Preparation of Fe3+ Doped High-ordered TiO2 Nanotubes Arrays with Visible Photocatalytic Activities

In this paper, the Fe3+ doped rutile phase TiO2 nanotubes arrays (NTAs) were prepared in a low temperature water-assistant crystallization method. It is noteworthy that the Fe3+ doping hardly hinders either the crystallization of rutile TiO2 NTAs or the highly-ordered nanotubular morphologies. Moreover, Fe3+ did not form other compound impurities, which indicated that Fe3+ substitute Ti4+ into the lattice of TiO2. With the introduction of Fe3+, the light absorption range of TiO2 NTAs extends from the ultraviolet band to the visible light range. Photocatalytic testing results indicate that Fe3+ doped TiO2 NTAs can effectively improve the degradation rate of methyl orange aqueous solution in visible light, and the TiO2 NTAs with 0.2 mol/L Fe3+ doping exhibits the highest photocatalytic degradation efficiency.

Metal ion doping modification mainly includes 3d transition metal ions into the crystals of titanium dioxide, taking advantage of the instability of the outer electrons of 3d transition metals [29][30][31]. On one hand, when these unstable and unsaturated electrons are presented inside titanium dioxide, it helps to reduce the recombination rates of photogenerated electron-hole pairs on the surface of TiO 2 and improve the utilization efficiency of photogenerated electrons [32][33][34][35]. On the other hand, some metal ions would form TiO 2 electron traps in the band gap, and then expand the photo response of TiO 2 NTAs to visible light [36,37]. Among various transition metal ions, Fe 3+ has demonstrated excellent properties when doped with TiO 2 NTAs [26,38,39]. However, preparation and usage of Fe 3+ doped TiO 2 NTAs reported in the literature does not achieve low power consumption, low cost and effective recycling. For example: Sun et al. used a high temperature annealing process to prepare TiO 2 NTAs [26], Yu et al. adopted a high temperature alloy process to fabricate the Fe doped Ti foil [38], and Pang et al. prepared discrete Fe 3+ doped TiO 2 NTAs, which is not conducive to rapid recycling [39].
In order to facilitate recycling, we proposed a simple and low-cost preparation process of Fe 3+ -doped TiO 2 NTAs on Ti foil under mild environment. In this study, anodic TiO 2 NTAs were crystallized at low temperature induced by Fe 3+ aqueous solution and the conditions and preparation process were optimized through multiple experiments. Various characterization and experimental results indicate that the prepared Fe 3+ doped TiO 2 NTAs have good photocatalytic activities for the degradation of organic substance under visible light irradiation.

Materials and Methods
The initial amorphous TiO 2 NTs were synthesized by electrochemical anodization of Ti foil (1.6 cm × 3 cm, >99.8%, 0.25 mm, Alfa Aesar, Shanghai, China) in glycol containing electrolyte (490 mL glycol (Alfa Aesar), 10 mL H 2 O and 1.66 g NH 4 F (Alfa Aesar)) at room temperature under a constant voltage of 50 V in a two-electrode configuration (a Ti foil as the working electrode and a platinum foil as the counter electrode), using a DC programmable power supply (Model IT6154, ITECH Electronics Co., Ltd. Nanjing, China). After anodization for 1.5 h, highly ordered TiO 2 nanotube arrays on the Ti sheet substrate were obtained. Then, the as-prepared TiO 2 NTs were thoroughly washed with a large amount of anhydrous alcohols and dried in air.
The Fe 3+ doped TiO 2 NTs were prepared by the water induced low temperature crystallization technology. The amorphous titanium dioxide nanotube films were immersed in mixed solutions which contained an equal volume of water and absolute alcohol, and 0.05 mol/L, 0.10 mol/L, 0.20 mol/L FeCl 3 (Alfa Aesar) were added into the solvent, separately. Then, the mixed solutions were heated to 70 o C and held for 10 h to fully crystallize the TiO 2 NTAs. After that, the samples were dried naturally at 20 o C room temperature to ensure the stability of the nanotube. It must be mentioned that we also tried to prepare higher doping concentration samples, but the TiO 2 NTAs layers broke into pieces due to high internal stress caused by the excessive doping concentration.
