Photocatalytic Degradation of Methyl Orange in Wastewater Using TiO₂-Based Coatings Prepared by Plasma Electrolytic Oxidation of Titanium: A Review

Abstract

This review analyzes TiO₂-based coatings formed by the plasma electrolytic oxidation (PEO) process of titanium for the photocatalytic degradation of methyl orange (MO) under simulated solar irradiation conditions. PEO is recognized as a useful technique for creating oxide coatings on various metals, particularly titanium, to assist in the degradation of organic pollutants. TiO₂-based photocatalysts in the form of coatings are more practical than TiO₂-based photocatalysts in the form of powder because the photocatalyst does not need to be recycled and reused after wastewater degradation treatment, which is an expensive and time-consuming process. In addition, the main advantage of PEO in the synthesis of TiO₂-based photocatalysts is its short processing time (a few minutes), as it excludes the annealing step needed to convert the amorphous TiO₂ into a crystalline phase, a prerequisite for a possible photocatalytic application. Pure TiO₂ coatings formed by PEO have a low photocatalytic efficiency in the degradation of MO, which is due to the rapid recombination of the photo-generated electron/hole pairs. In this review, recent advances in the sensitization of TiO₂ with narrow band gap semiconductors (WO₃, SnO₂, CdS, Sb₂O₃, Bi₂O₃, and Al₂TiO₅), doping with rare earth ions (example Eu³⁺) and transition metals (Mn, Ni, Co, Fe) are summarized as an effective strategy to reduce the recombination of photo-generated electron/hole pairs and to improve the photocatalytic efficiency of TiO₂ coatings.

1. Introduction

Methyl orange (MO, C₁₄H₁₄N₃NaO₃S) is a sulfonated azo dye containing sulphonic (–SO₃H) and azoic (–N = N–) groups. The sulphonic group in the MO structure is responsible for the high solubility of MO in water. The presence of the azoic group in MO is responsible for high toxicity and carcinogenic and harmful effects on the environment and human and aquatic life at low concentrations. MO dye is widely used in a variety of industries, including paper, textiles, leather, cosmetics, pharmacy, food, etc. [1]. To reduce the harmful effects of MO pollutants while also protecting the environment, they must be removed from wastewater. In recent years, semiconductor photocatalysis has proven to be amongst the best technologies for degrading a wide range of dye pollutants, as it achieves complete mineralization of organic dyes into H₂O and CO₂ without the formation of secondary hazardous products [2,3].

A variety of semiconductors, such as TiO₂, ZnO, CdS, Fe₂O₃, WO₃, V₂O₅, ZrO₂, ZnS, Nb₂O₅, and others, have been utilized as photocatalysts for the photocatalytic degradation of MO [4,5]. Among them, TiO₂ is probably the most studied photocatalyst because of its low cost, non-toxicity, high stability, commercial availability, environmental friendliness, strong oxidizing power, and long-term stability against photo-corrosion [6,7]. The two main disadvantages of TiO₂ that prevent its practical use in photocatalysis are its wide energy gap (3–3.2 eV), which restricts its application to the ultraviolet region, and the rapid recombination of photo-induced electron/hole pairs [8,9]. Numerous tactics have been used to increase the photocatalytic efficiency of TiO₂. They can be summed up as either lowering the electron/hole recombination rate or altering the energy structure of TiO₂, i.e., extending the optical absorption range from the ultraviolet to the visible region [10,11]. To address the shortcoming above, TiO₂ can be coupled with narrow band gap semiconductors and doped with metallic and nonmetallic species, rare earth elements, etc. [12,13,14].

TiO₂-based photocatalysts are obtainable in two forms: powder and supported on stable substrates (film/coating) [15]. There are some disadvantages to using powder TiO₂ photocatalysts, even though they are more photocatalytically efficient due to their higher surface-to-volume ratio than film/coating TiO₂-based photocatalysts. Powder photocatalysts reduce the efficiency of photocatalytic reactions by aggregating and scattering light. Furthermore, removing powder photocatalysts from treated wastewater is a complex, time-consuming, and costly process that is unsuitable for practical usage. Using film/coating photocatalysts makes the procedure easier because the photocatalyst only needs to be removed from the treated wastewater.

The plasma electrolytic oxidation (PEO) process has recently been recognized as a valuable technique for producing photocatalytically active oxide coatings for the degradation of organic pollutions [16], primarily on Ti [17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56] but also on Mg alloys [57,58,59,60,61,62,63,64,65], Al [66,67,68,69,70,71], Ta [72], Nb [73], Zr [74], Zn [75], and brass [76]. PEO is a straightforward, efficient, and environmentally friendly anodizing method that produces oxide coatings on various metals (such as Al, Mg, Ti, Zr, Hf, Ta, Zn, Nb, and W) and their alloys by applying a high voltage beyond the dielectric breakdown threshold. This induces numerous short-lived, transient micro-discharges on the coating surface, accompanied by gas emission [77,78,79]. During these micro-charges, localized high temperatures and pressures are generated, facilitating the formation of oxide coating, mainly composed of substrate oxides, although complex oxides with electrolyte components can also be formed.

In this paper, we have reviewed current research on the photocatalytic application of TiO₂-based coatings formed on a titanium substrate by the process of PEO for the degradation of MO. The primary benefits of PEO usage in the production of PEO TiO₂-based photocatalysts include low costs, ease of use, and notable energy and time savings, as the processing time is very short (just a few minutes). This is due to the elimination of the annealing step, which is typically required to convert amorphous TiO₂ into crystalline phases. PEO allows for the controlled incorporation of electrolyte species into TiO₂ coatings during growth, which improves photocatalytic performance. All of these are highly desirable for the purpose of the industrial application of PEO to create TiO₂-based photocatalysts.

