Introduction

Titanium dioxide is the most widely used semiconductor as a chemically stable, highly efficient, and relatively inexpensive photocatalyst [1]. The fabrication of TiO2 with interesting morphologies and properties has recently attracted considerable attention. Various nanostructures of TiO2 such as nanowire, nanoparticles, nanorods, nanosheets, nanotubes, and microspheres have been successfully synthesized [2,3,4,5,6,7,8]. Among these nanostructures, it is well known that ordered TiO2 nanotube arrays with larger specific surface areas are a suitable structure in photocatalytic applications. However, the photocatalytic efficiency of TiO2 is mainly limited by its large bandgap, the recombination of photogenerated electron–hole pairs, and their small lifetime [9, 10]. Recently, many efforts have been made, and different approaches have been developed to improve these limitations, such as doping of TiO2 [11,12,13], coupling the TiO2 with a semiconductor material with a narrow bandgap [14,15,16,17], decoration of TiO2 with different noble metals [18], and surface photosensitization [19]. The doping process is the typical approach that has been widely applied to enhance the photocatalytic efficiency of TiO2 [20]. A previous study revealed that doping with metal elements reduces the bandgap of TiO2 to increase the number of photogenerated electron–hole pairs and the recombination rate of the photogenerated electron–hole pairs [21, 22]. Park et al. [23]. doped TiO2 by divalent metals (Co2+, Ni2+, Cu2+, and Zn2+) and showed that Cu-doped TiO2 and Zn-doped TiO2 were highly interesting materials in the photodecomposition of methylene blue. Sangpour et al. [24] showed that doping with Ag, Au, and Cu metal increases the photocatalytic activity of TiO2 by improving the radical formation. They found that the photocatalytic efficiency increased from 40% for pure TiO2 to 90%, 75%, and 50% for Cu-doped TiO2, Au-doped TiO2, and Ag-doped TiO2, respectively. Only a few studies focused on the doping of TiO2 NTs during the anodizing process in the literature [25,26,27,28,29].

This study reports a simple approach to fabricating Cu and Ag-doped TiO2 NTs by anodizing of pure titanium in glycerol electrolyte containing ammonium fluoride. The effects of incorporating ions into TiO2 NTs on their photocatalytic efficiency were investigated. The morphology and structure were studied by scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX), X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), and Raman spectroscopy. The electrical properties were studied by Mott-Schottky measurement. The effect of Cu and Ag doping on the photocatalytic activity of doped samples was evaluated by methylene blue (MB) degradation under UV irradiation.

Materials and method

Materials

Titanium foil (99.99% pure, 1 mm thick), glycerol (99.8%, anhydrous), NH4F (98%), HNO3 (99.98%), HF (40%), silver nitrate (AgNO3), copper (II) sulfate (CuSO4, 5H2O), benzoquinone (99.5%), isopropanol (99.0%), triethanolamine (98%), and methylene blue were purchased from Sigma-Aldrich (St. Louis, USA).

Fabrication of pure and doped TiO2 NTs

The titanium foils were first polished with different emery paper sizes (from #1000 to #5000 grade), rinsed with distilled water, and then chemically etched by immersion in a mixture of HF and HNO3 acids for 10 s. The mix’s HF/HNO3/H2O ratio was 1:1:2 in volume [30]. The pure and doped TiO2 NTs were synthesized by electrochemical anodization of titanium foils in a glycerol-based electrolyte solution containing NH4F and distilled water. The conditions of anodization are described in detail in our previous paper [31]. Briefly, the anodization was performed in a two-electrode electrochemical device with a platinum foil as the cathode and pure titanium as the anode. The anodization was carried out under a constant voltage of 60 V for 60 min at a temperature of 25 ± 1 °C. The distance between anode and cathode was kept at 1 cm. The electrolyte bath composition used to elaborate the different samples is listed in Table 1. The electrolyte solution was magnetically stirred during the electrochemical reaction (250 tr/min). After anodization, the anodized samples were immediately washed in distilled water. To improve the crystallinity of the as-synthesized NTs, the as-formed samples were heat-treated in a muffle furnace at 600 °C for 2 h with a heating ramp of 10 °C/min [32].