X-ray diffraction (XRD) analysis (D8 ADVANCE, Bruker AXS, Karlsruhe, German) was employed to characterize the crystal structure and composition of the samples. The X-ray radiation source was Cu Kα, obtained at 40 kV, 30 mA, the scanning speed was 3.6 • /min at a step of 0.03 • , and the range of 2θ is 20-80 • . The morphological property and energy dispersive spectrometry (EDS) were observed with a field-emission scanning electron microscopy (GeminiSEM 500, Zeiss, Aalen, German). Transmission electron microscopy (TEM) and EDS mapping were performed on a transmission electron microscope (JEOL JEM-2100). Raman testing and analysis were used a laser Raman spectroscope (LabRAM HR Evolution, HORIBA, Fukuoka, Japan) with a 532 nm line for the excitation. The absorption spectra were collected by using integrating sphere accessory of the UV−vis spectrometry (Ultrospec 2100 pro, Harvard Apparatus, Cambridge, MA, USA).
The photocatalytic activities of Fe 3+ doped TiO 2 NTAs were evaluated on the basis of the degradation of methyl orange (MO) as model organic pollutants in aqueous solutions and measured by the UV-vis spectrometer (JASCO V-570 UV/vis/NIR, JASCO, Tokyo, Japan). The surface area (3.0 × 1.6 cm 2 ) of the samples were immersed in the 5 × 10 −5 mol/L MO aqueous solution and irradiated with eight 4W UV bulbs (Toshiba, Black Light Blue, FL4W, Tokyo, Japan), eight 4W visible bulbs (Toshiba, Cool white, FL4W, Tokyo, Japan) with UV filter (>480 nm), for UV photoactivities test and visible photoactivities test, respectively. Before measurement, the samples were soaked for 30 min in the dark with magnetic stirring to reach adsorption/desorption equilibrium. The photo-degradation measurement was carried out in an open quartz photoreaction vessel with rapid stirring for 120 min under room temperature. The concentration of the residual MO was measured by the UV-vis spectrometer at about 460 nm. The photocatalytic performance data of each sample were tested for 10 times, and the mean value and error were calculated and plotted. Figure 1 shows a schematic synthesis procedure of Fe 3+ doped TiO 2 NTAs. In step 1, self-organized TiO 2 NTAs were firstly prepared by electrochemical anodization. A yellowish film could be observed, which is attributed to the amorphous TiO 2 NTAs coated on the Ti foil. Then the as-prepared amorphous TiO 2 NTAs film was immersed into aqueous solution of Fe 3+ at 70 • C to induce crystallization and doping. After this step, the film turned a reddish color, which could be attributed to the doping of Fe 3+ ion in TiO 2 NTAs. Nanomaterials 2020, 10, x FOR PEER REVIEW 3 of 13 spectrometer at about 460 nm. The photocatalytic performance data of each sample were tested for 10 times, and the mean value and error were calculated and plotted. Figure 1 shows a schematic synthesis procedure of Fe 3+ doped TiO2 NTAs. In step 1, selforganized TiO2 NTAs were firstly prepared by electrochemical anodization. A yellowish film could be observed, which is attributed to the amorphous TiO2 NTAs coated on the Ti foil. Then the asprepared amorphous TiO2 NTAs film was immersed into aqueous solution of Fe 3+ at 70 °C to induce crystallization and doping. After this step, the film turned a reddish color, which could be attributed to the doping of Fe 3+ ion in TiO2 NTAs. In order to detect the effect of Fe 3+ doping on the morphology of TiO2 NTAs, SEM was used to observe and characterize the morphology of titanium dioxide nanotubes. As can be observed from Figure 2a, the TiO2 NTs are top-opened and have an inner diameter of about 80-90 nm with a wall thickness of ~20 nm. After doping with Fe 3+ , no significant difference in the nanotube diameter can be observed, while the wall thickness gets slightly increased about 2-5 nm (see Figure 2b). The cross section of Fe 3+ doped TiO2 NTAs is exhibited in Figure 2c; it can be seen that the inner surface of the nanotubes is clean and not blocked by particles. The SEM test results show that the crystallization/doping process did not damage the top-open and highly oriented morphologies of TiO2 NTAs, which is advantageous for the organic solvent to be fully filled into the nanotubes during the photocatalytic process. The weight ratios of Ti, O, and Fe in samples with different doping concentrations were