2. Materials and Methods

The starting material for the preparation of TiO₂-based coatings was a rectangular titanium sample (99.95% purity, Alfa Aesar, Haverhill, MA, USA) with dimensions of 25 mm × 10 mm × 0.25 mm. The preparation of titanium samples for PEO includes their ultrasonic cleaning with acetone, drying in a warm air stream, and coating with insulating resin, leaving a surface of 10 mm × 15 mm accessible to the electrolyte. Most of the TiO₂-based coatings were formed in a water solution of 10 g/L Na₃PO₄·12H₂O (basic electrolyte-BE) with the addition of SnO₂, CdS, Bi₂O₃, Sb₂O₃, MnO, NiO, Co₃O₄, and Eu₂O₃, particles as well as FeSO₄ and NaAlO₂ in different concentrations. TiO₂/WO₃ coatings were formed in a water solution of 10⁻³ M H₄SiW₁₂O₄₀ (12-tungstosilicic acid, WSiA, Merck, Darmstadt, Germany). A current density of 150 mA/cm² was applied for PEO. The PEO experimental setup is detailed in Ref. [80].

The TiO₂-based coatings were characterized by scanning electron microscopy (SEM, JEOL 840A, Tokyo, Japan) with energy-dispersive X-ray spectroscopy (EDS, Oxford INCA, Abingdon, UK), X-ray diffraction (XRD, Rigaku Ultima IV, Tokyo, Japan), Raman spectroscopy (TriVista 557 Raman system, S&I GmbH, Warstein, Germany), X-ray photoelectron spectroscopy (XPS, Kratos AXIS Supra, Kratos Analytical Limited, Manchester, UK), X-ray fluorescence (XRF, Shimadzu XRF-1800, Shimadzu Corporation, Kyoto, Japan), photoluminescence (PL, Horiba JobinYvon, Fluorolog FL3-22, Edison, NJ, USA), and diffuse reflectance spectroscopy (DRS, Shimadzu UV-3600, Tokyo, Japan).

The photocatalytic potential of the TiO₂-based coatings was assessed by the degradation of 8 mg/L solution of MO. MO was chosen as a model for an organic dye because its degradation by various photocatalysts can only be monitored by UV–Vis spectroscopy and no sophisticated analytical techniques are required. The experimental setup for photocatalytic measurements is described in Ref. [74]. The photocatalytic reactor was a double-walled glass vessel cooled with water. The original MO solution and samples were placed in the dark for one hour to reach the equilibrium adsorption–desorption between the photocatalyst and the MO. Then, the TiO₂-based samples were placed on a stainless steel holder that was 5 mm above the bottom of the reactor. The 10 cm³ MO solution was mixed using a magnetic stirrer. The samples were exposed to a 300 W light source (OSRAM ULTRA-VITALUX UV-A, OSRAM, Garching, Germany) located 25 cm above the solution’s surface. The photocatalytic activity (PA) of the TiO₂-based samples was measured by the degradation of MO after appropriate light exposure. The maximum absorbance peak of MO at 464 nm was measured by the UV–Vis spectrometer (Thermo Electron Nicolet, Evolution 500, Thermo Fisher Scientific Inc., Cambridge, UK). A standard curve was employed to convert the absorbance into a MO concentration, demonstrating a linear correlation between the concentration and absorbance at 464 nm.

3. Results

3.1. PA of TiO₂ Coatings

Figure 1a displays the potential–time characteristics for titanium anodization in BE at a constant current density of 150 mA/cm². The fairly uniform growth of a compact oxide layer is correlated with a linear increase in the anodization potential in region I. A strong electric field (~10⁷ V/cm) facilitates the migration of O²⁻/OH⁻ and Ti⁴⁺ ions across the oxide, causing an oxide layer to grow at the interfaces between Ti/oxide and oxide/BE. The following general reaction leads to the formation of the initial oxide layer on the surface of titanium:

Ti + 2H₂O → TiO₂ + 4H⁺ + 4e⁻

(1)

A deviation from the linearity of the potential–time characteristic signifies the dielectric breakdown (the onset of PEO), which marks the end of the uniform thickening of the film. Following the breakdown, the potential of anodization rises steadily, and a significant number of tiny micro-discharges that are uniformly spaced across the sample surface emerge (region II). Further anodization yields a stable value for the anodization potential (region III) and spark micro-discharges. Oxide layers are formed in a series of steps during the micro-discharges [81]. The loss of dielectric stability in a low conductivity region leads to the formation of a series of separate micro-discharge channels in the oxide layer. Due to the high temperatures at the micro-discharge sites, the titanium melts away from the substrate, penetrates the micro-discharge channels, and reacts with the electrolyte components. The reaction products are ejected from the micro-discharge channels onto the coating surface, where they rapidly solidify in contact with the low-temperature electrolyte. The PEO involves repeating this process to gradually increase the thickness of the coating (Figure 1a).

Figure 1b displays the SEM images of the surface coatings formed at various PEO times. The surface of the coatings is embellished with numerous micro-discharge channels of varying sizes and shapes, along with areas formed by the rapid cooling of molten material. In the PEO of titanium, it is typical that as the PEO time and coating thickness increase, the number of micro-discharge channels and micro-pores decreases while their size increases [82].

The XRD patterns of the PEO coatings in Figure 1b are shown in Figure 1c. The coatings were crystalline and composed of an anatase phase of TiO₂ (PDXL DB Card No. 9008213). Elemental titanium originates from the substrate as a result of X-rays passing through the porous oxide coating and reflecting off it.

The optical absorption, PA after 8 h of irradiation, and a schematic representation of the photocatalytic degradation mechanism are shown in Figure 2. TiO₂ coatings manifest a broad absorption band in the mid-UV region below 382 nm (Figure 2a). This band gap arises from intrinsic optical transitions between the energy levels of Ti and O. The Kubelka–Munk function was employed to calculate the band gap energy of the TiO₂ coatings:

$$F(R)={\displaystyle \frac{{(1-R)}^2}{2R}}={\displaystyle \frac{\alpha }{S}}$$

(2)

where R = 10^A is the reflectance, A is the absorbance, α and S are the absorption and scattering coefficients, respectively. The following equation describes the energy dependence of the α in semiconductors near the absorption edge:

$$\alpha \propto {\displaystyle \frac{{(h\nu -E_g)}^{\eta }}{h\nu }}$$

(3)

In this equation, ν represents the incident photon frequency, h is the Planck constant, and E_g is the optical absorption edge energy. The exponent η indicates the type of optical transition caused by photon absorption (η = 1/2 for the direct transition and η = 2 for the indirect transition). To calculate the band gap values in anatase TiO₂ coatings, we used a plot [F(R) · hν]^1/2 vs. hν [83], extrapolating the linear region of the Tauc plot to the intersection with the hν—axis (Figure 2b). The calculated E_g is around 3.15 eV.