Table 1 The electrolyte used for the preparation of the samples
Full size table

Characterization

The morphology and composition of the TiO2 NTs and modified TiO2 NTs were characterized under a field emission scanning electron microscope (TESCAN VEGA3 SEM) coupled with an energy dispersive X-ray microanalysis system (EDAX, EDX) operated at an accelerating voltage of 20 kV and under a pressure of 1.3 × 10−4 Pa. The samples’ surface properties and oxidation states were analyzed using X-ray photoelectron spectroscopy (XPS). The XPS peaks are deconvoluted using CasaXPS software (version 2.3.23) with Lorentzian Asymmetric LA (1.53, 243) peak fitting. The TiO2 crystalline structure was determined by X-ray diffraction (XRD) using a Rigaku, SmartLab SE, operated at 40 kV and 50 mA. The X-ray source consists of Cu Kα radiation (1.54184 Å) selected with a Cu Kβ filter. The data were collected between 10 and 70° with a scan speed of 5°/min. The Raman spectrum was recorded using a confocal Raman spectrometer (Confotec MR520 microscope) instrument with a laser (λ = 532 nm) and analyzed with 1200 lines/in the grating. The acquisition time of 30 s is used with 10 mW incident power. The functional groups and their modes of vibrations are analyzed by using a VERTEX 70 FT-IR Spectrometer with ATR attachment.

The Mott-Schottky measurements were performed using a conventional three-electrode cell in a 0.1 M Na2SO4 aqueous solution [33]. The anodized TiO2 NT electrode (working area = 1 cm2), a saturated calomel electrode (SCE), and a platinum sheet (4 cm2) were used as the working, reference, and counter electrodes, respectively. The measurements were performed using a VoltaLab potentiostat (PGZ301) controlled by the VoltaMaster 4 software. The temperature was controlled in jacketed glass at 293 K using a bath thermostat. The analysis was performed at a frequency of 1 kHz in a wide voltage range (from − 1 V to 1 V/SCE) [34].

Photocatalytic study

Photocatalytic activities of pure and doped TiO2 samples were evaluated by degradation of the MB dye in an aqueous solution under UV irradiation. The initial concentration of MB was 2.5 mg/L. Before illumination, the TiO2 NT photocatalyst was immersed in the reactor containing 50 mL of MB and magnetically stirred for 20 min in the dark to establish an adsorption–desorption equilibrium between the photocatalyst and MB. Then, the solution was irradiated under UV light using an Ultra-Vitalux lamp at 300 W with a high-pressure tungsten filament source for 4 h. The reactor temperature was kept at 25 °C by a water flow. At each 30-min interval, a volume of 3 mL of solution was sampled for analysis, and a UV–vis spectrophotometer measured the absorbance of MB at 664 nm. The reusability test was performed using the same protocol described above by reutilizing the used TiO2 NT photocatalyst for the next experiment.

Results and discussion

Morphology and composition characterizations

Figure 1 shows a representative SEM image and EDX spectra of the pure TiO2 NTs (Fig. 1a, d), Cu-TiO2 NTs (Fig. 1b, e), and Ag-TiO2 NTs (Fig. 1c, f). From SEM images, it can be seen that the morphology of the doped TNs is similar to that of pure TiO2 NTs, indicating that these elements’ doping does not influence the morphologies of the TiO2. The highly ordered NT arrays with an inner diameter and wall thickness of about 127 and 52 nm, respectively, were produced on the titanium surface by anodization.

Fig. 1
figure 1

SEM images and EDX spectra of pure TiO2 NTs a, d, Cu-TiO2 NTs b, e, and Ag-TiO2 NTs c, f

Full size image

The EDX analysis was performed on the sample to verify the doping process and determine its surface’s chemical composition. The EDX analysis indicates qualitatively the presence of the Ti, O, F, and C in all anodized samples. The peaks of copper (at 0.94 kV) and silver (at 2.98 kV) were also detected in the doped NTs, as revealed in Fig. 1e and f, respectively, indicating clearly that the doping has been carried out successfully.