also tested by using EDS, and the results were presented in Table 1. It can be observed that with the increase of Fe 3+ doping concentration, the Fe content gradually rises, the oxygen content basically remains unchanged, and the Ti content gradually decreases in these samples, indicating that the Fe 3+ is likely substituting Ti 4+ into the lattice of TiO2. In order to detect the effect of Fe 3+ doping on the morphology of TiO 2 NTAs, SEM was used to observe and characterize the morphology of titanium dioxide nanotubes. As can be observed from Figure 2a, the TiO 2 NTs are top-opened and have an inner diameter of about 80-90 nm with a wall thickness of~20 nm. After doping with Fe 3+ , no significant difference in the nanotube diameter can be observed, while the wall thickness gets slightly increased about 2-5 nm (see Figure 2b). The cross section of Fe 3+ doped TiO 2 NTAs is exhibited in Figure 2c; it can be seen that the inner surface of the nanotubes is clean and not blocked by particles. The SEM test results show that the crystallization/doping process did not damage the top-open and highly oriented morphologies of TiO 2 NTAs, which is advantageous for the organic solvent to be fully filled into the nanotubes during the photocatalytic process. The weight ratios of Ti, O, and Fe in samples with different doping concentrations were also tested by using EDS, and the results were presented in Table 1. It can be observed that with the increase of Fe 3+ doping concentration, the Fe content gradually rises, the oxygen content basically remains unchanged, and the Ti content gradually decreases in these samples, indicating that the Fe 3+ is likely substituting Ti 4+ into the lattice of TiO 2 .  For the sake of observing the distribution of Fe 3+ in TiO2 nanotube more intuitively, TEM and EDS mapping were used to test the distribution of Ti, O, and Fe elements in a single 0.2 mol/L Fe 3+ doped TiO2 nanotube as shown in Figure 3. We can observe that there is no cladding layer and enrichment of the Fe element is not present on the surface of the TiO2 nanotube, and the Fe element uniformly distributes in the TiO2 nanotube, indicating that the Fe did not form compounds coated on the surface of TiO2 nanotube.  For the sake of observing the distribution of Fe 3+ in TiO 2 nanotube more intuitively, TEM and EDS mapping were used to test the distribution of Ti, O, and Fe elements in a single 0.2 mol/L Fe 3+ doped TiO 2 nanotube as shown in Figure 3. We can observe that there is no cladding layer and enrichment of the Fe element is not present on the surface of the TiO 2 nanotube, and the Fe element uniformly distributes in the TiO 2 nanotube, indicating that the Fe did not form compounds coated on the surface of TiO 2 nanotube.  The XRD patterns of TiO2 NTAs samples with different Fe 3+ doping concentrations are shown in Figure 4. It can be seen that the diffraction peaks of the three samples are quite similar. After comparison with standard PDF cards, we find that these main diffraction peaks well match (110),  The XRD patterns of TiO 2 NTAs samples with different Fe 3+ doping concentrations are shown in Figure 4. It can be seen that the diffraction peaks of the three samples are quite similar. After comparison with standard PDF cards, we find that these main diffraction peaks well match (110), (101), (211), (002), (311) planes of rutile phase TiO 2 and (100), (002), (101) planes of titanium. In addition, no other impurity peaks are found, and whether Fe 3+ is doped into TiO 2 nanotubes in alternative forms needs further discussion. The lattice parameters and interplanar crystal spacing (d values) of pure and Fe 3+ doped TiO 2 NTAs samples are outlined in Table 2. With the increase of Fe 3+ doping concentration, the lattice constant and d values decrease slightly; the main reason is that the radius of Fe 3+ (64 Å) is slightly smaller than that of Ti 4+ (68 Å). This can also be proved from the side that Fe 3+ replaces Ti 4+ and enters into the lattice of TiO 2 .  