Although the coatings contain the anatase phase of TiO₂, the most photocatalytically active phase, they show low efficiency in the photodegradation of MO. After 120 s of TiO₂ formation, PA increases slowly with an increasing PEO time. Relative to the normal hydrogen electrode (NHE), the edges of the valence band (VB) and conduction band (CB) for anatase TiO₂ are approximately 2.83 V and −0.37 V, respectively [84]. The mechanism of MO degradation by the process of photocatalysis is shown in Figure 2d. The absorption of photons with energies exceeding the band gap of TiO₂ activates the material. The electrons (e⁻) move from the VB to the CB of the TiO₂ as a result of the absorption of the photons and form positive holes (h⁺) in the VB (Equation (4)):

$${\mathrm{TiO}}_2+h\nu \to {\mathrm{TiO}}_2 (e^{-}+h^{+})$$

(4)

Photo-generated electron/hole pairs can recombine, releasing heat or photons:

$${\mathrm{TiO}}_2(e^{-}+h^{+})\to {\mathrm{TiO}}_2+\mathrm{heat} \left(\mathrm{photons}\right)$$

(5)

or they can migrate to the surface of the catalyst and participate in redox reactions [7]:

$$h^{+}+{\mathrm{H}}_2\mathrm{O}\to {\mathrm{H}}^{+}+{\mathrm{H}\mathrm{O}}^{•}$$

(6)

$$h^{+}+{\mathrm{OH}}^{-}\to {\mathrm{H}\mathrm{O}}^{•}$$

(7)

$$e^{-}+{\mathrm{O}}_2\to {\mathrm{O}}_2^{•-}$$

(8)

$${\mathrm{O}}_2^{•-}+{\mathrm{H}}^{+}\to {\mathrm{HO}}_2^{•}$$

(9)

$${\mathrm{HO}}_2^{•}+{\mathrm{HO}}_2^{•}\to {\mathrm{H}}_2{\mathrm{O}}_2+{\mathrm{O}}_2$$

(10)

$$e^{-}+{\mathrm{H}}_2{\mathrm{O}}_2\to {\mathrm{H}\mathrm{O}}^{•}+{\mathrm{OH}}^{-}$$

(11)

$${\mathrm{H}}_2{\mathrm{O}}_2+{\mathrm{O}}_2^{•-}\to {\mathrm{H}\mathrm{O}}^{•}+{\mathrm{OH}}^{-}+{\mathrm{O}}_2$$

(12)

As a result, hydroxyl radicals (HO•) are generated. These powerful, non-selective oxidizing agents can lead to the partial or complete degradation of organic compounds when the VB holes react with water or hydroxyl ions. At the same time, excited electrons in the CB react with oxygen molecules and form reactive anion superoxide radicals (${\mathrm{O}}_2^{•-}$) along with hydrogen peroxide radicals (${\mathrm{HO}}_2^{•}$) and hydrogen peroxide (H₂O₂). Highly reactive species, such as HO•, ${\mathrm{HO}}_2^{•}$ and ${\mathrm{O}}_2^{•-}$, react with MO and lead to their degradation.

3.2. PA of TiO₂ Coupled with Narrow Band Gap Semiconductors

The coupling of TiO₂ with semiconductors with a suitable band edges position and lower band gap compared to TiO₂ (WO₃, SnO₂, CdS, Sb₂O₃, Bi₂O₃, etc.) is a very effective way to improve the photocatalytic efficiency of TiO₂ due to two factors: (i) semiconductors with a narrow band gap are employed as photosensitizers for visible light; (ii) the band level positions of the coupled semiconductors minimize the recombination of the electron/hole pairs generated by light and facilitate the transfer of electrons and holes between the two semiconductors [85].

3.2.1. PA of TiO₂/WO₃ Coatings

WO₃ is an inexpensive n-type semiconductor with a relatively small band gap (2.4–2.8 eV), low toxicity, and stability, and it is one of the most significant materials that can be coupled with TiO₂ to enhance the photocatalytic activity of TiO₂ [86,87]. The small band gap of WO₃ enables more efficient utilization of solar light but also leads to an increase in the recombination rate of photo-generated electron/hole pairs, which affects the PA of WO₃. It has been shown that TiO₂/WO₃ mixed oxides exhibit higher PA compared to their single components. When the TiO₂/WO₃ composites are excited by solar light, photo-generated electrons from the CB of the TiO₂ can move to the CB of WO₃. Simultaneously, the holes can move from the VB of WO₃ to the VB of TiO₂. The separation effect enhances the photocatalytic efficiency by minimizing the recombination of the photo-generated electron/hole pairs.

TiO₂/WO₃ coatings are the most studied system formed by PEO of titanium for the photocatalytic degradation of various organic dyes, most commonly methylene blue and MO [36,42,46,50,51,52,54,56]. Alkaline electrolytes supplemented with sodium tungstate are the most commonly used for the creation of TiO₂/WO₃ coatings. PEO on titanium in WSiA also allows the formation of very efficient TiO₂/WO₃ photocatalysts in a very short time [50]. Figure 3 shows SEM micrographs, the EDS composition, XRD patterns, and the light-harvesting properties of the coatings formed in WSiA at different PEO times.

The PEO time has a significant influence on the morphology of the coatings formed (Figure 3a). With an increasing PEO time, the number of micro-discharge channels decreases while their diameter increases. A longer PEO time causes a larger amount of molten oxidized metal to be ejected from the micro-discharge channels and solidified when it comes into contact with the electrolyte, resulting in rougher surfaces. The coatings consist of Ti from the substrate and O and W from the electrolyte (Figure 3b). The proportion of W in the coatings increases with PEO time while the proportion of Ti decreases.