XPS analysis

The XPS analysis was performed to determine the composition and chemical bonding states of all samples. Figure 2 shows the survey spectra for pure and doped TiO2 NTs. The photoelectron peaks for Ti, O, F, and C were clearly recorded for all three samples. The peaks of copper (Cu2p3/2 and Cu2p1/2) and silver (Ag3d3/2 and Ag3d5/2) were detected in the doped NTs, as shown in Fig. 2b and c, respectively, indicates clearly that Ag and Cu species are successfully incorporated into the oxide lattice during its growth.

Fig. 2
figure 2

Full XPS spectrum of a pure TiO2 NTs, b Cu doped TiO2 NTs, and c Ag-doped TiO2 NTs

Full size image

Figure 3a shows high-resolution region scan spectra of the Ti2p for pure TiO2 NTs. This spectrum is fitted with three peaks corresponding to titanium dioxide (Ti4+) and titanium suboxide (Ti3+) in Ti2p1/2 and Ti2p3/2, respectively. These peaks are fitted as Ti4+2p3/2 at 459.38 eV, Ti4+2p1/2 at 465.11 eV, and Ti3+2p3/2 at 460.87 eV. The position and the FWHM (~ 1.5 eV) of the observed peak at 460.87 eV are consistent with Ti4+ in the TiO2 lattice [34, 35]. The difference between the energy of Ti4+2p1/2 and Ti4+2p3/2 is 5.73 eV, which is consistent with the standard binding energy of TiO2 [34, 36, 37]. The Ti3+2p3/2 is attributed to the formation of Ti3+ and the presence of oxygen deficiency in TiO2 NTs [22, 38]. After doping with Ag, the high-resolution XPS spectrum (Fig. 3b) shows a slight shift in the position along with a variation in the area of the peaks. The Ti2p spectrum is fitted with three peaks located at binding energies 458.65 eV (Ti4+2p3/2), 464.38 eV (Ti4+2p1/2), and 460.87 eV (Ti3+2p3/2), respectively. Figure 3c shows the high-resolution spectrum of Cu-TiO2 NTs. Similar to Ag-doped TiO2 NTs, the Ti2p spectrum is fitted with peaks at 459.04 eV (Ti4+2p3/2), 464.78 eV (Ti4+2p1/2), and 459.11 eV (Ti3+2p3/2), respectively. The shift of Ti4+2p peaks to lower binding energy for doped TiO2 NTs is widely discussed in the literature. It has been explained by the formation of Ti3+ species or substitution of Ti+4 by Cu+ and Ag+ ions in Cu-TiO2 NTs and Ag-TiO2 NT samples, respectively [39,40,41,42]. In comparison to the pure TiO2, the area of the Ti3+ peak in Ag-TiO2 NTs and Cu-TiO2 NT samples decreased, and that of the Ti4+ peak increased (Table 2). The increase in the area of Ti4+ peak after doping indicates that the doping improves the formation of TiO2. These can be explained by the interaction between oxygen vacancies in TiO2 NTs and the doped element.

Fig. 3
figure 3

Ti2p spectra: a pure TiO2 NTs, b Cu-doped TiO2 NTs, and c Ag-doped TiO2 NTs

Full size image
Table 2 Chemical composition of the Ag- and Cu-doped TiO2 thin film surface
Full size table

The oxygen O1s peak of pure TiO2 NTs (Fig. 4a) is prominent and can be deconvoluted into two subpeaks, centered at 530.67 eV and 532.41 eV. The first peak is attributed to the oxygen engaged in a Ti–O bond in TiO2 or Ti2O3 lattice (OL), while the second one is related to the H–O bond or the adsorbed oxygen type O2 on the surface of TiO2 (OH) [34, 43,44,45]. Similarly, for the doped sample, the O1s spectrum of Cu and Ag-doped TiO2 NTs fitted with two peaks is shown in Fig. 4b and c, respectively. However, in this case, the first peak air is higher than that of the second one. The increase in the area of the original peak at 530.67 eV after doping indicates that the doping may improve the formation of TiO2.

