Growth, Structure, and Photocatalytic Properties of Hierarchical V2O5–TiO2 Nanotube Arrays Obtained from the One-step Anodic Oxidation of Ti–V Alloys

V2O5-TiO2 mixed oxide nanotube (NT) layers were successfully prepared via the one-step anodization of Ti-V alloys. The obtained samples were characterized by scanning electron microscopy (SEM), UV-Vis absorption, photoluminescence spectroscopy, energy-dispersive X-ray spectroscopy (EDX), X-ray diffraction (DRX), and micro-Raman spectroscopy. The effect of the applied voltage (30–50 V), vanadium content (5–15 wt %) in the alloy, and water content (2–10 vol %) in an ethylene glycol-based electrolyte was studied systematically to determine their influence on the morphology, and for the first-time, on the photocatalytic properties of these nanomaterials. The morphology of the samples varied from sponge-like to highly-organized nanotubular structures. The vanadium content in the alloy was found to have the highest influence on the morphology and the sample with the lowest vanadium content (5 wt %) exhibited the best auto-alignment and self-organization (length = 1 μm, diameter = 86 nm and wall thickness = 11 nm). Additionally, a probable growth mechanism of V2O5-TiO2 nanotubes (NTs) over the Ti-V alloys was presented. Toluene, in the gas phase, was effectively removed through photodegradation under visible light (LEDs, λmax = 465 nm) in the presence of the modified TiO2 nanostructures. The highest degradation value was 35% after 60 min of irradiation. V2O5 species were ascribed as the main structures responsible for the generation of photoactive e− and h+ under Vis light and a possible excitation mechanism was proposed.


Introduction
Over the past few decades, photocatalytic processes on the surface of TiO 2 have been intensively studied due to a wide range of industrially oriented applications based on the conversion of sunlight into usable chemical energy [1][2][3][4][5][6]. Being non-toxic, abundant, chemically and physically In view of this, this work focuses on the synthesis of visible-light photoactive V 2 O 5 -TiO 2 NTs through the one-step anodic oxidation of Ti-V alloys in an ethylene glycol-based electrolyte, and their application in the photocatalytic degradation of toluene. The effect of the vanadium content in the alloy, applied voltage, and electrolyte composition (water content) was systematically studied to determine the influence of these parameters on the morphology and gas phase photoactivity, evaluated for the first time, of the obtained nanomaterials. The as-prepared NTs were characterized by using scanning electron microscopy (SEM), X-ray diffraction (XRD), energy-dispersive X-ray spectroscopy (EDX), micro-Raman spectroscopy, UV-Vis absorption, and photoluminescence spectroscopy. A possible mechanism of toluene degradation at the surface of V 2 O 5 -TiO 2 NTs under the influence of visible light was also proposed.