To further verify that Fe 3+ did not form a compound but entered into the TiO2 lattice, the Raman spectrum of the 0.2 mol/L Fe 3+ doped sample was further investigated and analyzed. The Raman spectrum and peak fitting curves of the sample before photocatalytic experiments are shown in Figure 5a. After fitting the spectral line, we find that there are only three Raman peaks of rutile phase TiO2 (234 cm −1 (Eg mode), 445 cm −1 (Eg mode), and 609 cm −1 (A1g mode)) [26] and no other impurity peaks exist, which also confirmed the XRD test results that Fe was doped into the lattice of TiO2 and no impurity was formed.  To further verify that Fe 3+ did not form a compound but entered into the TiO 2 lattice, the Raman spectrum of the 0.2 mol/L Fe 3+ doped sample was further investigated and analyzed. The Raman spectrum and peak fitting curves of the sample before photocatalytic experiments are shown in Figure 5a. After fitting the spectral line, we find that there are only three Raman peaks of rutile phase TiO 2 (234 cm −1 (Eg mode), 445 cm −1 (Eg mode), and 609 cm −1 (A1g mode)) [26] and no other impurity peaks exist, which also confirmed the XRD test results that Fe was doped into the lattice of TiO 2 and no impurity was formed.  Figure 6 shows the UV-Vis absorption spectra of the pure and Fe 3+ doped TiO2 NTAs. As can be seen from the spectra, the TiO2 NTAs samples without Fe 3+ doping shows an absorption characteristic only in the ultraviolet region (<380 nm), but not in the visible region. Once doped with Fe 3+ , the red shift of absorption edge can be observed clearly in the Figure. 6. The absorption capacity of the ultraviolet region was also significantly enhanced. This result, along with the XRD and Raman analysis above, reveal that the Fe 3+ ions were indeed introduced into the lattice of TiO2 and formed the impurity energy level, thus expanding the absorption spectrum range of TiO2 nanotubes.  Figure 6 shows the UV-Vis absorption spectra of the pure and Fe 3+ doped TiO 2 NTAs. As can be seen from the spectra, the TiO 2 NTAs samples without Fe 3+ doping shows an absorption characteristic only in the ultraviolet region (<380 nm), but not in the visible region. Once doped with Fe 3+ , the red shift of absorption edge can be observed clearly in the Figure 6. The absorption capacity of the ultraviolet region was also significantly enhanced. This result, along with the XRD and Raman analysis above, reveal that the Fe 3+ ions were indeed introduced into the lattice of TiO 2 and formed the impurity energy level, thus expanding the absorption spectrum range of TiO 2 nanotubes. The photocatalytic activities of samples were measured by degradation of methyl orange (MO). Figure 7 shows the concentration change of MO as a function of irradiation time under ultraviolet light for pure and doped TiO2 NTAs with a different amount of Fe 3+ . First, we find that the photocatalytic activity of the sample doped with Fe 3+ in 0.05 mol/L precursor solution can only degrade MO about 33% in 2 h, which is much lower than that of pure titanium dioxide nanotubes. This is might due to the fact that the UV light has provided enough energy to generated electron-hole pairs, and a small amount doping of Fe 3+ tends to act as a recombination center, which will decrease the photoactivities [40]. However, with the increase of the Fe 3+ concentration, the catalytic activity increased gradually and reached the best level when the concentration was 0.2 mol/L. Under this condition, the major effect of the Fe 3+ doping is to introduce impurity levels in the band gap of TiO2, and become a shallow trap of photogenerated electrons or holes, for reducing the recombination of electron-hole pairs [41]. The photocatalytic activities of samples were measured by degradation of methyl orange (MO). Figure 7 shows the concentration change of MO as a function of irradiation time under ultraviolet light for pure and doped TiO 2 NTAs with a different amount of Fe 3+ . First, we find that the photocatalytic activity of the sample doped with Fe 3+ in 0.05 mol/L precursor solution can only degrade MO about 33% in 2 h, which is much lower than that of pure titanium dioxide nanotubes. This is might due to the fact that the UV light has provided enough energy to generated electron-hole pairs, and a small amount doping of Fe 3+ tends to act as a recombination center, which will decrease the photoactivities [40]. However, with the increase of the Fe 3+ concentration, the catalytic activity increased gradually and reached the best level when the concentration was 0.2 mol/L. Under this condition, the major effect of the Fe 3+ doping is to introduce impurity levels in the band gap of TiO 2 , and become a shallow trap of photogenerated electrons or holes, for reducing the recombination of electron-hole pairs [41]. For practical application, it is necessary to estimate the photocatalytic activity of these samples under visible light irradiation, which is shown in Figure 8.  