The XRD patterns of the coatings are shown in Figure 3c. The coatings are crystallized and consist of WO₃ and anatase TiO₂. It is evident that an increase in PEO enriches the coatings with WO₃. The elemental Ti originates from the substrate, and its intensity decreases with the duration of PEO, indicating an increase in coating thickness. These results are consistent with the EDS data (Figure 3b). Raman measurements confirm the presence of WO₃ and anatase TiO₂ in the coatings formed in WSiA (see Figure 3 in Ref. [42]). Figure 3d shows the DRS spectra for TiO₂/WO₃ coatings. Compared to pure TiO₂, the TiO₂/WO₃ coatings show a distinct redshift, with the shift to higher wavelengths correlating with the duration of the PEO. The redshifts in the DRS of TiO₂/WO₃ coatings are attributed to the formation of new energy levels resulting from the W-O-Ti bonds created during PEO [54].

Figure 4a shows the PA during MO photodegradation with TiO₂/WO₃ and pure TiO₂ coatings. The PA of the TiO₂/WO₃ coatings is influenced by the PEO time, with a decrease in PA as the PEO time is extended. Some TiO₂/WO₃ coatings exhibit higher PA than pure TiO₂, particularly with shorter PEO times (30 s, 60 s, and 90 s), which are linked to these increased activities. However, as the PA process time increases (180 s, 300 s, and 600 s), the PA drops significantly, falling below that of pure TiO₂.

Numerous factors influence the effectiveness of photocatalytic PEO materials, including their morphology, crystal structure, chemical composition, light absorption capacity, and the ability of the photocatalyst to generate long-lived electrons and holes that enable reduction and oxidation reactions, respectively [16]. Clearly, the decrease in the number of micro-discharge channels and/or the agglomeration of the molten materials on the surface (Figure 3a) leads to a decrease in PA. In addition, the decrease in PA with an increasing PEO process time can be attributed to an increase in the WO₃ content in the coatings (Figure 3b,c). The DRS results show that TiO₂ coatings containing WO₃ exhibit a significant redshift compared to pure TiO₂ (Figure 3d). The shift to higher wavelengths corresponds to the duration of the PEO process. Although the useful spectral range for photocatalysis was extended, TiO₂/WO₃ coatings formed by the PEO process show weak photocatalytic activity for 180 s, 300 s, and 600 s, suggesting that WO₃ is the major contributor to the photocatalytic reaction. It is likely that the prolonged PEO leads to thicker TiO₂/WO₃ coatings with a low effectiveness factor, making deeper layers of the coatings inaccessible to light.

The PL spectral measurements confirm this assumption (Figure 4b). PL is a useful technique to explain the mechanism of photocatalysis, as it can reveal the recombination rate of photo-generated electron/hole pairs [88]. The increase in PL intensity corresponds to a decrease in the PA of the coatings, indicating rapid electron/hole recombination. The photocatalytic activities (Figure 4a) align with the PL results. Coatings formed at PEO times of 30 s, 60 s, and 90 s show a lower PL intensity and follow the order of PA. The improvement in PA is achieved by the significant contribution of TiO₂/WO₃ centers, which enables the transfer of photo-generated electrons from the CB of TiO₂ to the lower-lying CB of WO₃ (Figure 4c). The positive holes move towards the VB of TiO₂, leading to a decrease in the electron/hole recombination rate. The high PL intensity corresponds to the coatings formed at PEO times of 180 s, 300 s, and 600 s, and the PA is lower than that of pure TiO₂. This result, together with the DRS data, confirms a significant involvement of WO₃ centers in the reaction since the CB and VB of WO₃ are close to each other, leading to a high electron/hole recombination rate.

3.2.2. PA of TiO₂/CdS Coatings

Due to relatively low band gap energy (~2.3 eV), CdS enhances the separation of photo-generated electron/hole pairs and improves visible light absorption when combined with TiO₂. These factors collectively enhance the photocatalytic activity of the TiO₂/CdS system [89]. SEM micrographs, XRD patterns, Raman spectra, and the Cd/Ti ratio of the coatings formed by PEO for 2 min in BE with added CdS particles at different concentrations are shown in Figure 5.

The amount of CdS particles in BE does not significantly affect the morphology of the coatings (Figure 5a). The well-defined diffraction peaks in the well-crystallized coatings represent the anatase phase of TiO₂ (Figure 5b). Since the CdS are present in low concentrations and uniformly distributed in the TiO₂ coatings, no CdS peaks are observed in the XRD patterns. Figure 5c shows the Raman spectra of the CdS particles and the coatings formed in BE and in BE + 8 g/L CdS particles. The Raman spectrum of the CdS particles shows a strong band at 296 cm⁻¹ for the first-order longitudinal optical phonon (1LO) and a band at 592 cm⁻¹ for the second-order optical phonon (2LO) [90]. The Raman spectrum of the pure TiO₂ coating shows dominant modes at 144 cm⁻¹ (Eg(1)), 197 cm⁻¹ (Eg(2)), 395 cm⁻¹ (B1g(1)), 514 cm⁻¹ (A1g, B1g(2)), and 637 cm⁻¹ (Eg(3)), which correspond to the active Raman modes of the anatase phase [91]. The Raman spectrum of the coatings prepared in BE with the added CdS confirms the presence of CdS particles in the TiO₂ coatings, displaying bands from both the TiO₂ coating and CdS particles. The Ti/Cd ratio, obtained from wavelength dispersive XRF measurements (Figure 5d), shows that the CdS content in the coatings increases with the concentration of CdS particles in BE. This suggests that the CdS particles’ concentration in BE regulates the amount of CdS particles incorporated into the TiO₂ coating.