Morphology and Growth Mechanism
Ti foils and Ti-V alloys of technical grade were anodically oxidized for 60 min, under the specific parameters summarized in Table 1. The effect of the applied potential (30,40, and 50 V), vanadium content in the alloy (5, 10, and 15 wt %), and water content in the electrolyte (2, 5, and 10 vol %) on the morphology of the as-prepared samples were studied by scanning electron microscopy. The top-view and cross-sectional scanning electron microscopy (SEM) images are presented in Figure 1. The anodization of Ti sheets led to the formation of uniform and self-organized NTs with an open tube top and smooth walls, and the tube diameter and length ranged from 81 to 120 nm and from 1.5 to 16.2 µm, respectively (Ti_30V, Ti_50V, respectively). The samples anodized from the Ti-V alloys presented a different morphology, depending on the preparation parameters. The series of samples synthesized from alloys with a 10 wt % vanadium content generally exhibited a sponge-like structure integrated by overlapped layers with a tubular appearance. The registered diameters of these structures varied from 61 to 101 nm and the average thickness of the mixed oxide layers was 0.3-0.8 µm. The samples prepared from alloys with 15 wt % of vanadium and using electrolytes with different water contents showed different morphologies. The Ti 85 V 15 _40V_2% and Ti 85 V 15 _40V_10% samples presented a sponge-like structure made up of interconnected disordered bundles. Conversely, the Ti 85 V 15 _40V_5% sample had a tubular structure with ripples on the tube wall, although the nanotubular layer was not highly organized. NTs presented a diameter (103 nm) similar to that of pristine TiO 2 NTs (100 nm) obtained at the same voltage (40 V), while the length (0.9 µm) was smaller than that of the analogous pristine sample (5 µm). The highest level of self-organization was achieved with the sample obtained from the anodization of the alloy with a 5 wt % of vanadium content (Ti 95 V 5 _40V), for which the synthesized NTs appeared to be composed of interconnected rings with a diameter of 86 nm and a length of 1 µm. As can be seen, the vanadium content in the alloy has a strong influence on the morphology of the samples. According to Yang,Kim,and Schmuki [55], the absence of a self-organized nanotube layer can be attributed to the low stability of the vanadium oxide, and therefore, the sample (Ti 95 V 5 _40V) synthesized from the alloy with the lowest vanadium content exhibited the best auto-alignment and self-organization. The influence of the other parameters, applied potential and water content, on the morphology of the samples was not clear due to the strong influence of the vanadium content in the alloy.    Considering these results, the SEM images of the Ti90V10_40V sample anodized during 4, 15, and 60 min (Figure 2d-f), together with literature data, a probable growth mechanism of V 2 O 5 -TiO 2 NTs has been described. As can be seen in Figure 2a-c, the shape of the current density-time curves recorded for the V 2 O 5 -TiO 2 samples were very similar to those of pristine TiO 2 NTs. During the first stage, the formation of the V 2 O 5 -TiO 2 oxide layer induced an exponential decrease in the current density, because of the reaction of Ti and V with the O 2 − and OH − ions from the water. The presence of this mixed oxide layer can be observed in Figure 2d, corresponding to the Ti90V10_40V sample after 4 min of anodization. Then, the current density progressively increased throughout the second stage due to the dissolution of the oxide layer, which led to an increase in the surface area of the electrode with the initiation of pore growth [64]. These soluble species correspond to the fluoride complexes, [TiF 6 ] 2− and [VF 6 ] − [65,66]. Figure 2e shows the initial pores in the sample after 15 min of anodic oxidation. Finally, a regular and self-ordered NT layer, which can be appreciated in Figure 2f, is formed under a quasi-steady state, which is stablished due to the equilibrium between the formation and dissolution of the oxide layer. During this stage, pores equally share the available current [45].   The elemental composition of the obtained samples was analyzed through energy-dispersive X-ray spectroscopy (EDX) and the results presented in Table 1 show that the mass ratios between Ti and V in the V 2 O 5 -TiO 2 mixed oxides nanostructures (NS) agree well with the nominal content of the alloy. In addition, no trace of elements other than Ti, V, C, and O, was observed. These findings confirm the chemical homogeneity of the nanotube layer. Furthermore, from the EDX mapping presented in Figure 3, it can be concluded that the aggregation of Ti and V was not observed. The elemental composition of the obtained samples was analyzed through energy-dispersive X-ray spectroscopy (EDX) and the results presented in Table 1 show that the mass ratios between Ti and V in the V2O5-TiO2 mixed oxides nanostructures (NS) agree well with the nominal content of the alloy. In addition, no trace of elements other than Ti, V, C, and O, was observed. These findings confirm the chemical homogeneity of the nanotube layer. Furthermore, from the EDX mapping presented in Figure 3, it can be concluded that the aggregation of Ti and V was not observed.

Optical Properties
The UV-Vis spectra of the obtained samples were compared with those of pristine TiO2 NTs. Figure 4a clearly shows that the samples prepared from Ti90V10 alloys exhibited a stronger absorbance in the broad visible range of 400-750 nm than TiO2 NTs. The spectra of the series with different vanadium contents, displayed in Figure 4b, indicated that an increase in the vanadium content in the alloy led to an increase in the absorbance intensity in the visible range, together with a red-shift. In particular, the spectrum of the sample Ti85V15_40V presented a peak of maximum absorbance near 500 nm, which, according to literature data, corresponds to V2O5 [42,67]. The spectra of the series of samples prepared in an electrolyte with different water contents and plotted in Figure 4c are consistent with the previous statements and no clear effect of the water content on the UV-Vis properties was found. All of the spectra for this series of samples showed a peak of absorption in the