For practical application, it is necessary to estimate the photocatalytic activity of these samples under visible light irradiation, which is shown in Figure 8. The results clearly revealed that doping with Fe 3+ does greatly enhance the photochemical activity of TiO 2 NTAs under visible light. Photogenerated charge carriers successfully flow to Fe 3+ energy level and form an electron trap which shortens the distance between conduction band and valence band [42,43]. The successful doping of Fe ion into TiO 2 makes it more liable for the visible radiation to excite the electrons in conduction band of TiO 2 . Along with the characterization and experiment above, it can be concluded that TiO 2 NTAs have the best photocatalytic performance doped with 0.2 mol/L of Fe 3+ . After photocatalytic experiments, the Raman spectrum and EDS of 0.2 mol/L Fe 3+ doped TiO2 NTAs were also investigated. Figure 5b shows the Raman spectrum and peak fitting curves of the sample after photocatalytic experiments. Still, only three characteristic peaks located at 236, 447, and 610 cm −1 can be observed, which is consistent with the Raman results before the photocatalytic experiments. In addition, after photocatalytic experiments the weight ratios of Ti, O, and Fe in the sample are 41.01, 50.23, and 8.76% w/w, respectively, which also coincides with the original results. The above test results indicate that Fe 3+ did not form other compound impurities and could not be released in the solution during the photoirradiation.

Conclusions
In summary, Fe 3+ doped rutile phase TiO2 NTAs have been successfully prepared by a simple and low-cost method which combined electrochemical anodization and low temperature waterassistant crystallization. The results of test and analysis show that Fe 3+ doping not only doesn't change the crystallization of TiO2 NTAs, but also no Fe-containing compounds impurities are produced. The Fe 3+ substitutes Ti 4+ into the lattice of TiO2 and forms the impurity energy level, which expands the absorption range of TiO2 NTAs from ultraviolet to the visible band. It is noteworthy that Fe 3+ doping significantly improves the photocatalytic activity of TiO2 NTAs in the visible light region, making it have a promising practical application prospect in the photocatalytic decomposition of organic pollutants. After photocatalytic experiments, the Raman spectrum and EDS of 0.2 mol/L Fe 3+ doped TiO 2 NTAs were also investigated. Figure 5b shows the Raman spectrum and peak fitting curves of the sample after photocatalytic experiments. Still, only three characteristic peaks located at 236, 447, and 610 cm −1 can be observed, which is consistent with the Raman results before the photocatalytic experiments. In addition, after photocatalytic experiments the weight ratios of Ti, O, and Fe in the sample are 41.01, 50.23, and 8.76% w/w, respectively, which also coincides with the original results. The above test results indicate that Fe 3+ did not form other compound impurities and could not be released in the solution during the photoirradiation.

Conclusions
In summary, Fe 3+ doped rutile phase TiO 2 NTAs have been successfully prepared by a simple and low-cost method which combined electrochemical anodization and low temperature water-assistant crystallization. The results of test and analysis show that Fe 3+ doping not only doesn't change the crystallization of TiO 2 NTAs, but also no Fe-containing compounds impurities are produced. The Fe 3+ substitutes Ti 4+ into the lattice of TiO 2 and forms the impurity energy level, which expands the absorption range of TiO 2 NTAs from ultraviolet to the visible band. It is noteworthy that Fe 3+ doping significantly improves the photocatalytic activity of TiO 2 NTAs in the visible light region, making it have a promising practical application prospect in the photocatalytic decomposition of organic pollutants.