficiency was substantially improved by increasing HRT from 2.5 to 10 h or by reducing initial AO7 concentration from 300 to 100 mg/L. Additionally, the degradation efficiency of AO7 increased as the pH decreased, with a maximum efficiency at pH 3.
Author Contributions: All the authors have contributed equally to the realization of work.
Funding: This study was supported by grant from Hsuteng Consulting International Co., Ltd.