Optical Properties
The UV-Vis spectra of the obtained samples were compared with those of pristine TiO 2 NTs. Figure 4a clearly shows that the samples prepared from Ti 90 V 10 alloys exhibited a stronger absorbance in the broad visible range of 400-750 nm than TiO 2 NTs. The spectra of the series with different  Figure 4b, indicated that an increase in the vanadium content in the alloy led to an increase in the absorbance intensity in the visible range, together with a red-shift. In particular, the spectrum of the sample Ti 85 V 15 _40V presented a peak of maximum absorbance near 500 nm, which, according to literature data, corresponds to V 2 O 5 [42,67]. The spectra of the series of samples prepared in an electrolyte with different water contents and plotted in Figure 4c are consistent with the previous statements and no clear effect of the water content on the UV-Vis properties was found. All of the spectra for this series of samples showed a peak of absorption in the Vis range near 500 nm, and the spectrum of the Ti 85 V 15 _40V_5% sample showed the highest absorbance intensity peak. It can be concluded that the presence of the V 2 O 5 in V 2 O 5 -TiO 2 matrix enhanced the light absorption in the range of 400-750 nm.

Optical Properties
The UV-Vis spectra of the obtained samples were compared with those of pristine TiO2 NTs. Figure 4a clearly shows that the samples prepared from Ti90V10 alloys exhibited a stronger absorbance in the broad visible range of 400-750 nm than TiO2 NTs. The spectra of the series with different vanadium contents, displayed in Figure 4b, indicated that an increase in the vanadium content in the alloy led to an increase in the absorbance intensity in the visible range, together with a red-shift. In particular, the spectrum of the sample Ti85V15_40V presented a peak of maximum absorbance near 500 nm, which, according to literature data, corresponds to V2O5 [42,67]. The spectra of the series of samples prepared in an electrolyte with different water contents and plotted in Figure 4c are consistent with the previous statements and no clear effect of the water content on the UV-Vis properties was found. All of the spectra for this series of samples showed a peak of absorption in the Vis range near 500 nm, and the spectrum of the Ti85V15_40V_5% sample showed the highest absorbance intensity peak. It can be concluded that the presence of the V2O5 in V2O5-TiO2 matrix enhanced the light absorption in the range of 400-750 nm.  It is known that photoluminescence (PL) spectroscopy is a powerful tool for determining the presence of surface defects, trap states, and sub-band states in the mid-gap level of photocatalysts [68]. The PL spectra of the obtained photocatalysts are presented in Figure 5. It should be noted that the same emission and position peaks were observed among all series. Notably, the emission peak at approximately at 420 nm can be ascribed to the existence of self-trapped excitons from the TiO 6 8− octahedron, while the two emission peaks at 450 and 485 nm could be assigned to the presence of surface defects, in the form of oxygen vacancies, which can create intermediate energy states located below the conduction band and which are able to trap electrons. The last peak at approximately 525 nm can be associated with the radiative recombination of the charge carriers [69,70]. It is known that photoluminescence (PL) spectroscopy is a powerful tool for determining the presence of surface defects, trap states, and sub-band states in the mid-gap level of photocatalysts [68]. The PL spectra of the obtained photocatalysts are presented in Figure 5. It should be noted that the same emission and position peaks were observed among all series. Notably, the emission peak at approximately at 420 nm can be ascribed to the existence of selftrapped excitons from the TiO6 8− octahedron, while the two emission peaks at 450 and 485 nm could be assigned to the presence of surface defects, in the form of oxygen vacancies, which can create intermediate energy states located below the conduction band and which are able to trap electrons. The last peak at approximately 525 nm can be associated with the radiative recombination of the charge carriers [69,70]. The results mentioned above confirm the presence of surface/structural defects, which can play a role in the photocatalytic degradation of pollutants.

Structural Properties
XRD patters of the obtained photocatalysts are presented in Figure 6. The calculated average crystallite size for pristine and modified TiO2 NTs are gathered in Table 1. The average crystallite size was calculated using the Scherrer equation, based on the (101) diffraction peak. In the registered region, peaks at 2θ values of 25.67°, 37.97°, 48.31°, 54.16°, and 55.30° can be ascribed to (101), (004), The results mentioned above confirm the presence of surface/structural defects, which can play a role in the photocatalytic degradation of pollutants.