Fig. 4
figure 4

Typical O1s spectra: a pure TiO2 NTs, b Cu-doped TiO2 NTs, and c Ag-doped TiO2 NTs

Full size image

Figure 5 shows the high-resolution XPS spectrum of Cu2p in Cu-TiO2 NTs. In this spectrum, the doublet Cu2p3/2 at 932.56 eV and Cu2p1/2 at 952.92 eV arises from spin orbit-splitting. These peaks are related to Cu+ [46, 47]. Also, the shoulders Cu2p3/2 at binding energy 934.20 eV and Cu2p1/2 at 953.23 eV are corresponding to Cu2+ [39, 46, 48]. The satellite peaks are also located at 943.93 eV [49]. This observation revealed that the doped Cu in the TiO2 lattice is in a mixture of Cu+ and Cu2+ oxidation states. The Cu2p3/2 peak transition is lower than the values reported in the literature (933.6 eV) [50]. The shift of binding energies of Cu2p3/2 peaks to the lower energy indicates the substitutional incorporation of Cu ions in the TiO2 lattice rather than the formation of Cu2O or/and CuO at the surface of TiO2 [51,52,53,54].

Fig. 5
figure 5

High-resolution XPS spectra of Cu2p region for Cu-TiO2 NTs

Full size image

For the Ag-TiO2 NT sample, the XPS diagram of high resolution for Ag3d (Fig. 6) has two peaks at 368.8 eV and 374.7 eV, belonging to the Ag3d5/2 and Ag3d3/2 orbits [55], respectively. The splitting of Ag3d doublet at about 5.9 eV confirms that the Ag element is present as metallic silver (Ag0) in the Ag-TiO2 NT sample [56, 57], while the weak peaks at 367.7 and 373.7 eV are attributed to Ag3d5/2 and Ag3d3/2 of silver ions (Ag+). These results further prove that Ag0 and Ag+ exist simultaneously on the surface of the Ag-TiO2 NT sample [58, 59]. The oxygen deficiency can explain the formation of Ag0 at the surface of TiO2 [24, 60].

Fig. 6
figure 6

High-resolution XPS spectra of Ag3d region for Ag-TiO2 NTs

Full size image

X-ray diffraction (XRD) pattern

Determination of crystal structure, phase composition, and crystallite size of the synthesized pure TiO2 NTs, Cu-doped TiO2 NTs, and Ag-doped TiO2 NTs is carried out using XRD, and the results are shown in Fig. 7. The X-ray diffraction spectra of all the samples showed well-defined peaks, which could be indexed to the anatase, rutile phases of TiO2, and titanium according to JCPDS file numbers 21–1272, 21–1276, and 44–1296, respectively. The peaks in pure TiO2 NTs are attributed to the reflection from (101), (200), and (105) planes of the anatase. In addition, the rutile phase is revealed by the apparition of peaks at 27.4°, 36.07°, 41°, 43.6°, and 56.9° corresponding to planes (110) (101), (111), (210), and (220), respectively. However, the Cu and Ag peaks are not visible in the XRD spectra of the doped TiO2 NTs. This is probably due to their low amount or their high dispersion in the samples with small dimensions below the detection limits of XRD. Xu et al. [61]. reported that the diffraction peaks of copper species disappeared when the copper component was highly dispersed in TiO2. Compared to pure TiO2 NTs, the anatase diffraction peak (200) shifts slightly to lower 2θ values in the Ag- and Cu-doped TiO2 NTs (inset in Fig. 7). In addition, the intensity of doped TiO2 NT peaks is higher than that of pure TiO2 NTs, which indicates that the doping enhances the crystallinity and improves the structural quality of TiO2 (according to XPS results). On the other hand, the peak intensity of (100) anatase plan in doped TiO2 NTs is higher than pure TiO2 samples, while the peak intensity of rutile decreases.