Figure 6a shows the PA of TiO₂/CdS coatings. TiO₂/CdS coatings formed in BE with the addition of CdS particles up to 0.5 g/L outperform the pure TiO₂ coating, with the highest PA observed for the coating formed in BE with 0.4 g/L CdS particles. A high concentration of CdS particles in BE significantly reduces PA, with values even lower than those of pure TiO₂. Since the CdS particles that are incorporated in the TiO₂ coatings have a negligible influence on the morphology and phase structure, the main contribution of CdS particles to PA may be either to extend the optical absorption range of TiO₂/CdS coatings or to prevent the fast recombination of photo-generated electron/hole pairs. The DRS spectra of the pure CdS particles and the resulting coatings are shown in Figure 6b. Although the band edge of the CdS particles is around 570 nm, the TiO₂/CdS coatings do not show a noticeable shift of the absorption band edge into the visible light region. The calculated band gap energy for all TiO₂/CdS coatings is approximately 3.16 eV, except for the coating prepared in BE with 8 g/L CdS particles, which has a slightly lower band gap energy of about 3.14 eV (Figure 6c). The lack of an adsorption shift could be due to the low concentration of CdS particles incorporated in the coatings, which means that the CdS particles inhibit the recombination of photo-generated electron/hole pairs. Conversely, an excessively high concentration of CdS particles in the TiO₂ coatings leads to an increase in electron/hole pair recombination centers, ultimately reducing the PA.

The PL emission of the prepared coatings excited at 350 nm and monitored at 425 nm is shown in Figure 6d [38]. When increasing the concentration of CdS particles in the BE to 0.4 g/L, a decrease in PL intensity is observed, while further increasing the concentration of CdS particles in the electrolyte to 1.0 g/L leads to an increase in PL intensity. These findings are consistent with PA, as the decrease in PL intensity corresponds to an increase in PA, indicating a slower recombination of electron/hole pairs. At higher concentrations of CdS particles in the BE (2 g/L or higher), a simultaneous decrease in PL intensity and PA may be attributed to the increased presence of CdS dopants, which act as capture centers for photo-induced electrons [88].

3.2.3. PA of TiO₂/SnO₂, TiO₂/Bi₂O₃, and TiO₂/Sb₂O₃ Coatings

SnO₂, Bi₂O₃, and Sb₂O₃ are good candidates for coupling with TiO₂ because the upper edge of the VB and the lower edge of the CB of SnO₂, Bi₂O₃, and Sb₂O₃ are lower than those of TiO₂ (Figure 7) [31,32,41]. The photo-generated electrons can move from the CB of TiO₂ to that of SnO₂, Bi₂O₃, or Sb₂O₃, and the photo-generated holes move in the opposite direction between the VBs. As a result, the overall PA of TiO₂/SnO₂, TiO₂/Bi₂O₃, and TiO₂/Sb₂O₃ coatings are higher than that of their individual components, as the recombination rate of electron/hole pairs in TiO₂, SnO₂, Bi₂O₃, or Sb₂O₃ semiconductors is reduced.

The influence of SnO₂, Bi₂O₃, and Sb₂O₃ particles on the morphology, chemical, and phase compositions of TiO₂/SnO₂, TiO₂/Bi₂O₃, and TiO₂/Sb₂O₃ coatings is similar to that of TiO₂/CdS [31,32,41]. The PA of the formed coatings is determined by the concentrations of SnO₂, Bi₂O₃, and Sb₂O₃ particles added to the BE, i.e., by their incorporation into TiO₂ coatings (Figure 8a). The highest PA of TiO₂/SnO₂, TiO₂/Bi₂O₃, and TiO₂/Sb₂O₃ was observed for the coatings formed in BE with the addition of 0.3 g/L SnO₂, 1.2 g/L Bi₂O₃, and 0.75 g/L Sb₂O₃ particles, respectively. The DRS spectra of the TiO₂/SnO₂, TiO₂/Bi₂O₃, and TiO₂/Sb₂O₃ coatings in Figure 8b show that the coatings exhibit no recognizable shift in the absorption band toward the visible light range. This result indicates that SnO₂, Bi₂O₃, and Sb₂O₃ particles incorporated into TiO₂ coatings prevent the fast recombination of photo-generated electron/hole pairs, consistent with the PL measurements (Figure 8c). Indeed, the PL emission spectra of TiO₂ and the most photocatalytically active TiO₂/SnO₂, TiO₂/Bi₂O₃, and TiO₂/Sb₂O₃ coatings excited with 350 nm show a decrease in the PL intensity of the TiO₂/SnO₂, TiO₂/Bi₂O₃, and TiO₂/Sb₂O₃ coatings compared to TiO₂.

3.2.4. PA of TiO₂/Al₂TiO₅ Coatings

Al₂TiO₅ is a pseudobrookite-type titanate with a wide band gap (~3.42 eV) and has been identified as a possible photocatalyst [92,93] and a suitable candidate for coupling with TiO₂ [94,95]. PEO of titanium in BE with the addition of NaAlO₂ is an effective method for the formation of TiO₂/Al₂TiO₅ coatings [26]. Figure 9 shows the SEM micrographs, EDS composition, XRD patterns, and the content of crystalline phases in the coatings formed in BE with the addition of different concentrations of NaAlO₂ for 10 min.

The micro-discharges become more intense with an increasing concentration of NaAlO₂ in BE [26], resulting in a larger amount of molten oxidized metal being ejected from the micro-discharge channels onto the coating surface and rapidly solidifying in contact with the electrolyte (Figure 9a). As a result, the number and diameter of the pores on the coating surface decreases, and it becomes rougher. The chemical composition of the coatings contains Ti from the substrate and O, P, and Al from the electrolyte (Figure 9b). The content of Al increases with an increasing concentration of NaAlO₂ in BE, while the content of Ti decreases. The coating obtained in the BE consists of anatase TiO₂, but when NaAlO₂ is added in the BE in addition to the anatase, the rutile phase of TiO₂ (PDF Card No.: 9001681) and Al₂TiO₅ (PDF Card No.: 2206495) were observed in the coatings. Figure 9d presents the weight fraction of the individual phases, estimated using the reference intensity ratio (RIR).

Figure 10a illustrates that the PA of the coatings formed in BE with the added NaAO₂ is superior to that of the anatase TiO₂ coating formed in BE. The PA improves with an increasing NaAlO₂ concentration, reaching its highest level by adding 2.0 g/L of NaAlO₂.