Structural Properties
XRD patters of the obtained photocatalysts are presented in Figure 6. The calculated average crystallite size for pristine and modified TiO 2 NTs are gathered in Table 1 can be ascribed to planes of metallic Ti substrate. As was mentioned above, the diffraction peaks corresponding to the pure anatase TiO 2 phase were found, but other phases assigned to V 2 O 5 were not observed. There are three possible explanations for this. Firstly, it could be because V 2 O 5 diffraction peaks exist; however, the intensity of peaks is too low for this to be true. The absence of peaks corresponding to V 2 O 5 in the XRD patters may be due to the low content and amorphous character of V 2 O 5 or the short-range crystalline. Eventually, the vanadium species are incorporated into the TiO 2 lattice. On the other hand, in modified samples, the bands assigned to the anatase phase had a smaller and wider intensity. In particular, the intensity of the pick ascribed to the characteristic (101) plane of anatase decreased with the increase in the vanadium content in the alloy. This is related to the smaller crystallite size of V 2 O 5 -TiO 2 NS than that of pristine TiO 2 NTs [71].
Furthermore, it can be seen that the intensity of the anatase reflexes increased, while those of the substrate decreased, with the increase of the anodizing voltage. This is caused by the increasing thickness of the nanotube layer.
The average crystallite size varied from 30 to 36 nm among Ti-V series, and from 33 to 38 nm for pristine TiO 2 NTs. The smallest crystallite size was found for the Ti 90 V 10 _50V sample, which reached 30 nm. A clear correlation between the crystallite size and (i) anodization potential; (ii) vanadium content in the alloy; and (iii) water content in the electrolyte, was not observed.  Micro-Raman spectroscopy was performed to determine the microstructure of the prepared samples. A 532 nm laser was used for the excitation. Figure 7 shows the Raman spectra of pristine TiO2 and V2O5-TiO2 NTs. The observed peaks at approximately 150, 396, 515, and 636 cm −1 are ascribed to the Eg, B1g, A1g + B1g, and Eg modes of the anatase phase, respectively, in agreement with previous reports [42,[72][73][74]. The Eg modes are assigned to TiO2 symmetry, B1g to O-Ti-O bending, and A1g + B1g to Ti-O stretching [75]. All of the spectra also registered a weak combination band at ca. 800 cm −1 , which is characteristic of the Raman signature of anatase [76]. No distinguishable crystalline V2O5 Raman bands were present at 703 and 997 cm −1 in any spectra, probably due to the low content of vanadium in the alloy precursors or to the highly dispersed state of V2O5 in V2O5-TiO2 NS. This was also reported by former publications for composites with the V2O5-TiO2 system [32,75,77]. Micro-Raman spectroscopy was performed to determine the microstructure of the prepared samples. A 532 nm laser was used for the excitation. Figure 7 shows the Raman spectra of pristine TiO 2 and V 2 O 5 -TiO 2 NTs. The observed peaks at approximately 150, 396, 515, and 636 cm −1 are ascribed to the E g , B 1g , A 1g + B 1g , and E g modes of the anatase phase, respectively, in agreement with previous reports [42,[72][73][74]. The E g modes are assigned to TiO 2 symmetry, B 1g to O-Ti-O bending, and A 1g + B 1g to Ti-O stretching [75]. All of the spectra also registered a weak combination band at ca. 800 cm −1 , which is characteristic of the Raman signature of anatase [76]. No distinguishable crystalline V 2 O 5 Raman bands were present at 703 and 997 cm −1 in any spectra, probably due to the low content of vanadium in the alloy precursors or to the highly dispersed state of V 2 O 5 in V 2 O 5 -TiO 2 NS. This was also reported by former publications for composites with the V 2 O 5 -TiO 2 system [32,75,77].
Micro-Raman spectroscopy was performed to determine the microstructure of the prepared samples. A 532 nm laser was used for the excitation. Figure 7 shows the Raman spectra of pristine TiO2 and V2O5-TiO2 NTs. The observed peaks at approximately 150, 396, 515, and 636 cm −1 are ascribed to the Eg, B1g, A1g + B1g, and Eg modes of the anatase phase, respectively, in agreement with previous reports [42,[72][73][74]. The Eg modes are assigned to TiO2 symmetry, B1g to O-Ti-O bending, and A1g + B1g to Ti-O stretching [75]. All of the spectra also registered a weak combination band at ca. 800 cm −1 , which is characteristic of the Raman signature of anatase [76]. No distinguishable crystalline V2O5 Raman bands were present at 703 and 997 cm −1 in any spectra, probably due to the low content of vanadium in the alloy precursors or to the highly dispersed state of V2O5 in V2O5-TiO2 NS. This was also reported by former publications for composites with the V2O5-TiO2 system [32,75,77].