Fig. 7
figure 7

XRD spectra of pure TiO2 NTs, Cu-TiO2 NTs, and Ag-TiO2 NTs

Full size image

It is well known that the grain size and phase composition of TiO2 are the most significant for a TiO2 photocatalyst [32, 62,63,64]. From XRD data, the grain size was calculated using the Scherrer equation [65]. The crystallite sizes of pure TiO2 NTs, Cu-doped TiO2 NTs, and Ag-doped TiO2 NTs ranged from 29.7 to 35.5 nm (pure TiO2: 35.5 nm; Ag-TiO2 NTs: 29.71 nm, and Cu-TiO2 NTs: 29.71 nm). In addition, the relative anatase/rutile ratio was estimated at 56.25%/43.75%, 83%/17%, and 83%/17% in pure TiO2 NTs, Cu-TiO2 NTs, and Ag-TiO2 NTs, respectively. These results indicate that Ag and Cu inhibit the anatase–rutile phase transformation and stabilize the anatase phase significantly compared to the rutile phase [66].

Raman spectroscopy

The Raman spectra obtained for pure and doped TiO2 NTs are presented in Fig. 8. For all samples, the Raman spectra show six Raman-active modes (Eg, Eg, B1g, B1g + A1g, and Eg), which indicates that the samples consist of a mixture of anatase and rutile phases [31, 35, 67]. The characteristic Raman peaks at around 144.56, 197.8, 396.1, 516, and 638 cm−1 are indexed to Eg (1), Eg (2), B1g, A1g, and Eg (3) of anatase, respectively [31, 68]. Furthermore, the characteristic peaks of the rutile phase are observed at around 452 and 615 cm−1 [69]. The anatase band Eg (1) is shifted, from 143 cm−1 (for pure TiO2) to 143.7 cm−1 for the doped samples (inset in Fig. 8). The shift of the anatase band (Eg (1)) has been reported in previous studies and attributed to the formation of Ti3+, change of crystal size, and oxygen deficiencies in TiO2 lattice [35, 70,71,72,73,74]. These results are consistent with those of XPS and XRD results.

Fig. 8
figure 8

Raman spectra of pure TiO2 NTs, Cu-TiO2 NTs, and Ag-TiO2 NTs

Full size image

FTIR analysis

The FT-IR spectra of pure and doped TiO2 NTs calcined for 2 h at 600 °C are presented in Fig. 9. Compared with pure TiO2 NTs, there are no differences in the FT-IR spectra of doped TiO2 NTs. The figure shows the presence of some bands between 560 and 800 cm−1 which are attributed to different vibrational modes of anatase and rutile phases of TiO2 [19, 75]. Especially, the intense band observed below at 620 cm−1 is due to Ti–O vibrations [19, 76, 77]. The band at around 1631 cm−1 is attributed to the stretching vibrations of the O–H groups at the TiO2 surface. On the other hand, a slight shift in the position and the change in the intensity of bands are observed. These shift may be due to the presence of the dopants in the interstitials of the lattices of the doped samples [78]. These results are in agreement with those obtained by XRD and Raman analysis.

Fig. 9
figure 9

FT-IR spectra of pure and doped TiO2 NTs

Full size image

Mott-Schottky (MS) analysis

Determination of the semiconductor nature, the flat band positions (EFB), and the carrier density (ND) provide the first test that the material may be an effective photocatalyst. The most often used method for determining all these proprieties is based on measuring the potential dependence of the space charge region capacity (Csc). Figure 10 shows the variation of Csc for all samples as a function of the applied voltage in Mott-Schottky representation. The slope of all lines is positive, indicating that all investigated samples are an n-type semiconductor. Using the slope and intercept of the linear region, the ND and EFB were calculated according to the following Mott-Schottky relation [34, 79, 80].