Compared to the anatase TiO₂ coating, TiO₂/Al₂TiO₅ coatings show no noticeable shift in the absorption band edge toward the visible spectrum (Figure 10b). The estimated E_g ranges from 3.15 eV for anatase TiO₂ to 3.07 eV for the TiO₂/Al₂TiO₅ coating prepared in BE with the addition of 2.0 g/L NaAlO₂ (Figure 10c). This shift is due to a change in the anatase/rutile phase ratio in the coatings (Figure 9d).

The exceptional PA of TiO₂/Al₂TiO₅ is most likely attributed to the high concentration of various oxygen vacancies and other defects in the formed coatings, which facilitate the regulation of electron transfer between reactants and photocatalysts. These defects are closely associated with the active sites involved in the heterogeneous photocatalytic reactions [96]. Figure 10d displays the PL emission spectra of the coatings excited at 260 nm. The PL centers in anatase TiO₂ are defect centers related to the oxygen vacancies in the coating, and the PL intensity is low [97]. Al₂TiO₅ coatings have formed a significant number of oxygen vacancies and other defects on the surface due to the strong micro-discharges in BE containing NaAlO₂ [98]. Since oxygen vacancies and defects can bind photo-induced electrons, forming free or binding excitons, a PL signal is easily generated when the coatings are exposed to radiation. The PL intensity increases with the number of oxygen vacancies and defects in the coatings [88]. However, oxygen vacancies and defects act as trapping sites for photo-induced electrons during photocatalysis. This effectively prevents the recombination of photo-generated electron/hole pairs, which increases the PA of TiO₂/Al₂TiO₅ coatings.

3.3. PA of TiO₂ Coatings Doped with Rare Earth Ions

Doping TiO₂ with rare earth (RE) ions is an effective method for enhancing its photocatalytic properties. A number of factors have been identified that contribute to the improved PA of RE-doped TiO₂: (i) the recombination rate of photo-generated electron/hole pairs decreases due to the formation of oxygen vacancy charge trapping centers; (ii) the response of TiO₂ is extended to the visible region; (iii) the conversion temperature from anatase to rutile is increased; and (iv) the pollutant interacts with the RE metals doped in TiO₂ via f-orbitals, which increases the capacity to absorb the pollutant molecules and improves the PA [99,100,101,102].

In recent years, it has been shown that PEO on titanium in BE with the addition of RE oxide is a very practical way to form TiO₂:RE³⁺ coatings for photocatalytic applications [21,40,43]. In this section, the results of PA for the TiO₂:Eu³⁺ coatings formed in BE with the addition of 2 g/L of Eu₂O₃ particles for different PEO times will be presented. Similar results were obtained for the TiO₂:Tb³⁺ and TiO₂:Gd³⁺ coatings formed in BE with the addition of 2 g/L of Tb₃O₄ [40] and 2 g/L of Gd₂O₃ particles [21].

The TiO₂:Eu³⁺ coatings formed in BE with the addition of Eu₂O₃ particles have a comparable morphology to the pure TiO₂ coatings formed in BE (Figure 1a) [43]. The elements of the coatings are Ti, P, O, and Eu. All elements are uniformly distributed in the coatings (Figure 11a). The concentration of Eu is low and increases with the PEO time (Figure 11b). The XRD patterns of the TiO₂:Eu³⁺ coatings formed after different PEO times are shown in Figure 11c. The coating formed for the 60 s consists of anatase TiO₂, while the anatase phase transforms into the rutile phase about 180 s after the start of PEO.

Figure 11d shows the weight fractions of the anatase (W_A) and rutile (W_R) phases, calculated using Equations (13) and (14):

$$W_R={\displaystyle \frac{I_R}{0.884I_A+I_R}}$$

(13)

$$W_A=1-W_R$$

(14)

where I_A is the integrated intensity of the anatase peak (101) at a 2θ angle of 25.4°, and I_R is the integrated intensity of the rutile peak (110) at a 2θ angle of 27.5^o. The low concentration of uniformly distributed Eu₂O₃ particles in the TiO₂ coatings or the substitution of Eu ions in the TiO₂ crystal lattice could be the reason for the lack of visible peaks of Eu species in the XRD patterns.

The chemical state of Eu in TiO₂ coatings can be analyzed using XPS (Figure 12) and PL (Figure 13). Figure 12a shows the high-resolution Eu 3d spectra of the Eu-doped TiO₂ coatings formed after varying PEO times. The spectra display two peaks resulting from the spin-orbit splitting of the Eu 3d levels (3d_5/2 and 3d_3/2). The Eu³⁺ state is indicated by the Eu 3d_5/2 peak, located at 1134.7 eV [103]. The spin-orbit components (2p_3/2 and 2p_1/2) of the Ti 2p XPS spectra for the TiO₂ coating appear at 459.0 eV and 464.9 eV, respectively, and are attributed to the Ti⁴⁺ ions in TiO₂ (Figure 12b). In contrast, the XPS peaks corresponding to the Ti 2p spin-orbit components are shifted by 0.2 eV towards higher binding energies in all TiO₂:Eu³⁺ coatings. This shift in binding energy suggests the incorporation of Eu³⁺ into the TiO₂ lattice and indicates an intermediate oxidation state of Ti.