Photocatalytic Performance
The effect of the anodization voltage, vanadium content in the alloy, and water content in the electrolyte on the photocatalytic activity was evaluated through the degradation of toluene from an air mixture (200 ppmv of toluene) under Vis irradiation (LEDs array, λmax = 465 nm). Figure 8 presents the degradation curves for the above-mentioned series and their comparison with the photoactivity of reference pristine TiO2 NTs. These plots show that V2O5-TiO2 samples from all series were active in the photodegradation reaction, in contrast with pristine TiO2 NTs which exhibited negligible toluene removal (ca. 5%). The highest degradation of toluene in the presence of samples prepared from the Ti90V10 alloys (see Figure 8a), after 60 min of irradiation, was observed for the sample

Photocatalytic Performance
The effect of the anodization voltage, vanadium content in the alloy, and water content in the electrolyte on the photocatalytic activity was evaluated through the degradation of toluene from an air mixture (200 ppmv of toluene) under Vis irradiation (LEDs array, λ max = 465 nm). Figure 8 presents the degradation curves for the above-mentioned series and their comparison with the photoactivity of reference pristine TiO 2 NTs. These plots show that V 2 O 5 -TiO 2 samples from all series were active in the photodegradation reaction, in contrast with pristine TiO 2 NTs which exhibited negligible toluene removal (ca. 5%). The highest degradation of toluene in the presence of samples prepared from the Ti 90 V 10 alloys (see Figure 8a), after 60 min of irradiation, was observed for the sample anodized under 40 V (34%). The toluene removal reached by samples anodized under 30 V and 50 V were not that different from the best one (27% and 33%, respectively). In view of this, 40 V was selected as the potential for further synthesis, to determine the vanadium content in the alloy and the composition of the electrolyte solution, which are favorable for the photodegradation reaction. Figure 8b presents similar results, for the samples obtained from alloys with different vanadium contents. It can be observed that the vanadium content in the alloy slightly affected the photoactivity of the samples. The maximum toluene removal was found to be achieved for the sample with 10 wt % of vanadium in the alloy (Ti 90 V 10 _40V, 34% of degradation). The analysis of the effect of water content in the electrolyte was carried out with NS obtained from Ti 85 V 15 alloys. As can be seen in Figure 8c, there is a slight difference in the photocatalytic performance between these samples, among this series. The highest degradation of toluene was exhibited by the sample anodized in the electrolyte containing 5% of water and it corresponded to 35% of toluene removal (Ti 85 V 15 _40V_5%). For a more detailed comparison of the obtained results, the initial reaction rate and reaction rate constants were calculated and presented in Table 1. The highest value for the initial reaction rate, among all series, was achieved in the presence of the Ti 85 V 15 _40V_5% sample (7.08 × 10 −2 µmol·dm −3 ·min −1 ), which also exhibited the highest absorbance intensity peak near 500 nm and consisted of a NT layer which was not highly organized. This suggests that this NT composite effectively enhanced visible light harvesting and the consequent photocatalytic reaction, owing to the presence of V 2 O 5 [35,43]. Furthermore, no correlation between the morphology and the photocatalytic performance of the samples was observed.
vanadium content, based on EDX analysis, was 9.08 wt %. On the other hand, the Ti85V15_40V_10% sample showed the lowest photoactivity. It had a sponge-like morphology with a vanadium content of 8.91%, based on EDX analysis, which is lower than the content of the sample with the highest photoactivity, considering that both were prepared from Ti85V15 alloys. Its crystallite size was 32 nm, smaller than that of the Ti85V15_40V_5% sample. The initial reaction rate achieved in the presence of this sample was 4.50 × 10 −2 μmol·dm −3 ·min −1 , which is 1.6 times lower than that reported for the most photoactive one (7.08 × 10 −2 μmol·dm −3 ·min −1 ). To further analyze the photocatalytic properties of the synthesized composites, the effect of different irradiation wavelengths was studied using the most photoactive sample (Ti85V15_40V_5%). The gas phase degradation of toluene was tested under 375, 415, and 465 nm and the obtained results are displayed in Figure 9. It can be observed that the highest degradation (52%) after 60 min of irradiation was achieved under UV light (375 nm). This can be explained by the presence of TiO2 in the NT matrix, which is the main active species under UV light irradiation. On the other hand, under the influence of visible light irradiation, 415 and 465 nm, the photocatalytic degradation reached almost the same level, in both cases, with values of 34% and 35%, respectively. This indicates that In conclusion, the highest photoactivity under visible light (465 nm) was observed in the presence of the Ti 85 V 15 _40V_5% sample. This sample not only exhibited the highest absorbance intensity at a wavelength of about 500 nm, but also reported the highest diameter (103 nm), the second longest NTs (0.9 µm), and the largest crystallite size (36 nm), from the modified samples. Its vanadium content, based on EDX analysis, was 9.08 wt %. On the other hand, the Ti 85 V 15 _40V_10% sample showed the lowest photoactivity. It had a sponge-like morphology with a vanadium content of 8.91%, based on EDX analysis, which is lower than the content of the sample with the highest photoactivity, considering that both were prepared from Ti 85 V 15 alloys. Its crystallite size was 32 nm, smaller than that of the Ti 85 V 15 _40V_5% sample. The initial reaction rate achieved in the presence of this sample was 4.50 × 10 −2 µmol·dm −3 ·min −1 , which is 1.6 times lower than that reported for the most photoactive one (7.08 × 10 −2 µmol·dm −3 ·min −1 ).
To further analyze the photocatalytic properties of the synthesized composites, the effect of different irradiation wavelengths was studied using the most photoactive sample (Ti 85 V 15 _40V_5%). The gas phase degradation of toluene was tested under 375, 415, and 465 nm and the obtained results are displayed in Figure 9. It can be observed that the highest degradation (52%) after 60 min of irradiation was achieved under UV light (375 nm). This can be explained by the presence of TiO 2 in the NT matrix, which is the main active species under UV light irradiation. On the other hand, under the influence of visible light irradiation, 415 and 465 nm, the photocatalytic degradation reached almost the same level, in both cases, with values of 34% and 35%, respectively. This indicates that under Vis light irradiation, V 2 O 5 are the main species responsible for the generation of e − and h + (as presented in Figure 10, excitation mechanism) over the surface of NTs, which led to the photodegradation of toluene, and this is supported by the negligible degradation reported for pristine TiO 2 NTs under Vis light. under Vis light irradiation, V2O5 are the main species responsible for the generation of e − and h + (as presented in Figure 10, excitation mechanism) over the surface of NTs, which led to the photodegradation of toluene, and this is supported by the negligible degradation reported for pristine TiO2 NTs under Vis light.