$$\frac{1}{{\mathrm{C}}_{\mathrm{S}}^{2}}=\frac{2}{{\varepsilon }_{0}{\upvarepsilon }_{\mathrm{S}}{\mathrm{eN}}_{\mathrm{D}}}(E-{\mathrm{E}}_{\mathrm{FB}}-\frac{kT}{e})$$
(1)

where CS is the space charge layer capacitance, e is the electron charge (1.60 10−19 C), \({\upvarepsilon }_{0}\) is the permittivity of free space (8.85 10−14 F cm−1), \({\upvarepsilon }_{\mathrm{S}}\) is the dielectric constant of TiO2, which is assumed to be 100 F cm−1 [81], E is the applied potential, k is the Boltzmann constant (1.38 1023 J K−1), and T is the absolute temperature.

Fig. 10
figure 10

MS plots of pure TiO2 NTs, Cu-TiO2 NTs, and Ag-TiO2 NTs

Full size image

The potential of the flat band (EFB) shifts from − 0.19 V/SCE for pure TiO2NTs to − 0.31 and − 0.46 V/SCE for Ag-TiO2 NTs and Cu-TiO2 NTs, respectively. On the other hand, the donor concentration ND of about 4.50 × 1017, 1.97 × 1017, and 4.20 × 1017 cm−3 for pure TiO2 NTs, Cu-TiO2 NTs, and Ag-TiO2 NTs, respectively.

Photocatalytic activity

The photocatalytic capacity of all samples was evaluated by the photodegradation of MB under UV irradiation. Figure 11a shows the evolution of the methylene blue concentration ([MB]) versus the irradiation time in the presence of a different catalyst. Under UV irradiation and in the absence of TiO2 (photolysis), the concentration of MB is kept constant, suggesting that the MB is stable under UV irradiation. However, in the simultaneous presence of TiO2 and UV, the MB concentration decreases with the irradiation time, demonstrating it is degradation.

Fig. 11
figure 11

Degradation of MB by photocatalysis on the pure TiO2NTs, Ag-TiO2 NTs, and Cu-TiO2 NT surface, [MB] = 2.5 mg/L, Xe lamp 300 W

Full size image

The experimental data of Fig. 11a were found to fit approximately a pseudo-first-order kinetic model by the linear transforms ln(C/C0) = f(t) = kt [82, 83], as shown in Fig. 11b. The values of the rate constant (k), regression coefficient, and degradation efficiency (%) are listed in Table 3. The higher R2 reveals that the decomposition kinetics essentially follows first-order kinetics. The results demonstrated that the Cu- and Ag-doped TiO2 nanotube arrays exhibited a higher photocatalytic activity than the pure TiO2 sample. From Table 3, only ∼2.5% of MB is degraded by the photolysis process, about 79.6 9%, 88.23%, and 92.61% after 4 h of irradiation using pure TiO2 NTs, Ag-TiO2 NTs, and Cu-TiO2 NTs, respectively. The higher efficiency of doped TiO2 can be explained by the enhancement of charge separation of photogenerated electrons and holes on TiO2 by doping.

Table 3 Photodegradation kinetics data of MB by TiO2
Full size table

A recent study showed that the inclusion of doped elements increased the photocatalytic activity of TiO2 under UV irradiation. For comparison, a summary list of very recent studies on photodegradation of MB by different based TiO2 photocatalyst is listed in Table 4. The degradation percentages obtained by doped TiO2 NTs prepared in our conditions are very near to those reported in the literature for TiO2 nanoparticles.