Figure 13 shows the PL excitation and emission spectra of the TiO₂:Eu³⁺ coatings formed after varying PEO times. In the PL excitation spectra monitored at 613 nm (⁵D₀→⁷F₂ transition of Eu³⁺) (Figure 13a), a broad band appears between 250 nm to 350 nm, peaking around 275 nm. Additionally, a series of sharp peaks in the 350 nm to 550 nm range are observed, which correspond to the direct excitation of the Eu³⁺ ground state ⁷F₀ into higher levels of 4f-manifold [104]. The broad band is associated with the charge transfer state (CTB) of Eu³⁺, which arises from the electronic transition between the empty 4f orbital of Eu³⁺ ions (4f⁶) and the fully occupied 2p orbital of O²⁻ ions (2p6) [105]. The most intense excitation bands assigned to 4f-4f transitions of Eu³⁺ are ⁷F₀→⁵L₆ at 393 nm and ⁷F₀→⁵D₂ at 463 nm. The PL emission spectra of the TiO₂:Eu³⁺ coatings, excited at 275 nm (Figure 13b) and 393 nm (Figure 13c), display sharp emission bands in the orange-red spectral range between 560 nm and 720 nm. These bands correspond to the 4f-4f transition of Eu³⁺ from the excited level ⁵D₀ to the lower levels ⁷F_J (J = 0, 1, 2, 3, and 4) [106]. The electric dipole transition ⁵D₀→⁷F₂, which is hypersensitive (to the local environment of the Eu³⁺ ion) and is only allowed when the Eu³⁺ ion occupies a site without an inversion center, is associated with the strongest PL peak at 613 nm [107]. In contrast, the magnetically allowed dipole ⁵D₀→⁷F₁ transitions at 591 nm and 596 nm are not affected by the symmetry of the site [107]. To measure the degree of distortion of the inversion symmetry of the local environment of the Eu³⁺ ions in the host matrix, a spectroscopic method involves calculating the ratio of the emission intensity between the ⁵D₀→⁷F₂ and ⁵D₀→⁷F₁ transitions, known as the asymmetric ratio—R.

The symmetry of the Eu³⁺ ions in the TiO₂ coatings remains largely unchanged with varying Eu³⁺ concentrations, as evidenced by the R value, which shows only a slight increase with a longer PEO time (Figure 13d) [108]. High values of R indicate a highly asymmetric environment of Eu³⁺ ions in TiO₂. The site symmetry of Ti⁴⁺ ions in the anatase lattice is D_2d. According to the branching rules of 32-point groups, the substitution of Ti⁴⁺ by Eu³⁺ ions leads to the formation of oxygen vacancies and lattice distortion due to the significant disparity of ionic radii and charge imbalance between Ti⁴⁺ (68 pm) and Eu³⁺ (95 pm), which changes the site symmetry of Eu³⁺ ions from the exact D_2d symmetry to a lower site symmetry [109].

Figure 14 presents the PA and DRS spectra of the TiO₂ coatings formed for 600 s, as well as the TiO₂:Eu³⁺ coatings created after varying PEO durations. Compared to pure TiO₂, the PA of the TiO₂:Eu³⁺ coatings is noticeably higher (Figure 14a). As the amount of incorporated Eu³⁺ increases, the absorption edge of the pure TiO₂ progressively shifts from the ultraviolet to the visible spectrum. The redshift is attributed to a charge transfer between the f electrons of Eu³⁺ ions and the CB or VB of TiO₂ [110]. While TiO₂:Eu³⁺ coatings formed after longer PEO times broaden the spectral range for photocatalysis, the TiO₂:Eu³⁺ coating formed after 60 s of PEO exhibits the best PA.

The TiO₂:Eu³⁺ coating that forms after 60 s of PEO contains only the anatase phase of TiO₂, which has a high degree of structural organization around the incorporated Eu³⁺ ions. The conversion of anatase to rutile by a prolonged PEO process time probably causes the distortion of the Eu³⁺ surrounding lattice (Figure 13d). The increase in R, the measure of the degree of distortion, also increases with the PEO time. This indicates that the Eu³⁺ ions, in this case, could serve as centers for the recombination of electron/hole pairs, suggesting that the Eu³⁺ environment is more important than the Eu³⁺ concentration for the photocatalytic application of TiO₂:Eu₃₊ coatings.

3.4. PA of TiO₂ Coatings Doped with Transition Metals

Doping of TiO₂ with transition metals is often used to improve PA by shifting the absorption edge of TiO₂ towards the visible light and preventing the recombination of photo-generated electron/hole pairs by creating new electron states within the electronic structure of TiO₂ [111,112,113]. In recent years, there has been significant research on the synthesis of TiO₂ coatings modified with various transition metals for photocatalytic applications [18,19,22,30,37,39,47]. The most common electrolytes used for the formation of such coatings contain acetate, nitrate, sulfate, and oxalate forms of transition metal salts. This section presents the results of PA for Mn-, Ni-, and Co-modified TiO₂ coatings formed for 60 s in BE with the addition of MnO, NiO, and Co₃O₄ particles at different concentrations, respectively. TiO₂:Fe [39], TiO₂:V [47], and TiO₂:Ag [22] photocatalysts for the degradation of MO can also be formed by PEO on titanium in BE with the addition of FeSO₄, NH₄VO₃, and colloidal Ag nanoparticles, respectively.

Figure 15a shows the SEM micrographs of coatings formed in BE and in BE with the addition of MnO, NiO, or Co₃O₄ particles at the highest concentrations used. The addition of MnO, NiO, or Co₃O₄ particles to the BE does not significantly affect the morphology of the coatings formed. The EDS composition of the coatings in Figure 15a is shown in Figure 15b. The content of Mn, Ni, and Co detected in the coatings is low, especially for the coatings formed in BE with small amounts of MnO, NiO, or Co₃O₄ particles. The Ti/Mn, Ti/Ni, and Ti/Co ratios from the wavelength dispersive XRF measurements show that the Mn, Ni, and Co contents in the coatings increase with the concentration of MnO, NiO, and Co₃O₄ particles in BE, respectively (Figure 15d). As shown in the XRD patterns of the coatings in Figure 15a, the coatings are predominantly composed of the anatase phase of TiO₂, with no indication of Mn-, Ni-, or Co-containing phases (Figure 15c).

High-resolution XPS spectra of the 2p region for the Mn, Ni, and Co-doped TiO₂ coatings are shown in Figure 16. The Mn species are in the Mn²⁺ oxidation state, as evidenced by the two peaks in the Mn 2p XPS spectrum (Figure 16a) at approximately 653.4 eV and 641.6 eV, corresponding to Mn 2p_1/2 and Mn 2p_3/2, respectively [114]. The XPS spectrum of Ni 2p, shown in Figure 16b, reveals satellite peaks at 878.0 eV and 860.2 eV, with the Ni 2p_1/2 and 2p_3/2peaks located around 872.7 eV and 855.0 eV, respectively. These results confirm the presence of Ni²⁺ [115]. The Co 2p XPS spectrum (Figure 16c) shows Co 2p_1/2 and Co 2p_3/2 peaks at binding energies of about 795.4 eV and 779.5 eV, respectively. The satellite peaks, along with the position and spin-orbit splitting between the Co 2p_1/2 and Co 2p_3/2 peaks of 15.9 eV [116], suggest that Co is mainly in the Co²⁺ oxidation state.