Synthesis of Pristine TiO2 and V2O5-TiO2 Nanotubes
Ti foils and Ti-V alloys were ultrasonically cleaned in acetone, isopropanol, methanol, and deionized water for 10 min. Then, the foils were dried in an air stream. The anodization processes were carried out at room temperature, in an electrochemical cell consisting of a platinum mesh as the counter electrode, and the Ti-V alloy (2.5 cm × 2.5 cm) as the working electrode. A reference electrode of Ag/AgCl connected to a digital multimeter (BRYMEN BM857a, New Taipei City, Taiwan) was used to control and record information about the actual potential and current on the alloy. The

Synthesis of Pristine TiO 2 and V 2 O 5 -TiO 2 Nanotubes
Ti foils and Ti-V alloys were ultrasonically cleaned in acetone, isopropanol, methanol, and deionized water for 10 min. Then, the foils were dried in an air stream. The anodization processes were carried out at room temperature, in an electrochemical cell consisting of a platinum mesh as the counter electrode, and the Ti-V alloy (2.5 cm × 2.5 cm) as the working electrode. A reference electrode of Ag/AgCl connected to a digital multimeter (BRYMEN BM857a, New Taipei City, Taiwan) was used to control and record information about the actual potential and current on the alloy. The anodization was conducted in an electrolyte composed of EG, water, and NH 4 F 0.09 M, during 60 min, with a voltage in the range of 30-50 V which was applied with a programmable DC power supply (MANSON SDP 2603, Hong Kong, China). Three electrolyte solutions with different water contents were used (volume ratios of EG:water of 98:2, 95:5, and 90:10). The obtained samples were rinsed with deionized water, sonicated in deionized water (1 min), dried in air (80 • C for 24 h), and calcined (450 • C, heating rate 2 • C/min) for 1 h.