Table 4 Photocatalytic activity of recently studied TiO2-based photocatalysts for MB degradation under UV irradiation
Full size table

Previous studies demonstrate that the photocatalytic reaction pathway is believed to involve the reaction of MB with the generated OH• radicals producing a range of intermediate products to reach complete mineralization with the formation of CO2 and H2O [82, 87,88,89,90]. To study the main active components in the degradation process of MB on the TiO2 surface and to understand the degradation mechanism in more detail, trapping experiments were carried out. Isopropanol [91], triethanolamine [92], and benzoquinone [93, 94] are used as scavengers to capture hydroxyl radicals (OH•), photogenerated holes (h+), and superoxide anion radical (O2•−), respectively. To demonstrate the involvement of these radicals, a mixture of MB and isopropanol (2% v/v) or triethanolamine (2% v/v) or benzoquinone (2% v/v) was irradiated under the same conditions. The results obtained are shown in Fig. 12. As can be observed in Fig. 12a, the addition of a radical scavenger inhibits the degradation of MB. These results indicate that holes are the primary active species in the degradation of MB, while OH• and O2•− radicals are likely of secondary importance in photodegradation.

Fig. 12
figure 12

Effects of radical scavenger addition on the photodegradation of MB a and schematic diagram of the photocatalytic mechanism of TiO2 photocatalyst b

Full size image

Based on the above results and literature reports [95, 96], the possible photocatalytic mechanism of the TiO2 NT photocatalyst was plotted as shown in Fig. 12b. The photodegradation process depends on the generation and separation of carriers; under light conditions, once the semiconductor absorbing energy is higher than the energy of its energy band, electrons will be excited from the valence band to the conduction band. At the same time, holes remain on the former (Eq. 2). According to the following equations, the formed holes reacted directly with MB molecules, H2O, or OH− adsorbed in the TiO2 surface to form the active species (OH•) [6, 97,98,99,100,101,102,103].

$$\mathrm{TiO}_2 +\mathrm{ h\upsilon }\to {\mathrm{e}}_{\mathrm{CB}}^{-}+ {\mathrm{h}}_{\mathrm{VB}}^{+}$$
(2)
$${\mathrm{h}}_{\mathrm{VB}}^{+}+{\mathrm{H}}_{2}\mathrm{O}\to +{\mathrm{H}}^{+}+\mathrm{OH^\cdot}$$
(3)
$${\mathrm{h}}_{\mathrm{VB}}^{+}+{\mathrm{OH}}^{-}\to +{\mathrm{TiO}}_{2}+\mathrm{OH^\cdot}$$
(4)
$$\mathrm{MB}\;\mathrm{Dye}+\mathrm h_{\mathrm{VB}}^+\rightarrow\mathrm{Oxidation}\;\mathrm{products}$$
(5)

Reusability efficiency

In addition to the photocatalytic efficiency of the TiO2 NTs, the stability and the reusability of the TiO2 NT photocatalyst are significant parameters to assess its practical application in wastewater treatment. Therefore, we carried out studies on the reutilization of the photocatalysts. After every use, the photocatalyst was washed with distilled water and ethanol and then dried for 30 min at 60 °C. The same procedure was repeated six times with the already used photocatalyst. The results obtained are represented in Fig. 13. The activity of all photocatalysts showed a gradual decrease. The degradation efficiencies of 64.35, 79.83, and 83.07% were achieved for pure TiO2 NT, Ag-TiO2 NT, and Cu-TiO2 NT photocatalysts, respectively, after 6 reutilizations. These results indicate that the photocatalysts are globally stables and efficient for water treatment. This decrease in photocatalytic activity can be explained by poisoning of the active surface of the photocatalyst or the occupation of active sites by intermediates that adsorbed strongly on the surface of the photocatalyst [104, 105].

Fig. 13
figure 13

Photocatalytic reusability efficiency of all photocatalysts up to six uses

Full size image

Conclusion

A facile and novel one-step anodization process fabricated photocatalytic materials based on copper- and silver-doped TiO2 NTs. The obtained NTs have a diameter and wall thickness of 128 and 56 nm, respectively. The Cu-doped TiO2 NT composites have excellent photocatalytic performance. Copper doping improved the photocatalytic performance of TiO2 NTs. These NTs are exciting candidates for photocatalytic reactions, such as MB oxidation. The new anodization process developed in this study is a simple and efficient method that can be easily scaled up, thereby pioneering the fabrication of high-performance metal-doped TiO2 NT photocatalysts with promising environmental applications.