Figure 17a illustrates the PA of the coatings formed in BE with varying concentrations of MnO, NiO, and Co₃O₄ particles. The concentrations of MnO, NiO, and Co₃O₄ particles significantly impact the PA of the coatings. In terms of PA, the Mn-, Ni- and Co-doped TiO₂ coatings outperformed the pure TiO₂ coatings, except for the coatings produced in BE by adding 3.0 g/L of NiO and 3.0 g/L of Co₃O₄ particles. The highest PA was observed in the coatings formed in BE with 0.75 g/L of MnO, 1.5 g/L of NiO, and 2.0 g/L of Co₃O₄ particles.

Since all formed anatase TiO₂ coatings modified with Mn, Ni, and Co exhibit nearly identical morphologies, the observed PA can be attributed to the varying amounts of Mn, Ni, and Co in the coatings. These transition metals (Mn, Ni, and Co) enhance the PA primarily by either slowing down the fast recombination of photo-generated electron/hole pairs or by extending the optical absorption spectrum of the TiO₂ coatings into the visible region.

Mn- and Ni-doped TiO₂ coatings show no noticeable shift of the absorption band in the direction of visible light, as the Tauc plots of the coatings formed show (Figure 17b). In contrast, Co-doped TiO₂ coatings show a slight shift in the absorption band, which is consistent with previous results [117]. These results indicate that the main function of the Mn, Ni, and Co in TiO₂ coatings is to prevent the rapid recombination of photo-generated electron/hole pairs, which is in accordance with the PL measurement (Figure 17c). The PL intensity decreases with an increasing concentration of MnO, NiO, and Co₃O₄ particles in the BE. A decrease in PL intensity is associated with an increase in PA, indicating improved separation of electron/hole pairs when MnO, NiO, and Co₃O₄ are present in BE at concentrations of 0.75 g/L, 1.5 g/L, and 2.0 g/L, respectively. However, both PL intensity and PA decrease simultaneously as the concentrations of MnO, NiO, and Co₃O₄ in the BE are further increased. This is because the transition metal dopants act as capture centers for photo-induced electrons [88].

3.5. The Chemical and Physical Stability of TiO₂-Based Photocatalysts

An important factor for the applicability of the photocatalyst is its chemical and physical stability, which determines the lifetime and operating costs of the catalyst. TiO₂-based photocatalysts formed by PEO of titanium have high durability and stability [26,32,43]. In this section, the results of the TiO₂:Fe photocatalyst stability tests are presented. The PA after 8 h of irradiation following ten consecutive runs with a TiO₂:Fe coating formed for 120 s in BE with the addition of 2 g/L of FeSO₄ is shown in Figure 18a. The sample was cleaned with water after each run, dried, and then used for the next catalytic cycle. The PA clearly did not decrease. The crystal structure (Figure 18b) and morphology (Figure 18c) also did not change after 10 runs, which proves that there is neither significant wear of the photocatalyst from the support nor poisoning of the catalyst.

4. Conclusions

In summary, this review focuses on introducing different strategies to enhance the photocatalytic activity (PA) of TiO₂-based coatings formed on titanium substrate by the plasma electrolytic oxidation (PEO) process for the degradation of methyl orange (MO). Pure TiO₂ coatings formed in BE (10 g/L Na₃PO₄·12H₂O) have a PA that does not exceed 35% after 8 h of simulated solar irradiation. Coupling TiO₂ with suitable semiconductors such as WO₃, SnO₂, CdS, Sb₂O₃, Bi₂O₃, and Al₂TiO₅ increases PA due to a reduction in the recombination rate of photo-generated electron/hole pairs. For each system, there is an optimal concentration of coupled semiconductors with TiO₂. When their concentration in the formed coatings is high, they oppose the center for trapping photo-induced electrons, leading to a decrease in PA. The TiO₂/Al₂TiO₅ coatings showed the best photocatalytic performance. The TiO₂/Al₂TiO₅ coating formed for 600 s in BE + 2 g/L of NaAlO₂ has a PA of about 90% after 8 h of irradiation. The TiO₂/WO₃ coatings formed for a very short time in WSiA also show very good PA. For example, the TiO₂/WO₃ coating formed for 30 s in WSiA has a PA of about 75%. All other systems (TiO₂/SnO₂, TiO₂/CdS, TiO₂/Sb₂O₃, and TiO₂/Bi₂O₃) have a PA of less than 60% after 8 h of irradiation.

The TiO₂ coatings doped with rare earth ions also have a higher PA than the pure TiO₂ coatings. For the most photocatalytically active coatings, it is around 60% after 8 h of irradiation. The TiO₂ coatings doped with transition metals (Mn, Ni, Co) also show similar photocatalytic performance, with the exception of the TiO₂:Fe coating, which was formed for 120 s in BE with the addition of 2 g/L of FeSO₄ and has a PA of about 78%. Incorporated ions of rare earths and transition metals in TiO₂ coatings prevent the fast recombination of photo-generated electron/hole pairs.

TiO₂-based photocatalysts can be formed by PEO in a very short time in environmentally compatible electrolytes on a titanium substrate. They exhibit excellent physical and chemical stability, which makes them good candidates for use in the continuous photocatalytic degradation of organic pollutants.

Due to the advantages of photocatalysts on solid substrates compared to photocatalysts in the form of particles and because PEO allows for the very easy incorporation of particles from the electrolyte into the formed coatings, future attempts should be made to incorporate photocatalytically active particles into the TiO₂ coatings by PEO to further improve the photocatalytic performance and extend the activation range to the visible range.