Characterization of Pristine TiO 2 and V 2 O 5 -TiO 2 Nanotubes
The morphology of the synthesized pristine TiO 2 and V 2 O 5 -TiO 2 nanotubes was determined by using scanning electron microscopy (SEM, FEI QUANTA 3D FEG, FEI Company, Brno, Czech Republic). Energy-dispersive X-ray spectroscopy (EDX) analysis was performed with a scanning electron microscope (SEM, Zeiss, Leo 1430 VP, Carl Zeiss, Oberkochen, Germany). The crystal structure of the samples was determined from X-ray diffraction patterns recorded in the range of 2θ = 20 • -90 • , using an X-ray diffractometer (X'Pert Pro, Panalytical, Almelo, The Netherlands) with Cu Kα radiation. The crystallite size was calculated based on the Scherrer formula. Raman spectra were measured with a micro-Raman spectrometer (Senterra, Bruker Optik, Billerica, MA, USA) with a 532 nm excitation laser.
The UV-Vis absorbance spectra were registered on a SHIMADZU (UV-2600) UV-VIS Spectrophotometer (SHIMADZU, Kioto, Japan) equipped with an integrating sphere. The measurements were carried out in the wavelength range of 300-800 nm, the baseline was determined with barium sulfate as the reference, and the scanning speed was 250 nm/min at room temperature. The photoluminescence (PL) spectra were recorded at room temperature with a LS-50B Luminescence Spectrometer equipped with a Xenon discharge lamp as an excitation source and a R928 photomultiplier (HAMAMATSU, Hamamatsu, Japan) as detector. The excitation radiation (300 nm) was directed onto the surface of the samples at an angle of 90 • .

Measurement of Photocatalytic Activity
The photocatalytic activity of the as-prepared NTs was analyzed, for the first time, in the purification of air from toluene, which was used as a model pollutant. The photodegradation experiments were carried out in a stainless-steel reactor with a volume of ca. 35 cm 3 . The reactor included a quartz window, two valves, and a septum. The light source consisting of an array of 25 LEDs (λ max = 375, 415 and 465 nm, Optel, Opole, Poland) was located above the sample. The anodized foil was placed at the bottom side of the reactor and it was closed with the quartz window. A gas mixture (200 ppmv) was passed through the reactor during 1 min, the valves were then closed, and the reactor was kept in the dark for 30 min in order to achieve the equilibrium. Before starting the irradiation, a reference toluene sample was taken. The concentration was determined by using a gas chromatograph (TRACE 1300, Thermo Scientific, Waltham, MA, USA), equipped with an ionization flame detector (FID) and an Elite-5 capillary column. The samples (200 µL) were dosed with a gas-tight syringe for 10 min.

Conclusions
In summary, V 2 O 5 -TiO 2 mixed oxide layers were successfully synthesized through the one-step anodization of Ti-V alloys in a fluoride-containing EG-based electrolyte. The obtained layers exhibited a sponge-like and nanotubular structure with highly enhanced optical and visible-light-photocatalytic properties, in contrast with pristine TiO 2 NTs. The photoactivity of these anodically-obtained composites was evaluated for the first time in the degradation of toluene (200 ppmv) in the gas phase under visible light, with a twenty-five-LED array as the irradiation source (λ max = 465 nm). All of the V 2 O 5 -TiO 2 samples were reported as photoactive and the initial degradation reaction rate was in the range of 4.50-7.08 × 10 −2 µmol·dm −3 ·min −1 . The visible light harvesting was attributed to the presence of the narrow-bandgap V 2 O 5 species in the matrix of the V 2 O 5 -TiO 2 composites. A morphological study was also reported and the vanadium content in the alloy was found as the key factor limiting the self-ordering of the electrochemically prepared thin layers. The highest photoactivity under visible light (465 nm) was observed in the presence of the Ti 85 V 15 _40V_5% sample. This sample not only exhibited the highest absorbance intensity at about 500 nm, but also reported the highest diameter (103 nm), the optimum length (0.9 µm), and the largest crystallite size (36 nm) among all of the modified samples. EDX analysis revealed that the vanadium content in this sample was equal to 9.08 wt %. In sum, the photocatalytic properties of these highly efficient nanocomposites, obtained through the most suitable method (electrochemical technique), permit new insights into the exploitation of industrially oriented applications, for instance, photocatalytic devices for air purification. The presented materials are photoactive under a low powered light source, and thus, the use of low cost light-emitting diodes (LEDs) as an irradiation source can significantly reduce the cost of photocatalytic air treatment processes, which is consistent with the principles of green chemistry.